I wanted to pin a comment to the top here to address the most common question that's been coming up in the comments. That's, "You can't use EQ to correct the frequency response when you have SBIR because you'd just be boosting the drive signal and still causing destructive interference at the same frequency based on the 1/4 wavelength" or something to that effect. The answer is.....YES, you absolutely can. Using standard analog EQ, it's not possible. But when you employ digital adaptive EQ that uses a modelling delay, you can achieve a near perfect response correction. Here are the details on that, and my sources. I had thought the Sonarworks SoundID Reference platform was using this technology, but I was incorrect. I apologize. Sonarworks cannot correct phase distortion from boundaries, yet, although they tell me they're working on it and hope to deploy this in a future version. Take a look at the Journal of the Audio Engineering Society / JAES Volume 33 Issue 3 pp. 127-132; March 1985. Spectral, Phase, and Transient Equalization for Audio Systems, P. M. CLARKSON, J. MOURJOPOULOS, AND J. K. HAMMOND, Institute of Sound and Vibration Research, University of Southampton, Highfield, Southampton, UK. www.aes.org/e-lib/browse.cfm?elib=4461 If you want to see evidence of this technology that is currently available as a commercial product, look at the Trinnov system. In their technical literature, you'll find confirmation, complete with visual graphics, of how this technology works to correct phase distortion by using a modelling delay to time-align specific frequencies, thus correcting for phase shifts in a non-minimum phase or mixed phase system. www.trinnov.com/en/technologies/. Also the AES book Recording Studio Design by Philip Newell, page 332-334, which states "Near perfect response at optimum listening position after correction (with digital adaptive EQ)". This example was based on a test where a loudspeaker was positioned on a stand, part way through a room (exactly as I did in position 2 and 3), after which digital adaptive EQ was applied. This section of the book is also key to understanding why and how the correction can occur: "A non-minimum phase effect (such as when a loudspeaker is freestanding in a room, and not mounted at or in the wall) cannot be universally corrected by (analog) equalization (alone) of the loudspeaker drive signal. Even perfect amplitude correction could not restore the original phase response by any known practical analogue means. Only by digital signal processing could an almost perfect response be restored. Adaptive digital filtering can model, very accurately, the inverse phase responses necessary to correct either the minimum phase or non-minimum phase components of a transient or steady state response anomaly. By means of measuring microphones, modelling delays, and adaptive filtering processes, the digital systems can be made to 'learn' what a given room will do to a loudspeaker response, and apply acausal corrections to cancel the disturbances [acausal, in this sense, means that by means of a signal delay, the measured response error is fed into the drive-signal, in inverted form, to precondition the output to anticipate the error." -Pg 334, 11.5.2 Digital Correction Techniques, Recording Studio Design, Philip Newell, Audio Engineering Society. I followed up by speaking with Philip Newell directly on this matter. He personally watched this video, reviewed the comments, and confirmed this is a correct presentation of the information. Here are his comments on this specific topic: "I stand by what I said in the section that you quoted from Recording Studio Design, but it has to be taken in context. Acausal filtering and signal processing, WITH A MODELLING DELAY, can correct for the room boundary effects, but only for a very small region of a room -- say a very restricted 'sweet spot'. Elsewhere in the room, the response may then be much worse than before the correction. So, for example, if someone must work in an acoustically-poor room, this type of processing could greatly improve things at a restricted listening position, which may be beneficial to a single user, or it may 'somewhat' improve things over a slightly wider area, but it cannot make an overall improvement to the room acoustics. When one place is improved, another place will suffer." Furthermore, I spoke with the team at Sonarworks about their platform and this is what they've said on the topic: "We align the time between channels by phase measurements currently. The boundary nulling question: It will not be the same as it will still be there, but at a lower energy level. In short, phase-induced dips cannot be corrected by amplitude EQ only, only their impact lessened. For proper handling of phase effects, phase shaping is required." This parallels exactly what Philip Newell has stated. Sonarworks does not currently do phase shaping (I got this wrong, so sorry about that. I had thought their platform was doing this but I was incorrect.), but I hear they may be planning to roll this out in a future release. Also, to be clear, the audience for this is music studios, not audiophile hifi systems. In audio engineering control rooms, the flattest response possible is most certainly a primary goal. My comments to audiophiles are: Do what sounds good to you. You're the end user of your media and you should optimize it however you want to. Audio engineers and music producers, like myself, and many of the subscribers on my channel, have a different set of needs, which include accuracy and translation. Cheers!
As a recovering audiophile and budding mix engineer, there are things both can learn from one another. Go to an audio show and listen to the top 3 rigs, you'll understand "audiophile sound character." In this priority: It's a recessed soundstage with 1) emphasis on depth and 2) solidity of apparent sounding body and 3) frequency accuracy. Engineers hear frequency accuracy 1st. Sounding body floats close to the listener, mostly L-R-C, intimately 2nd and 3rd low end punch is controlled, but not excessively tactile. Overall, engineers prefer a 3D soundstage close with forward reverberation and audiophiles prefer a 3D soundstage gigantic and recessed back and away. These are 2 unique sonic perspectives, achieved in slightly different ways and here's the crazy part: with the same source material. In other words, go ahead and mix for near-field, studio monitoring, but don't be surprised to hear this legitimately interpreted differently in a living-room with real-world room conditions. Peace.
Hey hey. Interesting insights. For engineers, accuracy and translation are paramount, not just frequency response. Accuracy is more all encompassing (such as accurate transient response). Translation, as I'm sure you're aware, is ensuring that your engineering decisions still sound correct when monitored in many other listening environments & playback systems. This has been a much studied subject in audio engineering journals and acoustics research. Many (most?) audiophile listening environments do virtually nothing about 2 of the main obstacles to accuracy and hence the ability to translate: 1) Room reflections color the sound of the room - speaker system as much as (or sometimes more than) the speakers themselves. And because room reflections are unique only to that specific acoustic space, then any system that has much in the way of reflections has a pattern of constructive and destructive interference in the room and resonances that will dramatically skew the sound from anything resembling accurate. IE the music will sound wildly different from room to room. If there is some reflected energy in a control room, it will be scattered by the use of diffusors so the energy is dispersed in a hemi-disc (typically) rather than a hard, distinct, specular reflection. 2) Decay time. Most residential (and audiophile) listening environments have decays times (varying by frequency) of 500-1000 ms. This has a significant masking effect as the decaying frequencies obscure the perception of the next transient, or word in a vocal, or note in a melody. Long decays times are generally desirable in large chambers / theatres and halls of worship as they add a sense of lushness, but never for engineering purposes. Over the last 4 decades, the effect of decay times have been studied. Even decay times of 300-700 ms, which were common for old style control rooms, have now been proven inadequately long. Audiophile spaces have a completely different set of priorities, which is why I carefully set the context of this video before discussing the acoustics. Control rooms are not designed for enjoyable listening; they're designed to turn out mixes and masters that translate well, and to do that as quickly and efficiently as possible.
Great info , Thanks . I had a Yamaha NS10 on the left and a similar type speaker on the right but front ported . I was shocked when I realized after comparing the two each by themselves , the front port had 300 hz - ish info coming out of the port , most noticeable on vocals . Something you don't notice when using the ported pair but having an NS10 to compare , it was very interesting to discover .
Very clear, love the drawings to illustrate the physics. I already knew about SBIR but it's always cool to see how other people explain it ! Such a nightmare to deal with ! The pos 1 is definitely the best if you don't mind the listening position. In my case i really wanted that optimised listening position and didn't want to have my monitors far away from me close to the walls. I heard subwoofer could help and it really did ! Because it lets you position your monitors freely while the low frequencies are carried by the subwoofer which i positionned right against the wall (since it's omnidirectional it doesn't even sound like the subwoofer is far) 👌
Nice one. Thanks for sharing a bit about your experience. Yes, that's certainly one of the advantages of using a sub. You can position it separately from your monitors. Although subs have several substantial drawbacks IMO: 1) Because they're positioned in the bottom front wall/floor seam or a lower tri-corner they excite room modes far more than speakers up on stands. That's an issue in most rooms that have inadequate LF acoustic treatment. 2) You have to carefully phase align your sub with your tops. Subs have very rough phase alignment options (usually 0, 180, or sometimes in 90 degree increments) so usually you need to manually play with the distances (string test) or use phase-alignment software / hardware like a Trinnov. 3) And this is the biggest one...if you don't have 2 subs configured in stereo, now your entire low end is in mono and you cannot effectively mix your low frequencies as an engineer. Any professional studio has full range stereo monitoring, not a single mono sub. Of course, you could always use headphones to check stereo sub information. Those are just some issues I find. You may find them totally acceptable and that's all good. It's about getting good enough for you. Thanks for the comment!
Thanks so much for the in depth video and real world measurements of your tests. These kinds of tests, results and explanations are INVALUABLE and I’ve struggled to find many videos documenting these topics to this degree. Well done as always, cheers!
Thanks Gabriel. Glad you got a lot out of this one. Opinions are plentiful, but facts are rare. I like the data-driven approach with actual acoustic tests. Cheers and best of luck with your music!
You're absolutely right, flush mounted or inside of two feet. It's worth noting doing so elevates the importance of rear wall treatment, ie., all modes additively sum up and at a pressure maximum. Obviously you're on your of it with three feet of fluffy or whatever.
Hey mate. Thanks for the comment. Yes, that font back axial room mode is down around 35 hz with our room dims. So the depth of treatment we have is absolutely necessary to reign in that mode.
I’d love a video on the role of ceiling corner traps and what they typically address (assuming near-wall placement and basic acoustic treatment: vertical corner traps and first reflection on side and cloud).
no its not just you saying this. it stands for example in the genelec manual for their SAM Monitors. I did it that way - pretty happy with the results.
@@warpacademy thanks for the video. And I can see it's relevance, because when I first read it 2 years back, I was a bit confused, why they told me to place the monitors so close at my walls 😀
That's one of the rare examples I've heard of where a speaker manufacturer actually gets it right. Quite often they say to gap them from the walls, which makes no sense when you know the physics.
So much valuable information! Thanks a lot for sharing. I can confirm all your test as I have both a Treated Mixing Studio and a Home Studio both calibrated with Sonarworx and I did get these curves as I tested different speaker positions. You put years of test in such a clear and concise video. Wonderfull work.
Good video and testing methods. 🙌 I also like this placement for the same reasons you mentioned. A few questions/comments: the 160hz dip you saw in the measurement you've attributed to floor bounce, but I didnt see it in the other measurements. I think that might be related to rear wall reflection. Another thing to note is that a dip in response caused by destructive cancellation cannot be fixed by boosting the signal since its also boosting the destructive reflected sound wave. Let me know if you disagree. Im open to being wrong.
Hey there. Thanks for watching and commenting. You've got a great channel as well! Regarding floor dip, good eye. In the floor dip calculator the dip frequency works out to be 158 hz, which is bang on where we're seeing the dip. You're correct in that it doesn't show as much in the other measurements, but it can easily be getting blurred by constructive interference from the front wall SBIR when speakers are offset from the wall. If the other boundaries, such as the back wall, were not heavily treated then it would be conceivable that it could be a reflection from one of them. In this case, however, we have 3.5 feet of fibreglass in the back wall, 2 feet in the ceiling and 1.5 feet in the side walls so far. The absorption coefficient for that depth of treatment is high enough (0.85 to 0.9 at 150 Hz) that it should eliminate those treated areas (which are floor to ceiling) as issues. But we'll only know for sure once we finish the room and get a desk in there. Very insightful observation on this one. Regarding the EQing of the drive signal, this is what I'm basing my comments on... It's not information coming from me. I haven't tested this personally and I'm relying on research from the AES. Take a look at the Journal of the Audio Engineering Society, volume 33, page 127-132 (Clarkson, Mourjopoulos and Hammond). Also the AES book Recording Studio Design by Philip Newell, page 332-334, which states "Near perfect response at optimum listening position after correction (with digital adaptive EQ)". This example was based on a test where a loudspeaker was positioned on a stand, part way through a room (exactly as I did in position 2 and 3), after which digital adaptive EQ (like Sonarworks) was applied. This section of the book is also key to understanding why and how the correction can occur, "A non-minimum phase effect (such as when a loudspeaker is freestanding in a room, and not mounted at or in the wall) cannot be universally corrected by equalization (alone) of the loudspeaker drive signal. Even perfect amplitude correction could not restore the original phase response by any known practical analogue means. Only by digital signal processing could an almost perfect response be restored. Adaptive digital filtering can model, very accurately, the inverse phase responses necessary to correct either the minimum phase or non-minimum phase components of a transient or steady state response anomaly. By means of measuring microphones, modelling delays, and adaptive filtering processes, the digital systems can be made to 'learn' what a given room will do to a loudspeaker response, and apply acausal corrections to cancel the disturbances [acausal, in this sense, means that by means of a signal delay, the measured response error is fed into the drive-signal, in inverted form, to precondition the output to anticipate the error." -Pg 334, 11.5.2 Digital Correction Techniques, Recording Studio Design, Philip Newell, Audio Engineering Society. This is exactly what Sonarworks is doing under the hood. Let me know it that makes sense. Cheers!
@@warpacademy The Quote from P.Newell doesn’t address the SBIR issue, it really doesn’t matter what leaves the loudspeaker at a given time, as the reflected wave will itself be changing with the generated wave, always in antiphase at that centre frequency that is a 1/4 wavelength, and frequencies either side in decreasing amounts.
Hey mate. The quote from Newell does indeed address this. He's speaking about the ability of non-minimum phase response modifications (speaker on a stand away from a boundary) to be corrected by digital adaptive EQ. Take another look over the quote, or read the book as it seems quite clear and discusses the mechanism of action (correction). But seeing as it's not seeming clear to several people on this thread (and because I'm always up for understanding a topic more deeply, or open to being wrong), I followed up by speaking with Philip Newell directly on this matter, and here are his comments on this specific topic: "I stand by what I said in the section that you quoted from Recording Studio Design, but it has to be taken in context. Acausal filtering and signal processing, WITH A MODELLING DELAY, can correct for the room boundary effects, but only for a very small region of a room -- say a very restricted 'sweet spot'. Elsewhere in the room, the response may then be much worse than before the correction. So, for example, if someone must work in an acoustically-poor room, this type of processing could greatly improve things at a restricted listening position, which may be beneficial to a single user, or it may 'somewhat' improve things over a slightly wider area, but it cannot make an overall improvement to the room acoustics. When one place is improved, another place will suffer." The key thing to understand here is that this is not just amplitude correction from a standard EQ. A platform must have a modelling delay feature to compensate for boundary effects and phase. So it is indeed possible and Philip has specifically confirmed this. Now, I also followed up with Sonarworks, who also confirmed this is possible with phase shaping. I was mistaken in that their current platform doesn't do phase shaping, so I stand corrected on that. They plan to roll this out in a future release.
Hey David. Yes, there are a lot of people giving speaker setup and room orientation advice who have never studied acoustics. Acoustics is a science and all of the books I've read agree and are consistent on this topic. But no book or theory can replace real world experience with acoustic testing. So that is what I did in this video. Real world situations, real testing, real data.
Hi this is a very informative video on speaker placement, thankyou for the detailed explanation, could you also give help for the addition of a studio sub, and where would this be placed? Is the best position parallel to front of monitor speakers but still very close to the wall if not against?
Hi there. Thanks for watching. With a sub, that complicates things for sure. Detailed acoustic testing and tuning is necessary. Some things to keep in mind. The sub is on the floor, so the distance from the listening position to you will be longer due to the downward angle to the sub. So typically you'd want the sub to be equidistant from your monitors, meaning the monitors would be in front of the sub. This keeps everything in phase. But many subs have phase adjustment so you can adjust that as well. Typically you'd want the sub loading against the lower front corner of the room (where the front wall meets the floor). This will increase the amount of LF energy it's creating. From there, you can move the sub left or right to change its distance (and phase) or adjust a phase parameter on the sub. The other consideration is if the sub is highpassing your monitors. Basically, you want to very carefully test the room and experiment while tuning the sub.
My experience is that you should not trust the graphs in Sonarworks. In my case the software was boosting 100 Hz by 7 dB and the correction graph showed a nearly flat line. I have double checked it with REW and the dip was still there, even a little worse than before. As far as I can understand this makes sense. If I raise the energy of a frequency coming out of the speaker, the energy of the reflected wave from the frontwall also has more energy, so as far as I can tell raising the level of a frequency to fix a destructive interference caused by SBIR doesn't work.
Hey mate. Thanks for joining the discussion. It's a smart idea to shoot the room with REW, if you know how to use it, to get more granular with your test results - if that's what you're looking for. I did the same thing. I shot the room with REW and Sonarworks SoundID Reference at each location to corroborate my results. If Sonarworks is boosting 100 Hz by 7 dB you have a substantial acoustics issue in your room. Are your speakers gapped from the front wall? Sonarworks can't correct for a incorrect room setup. It also can't correct for a lack of appropriate acoustic treatment. As I said in the video, don't create time-domain acoustics problems like SBIR and then expect to get good results. Sonarworks should be applied after room setup is optimized, and you've acoustically treated the room. It sounds like you have SBIR off your front wall, ceiling, side walls, or a combination thereof. Feel free to reply with information on your room and I'd be happy to advise on how to fix the 100 Hz dip. The issue of boosting an SBIR null has come up a lot and I've posted my thoughts about this at length in the pinned comment up top. Please refer to that. The Sonarworks platform can be a bit misunderstood if you don't know what's going on underneath the hood. It's not actually flattening the frequency response in the same way that REW would show it. It's taking into account the psychoacoustics of how we perceive a flat or balanced sound. You must account for the fact that the human ear is not linear and our ears or much less sensitive to certain frequencies. This is where equal loudness contours come in; they demonstrate that very well. There has been much research into what curve produces the perception of a balanced response. Sonarworks has done their own research and the "flat" target represents what we perceive as a flat frequency response in a a typical room. Typically that means a slightly exaggerated low end and a slightly depressed high end. It obviously depends on how the room is treated and the directivity of the HF drivers. Read more here: www.sonarworks.com/whitepaper
Thanks. Which part are you talking about? In the beginning I was using a lav mic (I really don't like them, but use them when necessary) and further through I was using a boom arm mounted condenser. In both cases I'm using 2-3 de-essers throughout the post-processing chain. Glad it sounds good. Cheers!
It's crazy, a bunch of my questions were covered! I just changed my old monitor speakers to a threeway Eve sc3070 monitor system and i wanted to make sure with measurements etc how to get the best sound efficiency in my room working with these monitors. Did a lot of research... and research created a lot of questions and then after two weeks this video pop up in my YT feed... and i did get my questions answered !!! Soooo big up to Warp !!! In a small room (treated thought) i got the best results when i placed my speakers as closes as it gets to the back wall. It's not as simple as it sounds but it's working and as soon as you understand what you are trying to achieve in a home studio situation it's a solid advice!🙂
Right on. Yeah, you definitely get the best results by doing the research, doing the testing, and following the physics of sound with all the best practices they dictate. Glad you're getting good results. Cheers!
Translating this to living room speaker placement, I guess the same placement advice applies to small bookshelf speakers ? ... but what about full range towers that reach to 25-30Hz ?
Hey hey. There's an important physics principle to get here. It's about the size of the driver relative to the size of the wavelength it's producing. If the driver is larger than the wavelength of the frequency, it will be more directional. For example, with these speakers th-cam.com/video/yzopMQBlTZY/w-d-xo.html. Their single driver is 5". A 5" wavelength corresponds to 2,720 Hz. At that frequency and above it, the speaker will be progressively more directional. Anything below that will be progressively less directional and eventually omnidirectional. With smaller bookshelf speakers, assuming they're full range, the driver producing the LF will be very small relative to the wavelength of the frequencies they'll be producing. 60 Hz is about 19 feet long. So that means that not only the LF will be omnidirectional, but the low mids will also be radiating spherically. So with small satellite style speakers, it's much better to have them highpassed with a sub handling bass management. Full range towers reaching down to 25-30 Hz will absolutely benefit from being right up against the wall. But remember, you will get front wall loading, which will broadly increase the LF and you will need to apply a low shelf EQ to attenuate that if you don't like the bass-lifted signal. Some people may find the bass lift pleasing as our ears are much less sensitive to bass in general and we need more bass to sound balanced, psychoacoustically. This is why Equal Loudness Contours exist. Hope that helps. Thanks for watching and commenting. Subscribe to the channel and stay in touch!
THX for this video. But one Question is still left. In my opiniion you can´t correct dips as long it is the result of a phase/comb filter effect. What I know is, if you push level up the frequency of the dip then you push the level of the signal that produces the dip, too. So the dip must be the same. Or is there something I don´t see?
