Thanks for the comment! It's easily the most viewed video I have. We crossed paths at the Atlanta Home Theater event a few months back, but were both so busy with things I couldn't find time to chat with you. I'm sure we'll meet at a similar event in the future. 🙂
Hey Ive watched your video 100 times but when i input the delay #s into MiniDSP it is WAY off the predicted curve when i run a measurement again. Any idea?
@@kringle924 The most common reason for this is that the measurements were taken without a timing reference. For USB microphones like the miniDSP UMIK, you need to tell REW to use an "acoustic timing reference" when the measurements are taken. This is on the measurement screen in REW. You will hear a chirp from the L speaker, then the sweep of the sub, then another chirp. Always use the same reference channel for all subs. It defaults to L which is fine. Once this is enabled, it should stay enabled which is a good thing. Always use a timing reference for every measurement you take. The second most common reason is that the delay values were not set to 0 ms on the output side of the miniDSP. You don't need to change the delay in the AVR / Processor.
Okay THIS video is probably the single most important one regarding REW! Delays are how you align you subs so that they work together well and holy shit I've learned that I've been going about this the wrong way.
I've been following Steven Smiths videos on Home Theater Gurus and happened to come across your videos. I'll now be going through all your stuff, cause this video specifically helped my wrap my head around getting everything aligned using the "Alignment Tool". Using the slider and getting a visual representation of "whats going on" made a world of difference. Thank you so much !!
Steve’s videos are great! They’re how started out myself. The Alignment Tool feature was added after he did those so I thought I’d try to give back a little 😊
Thanks so much for this fantastic tutorial. I'm wanting to get into REW this year and I'm doing as much homework as possible but I find so many people skip steps that to them are very simple but for me as someone who's not as savvy on a computer as others can leave me sort of stranded.. Really appreciate your effort to walk us through step by step. Really looking forward to seeing your videos on how to apply EQ through mini dsp and any other info you care to share. Cheers Mark.
Jeff, great video and explanation of how to use the alignment tool. Please keep up the great work and carry on with vidoes for next steps on EQ and alignment to mains etc. There is a lot of misinformation out there. Yours is one of the best I have seen. Well done sir!
Actually, I think that the next step before applying any EQ would be to capture an actual measurement. Would like to have seen a comparison between the predicted curve and the measured curve, but otherwise, a great and very, very helpful tutorial, thanks.
I completely agree. I didn't really plan on doing this video so I just winged it. ALWAYS take before measurements so you have a good baseline for comparison. ALWAYS take after measurements to verify things were implemented correctly, and that the results match the prediction. If the mic has not been touched between the before measurement and the verification measurement, I find the REW prediction to be extremely accurate.
@@jeffmery what i dont get is, now with this prediction curve, it will measure over 95db at lowest frequency... i think those sub can't handle it.. so a low pass filter is a u must hen?
@@Arvidje Maybe or maybe not. What you are seeing is the benefit of having phase better aligned. Phase alignment between two (or more) subs is free output. When phase is perfectly aligned, we get up to 6 dB of truly free output - meaning increased SPL that doesn't cost us any increased power. However, when phase is perfectly mis-aligned, we can have up to 100% cancellation - meaning that even with unlimited power, we can't fix the problem! Now, it's possible that with time alignment, your combined subs are now much louder than they should be. Normally, the LFE channel is +10 dB from the non-LFE channels per the specification. If it ends up super loud after alignment and EQ, then it just needs to be turned down a bit. Room correction or the controls built into your AVR are the right tools for the job. (I think you meant high-pass filter to protect the subs below their port tuning frequency). You might need a high-pass filter if your subs don't have one. Commercial subs almost always have one built into their amplifier, but it's best to be 100% sure and check with the manufacturer. For DIY ported subs, you will need some kind of high pass filter set at or near the tuning frequency of the port. The type and slope depends on the enclosure. Assemble-yourself boxes from the likes of GSG will tell you exactly what type of filter and the frequency. You'll need to make sure you implement this as required to avoid damaging the subs.