This is a very good question. If you were to have the speakers off the wall, such that there was a phase difference, and only use normal EQ to adjust the drive signal, then you'd be totally correct. Normal EQ would just boost the offending frequency, and still cause the dip. However, some forms of digital adaptive EQ is also taking phase into account. For evidence and corroboration, read Audio Engineering Society Recording Studio Design by Philip Newell, Page 333. He confirms this in the acoustic testing. I thought that the Sonarworks platform was doing this, but I was mistaken. Also read the Journal of the Audio Engineering Society Volume that I reference in the video on screen. This is the paper to which Philip Newell refers in the book.
I've updated things with a pinned comment about this at the very top of the comments. Please refer to that. You definitely can correct the response and the comment explains how. Cheers!
Interesting. Funny though that these audio shows have listening rooms and it looks like most of them are pulled their speakers 2 to 4ft sometimes even more away from the wall. Even some of these million dollar audiophile systems do the same? Why? If it would sound better closer to the wall. Just curious what your input on that is.
Yeah systems like that have done so due to some misguided interpretations of the physics behind things. As well as placing less priority on low end response.
Thanks ! Another super great video : clear, valuable & easy to understand. I feel less stupid today, thank you :-). I hope you will shoot another one about "where I should put this damned sub !!!".
You're most welcome. Thanks for taking the time to go through this deep dive and to invest in your knowledge. Learning this stuff is so key when you're a musician or engineer trying to get the best sound out of your equipment. Your question about the sub is a really good one. I definitely plan on shooting a video on that. I have a test room where I'll be showing how to build the best DIY acoustic treatment, how to position it in the room, and how to optimize a monitoring setup using Tannoy Reveal 8s and a 10" Genelec 7060B sub. I'll show all the acoustic testing along the way.
Thank you so much for this video! The most turorials are made by absolute m0r0ns who tell you that rear ported speakers can not be placed close to wall compared to front ported speakers (like this "genius" here: th-cam.com/video/ZjmqOjle2T0/w-d-xo.html ) FOKK EM! HAIL YOU! I am currently upgrading my studio, and looking for the best speaker placement. Yesterday after your video i finally püut my Yamahas H 80 S back to the wall and i fanylly enjoy bass again!
You're very welcome. I also used to hear that advice about gapping rear-ported speakers from the wall, that is until I studied acoustics and started to understand the actual physics. The issue is that just because you're a music producer or audio engineer, that doesn't mean you know anything about acoustics. It's a niche field, very technical, and not intuitive in the slightest. In order for a port to function, there must be a very small gap to the wall equivalent to at least the port size. So a 2" port will need a 2" gap, approximately. You just don't want to block the port or partially inactivate it. The thing is that speakers are toed in by 30 degrees, and even that little toe in will often create enough of a gap that a rear port will function normally. Even if you stuff something in the port to intentionally block it, the speaker will still function; it'll just reduce the port resonance and the speaker will function more like a sealed cabinet (IE less steep rolloff, but a higher cuttoff frequency). Glad you found the video useful. All the best.
Hi there Sir! This is great video that helped me a lot, but i have few questions that you kinda already answered but i need to confirm sry. :) So in a typical bedroom, 3,5meters l ,3.35meters w ,2.5meters h, i would place my speakers between 2 walls, and not between wall and window side, and speakers should be very close to the back for the best results right ? :) And i have a question how would objects in the room like bed, storage thing and couch affect the frequency response :/
Hey hey. Glad you liked the video. For more detailed questions about studio design and room setup join our free Discord server and post your question in the Acoustics forum: discord.gg/ZwNgZteGKw
Exactly. Most professional studios that are built around acoustics best practices, that is. Which is exactly what I did in my room. Here's the finished room: th-cam.com/video/XfO_btDeXjc/w-d-xo.html
@@warpacademy Im right now also in the planning phase of my studio. Is it possible to get it touch with you per email, maybe you can give me some tips for my room and the dimensions. I'm planning to make drywall too and make the sound and dimensions as symmetric as possible. Until now I only finished the framework of one wall ( front wall) but it's still easy to change things. Can you help me to plan my room ? Im also willing to pay a tribute. That would be nice! I'm also planning to make some videos and I would mention your videos and erpertice and help in them. That would be awesome.
My issue is that im using a sub. I have to make sure the speakers and sub are aligned so that they're in phase. Is it better to put the sub close to the front wall as well?
Hey Jeff. Having a sub is actually advantageous. You can (and should ) put your sub directly against the boundary, yes. That'll ensure you have a close to in-phase wall reflection that will add LF amplitude to the sub and allow it to operate more efficiently. Given that a lot of subs cross over at 80-120 Hz, you may still get boundary nulling with your tops, so it's also good to place them against the wall. Then for phase alignment, you have several tools at your disposal: 1) Manual phase alignment. You can use the tools on the back of the sub (phase switch) to roughly adjust phase. Then you can slide the sub to the left and right physically to fine tune phase. 2) You can use an automatic phase alignment system that's software or hardware based (Trinnov). Either way, it's pretty straightforward to tune the system. Make sure you're running tests all along the way with Room EQ Wizard or another room testing app.
But here you are talking about right against the wall WITHOUT the front wall being acoustically treated. I have four 4" thick GIK panels plus two 16" corner bass traps on my front wall.
Yes, exactly. I wanted to isolate the effect of a hard boundary to clearly demonstrate the nulling that happens when you gap the speakers from the wall. Of course you'd be adding treatment to your corners and front wall. It's good that you have thicker treatment in your front corners. I hope it's floor to ceiling. One of the most common acoustics mistakes is to use corner bass traps that leave the top and bottom tetrahedral corners untreated (IE they're only in the middle of the wall). Also, your 4" GIK panels are certainly helping, but they will be doing very little under 100 Hz. You'll most certainly still be getting a bass reflection off your front wall that will cause nulling. I expect you're seeing this in your acoustic tests. If I was going to place treatment on the front wall, I would use at least 5.5" panels made of rockwool. Ideally 2 layers of 5.5" so you're looking at 11" of trapping. And you'd want to build the treatment around the speakers, not behind them, like the studio I featured in the video.
@@warpacademy Yep, my bass traps are floor-to-ceiling. They are the Impression Pro Series Corner Bass Traps (ECOSE rockwool). The 4" panels (also ECOSE rockwool) coefficients are 0.7 at 100 Hz, 0.3 at 80 Hz, and 0.1 at 63 Hz. Up to 11" thick of trapping on a front wall would be impractical for a bedroom studio at 11' x 13'. I also have the 4" panels on the side walls. In the future, I could move up to 6" panels on the front and back walls though.
Sounds like a pretty smart setup for a bedroom studio. In terms of the 11" of trapping on the front wall, consider that the front baffles of your monitor speakers are likely to be that distance from the wall, or thereabouts. If you build the trapping so that the backs of your monitors are against the wall and the trapping modules sit flush or close to flush with the baffles, then you don't really sacrifice any space in the room. It's only when you try to put treatment behind the monitors that it becomes an issue because the monitor then needs to step forwards in the room. Perhaps you have yours mounted on a desk though, I'm not sure. Due to the near square shape of your room, you likely have some issues with modal spacing and stacked tangential and oblique room modes causing significant peaks and nulls, so some additional front wall trapping would be helpful if you can.
"Forever". Not really. It takes about 10-15 minutes to complete the testing and you have to do it once. It's far more streamlined and simple than using something like Room EQ Wizard and trying to manually create a parametric EQ curve. I've done both. It does all the work for you beneath the hood. All you need to do is watch the on screen tutorial and correctly position the mic for the testing points.
@@themattmcclellan Audiophiles almost universally suggest the speakers should be well into the room and here we have the suggestion to put them right against the front wall.
That's because those types of audiophiles are often more obsessed with smaller details and aren't paying attention to comprehensive acoustic testing. They also may have different goals from myself (as a mix engineer) - which is accuracy for making critical engineering decisions.
They definitely produce a lot of LF SPL, so you'd need very thick treatment. Yeah, with those room dims, you'd loose a lot of internal volume. Depends on your ceiling height too. If you're dealign with standard residential ceilings of 8' (
@@warpacademy Just like you, I've been rocking the DynAudio BM-12A MKII version for a bit more then a decade and a half. I still love them. Given the stated dimensions I'm considering the CORE 59. They do go low, but the LF roll off sets in quicker. I have to check the Freq roll off chart. Maybe those are more manageable given that they're DSP regulated and manual setting to further tweak them. I just want more low at speaking level.
Ah, nice one. Yeah I loved the Dynaudios while I had them. I did seriously consider the Core59s. Let me know what you thik of them if you end up putting them in your room. Cheers!
I've been sharing these videos with my clients, students and discord. Thanks for another fantastic video, you really go above and beyond on everything you do.
Don't position your monitors until you've identified the optimum listening position. Then position your monitors using the phantom centre method. Then using REW treat your room acoustics the very best you can. Finally use ARC 4 to tweak to taste. Bingo, did this for new mixing room and it sounds incredible. All details at Acoustic Insiders on TH-cam. 😎
Hey mate. I've watched those videos so I'm aware of what they advice; I just happen to disagree. Jesco talks about the importance of setting your priorities when designing a studio. Well, I place a very high priority on bass response / accuracy and you will never get a good bass response by placing your monitors out into the room. That's physics 101. SBIR / boundary nulling will shoot you in the foot. Do I mean that I don't care about phantom center, or a really solid sense of stereo field? Not at all. It's just not that hard to do. When you have enough side wall treatment to attenuate or absorb lateral reflections, the phantom center is amazing. It's reflected energy and diffraction that blurs your sense of phantom center, so you deal with that with absorption. Easily done. What is not easy is trying to "fix" your bass response after setting up your room in a way that ignores SBIR in the low end. You cannot simply place acoustic panels behind your monitors. That'll only partially reign the problem in, but not address it fully as the depth of treatment necessary to kill the reflection is enormous. And if you think I'm the only one who is saying this, I'm not. What I recommend is consistent with what Philip Newell and Wes Lachot say. They are professionals who've built some of the most important and accurate rooms in the world. As you may know, Wes Lachot is the creator of the "38% Rule". Here's what he has to say about the priority that bass response should take, and placing the speakers on the wall or out into the room. "All is won or lost in the bass." "You can buy your Focals or whatever, but you’ll just be guessing with your bass!" "You can move them around for years, but you’ll just create new problems." "You should always put speakers against the wall unless you have a very deep room! 9-10 feet from the wall in a 30 ft deep room, otherwise no."
Hey hey. Very good question. The infamous 38% rule was created by master acoustician and studio designer, Wes Lachot. Here is the post Wes made about it: repforums.prosoundweb.com/index.php?topic=21551.0 The vast majority of people misinterpret this "rule" that is actually just a guideline. That guideline was intended to locate the mix position away from the center of the room and to avoid pressure nodes and anti-nodes, especially overlapping ones, due to boundary spacing. It's important to note that Wes stated 38% for control rooms that are 20' deep at least and where the speakers have been flush mounted into the wall, like my control room, and not for speakers on stands: th-cam.com/video/XfO_btDeXjc/w-d-xo.html&pp=ygUad2FycCBhY2FkZW15IHN0dWRpbyByZXZlYWw%3D I've spoken to Wes directly about this. Studio design and room setup is all about priorities. Wes states, "It's all won or lost in the bass." If you gap your speakers from the wall like shown in this video, you get a huge hole in the bass, as I've demonstrated in this video. "You can buy your Focals or whatever, but you’ll just be guessing with your bass! You can move them around for years, but you’ll just create a new problem" So he is in full agreement with what I've presented in this video. Bass frequency response is one of the highest priorities in room setup. And this final quote from Wes sums it up and puts this issue to bed, "You would always put speakers against wall unless you have a very deep room! 9-10 feet from the wall in a 30 ft deep room, otherwise no."
I covered all surfaces of my listening room with 3.5 feet of bass trap. My room is now 18 inches tall. 11 feet long. 5 feet wide. But, it sounds phenomenal. Least ways I assume it does. I can’t get my 53 inch tall speakers in the room. At least my chihuahua has found a safe place, where my English Mastiff can’t terrorise him.
You point out a very common and definitely laughable aspect of incorrect room treatment. If you don't have much space to work with, then it makes little sense to use the type of treatment that would be ideal for a ground-up professional room. If you're working in a bedroom, it should be common sense that you'll be working with thinner treatment to preserve the working space.
Hey hey. Thanks for the comment. I would agree that your monitors, your desk, and you room all influence the final sound you hear in your studio without a doubt. And, at the same time, this is basic physics that will apply regardless of these things, universally. The only exception are cardioid speakers, and rooms that are oddly shaped to create an RFZ. All standard speakers will have rearward radiation of bass, all rectangular or square rooms will bounce back that reflection, and it will rejoin the direct signal causing boundary nulling. The desk is inconsequential as it does not impede the offending reflection from the wall behind the speakers; it can only impede floor bounce. The only variables are what frequency will be nulled (relating to the distance from the boundary) and how much it will be nulled (relating to the acoustic properties of the materials the boundary is made of). I myself have confirmed this phenomenon in every single studio I've tested (including many studios of other producers). This principle is discussed at length in acoustics books, such as Recording Studio Design by Philip Newell. This is established science. Once you grasp the physics involved, you can understand why this is universal.
Neumann all the way. The Adam XART accelerated ribbon tweeter design is so hyped and unnatural sounding. No one is going to be listening to the end result on that tech so don’t mix on it.
Thanks. Good question. A sub will behave exactly the same way. If you put it out into the room then you’ll get massive cancellation and boosts. The best place for a sub is built into the wall flush (and then you phase align it using digital time alignment software) or placed against the wall.
Thanks so much for this video. It's extremely informative and corroborates some other guidance I've seen on the subject of speaker placement. I'm looking forward to trying this method with my own speakers, in combination with Dirac Live room correction. Hopefully this technique works as well for stereo listening rooms as it does for studios.
Hey mate. Glad this was helpful for you. This is unpopular advice with a lot of living room / audiophile folks. There's a lot of outdated or just plain bad setup tips out there that aren't grounded in any kind of actual testing or physics. Gapping your speakers out from a wall, 1/3 of the way into the room to "solve" SBIR is the biggest one of them all and that's why I focused on it in this video. Good low end response accounts for 30% of the listening enjoyment experience, even if you're not a professional engineer (per Floyd Toole). And there's only one way to get a relatively flat low end - it's to put the rearward radiated bass energy from the speaker as in-phase as possible with the direct sound. Gapping a speaker from a wall, or putting 6" of porous absorption behind it does not stop the out-of-phase reflected bass energy from causing low end interference - the distance and the absorption characteristics of porous absorption are not capable of that - at least not with that amount of time of flight and that thickness of absorption. So really, you must put your speakers against the wall if you care about low end response.
Thanks for a great video! Gonna try the monitors up against the wall again, have experimented with different positions out from the wall, and having the exact problems you describe..
You're most welcome. Glad it was helpful. Yeah, this one is a very common issue in music studios, but easily resolved. I encourage you to run before and after acoustic tests in Sonarworks, if you have it, or Room EQ Wizard. Use the data to confirm the difference in room response. Feel free to post a comment or question if you need any advice going forwards. Cheers!
Hello, Do you want a flat target curve or a tilted or Floyd toole target curve? Or do people in studio always want a flat curve at listening position? When you experience some listening fatigue do you change the tuning? Regards Greg
Hey Greg. For me, I prefer the Bob Katz approach with a little low end lift. It can be interesting to experiment with a gradual, progressive HF rolloff as well (per Katz) but I don't find I'm a fan of that in this room and on these monitors. I don't make adjustments for listening fatigue. I take breaks. What I do make adjustments for is monitoring sound pressure level, because...Equal Loudness Contours. The ear is much less sensitive to extreme LF and the psychoacoustic perception of LF is very compressed as you increase SPL.
Thanks for your reply, interesting. I am just an audiophile, I design my own active speakers and I have a room which is half treated, some csd side walls and plenum plus corner traps behind the speakers. It helped a lot to measure the room plus speakers and for bass and clarity. Of course it is not perfect. I also have directive speakers with mid horn, hf horn, and ripole subwoofer at 80hz, this helps not sending energy to the room. I read about Bob Katz and I think this is a little bit what I do modestly. I also compare with my neumann ndh30 eqed with oratory1990 profile. Then it sounds pretty similar. If I want to retrieve the same amount of details with the speakers I have a more fatiguing experience. I don't know why. But we are talking of tiny differences. Regards Greg
Sounds like you have a strong interest in audio and speaker design. That's a neat passion. I know very little about designing speakers, but I do enjoy listening to good ones in a good room! All the best Greg.
Another heavy hitter of a video! One thing to consider with the boosted lows in a situation with monitors flush against a wall is that it's actually preferable to flat when mixing low, and is just more fun in general to mix that way as opposed to flat.
Thanks for the comment. You’re totally right. When mixing at low volumes the bass will be perceived as quieter due to equal loudness contours. So that would help avoid over mixing the bass. Good point!
Yeah, but up against the wall sounds gross everywhere else. You need bass, get some subs. I'm always near field or way out from the wall, via experience. Even my subs are away from the walls and lifted 3 feet off the ground.
To be clear, we're talking about music studio control rooms here. Not audiophile hifi systems. It's an industry-wide best practice to flush mount monitors in control rooms for a multitude of reasons, and barring that, to put the monitors right up against the wall. If you're optimizing in an environment that's not a studio control room for engineering, then just do what sounds best to you. Cheers!
@@Audfilethere is only a single position that matters in a studio and that is your listing position. If it sounds good in the listening position and terrible everywhere else you are winning 🎉
Well said. We generally optimize for the listening position. In my room - and many professional rooms - we needed a wider view though. You need to account for the fact that you'll also have clients and students in the room alongside you; not just a single engineer. In my room (it's a teaching studio where I sometimes have up to 8 additional people) it's a goal to have the frequency response as good as possible throughout a larger area. Of course it can't sound good everywhere, but by eventually flush mounting the monitors into the front wall (we're doing that right now) you can achieve that because it's minimum phase and no digital adaptive EQ correction is required. We can account for the LF lift (baffle step) by shelving down the monitor drive signal using analog EQ on the back of the KH420s and this will improve the response in all positions of the room. In most home and project studios though, you only need to have 1 area sound good. The mix position like you said. Thanks for watching and commenting!
Nice video. I might be on the wrong channel for this question, but is this video only applicable to recording rooms or should a regular listening music/theater room also have the speakers as close as possible to the front wall?
Oh that’s all good to ask here. If you have a bass managed system with a sub then this likely won’t be an issue for you as you can position the sub against the wall and the tops more freely. If the low end is also coming from the stereo mains and not a sub then you’ll benefit from putting them against the wall. But then you’ll want to shelve down the low end a bit to compensate for the front wall loading.
I think that’s because most people don’t take the time to understand the basic physics at work here. It’s not rocket science yet so many people get this wrong based on years of bad advice propagated on the internet that was never grounded in science or research.
Many good informations but still not the right conclusion. There are so many more different parameters who have an influence on the sound. If you talk about a true studio environment you may be right. But look around in all the studios worldwide - the near field speakers are mostly placed directly on the mixer in front of the engineer. So - are they in the wrong place? Of course not. To make a long story short - all the time we place a speaker in room we have to start right at the beginning to find the perfect place for them in the environment where they have to work and for what kind of task they are used. To tell all of us that there is only one perfect answer for all of the questions is stupid. And if you will show thousands of books that all tell the same it doesn't make it right. I have more than 40 years of experience with audio and studio equipment for hundred of thousands of cash for one setup and there was never one solution that fits them all.
Hey mate. Thanks for joining in the conversation. To clarify, my audience here is music producers with music studios, so I'm most certainly talking about a real studio environment where there is acoustic treatment. I'm not talking to audiophiles with living room sound systems, although many audiophile setups could benefit from taking these principles into account. You bring up a very important point about many professional studios and how they place nearfield monitors on the mixing console meter bridge. Despite the fact that you see this a lot, it leads to severely compromised acoustics. Just because lots of people do something doesn't make it a good idea. Any acoustician will tell you the same. It's obvious that this is not a good setup when you understand acoustics 101 and the rear radiation of bass. Also when you understand concepts like diffraction of audio and the consequent smearing of stereo image due to reflections off the mixing console, you'd never place monitors where their propagation of sound would contact a mixing console in such a close and direct way. So why do you see professional studios and engineers doing this, despite the fact that it violates many acoustics best practices? Because real education about acoustics is sadly quite rare and many engineers don't actually acoustically test their rooms, or they use room correction to attempt to solve problems that didn't need creating in the first place. Nearly all of these rooms have mains that are built into a flush mount setup by the original studio designer (who does understand acoustics). But in practice, so many engineers like to throw multiple sets of monitors into the room on stands or on the meter bridge, usually without carrying out testing or considering the full impacts. Read Philip Newell's book that I reference in the video. He weaves in countless stories of studio owners and engineers who don't properly understand acoustics and he's called in to fix avoidable problems. He talks a lot about mixing desks in general, and how big consoles (and other equipment) in the room actually negatively affect acoustics ("BDE" big desk energy) through diffraction. Look at really well designed studios, made by people who do high level acoustic testing, and you'll never see these mistakes. Look up any room made by Northward Acoustics. No nearfields. No mixing consoles. Just a very small vented desk with no reflections of diffraction sources. But outside of the studio control room, I'd agree with you. Find a good position and the room acoustics are less controlled so you have more experimentation to do. There isn't a one-size fits all approach for many (any) things in acoustics. In this video I wanted to demonstrate one simple thing; the effect of the distance from the front wall boundary on non-cardioid speakers. And in this case, there is only one position that works as I have demonstrated with testing. Right up against the wall.
Thanks for this video. I always used to have my speakers up against the wall, but for some reason, on Adam's instruction manual they say to put the speakers 30cm away from the wall ... So I did with my new Adams, and I get the exact dips and boosts you show in the vid! I'm so glad I can reclaim that 30cm of space I lost by moving my entire desk back off the wall haha. Only issue is I need to find my measurement mic so I can recalibrate my system!
Glad it helped! Yeah, much of the advice I've seen in speaker manuals is just dead wrong. Speaker designers care about frequency response in an anechoic chamber, and don't seem to give much useful advice for real-world situations that include putting them in a room. It's a fairly deep understanding of acoustics that's needed, but you'd think they'd be aware of this issue and the fact that once you create it, you cannot solve it using room EQ or realistic acoustic treatment. It's a damn shame so many people compromise their low end frequency response by following that wall gapping / speaker positioning advice without ever testing it themselves.
Many years ago I read a lot of literature on room acoustics and if I remember correctly the timing of the reflection from the original signal has an important effect on our ear's perception. I think it was something like a reflection 5-10 milliseconds delay after the original wave is perceived as a separate sound by our brains. Less and it's perceived as the same signal. It would be interesting to know the calculated delay for the three positions and room modes from other wall reflections. We can not localize bass frequencies like mid and high ones. Is optimizing the bass hurting the other frequencies? Is the effect of room modes caused by reflection from back and side walls more important? If my memory serves me correctly you should be able to calculate the best speaker and listening position which will be different depending on room dimensions to minimize room mode effects. Was the microphone positioned in a room mode?
Hey Dwayne. Thanks for the comment. I think what you're referring to is the human echo threshold. Although it can depend on the type of sound (transient vs steady state) and frequency, it's typically around 30 ms. But in some circumstances it could be shorter. You can test it yourself with a delay effect to see where you can start to discern a the delayed echo. I usually start to hear a distinct delay at 30 ms or so. In a music studio control room, you want to optimize all frequencies of course, and many other things, such as the decay rate of each frequency (addressing modal ringing), the stereo field and phantom center, the impulse response / ETC of the room (the liveness of the room as well as especially loud early reflections, desk reflections) etc. The low end just happens to be one of the most challenging aspects to optimize, which is why it's important not to create unnecessary problems through bad room setup and speaker placement. When your goal is a very controlled and accurate listening environment, you'll be using extensive acoustic treatment. Because of the frequency-dependent way that porous absorption works, mid and high frequencies are easier to control - hence the focus on getting the low end right from the start. All room modes are important, as is the spacing of room modes. The room modes are usually the most isolated and spaced out (IE uneven and lumpy looking in REW) in the low end. As you go higher and higher up, the room modes start to create a more even effect (IE the Bonello criterion). You typically would be designing a room based on one of the golden ratios using a room mode calculator to avoid stacking room modes on the same frequencies and ensuring the room meets Bonello. Although as you add more and more acoustic treatment those golden ratios get less and less important because you're damping them so heavily. In my room, we chose to maximize room volume rather than target a golden ratio. When you say, "was the microphone positioned in a room mode", do you mean in a node or anti-node? Cheers!
Glad you got a lot out of this. The general thinking in hi-fi circles is to move the speakers away from the boundaries in order to "reduce the effect of SBIR". My big question to that school of thought is, exactly HOW do you think that is going to reduce SBIR significantly enough to flatten out your bass response and avoid a large null. People might answer that when sound travels a further distance, it loses power. While that is correct, the losses you get due to increased time of flight are totally insufficient and you will still get a dramatic null from the out of phase reflection. And now the out of phase reflection (which always causes a null at the 1/4 wavelength frequency) is simply lower and even more problematic to address with any kind of acoustic treatment. It's a losing game that doesn't make sense when you fully grasp the physics involved. Also, I think that the people who follow this advice and gap their speakers either A) Aren't prioritizing an even low frequency response in their system, or B) Aren't actually testing their room with room testing software, or both. When you test your room, you will see the effects in the frequency response clear as day. It happens in every room that's a rectangle or square. And you'll hear the destructive effects of the null (and constructive effects of the peak) if you listen and care about the low end. 30% of the perception of quality in music is the low end, so you must get that right if you want an accurate high-quality (IE high-fidelity) experience. Reference on that is via Floyde Toole author of Sound Reproduction and former consultant to Harman.
Really great accurate information! Sonarworks is an excellent tool for finding problem areas in your room. I don't use Sonarworks to correct any problem areas, I think it's better to treat the room . Thank you for the info!
Definitely agree with you that your focus needs to be on treating the room and getting as optimal acoustics as you can. Then, you can use room EQ to address remaining issues.
Good video, thanks for making this. I do have a question though. What are your thoughts on the rule of thirds? How does that compare to the other approach you're suggesting which is to place the speakers right next to the wall? Thanks!
Hi Chris. Thanks for the comment. I think you're talking about the 38% rule yeah? Arqen answers this better than I could. Please see this: arqen.com/acoustics-101/room-setup-speaker-placement/ And this: arqen.com/acoustics-101/speaker-placement-boundary-interference/ Cheers!
Why the hell, math is not working at position no. 2. Wavelength/4 = 2"(0.61m) Wavelength= 8" (2.44m) Freq= 343/2.44 = 140~ HZ Sonarworks shows 50hz pulled back 9db instead of 140hz. Same is the case with position no.3
Theory and reality don't always mesh. I was quoting some of the information in the same way a theoretical acoustician would, but certain assumptions are made, such as infinitely rigid and massive boundaries that are perfectly reflective. In situ, this is not the case. Theory predicts there will be constructive and destructive interference in the bass range, resulting in a large cancellation notch and a corresponding peak. That is clearly the case in pos 1 and 2. Theory also predicts that the frequencies of the interference will shift, and again, clearly that's demonstrated. Don't get caught up on the math. It breaks down because the acoustics formulas involved are not sophisticated enough to take into account all the variables, and also input data such as transmission, TL, absorption and reflection coefficients are never available for all your materials.
Hi. I am a little confused about your math. You said that the 55 Hz dip is caused because the speaker is 2 ft off the wall. So if the wavelength of 55Hz is approximately 20ft, then 1/4 of that is 5 ft. The speaker in the video is 2 ft away, meaning 2ft is a quarter wave of 8ft, not 20 ft. Have I got something wrong.
That's theoretical based on a calculation. The calculation assumes perfectly rigid and perfectly reflective boundaries, which is never the case in actuality. What we're actually seeing is the front wall reflection, combine with the superimposition of side wall, ceiling, and floor reflections. The acoustic treatment I have on those boundaries was not very thick yet during the tests, so it was not sufficient to dramatically attenuate LF reflections. It effectively attenuated mid and high reflections, but the LF was still bouncing back. What can be easily seen though is the effect of moving the speaker in the room.
0:29 “…the data never lies.” If only it was that simple. 😊 My old Stats 101 course, among many other things, taught me about such hubristic nonsense. Good video, though. I will try it.
Well. This isn’t statistics where data can be cherry picked or manipulated by people is it? It’s a single measurement directly from the software. A data point. And it’s repeatable. And it’s been repeated in multiple rooms with different speakers. So in my books it’s reliable.
One question I have with that against the wall placement. With the Speaker responding the way they do against the wall, if you were to move the speaker a very small amount, would help to correct that rise on the low end? I'm thinking that it may do just that BUT, a small movement off the wall may create a problem in a different frequency area. Your thoughts?
Hey hey. I like where your head's at with this. Good to start to experiment and think things through. The concept to really get here is that of minimum phase. IE, the frequencies that are in phase completely or nearly completely with the direct sound. Very close to the wall will create a more well-behaved low shelf like boost. This can easily be corrected. However, even a relatively small distance from the wall will delay the bounce far enough that you get substantial nulling. Test the setup yourself with REW to see how far you can move it. But the key thing to realize is the bass boost is not a problem. The easy solution is simply to EQ with a subtractive low shelf. Or, you may actually like the lifted bass response. Many people these days seem to prefer a bass lift in the bottom end. I do (not a big boost, but 1-3 dB can sound great, given that our ears are less sensitive to SPL in that region).
Do you have any tips for doing that alongside a sub as well? Would you test them separately? I am planning on using Neumann’s MA1 for the eq, hopefully they can correct what you said and are linear phase
Hey hey. Adding a sub to the system will make tuning much more complex if it's by ear, but the MA1 system will make phase alignment very straightforward so that's good. Use of linear phase EQ in room EQ is of questionable benefit IMO. Mixed phase may give better results. IMO if you're doing anything other than bell attenuation in the LF, or broad spectral tilts, then you're potentially causing more harm than good. The acoustics of the room determine how helpful EQ will be. In the MA1 here are my tips: Do not allow it to "correct" any nulls by adding boosts. Like the Sonarworks system, it's not capable of addressing boundary nulling from SBIR. Restrict its correction so that it's only permitted to do subtractive bell EQ below 300-500 Hz to reduce the effect of constructive interference from boundary reflections and modal resonances. Then add any broad spectral tilts you desire using the house / custom curve. Bob Katz's personal preference for a "flat target" for a control room is flat from 20 Hz to 1 kHz and then a linear slope from 1 kHz to 20 kHz with 20 kHz being 7 dB down from 1 kHz. Then in many studios, engineers prefer a low shelf boost of 2 or 3 dB below 100 Hz, but otherwise Bob Katz pretty much got it right. It obviously depends on how the room is treated and the directivity of the HF drivers. If the room is less extensively treated and the HF drivers throw lots of energy out wide, then there will be more HF room gain.
I'm curious why some HIFI speaker manufacturer regularly recommend a minimum distance from the wall? they are not studio monitor, but still.. they should aim for a flat frequency response.
Simple. Because they aren't actually running room acoustic tests to verify that they're giving the right advice, or they have a bass-managed system and the sub is highpassing the tops so they're not radiating the frequencies responsible for SBIR. You cannot change physics. Anytime you have a full range speaker that's not a cardiod design, and you gap it from a reflective boundary, you will get SBIR that will null in your low end. Anyone who recommends this doesn't understand acoustics, or doesn't care about their bass / LF frequency response, or both. All you have to do is acoustically test the room and the data will not lie. You'll see the null clear as day.
Thank you, sir! So it's better to keep speakers closer to a wall with absorbers around than to have thick absorbers behind speakers and further from a wall?
Yes, correct. Absorbers behind the speakers will not do much for the sub range under 100 hz. Porous absorption performs very poorly there and it also performs more poorly when particle velocity is zero which is exactly what it’s at when you reach a boundary (wall).
Please elaborate on Floor Bounce. I'm attempting to optimize a room with couple significant dips (155 & 190hz) that does not have a "console" to block the destructive waves. I assume many people will find their way to this video looking to improve their Listening rooms that also do not have a console that is typical in a studio. Great vid BTW! Confirms my recent studies on the matter.
Hey Dom. Thanks for watching and commenting. Glad you found the video useful. While this video is primarily for music studios for critical listening and engineering, you could also block the floor bounce in a general listening room (like a hi-fi living room setup) by positioning a thick rug and coffee table in the reflection path. Although, for general listening, and not critical engineering of music, some floor bounce is generally acceptable to listeners. We're quite well adapted to hearing floor bounce as human beings. Cheers!
With all this focus on the non-directional freq interaction with the Front Wall could you please comment on the other room surfaces like Side Wall, Ceiling, & Floor. Would running the same experiment with a monitor up against the sidewall then pulled 2 & 4ft towards the center of room have similar results? I assume not as you would have just said stick em in the corners and be done with it...
Hey Dom. For sure. For that, please watch the detailed design video on my engineering control room: th-cam.com/video/5VrG2K_E7qI/w-d-xo.html The side walls, ceiling, and back wall are all treated with various thicknesses and densities of porous absorption and air gaps in strategic places. The floor is left hard and reflective, as is customary in control rooms, and helps out the function of the rear wall scattering slats adding some energy back into the ETC.
great video.... i have the exact frequency respond on time stamp 18:18. a dip around 1-2khz and a hump around 4-6khz. from your video, those frequency are 'flat' after the speaker is place up the back wall. 18:46 May i know why?
Thanks for the comment. The ranges in my video are much lower and they null and peak are due to reflected bass energy from the front wall. In my room the higher frequencies are flat because of the amount of absorption in the room. Those modes are heavily damped and the treatment easily absorbs at those wavelengths. By moving the speakers against the wall you get the reflected energy mostly in phase and you even out the bass response.
Wait why can't you block the port. I did that with a subwoofer way back which allowed me to turn it up louder and instead of being super boomy at low volume it was more even sound response and smoother. Cheap sub hack.
Hey hey. There are some reasons why you may want to intentionally block, or partially block, a port - you're right there. The purpose of the port is to extend the LF response of the speaker (or sub) by tuning it close to the frequency where the rolloff begins. So if you don't like the tuning of your speaker, and it's producing too much output at the port tuning frequency, then that may cause you to want to block it - at the expense of degraded LF extension. The speakers I had were very well tuned, so I did not want to play with their ports. In this case, the only reason it worked for us to block them was because the monitors were getting highpassed by the sub above their tuning frequency.
They are amazing headphones for detail, dynamics, sound stage. The frequency response is a touch dark on the top end and a touch bass-light for my taste, but that's my only slight criticism. I would just EQ them a bit with broad shelving filters and they're perfect.
Normally when doing acoustic tests that would be correct but that’s not how Sonarworks functions. You hand hold the mic and the software echo locates it and you test over 30 positions in a cloud around the mix position. It’s a very different testing method to using REW. When I use REW I carefully align the mic on a stand using a laser and don’t touch it during all the tests.
Very informative! Great job! But I still got a question. When you flush mounted the speakers, did you put more bass traps in the front wall around the speakers? If so, wouldn't it be absorbing all the reflected bass frequency that could've boosted the low end? Or you just simply didn't bass trap the front wall and filled it with normal broadband absorbers?
Very good question. Behind the front wall is basically an open cavity. In general, in acoustics, you usually damp any open cavity to tame up any resonance associated with it. We've use rock wool in the cavity. However, the bass performance is still increased as there is no rear radiation of bass. Because we've positioned the monitor baffle flush with the front wall, this extends the baffle. The speaker is now radiating into "half-space", which means any bass energy that would have been radiated rearward is now concentrated only in the half-space in front of the speaker baffle. This can theoretically produce a +6 dB LF effect. However in reality it's usually more like +3 dB. Similarly, placing a monitor in a corner forces it to radiate into quarter-space, which will have an even larger bass lift effect, although not usually desirable. Bass lift in a corner can be theoretically +9 dB, but practically more like +6 dB. Boundaries are not perfectly reflective, wall assemblies have low frequency resonances that act a resonant panel absorbers in the LF range, and there is some transmission loss.
As an acoustician, I'm going to offer some help to you and anyone else listening to this video!! (1) It's impossible to assume EQ is going to help with a 12dB swing on the low-end. At most 2dB is manageable and as you stated, moving the speakers forward can help. (2) Taking proper room dimensions out (for purposes of help) measuring room gain (knowing how is key) will tell you what treatments are needed and I can say no 4" panel is going to help with frequencies under 300Hz. I could go on however consulting is not free...
Thanks for joining in the discussion. Nice to have some professional acousticians in the mix. I agree completely that large peaks and nulls to the tune of 12 dB are not going to be magically corrected and sound flat with EQ. The approach of using EQ, as I said in the video, is an added final layer to be used, if necessary, after acoustic treatment, speaker optimization, etc. I also agree that no 4" panel is going to help with LF response and I don't think I made such a statement in the video. However, a floor to ceiling absorber that's 5.5" thick with Rockwool Comfortbatt R22, air gapped from the wall, will most certainly put a dent in the LF. Although that's only just a start, and in this room we went with much thicker treatment. The 5.5" thick absorbers are more for home project studios with DIY treatment, not professional rooms. Thanks for joining in the discussion. Cheers!
So, when we see a phase shift alongside the frequency response in programs like REW, does it mean that the exact drop in response is due to a reflection?
If your speaker has a relatively flat anechoic response, which any professional studio monitor should, then any peaks and nulls in your frequency response are due to room reflections (which by their nature are out of phase to varying degrees and are dependent on frequency). You don't need to see the phase in REW to analyze this, you'll see it in just the All SPL window. Phase analysis in REW is not as relevant to this because even the direct signal from the monitor will have phase shifts due to the fact that every frequency has a different period and simply time of flight will change the phase relationships of frequencies relative to each other.
REW's room sim is decent for sure, and certainly a valid approach. Although it's much more technical to use for beginners to acoustics and acoustic testing and it has some significant limitations in terms of data that can be input and calculated. It can never replace in-situ measurements. Room sims make certain assumptions, such as perfectly reflective boundaries, which are purely theoretical. They also exclude significant information, such as the resonant frequency and pressure-based absorption effects of drywall/gyprock wall assemblies. You can't get around room testing by plugging numbers into a calculator (at least not at this point in time). I use the room sim when I use REW, but only to see the theoretical / projected modal distribution over top of the actual in-situ measurements to compare them. The discrepancy between calculated modes and measured modes is known and discussed in a fair amount of detail in Rod Gervai's book Home Recording Studio.
Hey great video! Thank you for sharing the knowledge and sources. You mentioned back ports and the very low-end. However Im curious how does all that affect the midrange and high-end. I would bet there are some sacrifices (maybe not that excruciating in a standard box - however open baffles might be a different story). Perhaps is a bit more tricky to measure - but do you have any idea how does placing speakers near walls affect the perceived depth, width, and height of the soundstage? Open baffles that go down low with big drivers and some physical design optimization can achieve constant directivity throughout the whole spectrum because of the dipole configuration and CD horns. I’ve heard countless times that giving them lots of space is ideal. I wonder what is your take on this? Do same principles apply? What should one be wary about when playing with placement of such speakers? Great video once again! Keep’em coming:)
Glad you enjoyed it and thanks for the kind words and questions. I have no experience with open baffle speakers / dipoles. That's because my area of expertise is studio reference monitoring and I've never seen that design used. Have you? The closest I've seen is a hard-flush wall mounting (like I have) that approximates an infinite baffle, but still with front ports operational. Bass directivity is forwards only. Zero edge diffraction. As a mixer, I also work in a non-environment room, which is more or less a hemi-anechoic chamber with a hard floor. It's a very controlled environment with very short T30 times, and ruler flat low end response and full frequency decay times. Not everyone wants to work or listen in a room like that as there is a sense of "no room at all" in the perception of it, acoustically. You will always want to opt to keep the acoustic axis of your monitors at ear height for any control room / critical listening room. Keep them away from side walls, otherwise you'll get a change in the low end response (even bigger boost) and more SBIR to contend with. Even with fairly deeply treated side walls, you'd likely still get some boundary coupling effects. You cannot correct the destructive interference from SBIR with currently available room EQ systems, so don't create that problem in the first place IMO. In a setup like this, where you have much more of the direct sound and much less reflected energy, you will have vastly superior stereo field and imaging, as well as an accurate sense of depth and height. If you want an exaggerated sense of depth and height, then you'd likely want to aim for more of an audiophile listening room that is more "live" than a control room. To achieve that, you can add QRDs to the back wall, front wall, and even the ceiling, as well as porous absorption. This will give a larger sense of 3D envelopment, at the expense of longer decay times, more erratic frequency response, and masking of the direct sound within an interval that is unsuitable for audio engineering work, unless the room is huge. Read this article, it's very good: arqen.com/acoustics-101/room-setup-acoustic-treatment/
@@warpacademy thank you for the detailed answer and source. I will look into this further as I’m finishing a man-cave project. My listening room is still pretty bare but with high potential. However I have made some measurements and seen some simulations confirming my worries with placing full range dipole speakers too close to the wall. There are some abnormalities with the radiation pattern and directions. In the end what I’m aiming for is a slightly more reflective environment than it is acceptable for audio engineering work however I love to approach the problem from this very analytical perspective. Thank you once again and take care!