@@jeffmery thanks for your explanation. I understand it more. But I have one more question about that low frequence. For example if u have two different subs, one sub maxed out at 30hz, the other 18hz.. how does this play out after eq? Is it best to do a high or low pass in the output of that weakest sub to play safe or doesn't it matter, it doesn't try go lower?
@@Arvidje If you have different subs then you're most likely going to be limited by the less capable sub. It would be safest to high-pass the sub that plays to 18 Hz and match it to the other sub as best as possible. Do that before EQ. Then time-align the subs using the alignment tool. Then finally EQ the combined response.
Have a look at this video to see results of EQ in REW. It's based on a different set of measurements, but the result is the same. th-cam.com/video/ARztXSmoQbE/w-d-xo.html
Hey Jeff, I had just set up my Mini-DSP over the weekend and just watching/reading through all the other tutorials was daunting. This one was much more easier to consume so I'll give it a whirl this week.
Thanks for the feedback and let me know if you have any questions. It's definitely hard to find a balance between quick and consumable, while still giving enough information for people to understand what we're doing and why. Though .... some folks probably don't care about that at all.
@@michaelpiper6601 If you mean the sub distance in the AVR/Processor, Dirac will calculate those distances during the calibration. In my experience, the subs are usually (but not always) set to zero and then everything else is delayed from them. This is due to the lag most subs have due to their own internal DSP in their amp (i.e. crossover, delay, phase adjustments on the actual sub). If you mean aligning the 4 subs behind a miniDSP before presenting as a single sub to Dirac, then you should fully align and tune them in the miniDSP first. This means setting all delays, gains, and PEQs first in the miniDSP. Then measure to verify the results before running Dirac. Dirac will see a single sub instead of 4, but it will still set its own distance and gains as above. A further complication is DLBC. If you are not licensed for DLBC, then you should probably verify the integration with the center or mains and adjust the distance accordingly after running Dirac. If you are licensed for DLBC, it really takes care of all of that for you (at least in my experience and my theater). I hope this helps, but let me know if it creates more questions for you!
Thanks for such an informative video! I’ve been playing around with rew the last few days and using the alignment, time and SPL features I’ve increased my output significantly across the frequency spectrum with two subs. Upon eq and uploading to my MiniDSP I have a flat response with a small -3db dip at 53hz. Keep up the good work and keep the videos coming!😁 I’m subscribing now.
Three adittional tipps here: -You should also conciser trying negative delays. It might be that the woofer you are trying to align is already too late so you cant fix it with adding delay. You will instead need to "subtract delay" (which actually means adding delay to all the woofers from the previous measurements) -You can klick the slider and then move it with the arrow keys on your keyboard, which makes small adjustments way more convenient. -After aligning all woofers into one "virtual" woofer and integrating it with the mains, you should definitely check out the Group delay plot i REW to see if your subs are playing "in time" with your mains. In most cases they will be A LOT too late, even if the frequency response looks good in the crossover region.
All good points. I was trying to keep things as simple as possible for folks. I would also consider group delay a much more advanced topic than simple alignment. It's also not something that's easy to correct for the vast majority of people or the equipment they have. I know Trinnov's Optimizer will correct for group delay, but I'm not sure even Dirac with multi-sub DLBC will. I think I have an open weekend coming up so I might be able to experiment with this on my system. The upside is that at these frequencies, human hearing is much more sensitive to the magnitude response than the timing. It's why we can get away with sometimes significant delays and not notice. The best analogy I've seen is that bass frequencies are like a train. When does the signal arrive? When the first car crosses the road? The last? somewhere in the middle? This ^^ is more for less experienced folks that might read this, Ferdinand, and not directed at you in particular. 99.9% of folks out there have never heard of group delay :-). Thank you for your comment!
@@jeffmery I fall in this category since I'm a REW Beginner. Your videos are helping me a lot understanding and learning different REW methods. Thank you!