Thanks a LOT for those explanations! I will definitely give it a try because I have done DIY accoustic, I also have sonarworks and I feel my mixes so inaccurate within the low end. I was used to it my day to day routine but I figured out during my car test something is off in my studio under 100 hz, same feeling when I have gone to my best friend's home studio.
Best of luck! Keep in mind that most rooms also have a huge amount of LF ringing under 100 hz which masks things down there. That’s just due to those frequencies being harder to control. So that could be an issue too.
@@warpacademy Hi, I tested to place my monitors closer from the wall and I also moved my sweetspot a little bit by following the "bass hunter technique" from accoustic insider, the overall result before room eq correction seems a little bit more flat and low end is definitely more present. My room is fat from perfect but I can't do much with accoustic treatment as I am renting my flat. I will also test some other listening placement but I am already seeing some improvement. Thanks for your explanations :D
Stoked to hear that! Be very careful applying room EQ to a space that does not have adequate acoustic treatment. Room EQ will most likely be detrimental vs making an improvement if you do not have controlled acoustics. Only after a decent amount of wideband absorption has been applied should you consider using room EQ, otherwise you can do more harm than good as the room EQ cannot discriminate between a reflection causing interference, and an issue with your monitoring like a cabinet resonance.
Glad I could help. It's likely that your monitoring setup has a reflection that has an 8 foot longer time of flight, causing the cancellation. That could mean that you've gapped your monitors from the front wall by 4 feet, or that you have a floor or side wall or ceiling reflection causing the issue. Or a combination of them.
Hi thanks for the video! What if my front is window & I have -10db dip at 85 Hz, do I need to put acoustic panels behind my studio monitors or any solution?
Glad you enjoyed the video. Acoustics is a science. To set some good expectations, no acoustician could ever assess your room from a comment on social media. Room analysis requires dimensions, photos, information about monitor placement, monitor type and frequency response, and much more. It's involved. If you want a quality (and correct) answer, you should hire someone. I am available for consulting if you'd like to work with me in that capacity. You can inquire through the Warp Academy contact form on our site and I'd be happy to help. Regarding your question about placing acoustic panels behind your monitors, I have a deep dive video on porous absorption and its capabilities here: th-cam.com/video/SSn8HEsG8ro/w-d-xo.html. Watch this video and you'll quickly learn that acoustic panels cannot simply be placed behind monitors to solve a bass dip issue. It's more complex and will require a room analysis, including a modal analysis and room testing with REW. All the best!
I just repinned it. But here is the full comment. I wanted to pin a comment to the top here to address the most common question that's been coming up in the comments. That's, "You can't use EQ to correct the frequency response when you have SBIR because you'd just be boosting the drive signal and still causing destructive interference at the same frequency based on the 1/4 wavelength" or something to that effect. The answer is.....YES, you absolutely can. Using standard analog EQ, it's not possible. But when you employ digital adaptive EQ that uses a modelling delay, you can achieve a near perfect response correction. Here are the details on that, and my sources. I had thought the Sonarworks SoundID Reference platform was using this technology, but I was incorrect. I apologize. Sonarworks cannot correct phase distortion from boundaries, yet, although they tell me they're working on it and hope to deploy this in a future version. Take a look at the Journal of the Audio Engineering Society / JAES Volume 33 Issue 3 pp. 127-132; March 1985. Spectral, Phase, and Transient Equalization for Audio Systems, P. M. CLARKSON, J. MOURJOPOULOS, AND J. K. HAMMOND, Institute of Sound and Vibration Research, University of Southampton, Highfield, Southampton, UK. www.aes.org/e-lib/browse.cfm?elib=4461 If you want to see evidence of this technology that is currently available as a commercial product, look at the Trinnov system. In their technical literature, you'll find confirmation, complete with visual graphics, of how this technology works to correct phase distortion by using a modelling delay to time-align specific frequencies, thus correcting for phase shifts in a non-minimum phase or mixed phase system. www.trinnov.com/en/technologies/. Also the AES book Recording Studio Design by Philip Newell, page 332-334, which states "Near perfect response at optimum listening position after correction (with digital adaptive EQ)". This example was based on a test where a loudspeaker was positioned on a stand, part way through a room (exactly as I did in position 2 and 3), after which digital adaptive EQ was applied. This section of the book is also key to understanding why and how the correction can occur: "A non-minimum phase effect (such as when a loudspeaker is freestanding in a room, and not mounted at or in the wall) cannot be universally corrected by (analog) equalization (alone) of the loudspeaker drive signal. Even perfect amplitude correction could not restore the original phase response by any known practical analogue means. Only by digital signal processing could an almost perfect response be restored. Adaptive digital filtering can model, very accurately, the inverse phase responses necessary to correct either the minimum phase or non-minimum phase components of a transient or steady state response anomaly. By means of measuring microphones, modelling delays, and adaptive filtering processes, the digital systems can be made to 'learn' what a given room will do to a loudspeaker response, and apply acausal corrections to cancel the disturbances [acausal, in this sense, means that by means of a signal delay, the measured response error is fed into the drive-signal, in inverted form, to precondition the output to anticipate the error." -Pg 334, 11.5.2 Digital Correction Techniques, Recording Studio Design, Philip Newell, Audio Engineering Society. I followed up by speaking with Philip Newell directly on this matter. He personally watched this video, reviewed the comments, and confirmed this is a correct presentation of the information. Here are his comments on this specific topic: "I stand by what I said in the section that you quoted from Recording Studio Design, but it has to be taken in context. Acausal filtering and signal processing, WITH A MODELLING DELAY, can correct for the room boundary effects, but only for a very small region of a room -- say a very restricted 'sweet spot'. Elsewhere in the room, the response may then be much worse than before the correction. So, for example, if someone must work in an acoustically-poor room, this type of processing could greatly improve things at a restricted listening position, which may be beneficial to a single user, or it may 'somewhat' improve things over a slightly wider area, but it cannot make an overall improvement to the room acoustics. When one place is improved, another place will suffer." Furthermore, I spoke with the team at Sonarworks about their platform and this is what they've said on the topic: "We align the time between channels by phase measurements currently. The boundary nulling question: It will not be the same as it will still be there, but at a lower energy level. In short, phase-induced dips cannot be corrected by amplitude EQ only, only their impact lessened. For proper handling of phase effects, phase shaping is required." This parallels exactly what Philip Newell has stated. Sonarworks does not currently do phase shaping (I got this wrong, so sorry about that. I had thought their platform was doing this but I was incorrect.), but I hear they may be planning to roll this out in a future release. Also, to be clear, the audience for this is music studios, not audiophile hifi systems. In audio engineering control rooms, the flattest response possible is most certainly a primary goal. My comments to audiophiles are: Do what sounds good to you. You're the end user of your media and you should optimize it however you want to. Audio engineers and music producers, like myself, and many of the subscribers on my channel, have a different set of needs, which include accuracy and translation. Cheers!
I’m about to go move my speakers. Recently got stands and a smaller desk and doing testing last week had this exact dip. However the frequency of the biggest dip directly corresponds to one of my room modes so I thought that was the problem and I was about to build more bass traps. Maybe this is it.
Hey hey. Building more absorption into your room is probably a good idea regardless as most rooms are seriously under-treated. The more absorption you have, the more you're going to damp room modes and the more you will widen the Q of your modes - improving the overall sound of the room. Hard to know for sure without knowing about your room. If you want to share info on your room and get more detailed tips, join our free Discord Server ➤ discord.gg/ZwNgZteGKw and post in the acoustics-forum. You can test to see if it's likely a room mode by moving your listening (mic testing) position forwards and backwards on the room's line of symmetry. If it's a boundary reflection causing SBIR, the null will remain at the same frequency because it's related to the distance between the speaker and the wall which is fixed. If you move forwards and backwards on the line of symmetry in the room, and you notice that the dip is changing to flat and then a peak, then that indicates you're moving from a pressure node to an anti-node in a standing wave (room mode). But, this could also be back-wall reflection interaction causing destructive and constructive interference too. When your distance relative to the direct source and reflected source change, you will change the frequency of the interference. Are you using a room mode calculator to figure out your modal distribution?
@@warpacademy thanks for the detailed reply. Of course, I could probably use more trapping money is tight and it’s already pretty good, bass response improved greatly by moving my monitors to stands and off the top shelf of the desk, everything seems to be within 4db of flat. But in the problem area it’s -4 at 82Hz and +4 at 115Hz so I’m trying to flatten that out. Also, when I have the sub off it’s flat at 82Hz and a dip appears around 60Hz. As I’m typing this it makes it seem less like a mode bc if the mic was in a low pressure spot that probably wouldn’t change by adding/subtracting the sub- is that sound logic? I’m using REQ wizard for the testing and to calculate modes just did a quick and dirty 1130/room dimension but confirmed in the req room sim. Going to try moving the mic as you suggested before moving all my furniture. thanks again for the advice!
Hey hey. If you have tests in REW, please post them. I need to see your actual results. You can send the .mdat file, or you can screenshot showing the All SPL / Frequency Response full range (20 - 20,000 Hz), and then the waterfall showing 1,000 ms time range, and a 60 dB SPL range from 5 dB above peak level.
Glad you enjoyed it. The floor dip is addressed by placing the mixing desk in the reflection path and blocking / breaking it up. This is because it's not practical to have any acoustic treatment on the floor and carpet would do next to nothing for that frequency rance. My room will have a hardwood floor, partially for looks, and partially for function; the hardwood reflects some of the live scattering energy from the rear wall slats back at the mix pos. The ceiling is soft though, and will have extensive treatment plus a ceiling cloud. Thanks for watching and commenting!
It's unfortunately really common advice to move the monitors away from the wall and "solve" the rear radiated bass by putting absorption behind them. The problem with that is when you think it through, it doesn't work. 6" or 8" of rockwool won't take enough out of the bass energy to prevent the nulling that will still occur. It's best to leave the speaker right against the wall, putting the reflected energy as in-phase as possible, and to use your absorption around the speakers (above, below, side to side). A pseudo-soft-flush mount if you will.
@@teabreakbeats yes, this is very common advice. It is definitely a good idea to treat your corners of your room, as that's where modes terminate, but not only corners. For example the common advice to "bass trap your corners". In reality corner bass traps only cover perhaps 5-10% of your wall surface area. In order to get your decay times more even, without longer low frequency ringing which will mask the low end of your mix to the point where you can't effectively mix the low end, you need much more area of coverage. Which means floor to ceiling absorbers on your side walls, and a full-wall rear wall treatment system including both velocity and pressure-based absorbers.
Your assistant hand holds the testing mic in front of her body. Doesn't that invalidate the test? She's a barrier to sound propagation. I've always placed my measurement mic on a mic stand. Perhaps I'm wrong because you're performing a near field test? So, it doesn't matter?
Hey hey. You're totally correct that when you're testing a single mix position, with Room EQ Wizard, you would do that. However, Sonarworks operates differently and you test about 35 different positions by holding the mic by hand like this. When testing with REW I use a measurement delay and leave the room but Sonarworks has a different protocol.
Cheers! They're both great platforms. REW is more detailed, but also more difficult to use and more buggy as freeware. Sonarworks is more userfriendly and does the correction aspect for you.
I have those available actually. I shot the room at each position with both Sonarworks SoundID Reference and REW. Although I'll have to say that the waterfall does not provide any useful information about boundary nulling. The waterfall shows decay time, as I'm sure you're aware. Decay time has to do with modes, which will be active no matter where the sound source is located in the room. Boundary nulling is due to out of phase reflected energy superimposed on the direct signal; it has nothing to do with decay time. The waterfall in REW shows the exact same issue as Sonarworks and the REW frequency response all SPL and phase. Here is the REW frequency response for monitors 2' from the wall: www.dropbox.com/scl/fi/x3tjscjgfy6tzuks3lrez/2023-Control-Room-Speakers-2-Off-Wall-REW-All-SPL.png?rlkey=hz2ma7wyqzjlyo6kllvufhw1s&dl=0 Here's the waterfall for the same measurement, showing the exact same behaviour and no involvement of decay time: www.dropbox.com/scl/fi/41ak18hn553wxivqkcgn5/2023-Control-Room-Speakers-2-Off-Wall-REW-Waterfall.png?rlkey=vrl3y3cve3up3osplv2jxtcbf&dl=0
@@warpacademy Thank you 😊 Some people say, that you activate the modes more when you have your speakers close to a wall, and even more in a corner. I know the bass adds up 3dB for every plain you add, but some people say not only that, you would also activate more modes stronger
You're welcome. And you're 100% correct; you DO excite the modes more when the speaker is in the corner, When doing initial acoustic testing I intentionally put a speaker in the bottom tri corner of the room where the floor meets the front and side wall, to really set off the modes for analysis. And you're correct that *theoretically* you add 3 dB for each plane you add. Practically that's rarely the case, because boundaries are not perfectly reflective or rigid in reality. Theory breaks down in the real world. You should avoid ever putting a speaker in a corner when actually setting up the room for listening. But, you definitely should put the speaker close to the front wall like I said. Otherwise you will create the LF nulling issue and you will not be able to solve it. So your low end will be totally compromised. It's easy and straightforward to damp the modal activity with wide band porous absorption. So don't worry about exciting your room modes. Do your acoustic treatment correctly.
@@warpacademy Yes, regarding First-Reflection Issues, like discussed here for the front wall, it would be best to put the speakers like in the upper front corners directing down, then you only have the ground (or desk) reflection to deal with (which is ok cause I think human brains are used to it as we always have a ground reflection where ever we are normally). But then you have more Room-Modes ringing! Intresting you didn‘t have it at the front wall (with minor backwall bass-trapping)
@@warpacademy I saw you now have your Genelecs flush mounted with a hard surface… looks awesome 🤩 How did you plan the acoustics for that? I mean ‚try&error‘ is difficult for that 😄
GREAT video!! What happens to perceived soundstage depth with the monitors as close as possible to the front wall? Is there a degradation of the ability to perceive soundstage depth, and if so what can be done about it?
Glad you liked it. No there’s no degradation of sound stage. Stereo field is primarily perceived in the mid and high freqs. So as long as you’ve done a good job with left right room symmetry and side reflection treatment you should be doing quite well. Depth in a mix has more to do with psychoacoustics, digital reverbs or recorded room ambience in the recording. Mixing techniques like EQ, delays and such will create your depth staging. Reflections off other boundaries in the room will mess with you more in terms of that stuff as they can smear transients and detail in the mix if you haven’t attenuated and controlled the reflections.
I think there may be a primitive misconception about sound stage depth relating to the fore&aft position of speakers. Three things i think this could have stemmed from are: 1. Recognizing the undesirable coloration of too much low freq boost with speakers up against the wall, and SBIR issues with speakers not far enough away from the wall. Both of which are eliminated by pulling the speakers out a third into the room. It dosnt take a frequency sweep to tell something is desirable or undesirable, but current day easy access to microphones, analysis software, and EQ/DSP provide a detailed picture of what is going on and how to optimize. 2. The visual impact of a room's setup can alter ones perception of the sound. Simply seeing speakers up against a wall signals you brain to think there is no sound beyond it. This is not an issue if you have the speakers behind a transparent screen or simply listen with your eyes closed. As a former musician i am minimally effected by this because i naturally see&hear the music and not the speakers however i can understand how some people with differing backgrounds that are more fixated on the "gear" could be swayed by this effect. 3. The speaker & room's impact on reverb. Sure, desired reverb(and similar spatial effects) should be in the track whether recorded or added in the mix/master, but if your listening to a poor track lacking these elements then having a room that 'adds' reverb could lead you to believe that your setup is bringing the song to life when in fact the track was dead to begin with.
Hey Dom. Thanks for sharing your opinions. Thinking that you can "solve SBIR" by moving speakers into the room is incorrect and stems from old school, outdated HIFI recommendations. If you actually test a room as I did, you will clearly see that anytime you move speakers into the room on stands, you get substantial SBIR. The only way to address this is to move them so far into the room (IE 4 meters or more) that the cancellation dip moves below the audible threshold. BUT, who would do that? Who has 4 meters of room to work with? And where would that place you in terms of a listening position? The physics are basic here. As are the tests you need to run to demonstrate this effect. I'm willing to be proven wrong here though. If you have a room with speakers moved part way into it, shoot the room with REW and show me the results. If you get a flat frequency response (without use of EQ) I will eat my words. But I've tested this enough to know how it goes. In literally every acoustic test I've run in any room, this boundary nulling effect is present. You can't change the physics. Having a barrier behind the speaker will still reflect sound, no matter how far away it is, and having it just a few feet or meters away does very little to the strength of the reflection. The thinking that moving a speaker a short distance away from a wall (or a third into the room) somehow addresses the reflection is flawed logic and doesn't take into account how sound actually works.
I wanted to pin a comment to the top here to address the most common question that's been coming up in the comments. That's, "You can't use EQ to correct the frequency response when you have SBIR because you'd just be boosting the drive signal and still causing destructive interference at the same frequency based on the 1/4 wavelength" or something to that effect.
The answer is.....YES, you absolutely can. Using standard analog EQ, it's not possible. But when you employ digital adaptive EQ that uses a modelling delay, you can achieve a near perfect response correction. Here are the details on that, and my sources. I had thought the Sonarworks SoundID Reference platform was using this technology, but I was incorrect. I apologize. Sonarworks cannot correct phase distortion from boundaries, yet, although they tell me they're working on it and hope to deploy this in a future version.
Take a look at the Journal of the Audio Engineering Society / JAES Volume 33 Issue 3 pp. 127-132; March 1985. Spectral, Phase, and Transient Equalization for Audio Systems, P. M. CLARKSON, J. MOURJOPOULOS, AND J. K. HAMMOND, Institute of Sound and Vibration Research, University of Southampton, Highfield, Southampton, UK. www.aes.org/e-lib/browse.cfm?elib=4461
If you want to see evidence of this technology that is currently available as a commercial product, look at the Trinnov system. In their technical literature, you'll find confirmation, complete with visual graphics, of how this technology works to correct phase distortion by using a modelling delay to time-align specific frequencies, thus correcting for phase shifts in a non-minimum phase or mixed phase system. www.trinnov.com/en/technologies/.
Also the AES book Recording Studio Design by Philip Newell, page 332-334, which states "Near perfect response at optimum listening position after correction (with digital adaptive EQ)". This example was based on a test where a loudspeaker was positioned on a stand, part way through a room (exactly as I did in position 2 and 3), after which digital adaptive EQ was applied.
This section of the book is also key to understanding why and how the correction can occur:
"A non-minimum phase effect (such as when a loudspeaker is freestanding in a room, and not mounted at or in the wall) cannot be universally corrected by (analog) equalization (alone) of the loudspeaker drive signal. Even perfect amplitude correction could not restore the original phase response by any known practical analogue means. Only by digital signal processing could an almost perfect response be restored.
Adaptive digital filtering can model, very accurately, the inverse phase responses necessary to correct either the minimum phase or non-minimum phase components of a transient or steady state response anomaly. By means of measuring microphones, modelling delays, and adaptive filtering processes, the digital systems can be made to 'learn' what a given room will do to a loudspeaker response, and apply acausal corrections to cancel the disturbances [acausal, in this sense, means that by means of a signal delay, the measured response error is fed into the drive-signal, in inverted form, to precondition the output to anticipate the error."
-Pg 334, 11.5.2 Digital Correction Techniques, Recording Studio Design, Philip Newell, Audio Engineering Society.
I followed up by speaking with Philip Newell directly on this matter. He personally watched this video, reviewed the comments, and confirmed this is a correct presentation of the information. Here are his comments on this specific topic:
"I stand by what I said in the section that you quoted from Recording Studio Design, but it has to be taken in context. Acausal filtering and signal processing, WITH A MODELLING DELAY, can correct for the room boundary effects, but only for a very small region of a room -- say a very restricted 'sweet spot'. Elsewhere in the room, the response may then be much worse than before the correction. So, for example, if someone must work in an acoustically-poor room, this type of processing could greatly improve things at a restricted listening position, which may be beneficial to a single user, or it may 'somewhat' improve things over a slightly wider area, but it cannot make an overall improvement to the room acoustics. When one place is improved, another place will suffer."
Furthermore, I spoke with the team at Sonarworks about their platform and this is what they've said on the topic:
"We align the time between channels by phase measurements currently. The boundary nulling question: It will not be the same as it will still be there, but at a lower energy level. In short, phase-induced dips cannot be corrected by amplitude EQ only, only their impact lessened. For proper handling of phase effects, phase shaping is required."
This parallels exactly what Philip Newell has stated. Sonarworks does not currently do phase shaping (I got this wrong, so sorry about that. I had thought their platform was doing this but I was incorrect.), but I hear they may be planning to roll this out in a future release.
Also, to be clear, the audience for this is music studios, not audiophile hifi systems. In audio engineering control rooms, the flattest response possible is most certainly a primary goal.
My comments to audiophiles are: Do what sounds good to you. You're the end user of your media and you should optimize it however you want to.
Audio engineers and music producers, like myself, and many of the subscribers on my channel, have a different set of needs, which include accuracy and translation.
Cheers!
As a recovering audiophile and budding mix engineer, there are things both can learn from one another. Go to an audio show and listen to the top 3 rigs, you'll understand "audiophile sound character." In this priority: It's a recessed soundstage with 1) emphasis on depth and 2) solidity of apparent sounding body and 3) frequency accuracy. Engineers hear frequency accuracy 1st. Sounding body floats close to the listener, mostly L-R-C, intimately 2nd and 3rd low end punch is controlled, but not excessively tactile.
Overall, engineers prefer a 3D soundstage close with forward reverberation and audiophiles prefer a 3D soundstage gigantic and recessed back and away.
These are 2 unique sonic perspectives, achieved in slightly different ways and here's the crazy part: with the same source material. In other words, go ahead and mix for near-field, studio monitoring, but don't be surprised to hear this legitimately interpreted differently in a living-room with real-world room conditions. Peace.
Hey hey. Interesting insights. For engineers, accuracy and translation are paramount, not just frequency response. Accuracy is more all encompassing (such as accurate transient response). Translation, as I'm sure you're aware, is ensuring that your engineering decisions still sound correct when monitored in many other listening environments & playback systems. This has been a much studied subject in audio engineering journals and acoustics research.
Many (most?) audiophile listening environments do virtually nothing about 2 of the main obstacles to accuracy and hence the ability to translate:
1) Room reflections color the sound of the room - speaker system as much as (or sometimes more than) the speakers themselves. And because room reflections are unique only to that specific acoustic space, then any system that has much in the way of reflections has a pattern of constructive and destructive interference in the room and resonances that will dramatically skew the sound from anything resembling accurate. IE the music will sound wildly different from room to room. If there is some reflected energy in a control room, it will be scattered by the use of diffusors so the energy is dispersed in a hemi-disc (typically) rather than a hard, distinct, specular reflection.
2) Decay time. Most residential (and audiophile) listening environments have decays times (varying by frequency) of 500-1000 ms. This has a significant masking effect as the decaying frequencies obscure the perception of the next transient, or word in a vocal, or note in a melody. Long decays times are generally desirable in large chambers / theatres and halls of worship as they add a sense of lushness, but never for engineering purposes. Over the last 4 decades, the effect of decay times have been studied. Even decay times of 300-700 ms, which were common for old style control rooms, have now been proven inadequately long.
Audiophile spaces have a completely different set of priorities, which is why I carefully set the context of this video before discussing the acoustics. Control rooms are not designed for enjoyable listening; they're designed to turn out mixes and masters that translate well, and to do that as quickly and efficiently as possible.
You helped me solve 2 of my room’s biggest problems, 35hz and 80 hz. Thank you for this video 🙏🏼
Glad I could help!
Great info , Thanks .
I had a Yamaha NS10 on the left and a similar type speaker on the right but front ported .
I was shocked when I realized after comparing the two each by themselves , the front
port had 300 hz - ish info coming out of the port , most noticeable on vocals .
Something you don't notice when using the ported pair but having an NS10 to
compare , it was very interesting to discover .
Thanks for sharing. Glad you enjoyed the video. Cheers!
Very clear, love the drawings to illustrate the physics. I already knew about SBIR but it's always cool to see how other people explain it ! Such a nightmare to deal with !
The pos 1 is definitely the best if you don't mind the listening position. In my case i really wanted that optimised listening position and didn't want to have my monitors far away from me close to the walls. I heard subwoofer could help and it really did ! Because it lets you position your monitors freely while the low frequencies are carried by the subwoofer which i positionned right against the wall (since it's omnidirectional it doesn't even sound like the subwoofer is far) 👌
Nice one. Thanks for sharing a bit about your experience. Yes, that's certainly one of the advantages of using a sub. You can position it separately from your monitors. Although subs have several substantial drawbacks IMO: 1) Because they're positioned in the bottom front wall/floor seam or a lower tri-corner they excite room modes far more than speakers up on stands. That's an issue in most rooms that have inadequate LF acoustic treatment. 2) You have to carefully phase align your sub with your tops. Subs have very rough phase alignment options (usually 0, 180, or sometimes in 90 degree increments) so usually you need to manually play with the distances (string test) or use phase-alignment software / hardware like a Trinnov. 3) And this is the biggest one...if you don't have 2 subs configured in stereo, now your entire low end is in mono and you cannot effectively mix your low frequencies as an engineer. Any professional studio has full range stereo monitoring, not a single mono sub. Of course, you could always use headphones to check stereo sub information.
Those are just some issues I find. You may find them totally acceptable and that's all good. It's about getting good enough for you. Thanks for the comment!
Thanks so much for the in depth video and real world measurements of your tests. These kinds of tests, results and explanations are INVALUABLE and I’ve struggled to find many videos documenting these topics to this degree. Well done as always, cheers!
Thanks Gabriel. Glad you got a lot out of this one. Opinions are plentiful, but facts are rare. I like the data-driven approach with actual acoustic tests. Cheers and best of luck with your music!
You're absolutely right, flush mounted or inside of two feet.
It's worth noting doing so elevates the importance of rear wall treatment, ie., all modes additively sum up and at a pressure maximum.
Obviously you're on your of it with three feet of fluffy or whatever.
Hey mate. Thanks for the comment. Yes, that font back axial room mode is down around 35 hz with our room dims. So the depth of treatment we have is absolutely necessary to reign in that mode.
I’d love a video on the role of ceiling corner traps and what they typically address (assuming near-wall placement and basic acoustic treatment: vertical corner traps and first reflection on side and cloud).
Thanks for the suggestion. I'll keep that in mind. Cheers!
no its not just you saying this. it stands for example in the genelec manual for their SAM Monitors. I did it that way - pretty happy with the results.
Thanks for sharing that.
@@warpacademy thanks for the video. And I can see it's relevance, because when I first read it 2 years back, I was a bit confused, why they told me to place the monitors so close at my walls 😀
That's one of the rare examples I've heard of where a speaker manufacturer actually gets it right. Quite often they say to gap them from the walls, which makes no sense when you know the physics.
So much valuable information! Thanks a lot for sharing. I can confirm all your test as I have both a Treated Mixing Studio and a Home Studio both calibrated with Sonarworx and I did get these curves as I tested different speaker positions. You put years of test in such a clear and concise video. Wonderfull work.
Thanks for the comment. And I'm glad you've been able to optimize things in your studios using these principles. Cheers!
Good video and testing methods. 🙌 I also like this placement for the same reasons you mentioned. A few questions/comments: the 160hz dip you saw in the measurement you've attributed to floor bounce, but I didnt see it in the other measurements. I think that might be related to rear wall reflection. Another thing to note is that a dip in response caused by destructive cancellation cannot be fixed by boosting the signal since its also boosting the destructive reflected sound wave. Let me know if you disagree. Im open to being wrong.
Hey there. Thanks for watching and commenting. You've got a great channel as well!
Regarding floor dip, good eye. In the floor dip calculator the dip frequency works out to be 158 hz, which is bang on where we're seeing the dip. You're correct in that it doesn't show as much in the other measurements, but it can easily be getting blurred by constructive interference from the front wall SBIR when speakers are offset from the wall.
If the other boundaries, such as the back wall, were not heavily treated then it would be conceivable that it could be a reflection from one of them. In this case, however, we have 3.5 feet of fibreglass in the back wall, 2 feet in the ceiling and 1.5 feet in the side walls so far. The absorption coefficient for that depth of treatment is high enough (0.85 to 0.9 at 150 Hz) that it should eliminate those treated areas (which are floor to ceiling) as issues. But we'll only know for sure once we finish the room and get a desk in there. Very insightful observation on this one.
Regarding the EQing of the drive signal, this is what I'm basing my comments on... It's not information coming from me. I haven't tested this personally and I'm relying on research from the AES. Take a look at the Journal of the Audio Engineering Society, volume 33, page 127-132 (Clarkson, Mourjopoulos and Hammond). Also the AES book Recording Studio Design by Philip Newell, page 332-334, which states "Near perfect response at optimum listening position after correction (with digital adaptive EQ)". This example was based on a test where a loudspeaker was positioned on a stand, part way through a room (exactly as I did in position 2 and 3), after which digital adaptive EQ (like Sonarworks) was applied.
This section of the book is also key to understanding why and how the correction can occur,
"A non-minimum phase effect (such as when a loudspeaker is freestanding in a room, and not mounted at or in the wall) cannot be universally corrected by equalization (alone) of the loudspeaker drive signal. Even perfect amplitude correction could not restore the original phase response by any known practical analogue means. Only by digital signal processing could an almost perfect response be restored.
Adaptive digital filtering can model, very accurately, the inverse phase responses necessary to correct either the minimum phase or non-minimum phase components of a transient or steady state response anomaly. By means of measuring microphones, modelling delays, and adaptive filtering processes, the digital systems can be made to 'learn' what a given room will do to a loudspeaker response, and apply acausal corrections to cancel the disturbances [acausal, in this sense, means that by means of a signal delay, the measured response error is fed into the drive-signal, in inverted form, to precondition the output to anticipate the error."
-Pg 334, 11.5.2 Digital Correction Techniques, Recording Studio Design, Philip Newell, Audio Engineering Society.
This is exactly what Sonarworks is doing under the hood. Let me know it that makes sense. Cheers!
@@warpacademy The Quote from P.Newell doesn’t address the SBIR issue, it really doesn’t matter what leaves the loudspeaker at a given time, as the reflected wave will itself be changing with the generated wave, always in antiphase at that centre frequency that is a 1/4 wavelength, and frequencies either side in decreasing amounts.
Hey mate. The quote from Newell does indeed address this. He's speaking about the ability of non-minimum phase response modifications (speaker on a stand away from a boundary) to be corrected by digital adaptive EQ. Take another look over the quote, or read the book as it seems quite clear and discusses the mechanism of action (correction).
But seeing as it's not seeming clear to several people on this thread (and because I'm always up for understanding a topic more deeply, or open to being wrong), I followed up by speaking with Philip Newell directly on this matter, and here are his comments on this specific topic:
"I stand by what I said in the section that you quoted from Recording Studio Design, but it has to be taken in context. Acausal filtering and signal processing, WITH A MODELLING DELAY, can correct for the room boundary effects, but only for a very small region of a room -- say a very restricted 'sweet spot'. Elsewhere in the room, the response may then be much worse than before the correction. So, for example, if someone must work in an acoustically-poor room, this type of processing could greatly improve things at a restricted listening position, which may be beneficial to a single user, or it may 'somewhat' improve things over a slightly wider area, but it cannot make an overall improvement to the room acoustics. When one place is improved, another place will suffer."
The key thing to understand here is that this is not just amplitude correction from a standard EQ. A platform must have a modelling delay feature to compensate for boundary effects and phase. So it is indeed possible and Philip has specifically confirmed this.
Now, I also followed up with Sonarworks, who also confirmed this is possible with phase shaping. I was mistaken in that their current platform doesn't do phase shaping, so I stand corrected on that. They plan to roll this out in a future release.
Sir . Im without words..I have much to discuss even though Ive been in the audio business all of my life..time to hit the books..thanks..
Hey David. Yes, there are a lot of people giving speaker setup and room orientation advice who have never studied acoustics. Acoustics is a science and all of the books I've read agree and are consistent on this topic. But no book or theory can replace real world experience with acoustic testing. So that is what I did in this video. Real world situations, real testing, real data.
Super useful video thank you mate!
No problem 👍. Subscribe and stay in touch!
Hi this is a very informative video on speaker placement, thankyou for the detailed explanation, could you also give help for the addition of a studio sub, and where would this be placed? Is the best position parallel to front of monitor speakers but still very close to the wall if not against?
Hi there. Thanks for watching. With a sub, that complicates things for sure. Detailed acoustic testing and tuning is necessary. Some things to keep in mind. The sub is on the floor, so the distance from the listening position to you will be longer due to the downward angle to the sub. So typically you'd want the sub to be equidistant from your monitors, meaning the monitors would be in front of the sub. This keeps everything in phase.
But many subs have phase adjustment so you can adjust that as well. Typically you'd want the sub loading against the lower front corner of the room (where the front wall meets the floor). This will increase the amount of LF energy it's creating. From there, you can move the sub left or right to change its distance (and phase) or adjust a phase parameter on the sub.
The other consideration is if the sub is highpassing your monitors. Basically, you want to very carefully test the room and experiment while tuning the sub.
Great video, thank you for taking the time to explain all this. Love all your videos
My pleasure!
My experience is that you should not trust the graphs in Sonarworks. In my case the software was boosting 100 Hz by 7 dB and the correction graph showed a nearly flat line. I have double checked it with REW and the dip was still there, even a little worse than before. As far as I can understand this makes sense. If I raise the energy of a frequency coming out of the speaker, the energy of the reflected wave from the frontwall also has more energy, so as far as I can tell raising the level of a frequency to fix a destructive interference caused by SBIR doesn't work.
Hey mate. Thanks for joining the discussion. It's a smart idea to shoot the room with REW, if you know how to use it, to get more granular with your test results - if that's what you're looking for. I did the same thing. I shot the room with REW and Sonarworks SoundID Reference at each location to corroborate my results.
If Sonarworks is boosting 100 Hz by 7 dB you have a substantial acoustics issue in your room. Are your speakers gapped from the front wall? Sonarworks can't correct for a incorrect room setup. It also can't correct for a lack of appropriate acoustic treatment. As I said in the video, don't create time-domain acoustics problems like SBIR and then expect to get good results. Sonarworks should be applied after room setup is optimized, and you've acoustically treated the room.
It sounds like you have SBIR off your front wall, ceiling, side walls, or a combination thereof. Feel free to reply with information on your room and I'd be happy to advise on how to fix the 100 Hz dip.
The issue of boosting an SBIR null has come up a lot and I've posted my thoughts about this at length in the pinned comment up top. Please refer to that.
The Sonarworks platform can be a bit misunderstood if you don't know what's going on underneath the hood. It's not actually flattening the frequency response in the same way that REW would show it. It's taking into account the psychoacoustics of how we perceive a flat or balanced sound. You must account for the fact that the human ear is not linear and our ears or much less sensitive to certain frequencies. This is where equal loudness contours come in; they demonstrate that very well.
There has been much research into what curve produces the perception of a balanced response. Sonarworks has done their own research and the "flat" target represents what we perceive as a flat frequency response in a a typical room. Typically that means a slightly exaggerated low end and a slightly depressed high end. It obviously depends on how the room is treated and the directivity of the HF drivers.
Read more here: www.sonarworks.com/whitepaper
Thanks so much for this!
Glad it was helpful!
Are you de-essing your audio, or is it just you and the mic placement? Sounds so good. Thanks for this video!
Thanks. Which part are you talking about? In the beginning I was using a lav mic (I really don't like them, but use them when necessary) and further through I was using a boom arm mounted condenser. In both cases I'm using 2-3 de-essers throughout the post-processing chain. Glad it sounds good. Cheers!
@@warpacademypretty seamless settings!
@@emiliano9321 thanks!
this video was made for me :) Tranks
You're most welcome. Thanks for tuning in.
omg!!! What a great video !! Well done and thanks!
Our pleasure!
It's crazy, a bunch of my questions were covered! I just changed my old monitor speakers to a threeway Eve sc3070 monitor system and i wanted to make sure with measurements etc how to get the best sound efficiency in my room working with these monitors. Did a lot of research... and research created a lot of questions and then after two weeks this video pop up in my YT feed... and i did get my questions answered !!! Soooo big up to Warp !!! In a small room (treated thought) i got the best results when i placed my speakers as closes as it gets to the back wall. It's not as simple as it sounds but it's working and as soon as you understand what you are trying to achieve in a home studio situation it's a solid advice!🙂
Right on. Yeah, you definitely get the best results by doing the research, doing the testing, and following the physics of sound with all the best practices they dictate. Glad you're getting good results. Cheers!
Translating this to living room speaker placement, I guess the same placement advice applies to small bookshelf speakers ? ... but what about full range towers that reach to 25-30Hz ?
Hey hey. There's an important physics principle to get here. It's about the size of the driver relative to the size of the wavelength it's producing. If the driver is larger than the wavelength of the frequency, it will be more directional. For example, with these speakers th-cam.com/video/yzopMQBlTZY/w-d-xo.html. Their single driver is 5". A 5" wavelength corresponds to 2,720 Hz. At that frequency and above it, the speaker will be progressively more directional. Anything below that will be progressively less directional and eventually omnidirectional.
With smaller bookshelf speakers, assuming they're full range, the driver producing the LF will be very small relative to the wavelength of the frequencies they'll be producing. 60 Hz is about 19 feet long. So that means that not only the LF will be omnidirectional, but the low mids will also be radiating spherically. So with small satellite style speakers, it's much better to have them highpassed with a sub handling bass management.
Full range towers reaching down to 25-30 Hz will absolutely benefit from being right up against the wall. But remember, you will get front wall loading, which will broadly increase the LF and you will need to apply a low shelf EQ to attenuate that if you don't like the bass-lifted signal.
Some people may find the bass lift pleasing as our ears are much less sensitive to bass in general and we need more bass to sound balanced, psychoacoustically. This is why Equal Loudness Contours exist.
Hope that helps. Thanks for watching and commenting. Subscribe to the channel and stay in touch!
THX for this video. But one Question is still left. In my opiniion you can´t correct dips as long it is the result of a phase/comb filter effect. What I know is, if you push level up the frequency of the dip then you push the level of the signal that produces the dip, too. So the dip must be the same. Or is there something I don´t see?
This is a very good question. If you were to have the speakers off the wall, such that there was a phase difference, and only use normal EQ to adjust the drive signal, then you'd be totally correct. Normal EQ would just boost the offending frequency, and still cause the dip.
However, some forms of digital adaptive EQ is also taking phase into account. For evidence and corroboration, read Audio Engineering Society Recording Studio Design by Philip Newell, Page 333. He confirms this in the acoustic testing. I thought that the Sonarworks platform was doing this, but I was mistaken.
Also read the Journal of the Audio Engineering Society Volume that I reference in the video on screen. This is the paper to which Philip Newell refers in the book.
I've updated things with a pinned comment about this at the very top of the comments. Please refer to that. You definitely can correct the response and the comment explains how. Cheers!
@@warpacademy What EQ should you use instead of a "normal" EQ ?
Sonarworks is the EQ platform I'd recommend. Or perhaps a Trinnov if you have the budget.
Interesting. Funny though that these audio shows have listening rooms and it looks like most of them are pulled their speakers 2 to 4ft sometimes even more away from the wall. Even some of these million dollar audiophile systems do the same? Why? If it would sound better closer to the wall. Just curious what your input on that is.
Yeah systems like that have done so due to some misguided interpretations of the physics behind things. As well as placing less priority on low end response.
Mandy's got quite the bass traps
😉😉
Thanks ! Another super great video : clear, valuable & easy to understand. I feel less stupid today, thank you :-). I hope you will shoot another one about "where I should put this damned sub !!!".
You're most welcome. Thanks for taking the time to go through this deep dive and to invest in your knowledge. Learning this stuff is so key when you're a musician or engineer trying to get the best sound out of your equipment. Your question about the sub is a really good one. I definitely plan on shooting a video on that.
I have a test room where I'll be showing how to build the best DIY acoustic treatment, how to position it in the room, and how to optimize a monitoring setup using Tannoy Reveal 8s and a 10" Genelec 7060B sub. I'll show all the acoustic testing along the way.
@@warpacademy Yeah ! I can't wait. Thank again to take time for us. Now : let's move some stuff 🙂
You're most welcome. Happy music making!