Great job buddy 👍 Jeff (AKA Fattire @AVS) I was going suggest making a video about the Alignment Tool but it seem your all over it 😀 I think this will help a ton of people out as video tutorials make the process seem so much easier. I can see myself linking this video a lot 👍 I found it strange that you didn't need any inversion for the back subs, this is the same when aligning my two sub 1 front 1 back no inversion needed and little to no delay. I would for future videos turn off the the dark mode in REW, it comes on as default for newer versions and I find myself that it is harder to see the different sub FR's Also I found when fine tuning the delay I click the button once that moves the delay and then use the right and left on the keyboard to visually get what I want. Again Great Job Jim
Great feedback. Thank you! I call them "rear" subs for simplicity, but there not really behind the MLP. Our listening area is a large 3-4 person couch. The "rear" subs are on the left and right sides, directly in-line with the long axis of the couch. My room is fully enclosed, but it's an odd shape, and is split use (2/3 theater and 1/3 home office). The office part is at the rear and limits my sub placement options.
Thank you! 4 subs means 4 times the power and 4 times the cone area. 1sub 90dB will redult in 2 subs 96dB and 4 subs 102dB then. 12dB extra in a perfect world. You could have added the 4 without any delay and compare to your optimzed version to show the difference between simply 4 and 4 optimized with time alignment aswell.
Headroom is a wonderful thing isn't it?! ;-). That's an excellent point on a before/after comparison. However, here's how the sausage is actually made in my room with my gear ... Using the REW methods, I get the best results in my room *without any* individual sub delay. Call me lucky or spoiled or whatever. Adding any delay to any of my individual subs results in a worse combined response. A lot of it is due to placement of the subs and listening position in my multi-purpose room. So had I done a before/after with actual results, it would have looked worse! LOL! That is absolutely not the case for 99.9% of the people that need to do this. I wanted to be sure to show reality as opposed to the unusual case of my specific room. It was also not the case when I initially started out, so I know the process works very well and others would benefit from learning it. That being said, REW didn't give the best total/overall response for all listening positions. That was achieved with MSO resulting in a basically perfect flat response (before house curve) from 15 - 150 Hz. The MSO method, however, *did* apply different delays to the individual subs. Obviously that's a totally different method from pure REW. I thought it important to call out that detail and be fully transparent.
Awesome video. Always a pleasure to see a demo of a system from someone that clearly understands it. I haven't gotten into REW yet but I appreciate the primer even though this was a more advanced video I'm a genius so... 😉
Great video! +10 dB is usually perceived as a doubling of loudness, which is still really impressive considering you're just time aligning the subs. It beats doubling your wattage to get a 3 db increase
@@jeffmery Hi, Denon AVR user here. Just to confirm. You sweep your subs using LFE channel at 85dB each sub(30-40dB above noise floor method) then if the subs combined response is to loud when level matching you can use the gain controls in the miniDSP to lower the individual and combined subs volume. Our goal here is to get the most headroom possible right?
@@martinmares8998 I wouldn't adjust the levels of the subs until after they are aligned and EQ'd together. Each room correction tool is slightly different so it's best to adjust the levels as part of the room correction process. Sometimes that just means turning the subs down within the correction software. Sometimes the gain on the subs needs to be adjusted. If you need to adjust the gain on the subs, they all need to be adjusted identically either up or down as required by your room correction software.
@@jeffmery I just completed my subs time alignment and EQ'd them together using the time alignment tool as shown in your video and I was impressed how accurate this tool is not to mention all the time you save doing it this way. Do you only focus on positive summation of subs or do you check phase also? Great video and clear easy to follow instructions. Thank you!
The most important differences, in my opinion, between the 2x4 and 2x4 HD are 1) 7.5 ms vs 80 ms available delay 2) 6 PEQ filters vs 10 PEQ filters on each channel 3) 0.9V vs 2.0V output voltage and 4) 0.9/2.0V vs 2.0V/4.0V input voltage options. None of these alone are really enough to justify 2x the cost. However, when put together, the HD is a much more flexible device. The output voltage is probably the most important of those. Again ... just in my personal opinion.
This is good stuff. I've played around with the alignment tool, but maybe I didn't realize the power of it. Always thought minidsp was the easiest way...
miniDSP is probably the highest "value" way to achieve the most flexibility, but it's not the only way to implement things things :-). Just tweaking the distance in the AVR/Processor can provide huge benefits in the crossover region. Some AVRs/Processors and subs allow for manual EQ as well. My old Yamaha RX-A1080 allowed for 4 bands of manual PEQ on the sub. SVS "Pro" model subs have a 3 or 4 band PEQ built in to the DSP on the plate amp as well. A calibrated mic, REW, and a miniDSP are mentioned together so often that a lot of folks think you have to do all or nothing which definitely isn't the case.