Thank you so much for this video! The most turorials are made by absolute m0r0ns who tell you that rear ported speakers can not be placed close to wall compared to front ported speakers (like this "genius" here: th-cam.com/video/ZjmqOjle2T0/w-d-xo.html ) FOKK EM! HAIL YOU! I am currently upgrading my studio, and looking for the best speaker placement. Yesterday after your video i finally püut my Yamahas H 80 S back to the wall and i fanylly enjoy bass again!
You're very welcome. I also used to hear that advice about gapping rear-ported speakers from the wall, that is until I studied acoustics and started to understand the actual physics. The issue is that just because you're a music producer or audio engineer, that doesn't mean you know anything about acoustics. It's a niche field, very technical, and not intuitive in the slightest.
In order for a port to function, there must be a very small gap to the wall equivalent to at least the port size. So a 2" port will need a 2" gap, approximately. You just don't want to block the port or partially inactivate it.
The thing is that speakers are toed in by 30 degrees, and even that little toe in will often create enough of a gap that a rear port will function normally. Even if you stuff something in the port to intentionally block it, the speaker will still function; it'll just reduce the port resonance and the speaker will function more like a sealed cabinet (IE less steep rolloff, but a higher cuttoff frequency).
Glad you found the video useful. All the best.
Hi there Sir! This is great video that helped me a lot, but i have few questions that you kinda already answered but i need to confirm sry. :) So in a typical bedroom, 3,5meters l ,3.35meters w ,2.5meters h, i would place my speakers between 2 walls, and not between wall and window side, and speakers should be very close to the back for the best results right ? :) And i have a question how would objects in the room like bed, storage thing and couch affect the frequency response :/
Hey hey. Glad you liked the video. For more detailed questions about studio design and room setup join our free Discord server and post your question in the Acoustics forum: discord.gg/ZwNgZteGKw
Excellent, thank you sir!
Glad it was helpful!
Most Studios implement then directly into the walls
Exactly. Most professional studios that are built around acoustics best practices, that is. Which is exactly what I did in my room. Here's the finished room: th-cam.com/video/XfO_btDeXjc/w-d-xo.html
@@warpacademy Im right now also in the planning phase of my studio. Is it possible to get it touch with you per email, maybe you can give me some tips for my room and the dimensions. I'm planning to make drywall too and make the sound and dimensions as symmetric as possible. Until now I only finished the framework of one wall ( front wall) but it's still easy to change things. Can you help me to plan my room ? Im also willing to pay a tribute.
That would be nice! I'm also planning to make some videos and I would mention your videos and erpertice and help in them. That would be awesome.
You can get in touch with me for a consult via my sites: vespers.ca or warpacademy.com Use the contact forms. Cheers!
My issue is that im using a sub. I have to make sure the speakers and sub are aligned so that they're in phase. Is it better to put the sub close to the front wall as well?
Hey Jeff. Having a sub is actually advantageous. You can (and should ) put your sub directly against the boundary, yes. That'll ensure you have a close to in-phase wall reflection that will add LF amplitude to the sub and allow it to operate more efficiently. Given that a lot of subs cross over at 80-120 Hz, you may still get boundary nulling with your tops, so it's also good to place them against the wall.
Then for phase alignment, you have several tools at your disposal:
1) Manual phase alignment. You can use the tools on the back of the sub (phase switch) to roughly adjust phase. Then you can slide the sub to the left and right physically to fine tune phase.
2) You can use an automatic phase alignment system that's software or hardware based (Trinnov).
Either way, it's pretty straightforward to tune the system. Make sure you're running tests all along the way with Room EQ Wizard or another room testing app.
But here you are talking about right against the wall WITHOUT the front wall being acoustically treated. I have four 4" thick GIK panels plus two 16" corner bass traps on my front wall.
Yes, exactly. I wanted to isolate the effect of a hard boundary to clearly demonstrate the nulling that happens when you gap the speakers from the wall. Of course you'd be adding treatment to your corners and front wall.
It's good that you have thicker treatment in your front corners. I hope it's floor to ceiling. One of the most common acoustics mistakes is to use corner bass traps that leave the top and bottom tetrahedral corners untreated (IE they're only in the middle of the wall). Also, your 4" GIK panels are certainly helping, but they will be doing very little under 100 Hz. You'll most certainly still be getting a bass reflection off your front wall that will cause nulling. I expect you're seeing this in your acoustic tests.
If I was going to place treatment on the front wall, I would use at least 5.5" panels made of rockwool. Ideally 2 layers of 5.5" so you're looking at 11" of trapping. And you'd want to build the treatment around the speakers, not behind them, like the studio I featured in the video.
@@warpacademy Yep, my bass traps are floor-to-ceiling. They are the Impression Pro Series Corner Bass Traps (ECOSE rockwool). The 4" panels (also ECOSE rockwool) coefficients are 0.7 at 100 Hz, 0.3 at 80 Hz, and 0.1 at 63 Hz. Up to 11" thick of trapping on a front wall would be impractical for a bedroom studio at 11' x 13'. I also have the 4" panels on the side walls. In the future, I could move up to 6" panels on the front and back walls though.
Sounds like a pretty smart setup for a bedroom studio. In terms of the 11" of trapping on the front wall, consider that the front baffles of your monitor speakers are likely to be that distance from the wall, or thereabouts. If you build the trapping so that the backs of your monitors are against the wall and the trapping modules sit flush or close to flush with the baffles, then you don't really sacrifice any space in the room.
It's only when you try to put treatment behind the monitors that it becomes an issue because the monitor then needs to step forwards in the room. Perhaps you have yours mounted on a desk though, I'm not sure.
Due to the near square shape of your room, you likely have some issues with modal spacing and stacked tangential and oblique room modes causing significant peaks and nulls, so some additional front wall trapping would be helpful if you can.
Sonarworks is not painless and easy. It takes forever to finish all of the testing points!
"Forever". Not really. It takes about 10-15 minutes to complete the testing and you have to do it once. It's far more streamlined and simple than using something like Room EQ Wizard and trying to manually create a parametric EQ curve. I've done both. It does all the work for you beneath the hood. All you need to do is watch the on screen tutorial and correctly position the mic for the testing points.
Oh man .... Every audiophile on Earth just crossed his legs and looked away...
I'll take that as a compliment :P
No way - I threw up and passed out, but that could just be Thursday ;)
Huh? What a weird thing to say...
@@themattmcclellan
Audiophiles almost universally suggest the speakers should be well into the room and here we have the suggestion to put them right against the front wall.
That's because those types of audiophiles are often more obsessed with smaller details and aren't paying attention to comprehensive acoustic testing. They also may have different goals from myself (as a mix engineer) - which is accuracy for making critical engineering decisions.
The size of those 420's :))
They're beasts. Nearly 80 lbs. Deadlifting them is a workout.
@@warpacademy So cool. I want them for my 6x4 M2 room but I'm afraid I can not threat my room enough!
They definitely produce a lot of LF SPL, so you'd need very thick treatment. Yeah, with those room dims, you'd loose a lot of internal volume. Depends on your ceiling height too. If you're dealign with standard residential ceilings of 8' (
@@warpacademy Just like you, I've been rocking the DynAudio BM-12A MKII version for a bit more then a decade and a half. I still love them. Given the stated dimensions I'm considering the CORE 59. They do go low, but the LF roll off sets in quicker. I have to check the Freq roll off chart. Maybe those are more manageable given that they're DSP regulated and manual setting to further tweak them. I just want more low at speaking level.
Ah, nice one. Yeah I loved the Dynaudios while I had them. I did seriously consider the Core59s. Let me know what you thik of them if you end up putting them in your room. Cheers!
Please stop using stock footage. More Mandy though.
I'll pass that on to Mandy.
I've been sharing these videos with my clients, students and discord. Thanks for another fantastic video, you really go above and beyond on everything you do.
Thanks for sharing these around. Much appreciated! All the best.
Don't position your monitors until you've identified the optimum listening position. Then position your monitors using the phantom centre method. Then using REW treat your room acoustics the very best you can. Finally use ARC 4 to tweak to taste. Bingo, did this for new mixing room and it sounds incredible. All details at Acoustic Insiders on TH-cam. 😎
Hey mate. I've watched those videos so I'm aware of what they advice; I just happen to disagree. Jesco talks about the importance of setting your priorities when designing a studio. Well, I place a very high priority on bass response / accuracy and you will never get a good bass response by placing your monitors out into the room. That's physics 101. SBIR / boundary nulling will shoot you in the foot.
Do I mean that I don't care about phantom center, or a really solid sense of stereo field? Not at all. It's just not that hard to do. When you have enough side wall treatment to attenuate or absorb lateral reflections, the phantom center is amazing. It's reflected energy and diffraction that blurs your sense of phantom center, so you deal with that with absorption. Easily done.
What is not easy is trying to "fix" your bass response after setting up your room in a way that ignores SBIR in the low end. You cannot simply place acoustic panels behind your monitors. That'll only partially reign the problem in, but not address it fully as the depth of treatment necessary to kill the reflection is enormous.
And if you think I'm the only one who is saying this, I'm not. What I recommend is consistent with what Philip Newell and Wes Lachot say. They are professionals who've built some of the most important and accurate rooms in the world.
As you may know, Wes Lachot is the creator of the "38% Rule". Here's what he has to say about the priority that bass response should take, and placing the speakers on the wall or out into the room.
"All is won or lost in the bass."
"You can buy your Focals or whatever, but you’ll just be guessing with your bass!"
"You can move them around for years, but you’ll just create new problems."
"You should always put speakers against the wall unless you have a very deep room! 9-10 feet from the wall in a 30 ft deep room, otherwise no."
What about the 38% rule? is placing the speakers right against the wall a better solution? (for a small room)
Hey hey. Very good question. The infamous 38% rule was created by master acoustician and studio designer, Wes Lachot. Here is the post Wes made about it: repforums.prosoundweb.com/index.php?topic=21551.0
The vast majority of people misinterpret this "rule" that is actually just a guideline. That guideline was intended to locate the mix position away from the center of the room and to avoid pressure nodes and anti-nodes, especially overlapping ones, due to boundary spacing.
It's important to note that Wes stated 38% for control rooms that are 20' deep at least and where the speakers have been flush mounted into the wall, like my control room, and not for speakers on stands: th-cam.com/video/XfO_btDeXjc/w-d-xo.html&pp=ygUad2FycCBhY2FkZW15IHN0dWRpbyByZXZlYWw%3D
I've spoken to Wes directly about this. Studio design and room setup is all about priorities. Wes states, "It's all won or lost in the bass." If you gap your speakers from the wall like shown in this video, you get a huge hole in the bass, as I've demonstrated in this video. "You can buy your Focals or whatever, but you’ll just be guessing with your bass! You can move them around for years, but you’ll just create a new problem"
So he is in full agreement with what I've presented in this video. Bass frequency response is one of the highest priorities in room setup. And this final quote from Wes sums it up and puts this issue to bed, "You would always put speakers against wall unless you have a very deep room! 9-10 feet from the wall in a 30 ft deep room, otherwise no."
I covered all surfaces of my listening room with 3.5 feet of bass trap. My room is now 18 inches tall. 11 feet long. 5 feet wide. But, it sounds phenomenal. Least ways I assume it does. I can’t get my 53 inch tall speakers in the room. At least my chihuahua has found a safe place, where my English Mastiff can’t terrorise him.
You point out a very common and definitely laughable aspect of incorrect room treatment. If you don't have much space to work with, then it makes little sense to use the type of treatment that would be ideal for a ground-up professional room. If you're working in a bedroom, it should be common sense that you'll be working with thinner treatment to preserve the working space.
I watch every video from warp academy even tho I have an untreated room with the mackie CR4 :p waiting for your studio build update as well!
Glad to have you on the channel! Happy music making.
Your data and testing is also highly dependent on your monitors, your desk, your room etc. It's going to be different for everyone.
Hey hey. Thanks for the comment. I would agree that your monitors, your desk, and you room all influence the final sound you hear in your studio without a doubt. And, at the same time, this is basic physics that will apply regardless of these things, universally. The only exception are cardioid speakers, and rooms that are oddly shaped to create an RFZ.
All standard speakers will have rearward radiation of bass, all rectangular or square rooms will bounce back that reflection, and it will rejoin the direct signal causing boundary nulling. The desk is inconsequential as it does not impede the offending reflection from the wall behind the speakers; it can only impede floor bounce. The only variables are what frequency will be nulled (relating to the distance from the boundary) and how much it will be nulled (relating to the acoustic properties of the materials the boundary is made of).
I myself have confirmed this phenomenon in every single studio I've tested (including many studios of other producers). This principle is discussed at length in acoustics books, such as Recording Studio Design by Philip Newell. This is established science. Once you grasp the physics involved, you can understand why this is universal.
@warp….. Adam’s or Neumann speaks? Looking kh80 vs the av4 ?
Neumann all the way. The Adam XART accelerated ribbon tweeter design is so hyped and unnatural sounding. No one is going to be listening to the end result on that tech so don’t mix on it.
This video was PERFECT thank you
Thanks!
Great video. How does it work when a Sub is placed in the room? Also flush it up against the wall and acoustically treat around it?
Thanks. Good question. A sub will behave exactly the same way. If you put it out into the room then you’ll get massive cancellation and boosts. The best place for a sub is built into the wall flush (and then you phase align it using digital time alignment software) or placed against the wall.
@@warpacademy Thanks a lot for your response and your videos. It really helps! :)
My pleasure.
Excellent video. I think this will help viewers understand what happens when you have speakers close to or pulled away from walls.
Cheers! Nice channel BTW. Your videos look excellent.
Thanks so much for this video. It's extremely informative and corroborates some other guidance I've seen on the subject of speaker placement. I'm looking forward to trying this method with my own speakers, in combination with Dirac Live room correction. Hopefully this technique works as well for stereo listening rooms as it does for studios.
Hey mate. Glad this was helpful for you. This is unpopular advice with a lot of living room / audiophile folks. There's a lot of outdated or just plain bad setup tips out there that aren't grounded in any kind of actual testing or physics. Gapping your speakers out from a wall, 1/3 of the way into the room to "solve" SBIR is the biggest one of them all and that's why I focused on it in this video.
Good low end response accounts for 30% of the listening enjoyment experience, even if you're not a professional engineer (per Floyd Toole). And there's only one way to get a relatively flat low end - it's to put the rearward radiated bass energy from the speaker as in-phase as possible with the direct sound. Gapping a speaker from a wall, or putting 6" of porous absorption behind it does not stop the out-of-phase reflected bass energy from causing low end interference - the distance and the absorption characteristics of porous absorption are not capable of that - at least not with that amount of time of flight and that thickness of absorption.
So really, you must put your speakers against the wall if you care about low end response.
Wow amazed by the quality of your videos. You're like the Dr. Berg of audio engineering.
Thanks very much. Glad you've been enjoying them.
Thanks for a great video! Gonna try the monitors up against the wall again, have experimented with different positions out from the wall, and having the exact problems you describe..
You're most welcome. Glad it was helpful. Yeah, this one is a very common issue in music studios, but easily resolved. I encourage you to run before and after acoustic tests in Sonarworks, if you have it, or Room EQ Wizard. Use the data to confirm the difference in room response. Feel free to post a comment or question if you need any advice going forwards. Cheers!
Hello,
Do you want a flat target curve or a tilted or Floyd toole target curve?
Or do people in studio always want a flat curve at listening position? When you experience some listening fatigue do you change the tuning?
Regards Greg
Hey Greg. For me, I prefer the Bob Katz approach with a little low end lift. It can be interesting to experiment with a gradual, progressive HF rolloff as well (per Katz) but I don't find I'm a fan of that in this room and on these monitors.
I don't make adjustments for listening fatigue. I take breaks. What I do make adjustments for is monitoring sound pressure level, because...Equal Loudness Contours. The ear is much less sensitive to extreme LF and the psychoacoustic perception of LF is very compressed as you increase SPL.
Thanks for your reply, interesting.
I am just an audiophile, I design my own active speakers and I have a room which is half treated, some csd side walls and plenum plus corner traps behind the speakers. It helped a lot to measure the room plus speakers and for bass and clarity. Of course it is not perfect.
I also have directive speakers with mid horn, hf horn, and ripole subwoofer at 80hz, this helps not sending energy to the room.
I read about Bob Katz and I think this is a little bit what I do modestly. I also compare with my neumann ndh30 eqed with oratory1990 profile. Then it sounds pretty similar. If I want to retrieve the same amount of details with the speakers I have a more fatiguing experience. I don't know why. But we are talking of tiny differences.
Regards Greg
Sounds like you have a strong interest in audio and speaker design. That's a neat passion. I know very little about designing speakers, but I do enjoy listening to good ones in a good room! All the best Greg.
Another heavy hitter of a video! One thing to consider with the boosted lows in a situation with monitors flush against a wall is that it's actually preferable to flat when mixing low, and is just more fun in general to mix that way as opposed to flat.
Thanks for the comment. You’re totally right. When mixing at low volumes the bass will be perceived as quieter due to equal loudness contours. So that would help avoid over mixing the bass. Good point!
Yeah, but up against the wall sounds gross everywhere else. You need bass, get some subs. I'm always near field or way out from the wall, via experience. Even my subs are away from the walls and lifted 3 feet off the ground.
To be clear, we're talking about music studio control rooms here. Not audiophile hifi systems. It's an industry-wide best practice to flush mount monitors in control rooms for a multitude of reasons, and barring that, to put the monitors right up against the wall. If you're optimizing in an environment that's not a studio control room for engineering, then just do what sounds best to you. Cheers!
@@Audfilethere is only a single position that matters in a studio and that is your listing position. If it sounds good in the listening position and terrible everywhere else you are winning 🎉
Well said. We generally optimize for the listening position. In my room - and many professional rooms - we needed a wider view though. You need to account for the fact that you'll also have clients and students in the room alongside you; not just a single engineer. In my room (it's a teaching studio where I sometimes have up to 8 additional people) it's a goal to have the frequency response as good as possible throughout a larger area.
Of course it can't sound good everywhere, but by eventually flush mounting the monitors into the front wall (we're doing that right now) you can achieve that because it's minimum phase and no digital adaptive EQ correction is required. We can account for the LF lift (baffle step) by shelving down the monitor drive signal using analog EQ on the back of the KH420s and this will improve the response in all positions of the room.
In most home and project studios though, you only need to have 1 area sound good. The mix position like you said.
Thanks for watching and commenting!
Nice video. I might be on the wrong channel for this question, but is this video only applicable to recording rooms or should a regular listening music/theater room also have the speakers as close as possible to the front wall?
Oh that’s all good to ask here. If you have a bass managed system with a sub then this likely won’t be an issue for you as you can position the sub against the wall and the tops more freely. If the low end is also coming from the stereo mains and not a sub then you’ll benefit from putting them against the wall. But then you’ll want to shelve down the low end a bit to compensate for the front wall loading.
@@warpacademy Cool! This is one of the few channels that suggests putting speakers close to the wall. Thanks for taking the time to reply.
I think that’s because most people don’t take the time to understand the basic physics at work here. It’s not rocket science yet so many people get this wrong based on years of bad advice propagated on the internet that was never grounded in science or research.
Many good informations but still not the right conclusion.
There are so many more different parameters who have an influence on the sound.
If you talk about a true studio environment you may be right.
But look around in all the studios worldwide - the near field speakers are mostly placed directly on the mixer in front of the engineer.
So - are they in the wrong place?
Of course not.
To make a long story short - all the time we place a speaker in room we have to start right at the beginning to find the perfect place for them in the environment where they have to work and for what kind of task they are used.
To tell all of us that there is only one perfect answer for all of the questions is stupid.
And if you will show thousands of books that all tell the same it doesn't make it right.
I have more than 40 years of experience with audio and studio equipment for hundred of thousands of cash for one setup and there was never one solution that fits them all.
Hey mate. Thanks for joining in the conversation. To clarify, my audience here is music producers with music studios, so I'm most certainly talking about a real studio environment where there is acoustic treatment. I'm not talking to audiophiles with living room sound systems, although many audiophile setups could benefit from taking these principles into account.
You bring up a very important point about many professional studios and how they place nearfield monitors on the mixing console meter bridge. Despite the fact that you see this a lot, it leads to severely compromised acoustics. Just because lots of people do something doesn't make it a good idea. Any acoustician will tell you the same. It's obvious that this is not a good setup when you understand acoustics 101 and the rear radiation of bass. Also when you understand concepts like diffraction of audio and the consequent smearing of stereo image due to reflections off the mixing console, you'd never place monitors where their propagation of sound would contact a mixing console in such a close and direct way.