Great video. How about a video about the so called "sub distance tweak" as well, since that would be a logical next step after getting the subs aligned to one another.
I thought about this, but there are a bunch of "sub distance tweak" videos already available. That being said, it is on my list to do one "in my style" if that makes sense. There are a lot of people that appreciate the context of what is being done vs just following instructions blindly. This is still still on my list to do, but finding the time is unfortunately hard 😞
@@jeffmery there are a lot of mediocre ones out there that ramble around the topic. If you don't have time to make one (understandable) but know of one you believe is good, perhaps link to that?
Glad to hear it! The tape measure method just flat out doesn't work for subs. Below the Schroeder frequency (aka the transition frequency; typically somewhere between 200 - 500 Hz depending on the specific room), we're not hearing the direct sound from the speakers. We're hearing the resonances of the room they are in. Sub phase is a function of time from the individual driver (and port if ported) as well. The alignment tool takes phase into account when summing the responses giving MUCH better results than manual methods.
@@jeffmery Do you have any advice on best alignment of multiple subs with multiple seats taken into consideration? You recommend aligning for a single seat and then move on to EQ? EDIT: My bad. I see in the description you mention MSO. I'm back to trying to fully grok how to use that tool effectively :)
@@joker927 No worries! Yes, MSO is probably the best option there. I also think that Dirac's DLBC does a really, really good job and is about as easy as it gets for multi sub integration. However, it's only available on a Dirac-based processor and even then, there may be an up-charge for using it depending on the specific processor. Using MSO with your DSP of choice is a much more "universal" option. A miniDSP 2x4 HD is probably the most common and recommended device, but the filters from MSO can be used with any DSP. You would need to set the filter limits in MSO to match what the DSP is capable of, but that's about it. I recently helped someone implement MSO using the Trinnov-native filters. Those have slightly different min/max values for gain, Q, etc. We just set those limits in MSO and were good to go.
Thanks for the great walkthrough. I just subscribed to the channel. Hopefully on your channel I can find other videos of you teaching us how to use this EQ and other adjustments that you mentioned at the end of the video. If you do have those videos, you might want to add links to them in the description that you wrote above for this video. FYI, In my case I used MSO and a miniDSP 2x4HD since I have 4 subs and 2 rows of seating. With MSO I only used the sub alignment feature to get better bass across all seats, and now I will run MCACC PRO on my Pioneer Elite SC-99 to get the subs and mains aligned. My goal is to then use the time alignment tool in REW after the MCACC PRO auto calibration, to tweak the time alignment of the subs and mains again. Hopefully you have a video showing this process as well.
Thanks for the feedback! Finally got time to work on videos and will be posting new ones very soon. The alignment process is exactly the same. Instead of aligning the measurements of 2 subs, you align the measurements of the combined subs (post MCACC) and then either the Center or L+R channels (again, post MCACC). The center or L+R measurement goes in the top box, and the sub measurement goes in the bottom (since we are aligning the subs with the mains). If you get a positive delay number on the subs, you need to add distance to the subs in the AVR. If you get a negative delay number on the subs, you need to reduce sub distance in the AVR.
I would be interested in seeing what the results are if you change the order of sub introductions. In other words, with the subs labeled ABCD, if you added them in that order one time and then reversed to DCBA, would you still get the same answers? Have you ever done this? Maybe a dumb question but just a thought. This is not terribly applicable for my sub, a Velodyne servo 15, as the gain runs typically about one micron off the pot stop, and add about a 1/4" circumferential movement and windows/dishes/etc are rattling. My worst problem with this sub is that it produces standing wave resonance nodes in the room no matter where you point or place it, as it runs right down into the 32' organ pipe range very nicely. I definitely need some proper acoustical treatments, meaning some properly designed and placed bass traps for at least the first cut. Thx for the vid.
I have looked at changing the order as you describe. But to be perfectly honest, it's been so long I don't remember. The fact that I don't probably says there wasn't much of a difference. It's not much effort to try it out. I"ll see if I can find some time this weekend. My previous sub was a Velodyne CHT-15 which I believe was one of their early servo subs. Bought it in like 2001/2002? Good sub and my brother is using it now. It definitely had plenty of power, but wasn't tuned terribly low; maybe 28 Hz? Bass traps should help with that at least a little if you can get enough of them into the right places in the room. We also shouldn't expect too much out of a single sub when it comes to an even response. Unless it's (very) near field, physics just won't let us get even sound from one sub in a small room (to be clear, I'm talking acoustically small which is basically every room in every house ever built; smaller than an actual movie theater).
I should have to illustrate how accurate the alignment tool is. I recommend trying it yourself. The measurement will directly overlay the prediction (assuming the microphone wasn't touched or moved between the initial and then verification measurements).
Thank you very much - very useful video - best on the topic. Please do follow up with the sub eq and subs to main integration videos - really looking forward to those. Saw your MSO tutorial as well, but since I have only 2 subs and one MLP, this method should suffice, in my room. Best wishes..
Hello Jeif Great video easy to follow. "I’m not using MiniDSP because I have a processor with independent subwoofer control. Should I input the time delay result directly into the processor’s subwoofer delay settings?"
As a general rule, you should always use a timing reference. Generally speaking the acoustic timing reference is the easiest to use and has the fewest caveats as opposed to the loopback reference. I am not an expert in the nuances between the two methods, though. Based on what I’ve read while learning and experimenting with MSO as well as several online exchanges with Andy, the creator and maintainer of MSO, the acoustic timing is definitely the path of least resistance.
@@jeffmery I have been using a 1/4" cable on my scarlet Solo as loopback reference with success. A bit tricky tho to set the right output inputs in your preferences.
@@kewlbug Good to know! That's one of the reasons I recommend the acoustic reference over the loopback. The loopback requires additional hardware (like the mic preamp you're using). Folks coming from a recording/studio background will have that equipment, but the vast majority of folks coming from a home theater background/perspective won't (USB mic and go!). Figuring out the right output/input is also tricky as you mention. Those inputs could also vary depending on the specific mic preamp in use. All that being said, there may be technical advantages of a loopback over an acoustic reference. I haven't dug into it to verify that one way or the other, or under what conditions any advantages may apply.
Is this a pro upgrade? The alignment tool is not showing as an option for me under controls. Version 5.20.13 Update got it figured out and now the subs are working pretty good together besides one small area. Between 80 and 100hz it dips down even below some of just the single responses from a raw measurement and is the only area that isn’t close to the summed measurement from your video. Is there something I can do before moving on to eq. Following your video I definitely found the delay to give the best summed responses just hoping to fix this small area before I move to eq.
Glad you figured it out! The only thing the Pro upgrade adds is the ability to measure from more than 1 microphone at the same time. The most common use-case is taking spatially averaged measurements. This is an advanced topic, and I don't have any tutorials on it. However, assuming you're using a miniDSP, an all-pass filter on one of the subs might be able to resolve that for you. It's too complicated to go into here. This video from Meyer Sound does a good job explaining all-pass filters and some ideas on how to implement them: th-cam.com/video/_KAa39OcMoo/w-d-xo.html
@@jeffmery awesome! Thank you. So the raw measurement with alignment is definitely much better across the board and similar to the to the last summed line you get. But it’s definitely way better compared to just the single measurements I took. Before delay or anything.
Jerry great video What about using this approach to combine subwoofer and bass to help determine a better crossover frequency between the two drivers ? Thanks
I assume you mean the crossover between the subs and non-sub speakers? It's actually the exact same process and is commonly referred to as the "sub distance tweak". You just need to get the correct set of measurements. For home theater, subs are usually aligned with the C channel. 1. Set your crossover for C 2. Measure C from 20Hz-20kHz 3. Mute the subs, measure C from 20Hz-20kHz 4. Unmute the subs, temporarily disconnect the C speaker, measure C from 20Hz-20kHz You now have a "before" from step 2, a C-only measurement from step 3, and a sub-only measurement from step 4. In the alignment tool, you want to use the step 3 measurement in the top part of the alignment tool and the step 4 measurement in the bottom part. This will help to time align the subs to the C channel at the selected crossover frequency. Let me know if I missed the mark and you were actually referring to something else on the better crossover frequency.