So why do you see professional studios and engineers doing this, despite the fact that it violates many acoustics best practices? Because real education about acoustics is sadly quite rare and many engineers don't actually acoustically test their rooms, or they use room correction to attempt to solve problems that didn't need creating in the first place. Nearly all of these rooms have mains that are built into a flush mount setup by the original studio designer (who does understand acoustics). But in practice, so many engineers like to throw multiple sets of monitors into the room on stands or on the meter bridge, usually without carrying out testing or considering the full impacts.
Read Philip Newell's book that I reference in the video. He weaves in countless stories of studio owners and engineers who don't properly understand acoustics and he's called in to fix avoidable problems. He talks a lot about mixing desks in general, and how big consoles (and other equipment) in the room actually negatively affect acoustics ("BDE" big desk energy) through diffraction.
Look at really well designed studios, made by people who do high level acoustic testing, and you'll never see these mistakes. Look up any room made by Northward Acoustics. No nearfields. No mixing consoles. Just a very small vented desk with no reflections of diffraction sources.
But outside of the studio control room, I'd agree with you. Find a good position and the room acoustics are less controlled so you have more experimentation to do. There isn't a one-size fits all approach for many (any) things in acoustics. In this video I wanted to demonstrate one simple thing; the effect of the distance from the front wall boundary on non-cardioid speakers. And in this case, there is only one position that works as I have demonstrated with testing. Right up against the wall.
Thanks for this video. I always used to have my speakers up against the wall, but for some reason, on Adam's instruction manual they say to put the speakers 30cm away from the wall ... So I did with my new Adams, and I get the exact dips and boosts you show in the vid! I'm so glad I can reclaim that 30cm of space I lost by moving my entire desk back off the wall haha. Only issue is I need to find my measurement mic so I can recalibrate my system!
Glad it helped! Yeah, much of the advice I've seen in speaker manuals is just dead wrong. Speaker designers care about frequency response in an anechoic chamber, and don't seem to give much useful advice for real-world situations that include putting them in a room. It's a fairly deep understanding of acoustics that's needed, but you'd think they'd be aware of this issue and the fact that once you create it, you cannot solve it using room EQ or realistic acoustic treatment. It's a damn shame so many people compromise their low end frequency response by following that wall gapping / speaker positioning advice without ever testing it themselves.
Many years ago I read a lot of literature on room acoustics and if I remember correctly the timing of the reflection from the original signal has an important effect on our ear's perception. I think it was something like a reflection 5-10 milliseconds delay after the original wave is perceived as a separate sound by our brains. Less and it's perceived as the same signal. It would be interesting to know the calculated delay for the three positions and room modes from other wall reflections. We can not localize bass frequencies like mid and high ones. Is optimizing the bass hurting the other frequencies? Is the effect of room modes caused by reflection from back and side walls more important? If my memory serves me correctly you should be able to calculate the best speaker and listening position which will be different depending on room dimensions to minimize room mode effects. Was the microphone positioned in a room mode?
Hey Dwayne. Thanks for the comment. I think what you're referring to is the human echo threshold. Although it can depend on the type of sound (transient vs steady state) and frequency, it's typically around 30 ms. But in some circumstances it could be shorter. You can test it yourself with a delay effect to see where you can start to discern a the delayed echo. I usually start to hear a distinct delay at 30 ms or so.
In a music studio control room, you want to optimize all frequencies of course, and many other things, such as the decay rate of each frequency (addressing modal ringing), the stereo field and phantom center, the impulse response / ETC of the room (the liveness of the room as well as especially loud early reflections, desk reflections) etc. The low end just happens to be one of the most challenging aspects to optimize, which is why it's important not to create unnecessary problems through bad room setup and speaker placement.
When your goal is a very controlled and accurate listening environment, you'll be using extensive acoustic treatment. Because of the frequency-dependent way that porous absorption works, mid and high frequencies are easier to control - hence the focus on getting the low end right from the start.
All room modes are important, as is the spacing of room modes. The room modes are usually the most isolated and spaced out (IE uneven and lumpy looking in REW) in the low end. As you go higher and higher up, the room modes start to create a more even effect (IE the Bonello criterion). You typically would be designing a room based on one of the golden ratios using a room mode calculator to avoid stacking room modes on the same frequencies and ensuring the room meets Bonello. Although as you add more and more acoustic treatment those golden ratios get less and less important because you're damping them so heavily. In my room, we chose to maximize room volume rather than target a golden ratio.
When you say, "was the microphone positioned in a room mode", do you mean in a node or anti-node? Cheers!
Outstanding! Thank you for switching the light on for me on this topic 🙏
Glad it was helpful!
Wow. Great video. I learnt so much! Does your advice at the end apply to hifi systems for music listening, not mixing?
Glad you got a lot out of this. The general thinking in hi-fi circles is to move the speakers away from the boundaries in order to "reduce the effect of SBIR". My big question to that school of thought is, exactly HOW do you think that is going to reduce SBIR significantly enough to flatten out your bass response and avoid a large null. People might answer that when sound travels a further distance, it loses power. While that is correct, the losses you get due to increased time of flight are totally insufficient and you will still get a dramatic null from the out of phase reflection. And now the out of phase reflection (which always causes a null at the 1/4 wavelength frequency) is simply lower and even more problematic to address with any kind of acoustic treatment. It's a losing game that doesn't make sense when you fully grasp the physics involved.
Also, I think that the people who follow this advice and gap their speakers either A) Aren't prioritizing an even low frequency response in their system, or B) Aren't actually testing their room with room testing software, or both.
When you test your room, you will see the effects in the frequency response clear as day. It happens in every room that's a rectangle or square. And you'll hear the destructive effects of the null (and constructive effects of the peak) if you listen and care about the low end. 30% of the perception of quality in music is the low end, so you must get that right if you want an accurate high-quality (IE high-fidelity) experience. Reference on that is via Floyde Toole author of Sound Reproduction and former consultant to Harman.
Really great accurate information! Sonarworks is an excellent tool for finding problem areas in your room. I don't use Sonarworks to correct any problem areas, I think it's better to treat the room . Thank you for the info!
Definitely agree with you that your focus needs to be on treating the room and getting as optimal acoustics as you can. Then, you can use room EQ to address remaining issues.
Good video, thanks for making this. I do have a question though. What are your thoughts on the rule of thirds? How does that compare to the other approach you're suggesting which is to place the speakers right next to the wall?
Thanks!
Hi Chris. Thanks for the comment. I think you're talking about the 38% rule yeah? Arqen answers this better than I could. Please see this: arqen.com/acoustics-101/room-setup-speaker-placement/
And this: arqen.com/acoustics-101/speaker-placement-boundary-interference/
Cheers!
Why the hell, math is not working at position no. 2.
Wavelength/4 = 2"(0.61m)
Wavelength= 8" (2.44m)
Freq= 343/2.44 = 140~ HZ
Sonarworks shows 50hz pulled back 9db instead of 140hz.
Same is the case with position no.3
Theory and reality don't always mesh. I was quoting some of the information in the same way a theoretical acoustician would, but certain assumptions are made, such as infinitely rigid and massive boundaries that are perfectly reflective. In situ, this is not the case.
Theory predicts there will be constructive and destructive interference in the bass range, resulting in a large cancellation notch and a corresponding peak. That is clearly the case in pos 1 and 2. Theory also predicts that the frequencies of the interference will shift, and again, clearly that's demonstrated.
Don't get caught up on the math. It breaks down because the acoustics formulas involved are not sophisticated enough to take into account all the variables, and also input data such as transmission, TL, absorption and reflection coefficients are never available for all your materials.
Great video! Lot's of useful information for optimizing results :)
Glad it was helpful!
Hi. I am a little confused about your math. You said that the 55 Hz dip is caused because the speaker is 2 ft off the wall. So if the wavelength of 55Hz is approximately 20ft, then 1/4 of that is 5 ft. The speaker in the video is 2 ft away, meaning 2ft is a quarter wave of 8ft, not 20 ft. Have I got something wrong.
That's theoretical based on a calculation. The calculation assumes perfectly rigid and perfectly reflective boundaries, which is never the case in actuality. What we're actually seeing is the front wall reflection, combine with the superimposition of side wall, ceiling, and floor reflections. The acoustic treatment I have on those boundaries was not very thick yet during the tests, so it was not sufficient to dramatically attenuate LF reflections. It effectively attenuated mid and high reflections, but the LF was still bouncing back.
What can be easily seen though is the effect of moving the speaker in the room.
0:29 “…the data never lies.” If only it was that simple. 😊 My old Stats 101 course, among many other things, taught me about such hubristic nonsense. Good video, though. I will try it.
Well. This isn’t statistics where data can be cherry picked or manipulated by people is it? It’s a single measurement directly from the software. A data point. And it’s repeatable. And it’s been repeated in multiple rooms with different speakers. So in my books it’s reliable.
One question I have with that against the wall placement. With the Speaker responding the way they do against the wall, if you were to move the speaker a very small amount, would help to correct that rise on the low end? I'm thinking that it may do just that BUT, a small movement off the wall may create a problem in a different frequency area.
Your thoughts?
Hey hey. I like where your head's at with this. Good to start to experiment and think things through. The concept to really get here is that of minimum phase. IE, the frequencies that are in phase completely or nearly completely with the direct sound.
Very close to the wall will create a more well-behaved low shelf like boost. This can easily be corrected. However, even a relatively small distance from the wall will delay the bounce far enough that you get substantial nulling. Test the setup yourself with REW to see how far you can move it.
But the key thing to realize is the bass boost is not a problem. The easy solution is simply to EQ with a subtractive low shelf. Or, you may actually like the lifted bass response. Many people these days seem to prefer a bass lift in the bottom end. I do (not a big boost, but 1-3 dB can sound great, given that our ears are less sensitive to SPL in that region).
Great explanation
Glad it was helpful!
hands down the best video on this subject i've seen so far.
Glad it was helpful!
Do you have any tips for doing that alongside a sub as well? Would you test them separately?
I am planning on using Neumann’s MA1 for the eq, hopefully they can correct what you said and are linear phase
Hey hey. Adding a sub to the system will make tuning much more complex if it's by ear, but the MA1 system will make phase alignment very straightforward so that's good. Use of linear phase EQ in room EQ is of questionable benefit IMO. Mixed phase may give better results.
IMO if you're doing anything other than bell attenuation in the LF, or broad spectral tilts, then you're potentially causing more harm than good. The acoustics of the room determine how helpful EQ will be.
In the MA1 here are my tips: Do not allow it to "correct" any nulls by adding boosts. Like the Sonarworks system, it's not capable of addressing boundary nulling from SBIR. Restrict its correction so that it's only permitted to do subtractive bell EQ below 300-500 Hz to reduce the effect of constructive interference from boundary reflections and modal resonances.
Then add any broad spectral tilts you desire using the house / custom curve. Bob Katz's personal preference for a "flat target" for a control room is flat from 20 Hz to 1 kHz and then a linear slope from 1 kHz to 20 kHz with 20 kHz being 7 dB down from 1 kHz.
Then in many studios, engineers prefer a low shelf boost of 2 or 3 dB below 100 Hz, but otherwise Bob Katz pretty much got it right. It obviously depends on how the room is treated and the directivity of the HF drivers. If the room is less extensively treated and the HF drivers throw lots of energy out wide, then there will be more HF room gain.
I'm curious why some HIFI speaker manufacturer regularly recommend a minimum distance from the wall? they are not studio monitor, but still.. they should aim for a flat frequency response.
Simple. Because they aren't actually running room acoustic tests to verify that they're giving the right advice, or they have a bass-managed system and the sub is highpassing the tops so they're not radiating the frequencies responsible for SBIR. You cannot change physics.
Anytime you have a full range speaker that's not a cardiod design, and you gap it from a reflective boundary, you will get SBIR that will null in your low end. Anyone who recommends this doesn't understand acoustics, or doesn't care about their bass / LF frequency response, or both. All you have to do is acoustically test the room and the data will not lie. You'll see the null clear as day.
Thank you, sir! So it's better to keep speakers closer to a wall with absorbers around than to have thick absorbers behind speakers and further from a wall?
Yes, correct. Absorbers behind the speakers will not do much for the sub range under 100 hz. Porous absorption performs very poorly there and it also performs more poorly when particle velocity is zero which is exactly what it’s at when you reach a boundary (wall).
Please elaborate on Floor Bounce. I'm attempting to optimize a room with couple significant dips (155 & 190hz) that does not have a "console" to block the destructive waves. I assume many people will find their way to this video looking to improve their Listening rooms that also do not have a console that is typical in a studio.
Great vid BTW! Confirms my recent studies on the matter.
Hey Dom. Thanks for watching and commenting. Glad you found the video useful. While this video is primarily for music studios for critical listening and engineering, you could also block the floor bounce in a general listening room (like a hi-fi living room setup) by positioning a thick rug and coffee table in the reflection path. Although, for general listening, and not critical engineering of music, some floor bounce is generally acceptable to listeners. We're quite well adapted to hearing floor bounce as human beings. Cheers!
With all this focus on the non-directional freq interaction with the Front Wall could you please comment on the other room surfaces like Side Wall, Ceiling, & Floor. Would running the same experiment with a monitor up against the sidewall then pulled 2 & 4ft towards the center of room have similar results? I assume not as you would have just said stick em in the corners and be done with it...
Hey Dom. For sure. For that, please watch the detailed design video on my engineering control room: th-cam.com/video/5VrG2K_E7qI/w-d-xo.html The side walls, ceiling, and back wall are all treated with various thicknesses and densities of porous absorption and air gaps in strategic places. The floor is left hard and reflective, as is customary in control rooms, and helps out the function of the rear wall scattering slats adding some energy back into the ETC.
great video....
i have the exact frequency respond on time stamp 18:18.
a dip around 1-2khz and a hump around 4-6khz.
from your video, those frequency are 'flat' after the speaker is place up the back wall. 18:46
May i know why?
Thanks for the comment. The ranges in my video are much lower and they null and peak are due to reflected bass energy from the front wall. In my room the higher frequencies are flat because of the amount of absorption in the room. Those modes are heavily damped and the treatment easily absorbs at those wavelengths.
By moving the speakers against the wall you get the reflected energy mostly in phase and you even out the bass response.
Wait why can't you block the port. I did that with a subwoofer way back which allowed me to turn it up louder and instead of being super boomy at low volume it was more even sound response and smoother. Cheap sub hack.
Hey hey. There are some reasons why you may want to intentionally block, or partially block, a port - you're right there. The purpose of the port is to extend the LF response of the speaker (or sub) by tuning it close to the frequency where the rolloff begins. So if you don't like the tuning of your speaker, and it's producing too much output at the port tuning frequency, then that may cause you to want to block it - at the expense of degraded LF extension.
The speakers I had were very well tuned, so I did not want to play with their ports. In this case, the only reason it worked for us to block them was because the monitors were getting highpassed by the sub above their tuning frequency.
LCD-5 looks sexy! Are you happy with its micro-detail resolution? ;)
They are amazing headphones for detail, dynamics, sound stage. The frequency response is a touch dark on the top end and a touch bass-light for my taste, but that's my only slight criticism. I would just EQ them a bit with broad shelving filters and they're perfect.
Mandy seems nice but I think I'd go for a proper mic stand for constant position and no movements during measuring.
Normally when doing acoustic tests that would be correct but that’s not how Sonarworks functions. You hand hold the mic and the software echo locates it and you test over 30 positions in a cloud around the mix position. It’s a very different testing method to using REW. When I use REW I carefully align the mic on a stand using a laser and don’t touch it during all the tests.
Thanks for helping to dispel the tired audiophile myth about "correct" speaker distance to the rear wall.
Happy to. As Wes Lachot says “everything is won or lost in the bass”. If you don’t get that right (or hear that right) you get a bad foundation.
Very informative! Great job!
But I still got a question. When you flush mounted the speakers, did you put more bass traps in the front wall around the speakers? If so, wouldn't it be absorbing all the reflected bass frequency that could've boosted the low end? Or you just simply didn't bass trap the front wall and filled it with normal broadband absorbers?
Very good question. Behind the front wall is basically an open cavity. In general, in acoustics, you usually damp any open cavity to tame up any resonance associated with it. We've use rock wool in the cavity.
However, the bass performance is still increased as there is no rear radiation of bass. Because we've positioned the monitor baffle flush with the front wall, this extends the baffle. The speaker is now radiating into "half-space", which means any bass energy that would have been radiated rearward is now concentrated only in the half-space in front of the speaker baffle. This can theoretically produce a +6 dB LF effect. However in reality it's usually more like +3 dB.
Similarly, placing a monitor in a corner forces it to radiate into quarter-space, which will have an even larger bass lift effect, although not usually desirable. Bass lift in a corner can be theoretically +9 dB, but practically more like +6 dB. Boundaries are not perfectly reflective, wall assemblies have low frequency resonances that act a resonant panel absorbers in the LF range, and there is some transmission loss.
@ 4:18 What music is in the bg, I am always being a head hunter for those types of zen house tracks. Love the video!
Thanks. It's a song from Epidemic Sound, that's the service I use for royalty free music for backgrounds. Not sure what the track name is.
✨✨👏🏻👏🏻
Cheers!
As an acoustician, I'm going to offer some help to you and anyone else listening to this video!! (1) It's impossible to assume EQ is going to help with a 12dB swing on the low-end. At most 2dB is manageable and as you stated, moving the speakers forward can help. (2) Taking proper room dimensions out (for purposes of help) measuring room gain (knowing how is key) will tell you what treatments are needed and I can say no 4" panel is going to help with frequencies under 300Hz. I could go on however consulting is not free...
Thanks for joining in the discussion. Nice to have some professional acousticians in the mix.
I agree completely that large peaks and nulls to the tune of 12 dB are not going to be magically corrected and sound flat with EQ. The approach of using EQ, as I said in the video, is an added final layer to be used, if necessary, after acoustic treatment, speaker optimization, etc.
I also agree that no 4" panel is going to help with LF response and I don't think I made such a statement in the video. However, a floor to ceiling absorber that's 5.5" thick with Rockwool Comfortbatt R22, air gapped from the wall, will most certainly put a dent in the LF. Although that's only just a start, and in this room we went with much thicker treatment. The 5.5" thick absorbers are more for home project studios with DIY treatment, not professional rooms.
Thanks for joining in the discussion. Cheers!
So, when we see a phase shift alongside the frequency response in programs like REW, does it mean that the exact drop in response is due to a reflection?
If your speaker has a relatively flat anechoic response, which any professional studio monitor should, then any peaks and nulls in your frequency response are due to room reflections (which by their nature are out of phase to varying degrees and are dependent on frequency). You don't need to see the phase in REW to analyze this, you'll see it in just the All SPL window.
Phase analysis in REW is not as relevant to this because even the direct signal from the monitor will have phase shifts due to the fact that every frequency has a different period and simply time of flight will change the phase relationships of frequencies relative to each other.
Thks a lot !!!
My pleasure. Happy music making!
Just use REW and simulate the room.. takes all this work out of the equation lol
REW's room sim is decent for sure, and certainly a valid approach. Although it's much more technical to use for beginners to acoustics and acoustic testing and it has some significant limitations in terms of data that can be input and calculated. It can never replace in-situ measurements.
Room sims make certain assumptions, such as perfectly reflective boundaries, which are purely theoretical. They also exclude significant information, such as the resonant frequency and pressure-based absorption effects of drywall/gyprock wall assemblies.
You can't get around room testing by plugging numbers into a calculator (at least not at this point in time). I use the room sim when I use REW, but only to see the theoretical / projected modal distribution over top of the actual in-situ measurements to compare them. The discrepancy between calculated modes and measured modes is known and discussed in a fair amount of detail in Rod Gervai's book Home Recording Studio.
You are so right about that, which you discussed about in the beginning. It is so true. Thank you very much.
Glad it was helpful!
Hey great video! Thank you for sharing the knowledge and sources. You mentioned back ports and the very low-end. However Im curious how does all that affect the midrange and high-end. I would bet there are some sacrifices (maybe not that excruciating in a standard box - however open baffles might be a different story). Perhaps is a bit more tricky to measure - but do you have any idea how does placing speakers near walls affect the perceived depth, width, and height of the soundstage?
Open baffles that go down low with big drivers and some physical design optimization can achieve constant directivity throughout the whole spectrum because of the dipole configuration and CD horns. I’ve heard countless times that giving them lots of space is ideal. I wonder what is your take on this? Do same principles apply? What should one be wary about when playing with placement of such speakers?