Hi Jerry thanks for quick response I’ve built a fully active 3-way with sub. I’ve questioned whether I have selected the correct xover frequencies, ( sub to woofer, woofer to Mid, mid to tweeter). I’m thinking what I’ve learning from your video if I can start with woofer and align with mid ? Use the technique to see if I slid the freq up/ dwn if there is a sweet spot vs mfg xover freq?
@@zeekbruno4869 Ahh yes, I completely misunderstood. Yes, I think the alignment tool can be used for that use-case. But it doesn't have the ability to leave the timing fixed and just slide the crossover frequency around to see what works. I'm sure you know this, but crossover design gets complicated. It's about the frequency as well as the slopes on both sides, then gain, delay, inversion, and even AP filters can play a role. I honestly haven't done any driver-to-driver crossover work outside of implementing manufacturer supplied specifications for their active speakers. There are probably tricks in REW to make things go faster, but I just don't know them.
Excellent video, mate. I have been pondering if incorporating a MiniDSP while already having Dirac Live (and four independent sub outs on my AVR) is redundant. But if I’m understanding this process correctly, MiniDSP plays a role in establishing the most neutral base for Dirac to do its thing. Thoughts?
Thanks for the feedback! With Dirac it depends on the specific AVR and what Dirac features it supports. Not all AVRs/Processors support all levels of Dirac so that's something you want to find out. There's Dirac Full-Range, DLBC (with single and multi-sub options), and ART. A miniDSP can help Dirac specifically with the "Full-Range" version. This is the entry level Dirac license that doesn't do anything special with subs or crossovers. In this case, the miniDSP would be used to pre-align and EQ the subs as one before running Dirac. Things get a bit more interesting with DLBC (Dirac Live Bass Control). DLBC does two things: automatically aligns and integrates multiple subs together, aligns the non-sub speakers with the subs at the crossover. DLBC can be purchased in a single-sub or multi-sub license. With multi-sub DLBC, there's really no reason to have a miniDSP in the signal chain. It does a great job on its own. With the single-sub license, a miniDSP can, and probably should, be used to present a single big sub to Dirac. In this case you're really getting just the benefit of integration at the crossovers with non-sub speakers (still very good). Depending on your level of comfort and how much you like to tinker, DLBC is the closest thing to a multi-sub "easy button" around right now. It's not perfect and takes some tinkering, but it's less complicated and less costly than doing that plus a miniDSP. ART is the top of the line room correction from Dirac. It's only available on Storm processors right now but others have announced that they'll add it in the future. With ART, you absolutely do not want a miniDSP between the subs and AVR. ART needs direct control of all subs and speakers to work properly. I hope this helps, but if not, keep asking questions and I'll help out 🙂
The first time I aligned my subs was with the phase on my 2nd PB-1000Pro and REW sweeps, and it worked great. This time I used this method and got worse results. The actual response compared to the simulated one is pretty off for me. I tried adding each individually, and summing the front and rears then summing. I'm going to work on it again now.
The three biggest reasons for the prediction not matching the results are: 1) The mic was moved and is in a different position 2) An acoustic timing reference wasn't used 3) The measurements used for alignment were taken in what I'll call a "non-zero" or "non-reference" state. On that last one, we need to turn off all room correction, dynamic volume, and dynamic EQ in the AVR/processor. Gains, delays, inversion, and PEQs need to be zeroed out both the AVR/processor and the miniDSP (if one is in use). Finally, the settings on the sub itself need to be at a baseline of zero phase, no inversion, crossover bypassed or set to the highest value possible, any PEQ bypassed, and gains set to a known level. I use the midpoint of my gain knob, but some subs like your SVS, have digital gain with a default of -10 dB; that's fine. The phase setting on the SVS pro models is nice (I use the SB-1000 Pro in my desktop 2.1 setup). Single degree adjustment is awesome granularity if you don't have a miniDSP to provide extremely precise control over delay.
@@jeffmery 1. Mic never moved. 2. Not sure how you create a timing source with REW sweeps. 3. All DSP on all 4 PB-1000Pros, MiniDSP 2x4 HD, and JBL SDP-55 turned off. I know to do all that. I ended up just doing manual sweeps and aligning myself. Got a MUCH better and flatter response.
@@andrewskaterrr Good deal. The acoustic timing reference is on the measurement screen. On the right side near the top, there's a "Timing" field. This defaults to "No Timing Reference". Change this to "Use acoustic timing reference". When you do this, an additional field will pop up below "Output". "Output" is where you select the channel that you're going to measure. The new field below this is "Ref output". This is the channel that will be used to create the acoustic timing reference. It doesn't matter what channel you use. However, it's important that you use the same channel for all measurements that you're taking in that session. It's my personal habit to always use the left main, but that's just habit. That's all there is to it for enabling the acoustic timing reference. These settings should persist so you really only need to do this setup once. Now, when you take a measurement, you'll hear a short chirp out of the "Ref output" channel, and then the measurement sweep will run on the selected "Output" channel. You should always use an acoustic timing reference for every single measurement. Features like the alignment tool require them to work.
@@jeffmery thanks. I wondered if this existed but Ive never heard/saw anyone mention/show it in any of their videos on REW or MiniDSP, which blows my mind. i was actually trying to think of a way to do it myself, thank god I don’t have to lol. I googled it earlier and saw it can use the loop back input? I’m going to try when I get home in a bit.
@@andrewskaterrr 99% of home theater hobbyists, including me, use USB mics. The acoustic reference is the only option when using a USB mic. Loopback references have some specific hardware requirements due to the synchronization required between the sending and receiving sides of the actual loopback. The link below has a good explanation from Andyc56, the creator and maintainer of MSO: audiosciencereview.com/forum/index.php?threads/rew-loopback-timing-with-a-usb-mic.14273/post-438115 I can't speak to why so many videos don't discuss the use of an acoustic timing reference. Some features, like the alignment tool, are relatively new features released at the same time or shortly after some of the best, and most often referenced videos, were created. I've watched others, though, where it's clear the person is chasing views and parroting things without fully understanding the "why" behind the "what" they're doing. There's no malice in what they do. And some viewers just want to get "better" results as fast as possible. Some people don't care about the "why". That's all cool. I want to understand what's going on and get the best results I can, not just better results. I'm still learning a TON from a lot of people too. Just this week, someone pointed me to a video on room calibration that completely blew my mind and changed the way I think about things. I'm still trying to process everything talked about there (it was a 3-hour video!). Always be learning!!
Thanks for the video! Only problem is every time I use the adjustment tool and then measure to verify, they aren't similar. I've tried multiple times after rewatching your video & I get the same inconsistent results. Any ideas why?
Are you taking your measurements using an acoustic timing reference in REW? That's a hard requirement to use the alignment tool. If there's no timing reference then the alignment will never be correct or match the predicted response.
@@doublet147 Easy mistake. I've done it many times by accident myself! I personally haven't run across any circumstances where I *don't* want to use a timing reference. Using one for all measurements doesn't hurt anything at all, and can keep you from having to try and duplicate measurements if you find you need one. It's a good habit to get into for sure. Good luck, and let me know how it turns out!
Hi Jeff! Thank you for your video, very educational! I just have a question… I hope you can help me with this. When and what sub I have to put the delays? Every time after align 2 subs? Or at the end? Thank you!
If you're doing more than 2 subs, it has to be done in steps. We'll use sub A, B, C, and D as an example. Each time you align things, only one gets delay added. So if we start with aligning A+B, only B will get delay. We enter delay for B, then take a verification measurement. The verification should almost perfectly match the predicted alignment if the mic hasn't moved. Now we align the new A+B with C. Only C will get delay. Enter that for sub C then take a verification measurement to ensure it matches the the prediction. Finally, we align A+B+C with D. Only D will get delay. Enter that for sub D and take a final verification measurement.