Great video once again! Keep’em coming:)
Glad you enjoyed it and thanks for the kind words and questions. I have no experience with open baffle speakers / dipoles. That's because my area of expertise is studio reference monitoring and I've never seen that design used. Have you? The closest I've seen is a hard-flush wall mounting (like I have) that approximates an infinite baffle, but still with front ports operational. Bass directivity is forwards only. Zero edge diffraction.
As a mixer, I also work in a non-environment room, which is more or less a hemi-anechoic chamber with a hard floor. It's a very controlled environment with very short T30 times, and ruler flat low end response and full frequency decay times. Not everyone wants to work or listen in a room like that as there is a sense of "no room at all" in the perception of it, acoustically.
You will always want to opt to keep the acoustic axis of your monitors at ear height for any control room / critical listening room. Keep them away from side walls, otherwise you'll get a change in the low end response (even bigger boost) and more SBIR to contend with. Even with fairly deeply treated side walls, you'd likely still get some boundary coupling effects. You cannot correct the destructive interference from SBIR with currently available room EQ systems, so don't create that problem in the first place IMO.
In a setup like this, where you have much more of the direct sound and much less reflected energy, you will have vastly superior stereo field and imaging, as well as an accurate sense of depth and height.
If you want an exaggerated sense of depth and height, then you'd likely want to aim for more of an audiophile listening room that is more "live" than a control room. To achieve that, you can add QRDs to the back wall, front wall, and even the ceiling, as well as porous absorption. This will give a larger sense of 3D envelopment, at the expense of longer decay times, more erratic frequency response, and masking of the direct sound within an interval that is unsuitable for audio engineering work, unless the room is huge.
Read this article, it's very good: arqen.com/acoustics-101/room-setup-acoustic-treatment/
@@warpacademy thank you for the detailed answer and source. I will look into this further as I’m finishing a man-cave project. My listening room is still pretty bare but with high potential. However I have made some measurements and seen some simulations confirming my worries with placing full range dipole speakers too close to the wall. There are some abnormalities with the radiation pattern and directions.
In the end what I’m aiming for is a slightly more reflective environment than it is acceptable for audio engineering work however I love to approach the problem from this very analytical perspective. Thank you once again and take care!
Best of luck with the project. Cheers!
Thanks a LOT for those explanations! I will definitely give it a try because I have done DIY accoustic, I also have sonarworks and I feel my mixes so inaccurate within the low end.
I was used to it my day to day routine but I figured out during my car test something is off in my studio under 100 hz, same feeling when I have gone to my best friend's home studio.
Best of luck! Keep in mind that most rooms also have a huge amount of LF ringing under 100 hz which masks things down there. That’s just due to those frequencies being harder to control. So that could be an issue too.
@@warpacademy Hi, I tested to place my monitors closer from the wall and I also moved my sweetspot a little bit by following the "bass hunter technique" from accoustic insider, the overall result before room eq correction seems a little bit more flat and low end is definitely more present.
My room is fat from perfect but I can't do much with accoustic treatment as I am renting my flat.
I will also test some other listening placement but I am already seeing some improvement.
Thanks for your explanations :D
Stoked to hear that! Be very careful applying room EQ to a space that does not have adequate acoustic treatment. Room EQ will most likely be detrimental vs making an improvement if you do not have controlled acoustics. Only after a decent amount of wideband absorption has been applied should you consider using room EQ, otherwise you can do more harm than good as the room EQ cannot discriminate between a reflection causing interference, and an issue with your monitoring like a cabinet resonance.
Wow!! What a great video! Thanks for sharing. May I ask what headphones are you wearing?
Cheers Kevin. I'm wearing Audeze LCD-5 headphones. Amazing for mixing and mastering. They're my main cans.
Thanks so much now i know why i have a dip at around 70hz
Glad I could help. It's likely that your monitoring setup has a reflection that has an 8 foot longer time of flight, causing the cancellation. That could mean that you've gapped your monitors from the front wall by 4 feet, or that you have a floor or side wall or ceiling reflection causing the issue. Or a combination of them.
Hi thanks for the video! What if my front is window & I have -10db dip at 85 Hz, do I need to put acoustic panels behind my studio monitors or any solution?
Glad you enjoyed the video. Acoustics is a science. To set some good expectations, no acoustician could ever assess your room from a comment on social media. Room analysis requires dimensions, photos, information about monitor placement, monitor type and frequency response, and much more. It's involved. If you want a quality (and correct) answer, you should hire someone. I am available for consulting if you'd like to work with me in that capacity. You can inquire through the Warp Academy contact form on our site and I'd be happy to help.
Regarding your question about placing acoustic panels behind your monitors, I have a deep dive video on porous absorption and its capabilities here: th-cam.com/video/SSn8HEsG8ro/w-d-xo.html.
Watch this video and you'll quickly learn that acoustic panels cannot simply be placed behind monitors to solve a bass dip issue. It's more complex and will require a room analysis, including a modal analysis and room testing with REW. All the best!
@@warpacademy I have move my studio monitor close to the window/curtain. Now the 85 Hz dip is -4db but I got a huge boost on 50 Hz, 6db increased
That’s low frequency boundary loading most likely from in phase constructive interference. You can EQ that to reduce it.
Lol you can't boost the 54 Hz to fix your dip due to SBIR destructive interference. You can boost it all you want it will just cancel itself out lol.
Did you read the pinned comment? This has been discussed in incredible length and all of it is at the very top of the thread. Read that first.
No pinned comment@@warpacademy
I just repinned it. But here is the full comment.
I wanted to pin a comment to the top here to address the most common question that's been coming up in the comments. That's, "You can't use EQ to correct the frequency response when you have SBIR because you'd just be boosting the drive signal and still causing destructive interference at the same frequency based on the 1/4 wavelength" or something to that effect.
The answer is.....YES, you absolutely can. Using standard analog EQ, it's not possible. But when you employ digital adaptive EQ that uses a modelling delay, you can achieve a near perfect response correction. Here are the details on that, and my sources. I had thought the Sonarworks SoundID Reference platform was using this technology, but I was incorrect. I apologize. Sonarworks cannot correct phase distortion from boundaries, yet, although they tell me they're working on it and hope to deploy this in a future version.
Take a look at the Journal of the Audio Engineering Society / JAES Volume 33 Issue 3 pp. 127-132; March 1985. Spectral, Phase, and Transient Equalization for Audio Systems, P. M. CLARKSON, J. MOURJOPOULOS, AND J. K. HAMMOND, Institute of Sound and Vibration Research, University of Southampton, Highfield, Southampton, UK. www.aes.org/e-lib/browse.cfm?elib=4461
If you want to see evidence of this technology that is currently available as a commercial product, look at the Trinnov system. In their technical literature, you'll find confirmation, complete with visual graphics, of how this technology works to correct phase distortion by using a modelling delay to time-align specific frequencies, thus correcting for phase shifts in a non-minimum phase or mixed phase system. www.trinnov.com/en/technologies/.
Also the AES book Recording Studio Design by Philip Newell, page 332-334, which states "Near perfect response at optimum listening position after correction (with digital adaptive EQ)". This example was based on a test where a loudspeaker was positioned on a stand, part way through a room (exactly as I did in position 2 and 3), after which digital adaptive EQ was applied.
This section of the book is also key to understanding why and how the correction can occur:
"A non-minimum phase effect (such as when a loudspeaker is freestanding in a room, and not mounted at or in the wall) cannot be universally corrected by (analog) equalization (alone) of the loudspeaker drive signal. Even perfect amplitude correction could not restore the original phase response by any known practical analogue means. Only by digital signal processing could an almost perfect response be restored.
Adaptive digital filtering can model, very accurately, the inverse phase responses necessary to correct either the minimum phase or non-minimum phase components of a transient or steady state response anomaly. By means of measuring microphones, modelling delays, and adaptive filtering processes, the digital systems can be made to 'learn' what a given room will do to a loudspeaker response, and apply acausal corrections to cancel the disturbances [acausal, in this sense, means that by means of a signal delay, the measured response error is fed into the drive-signal, in inverted form, to precondition the output to anticipate the error."
-Pg 334, 11.5.2 Digital Correction Techniques, Recording Studio Design, Philip Newell, Audio Engineering Society.
I followed up by speaking with Philip Newell directly on this matter. He personally watched this video, reviewed the comments, and confirmed this is a correct presentation of the information. Here are his comments on this specific topic:
"I stand by what I said in the section that you quoted from Recording Studio Design, but it has to be taken in context. Acausal filtering and signal processing, WITH A MODELLING DELAY, can correct for the room boundary effects, but only for a very small region of a room -- say a very restricted 'sweet spot'. Elsewhere in the room, the response may then be much worse than before the correction. So, for example, if someone must work in an acoustically-poor room, this type of processing could greatly improve things at a restricted listening position, which may be beneficial to a single user, or it may 'somewhat' improve things over a slightly wider area, but it cannot make an overall improvement to the room acoustics. When one place is improved, another place will suffer."
Furthermore, I spoke with the team at Sonarworks about their platform and this is what they've said on the topic:
"We align the time between channels by phase measurements currently. The boundary nulling question: It will not be the same as it will still be there, but at a lower energy level. In short, phase-induced dips cannot be corrected by amplitude EQ only, only their impact lessened. For proper handling of phase effects, phase shaping is required."
This parallels exactly what Philip Newell has stated. Sonarworks does not currently do phase shaping (I got this wrong, so sorry about that. I had thought their platform was doing this but I was incorrect.), but I hear they may be planning to roll this out in a future release.
Also, to be clear, the audience for this is music studios, not audiophile hifi systems. In audio engineering control rooms, the flattest response possible is most certainly a primary goal.
My comments to audiophiles are: Do what sounds good to you. You're the end user of your media and you should optimize it however you want to.
Audio engineers and music producers, like myself, and many of the subscribers on my channel, have a different set of needs, which include accuracy and translation.
Cheers!
I’m about to go move my speakers. Recently got stands and a smaller desk and doing testing last week had this exact dip. However the frequency of the biggest dip directly corresponds to one of my room modes so I thought that was the problem and I was about to build more bass traps. Maybe this is it.
Hey hey. Building more absorption into your room is probably a good idea regardless as most rooms are seriously under-treated. The more absorption you have, the more you're going to damp room modes and the more you will widen the Q of your modes - improving the overall sound of the room. Hard to know for sure without knowing about your room. If you want to share info on your room and get more detailed tips, join our free Discord Server ➤ discord.gg/ZwNgZteGKw and post in the acoustics-forum.
You can test to see if it's likely a room mode by moving your listening (mic testing) position forwards and backwards on the room's line of symmetry. If it's a boundary reflection causing SBIR, the null will remain at the same frequency because it's related to the distance between the speaker and the wall which is fixed. If you move forwards and backwards on the line of symmetry in the room, and you notice that the dip is changing to flat and then a peak, then that indicates you're moving from a pressure node to an anti-node in a standing wave (room mode).
But, this could also be back-wall reflection interaction causing destructive and constructive interference too. When your distance relative to the direct source and reflected source change, you will change the frequency of the interference.
Are you using a room mode calculator to figure out your modal distribution?
@@warpacademy thanks for the detailed reply. Of course, I could probably use more trapping money is tight and it’s already pretty good, bass response improved greatly by moving my monitors to stands and off the top shelf of the desk, everything seems to be within 4db of flat. But in the problem area it’s -4 at 82Hz and +4 at 115Hz so I’m trying to flatten that out. Also, when I have the sub off it’s flat at 82Hz and a dip appears around 60Hz. As I’m typing this it makes it seem less like a mode bc if the mic was in a low pressure spot that probably wouldn’t change by adding/subtracting the sub- is that sound logic?
I’m using REQ wizard for the testing and to calculate modes just did a quick and dirty 1130/room dimension but confirmed in the req room sim. Going to try moving the mic as you suggested before moving all my furniture. thanks again for the advice!
Hey hey. If you have tests in REW, please post them. I need to see your actual results. You can send the .mdat file, or you can screenshot showing the All SPL / Frequency Response full range (20 - 20,000 Hz), and then the waterfall showing 1,000 ms time range, and a 60 dB SPL range from 5 dB above peak level.
Wow this is the most informative video for acoustics and sonar works thanks
Glad it was helpful!
Very good video! Looking forward for more on this topic. BTW - what about that 160Hz floor dip, how do you fix that?
Glad you enjoyed it. The floor dip is addressed by placing the mixing desk in the reflection path and blocking / breaking it up. This is because it's not practical to have any acoustic treatment on the floor and carpet would do next to nothing for that frequency rance. My room will have a hardwood floor, partially for looks, and partially for function; the hardwood reflects some of the live scattering energy from the rear wall slats back at the mix pos. The ceiling is soft though, and will have extensive treatment plus a ceiling cloud. Thanks for watching and commenting!
@@warpacademy thanks! Will you explain it further in a video? That would be great!
Sure. I'll likely do that when I get to the part of my room setup where I'm positioning the desk. Cheers.
My ears never lie
thx for all this good research. very well explained
Glad it was helpful!
damn - i had put the acoustic panels behind the speaker.
It's unfortunately really common advice to move the monitors away from the wall and "solve" the rear radiated bass by putting absorption behind them. The problem with that is when you think it through, it doesn't work. 6" or 8" of rockwool won't take enough out of the bass energy to prevent the nulling that will still occur. It's best to leave the speaker right against the wall, putting the reflected energy as in-phase as possible, and to use your absorption around the speakers (above, below, side to side). A pseudo-soft-flush mount if you will.
@@warpacademy thanks! yes, the (incorrect) advice I followed was put panels behind the speakers, first reflection points and in the corners.
@@teabreakbeats yes, this is very common advice. It is definitely a good idea to treat your corners of your room, as that's where modes terminate, but not only corners. For example the common advice to "bass trap your corners". In reality corner bass traps only cover perhaps 5-10% of your wall surface area.
In order to get your decay times more even, without longer low frequency ringing which will mask the low end of your mix to the point where you can't effectively mix the low end, you need much more area of coverage. Which means floor to ceiling absorbers on your side walls, and a full-wall rear wall treatment system including both velocity and pressure-based absorbers.
This is soooo pro and satisfying. TY
Cheers mate. Glad you found this one useful.
Your assistant hand holds the testing mic in front of her body. Doesn't that invalidate the test? She's a barrier to sound propagation. I've always placed my measurement mic on a mic stand. Perhaps I'm wrong because you're performing a near field test? So, it doesn't matter?
Hey hey. You're totally correct that when you're testing a single mix position, with Room EQ Wizard, you would do that. However, Sonarworks operates differently and you test about 35 different positions by holding the mic by hand like this.
When testing with REW I use a measurement delay and leave the room but Sonarworks has a different protocol.
@@warpacademy Great to know. I use REW but I will look into Sonarworks. Thanks!
Cheers! They're both great platforms. REW is more detailed, but also more difficult to use and more buggy as freeware. Sonarworks is more userfriendly and does the correction aspect for you.
this should have a million views at least
I support that intention! People need to know this when setting up speakers in studios.
This is a very well explained video, super helpful.
Glad it was helpful!
Would be intresting to see the waterfall-diagrams too, cause frequency-spectrum only tells halve the story
I have those available actually. I shot the room at each position with both Sonarworks SoundID Reference and REW. Although I'll have to say that the waterfall does not provide any useful information about boundary nulling. The waterfall shows decay time, as I'm sure you're aware. Decay time has to do with modes, which will be active no matter where the sound source is located in the room. Boundary nulling is due to out of phase reflected energy superimposed on the direct signal; it has nothing to do with decay time. The waterfall in REW shows the exact same issue as Sonarworks and the REW frequency response all SPL and phase.
Here is the REW frequency response for monitors 2' from the wall: www.dropbox.com/scl/fi/x3tjscjgfy6tzuks3lrez/2023-Control-Room-Speakers-2-Off-Wall-REW-All-SPL.png?rlkey=hz2ma7wyqzjlyo6kllvufhw1s&dl=0
Here's the waterfall for the same measurement, showing the exact same behaviour and no involvement of decay time: www.dropbox.com/scl/fi/41ak18hn553wxivqkcgn5/2023-Control-Room-Speakers-2-Off-Wall-REW-Waterfall.png?rlkey=vrl3y3cve3up3osplv2jxtcbf&dl=0
@@warpacademy Thank you 😊
Some people say, that you activate the modes more when you have your speakers close to a wall, and even more in a corner. I know the bass adds up 3dB for every plain you add, but some people say not only that, you would also activate more modes stronger
You're welcome. And you're 100% correct; you DO excite the modes more when the speaker is in the corner, When doing initial acoustic testing I intentionally put a speaker in the bottom tri corner of the room where the floor meets the front and side wall, to really set off the modes for analysis. And you're correct that *theoretically* you add 3 dB for each plane you add. Practically that's rarely the case, because boundaries are not perfectly reflective or rigid in reality. Theory breaks down in the real world.
You should avoid ever putting a speaker in a corner when actually setting up the room for listening. But, you definitely should put the speaker close to the front wall like I said. Otherwise you will create the LF nulling issue and you will not be able to solve it. So your low end will be totally compromised.
It's easy and straightforward to damp the modal activity with wide band porous absorption. So don't worry about exciting your room modes. Do your acoustic treatment correctly.
@@warpacademy Yes, regarding First-Reflection Issues, like discussed here for the front wall, it would be best to put the speakers like in the upper front corners directing down, then you only have the ground (or desk) reflection to deal with (which is ok cause I think human brains are used to it as we always have a ground reflection where ever we are normally).
But then you have more Room-Modes ringing!
Intresting you didn‘t have it at the front wall (with minor backwall bass-trapping)
@@warpacademy I saw you now have your Genelecs flush mounted with a hard surface… looks awesome 🤩
How did you plan the acoustics for that? I mean ‚try&error‘ is difficult for that 😄
GREAT video!! What happens to perceived soundstage depth with the monitors as close as possible to the front wall? Is there a degradation of the ability to perceive soundstage depth, and if so what can be done about it?
Glad you liked it. No there’s no degradation of sound stage. Stereo field is primarily perceived in the mid and high freqs. So as long as you’ve done a good job with left right room symmetry and side reflection treatment you should be doing quite well.
Depth in a mix has more to do with psychoacoustics, digital reverbs or recorded room ambience in the recording. Mixing techniques like EQ, delays and such will create your depth staging.
Reflections off other boundaries in the room will mess with you more in terms of that stuff as they can smear transients and detail in the mix if you haven’t attenuated and controlled the reflections.
I think there may be a primitive misconception about sound stage depth relating to the fore&aft position of speakers.
Three things i think this could have stemmed from are:
1. Recognizing the undesirable coloration of too much low freq boost with speakers up against the wall, and SBIR issues with speakers not far enough away from the wall. Both of which are eliminated by pulling the speakers out a third into the room. It dosnt take a frequency sweep to tell something is desirable or undesirable, but current day easy access to microphones, analysis software, and EQ/DSP provide a detailed picture of what is going on and how to optimize.
2. The visual impact of a room's setup can alter ones perception of the sound. Simply seeing speakers up against a wall signals you brain to think there is no sound beyond it. This is not an issue if you have the speakers behind a transparent screen or simply listen with your eyes closed. As a former musician i am minimally effected by this because i naturally see&hear the music and not the speakers however i can understand how some people with differing backgrounds that are more fixated on the "gear" could be swayed by this effect.
3. The speaker & room's impact on reverb. Sure, desired reverb(and similar spatial effects) should be in the track whether recorded or added in the mix/master, but if your listening to a poor track lacking these elements then having a room that 'adds' reverb could lead you to believe that your setup is bringing the song to life when in fact the track was dead to begin with.
Hey Dom. Thanks for sharing your opinions. Thinking that you can "solve SBIR" by moving speakers into the room is incorrect and stems from old school, outdated HIFI recommendations. If you actually test a room as I did, you will clearly see that anytime you move speakers into the room on stands, you get substantial SBIR. The only way to address this is to move them so far into the room (IE 4 meters or more) that the cancellation dip moves below the audible threshold. BUT, who would do that? Who has 4 meters of room to work with? And where would that place you in terms of a listening position?
The physics are basic here. As are the tests you need to run to demonstrate this effect. I'm willing to be proven wrong here though. If you have a room with speakers moved part way into it, shoot the room with REW and show me the results. If you get a flat frequency response (without use of EQ) I will eat my words.
But I've tested this enough to know how it goes. In literally every acoustic test I've run in any room, this boundary nulling effect is present. You can't change the physics. Having a barrier behind the speaker will still reflect sound, no matter how far away it is, and having it just a few feet or meters away does very little to the strength of the reflection. The thinking that moving a speaker a short distance away from a wall (or a third into the room) somehow addresses the reflection is flawed logic and doesn't take into account how sound actually works.