Yes it's absolutely needed, but the ability to implement will be questionable. That will depend on the specific capabilities of the gear in use. To implement this, you need to be able to set different delays for each sub. One needs to stay at zero, and the other needs to be set to the value from the alignment tool. You'll have to figure out if your AVR or processor has that capability. There's also the issue of the granularity of that setting. Many AVRs only allow 0.5 ms granularity where the miniDSP allows 0.01 ms granularity. This doesn't mean you can't do it, it just means that the actual results might not be quite what REW predicts. It should still allow for improvement provided the subs have independent delay as indicated above.
Good day. I just stumbled upon this video. I've watched your MSO videos for my subs but is using REW's alignment tool better than MSO. I ask because I used MSO and had great a response, but I changed out one of my subs, so I have to start over. I aligned the manual way through the MiniDSP, but I like the REW alignment tool to better see what each delay does. Should I use MSO after alignment?
This depends on your goals. If you only care about a single listening position, then the REW-only method using the alignment tool is great. It takes less time/effort and gives great results. If you care about more than one listening position, like an entire row or multiple rows, then MSO is probably the better choice. This all depends on the room and the placement of subs / seating in order to mitigate the room's modes. There aren't any absolute answers. I would not try to align the subs manually and then use MSO. A manual alignment for a single listening position won't usually be the best alignment for multiple listening positions.
I like to print the aligned sum for a few different options to be able to compare them. (e.g. my best results normal and best results of inverted). Also, you dont even need to have the original traces "on". Makes it a bit cleaner. I don't mind a little bit of smoothing, (at least to get a better idea of the summation) as I'm just looking for the best "average" kinda result.
Sort of related, I've started using the VAR smoothing more with eqs, and putting most of my eq bands in the lower regions. (I've read we are more sensitive to EQ issues in the lower freq. ) So far the results are great. I've discovered you will also get a lot different eq results depending on the smoothing you have.
Variable smoothing is what should be used when you're going to apply EQ to the curve that you're looking at. That's actually what it was designed for :-). I use it almost exclusively, but will sometimes flip to no smoothing (subs) or a specific smoothing (like 1/6) depending on what I'm looking at.
@@jeffmery Yes I've recently started to ditch my YPAO, looking at the YPAO result, it does some weird stuff. It might be good for most people, but we can get a lot better with REW.
@@kewlbug I don't remember the specific AVR you have. Up until the newest flagship AVRs from Yamaha, YPAO was terrible at optimizing subs. My previous AVR was an RX-A1080 in a 7.4 system. The subs were optimized with MSO and then presented to the AVR as a single sub. I always had to tweak the distance to eliminate the dip at the crossover point. There's also no way to limit the curtains or do house curves in the older versions of YPAO. The current flagship AVRs have a new version that adds this missing functionality. I haven't, and might not, get the chance to play with them myself. I've always liked the sound of Yamaha AVRs. I still have a 1080 running the living room system and patio on Zone 2. They've just needed help with the subs. Don't forget that YPAO volume and Dynamic EQ (think those are the right names, not home and going from memory), will change the contours of the frequency response. You might or might not like what they do. I always turned them off completely when measuring/calibrating. Then I'd turn them on after I was happy with things.
@@jeffmery Its a 2070, so yeah the older softwares. It is nice to jump on the PC in my office, and use the web interface to enter in the PEQ numbers. Not great though with only 7 bands per channel and 4 bands for each sub. The bands aren't fully adjustable, but you can move them around a little bit trying to squash those peaks.
I guess it should be noted that one can achieve this without a miniDSP as well, if one has subs with a variable phase knob, which effectively allows one to choose 0 to 14 ms of delay for that sub (0 to 180 degrees on the phase knob).
Correct if you need 14ms or less delay. There are problems with phase knobs too. The main issue is the lack of control since most knobs are analogue. Very few sub manufacturers include digital DSP on the sub amps that provide granular control. The second issue is that phase is highly dependent on frequency. How many degrees of delay are how many ms? At 100 Hz, 180° of phase is 5.64 ms. at 50 Hz, it's double that, or 11.28 ms, at 25 Hz, it's double again at 11.56 ms. I don't want to give the impression that the phase knob is useless. That's absolutely not the case. It just takes more work than DSP with extremely granular delay. That could be a miniDSP, other external DSP, or within the distance/delay capabilities of the AVR/Processor in question.
Amazing 🤩 so useful and well presented. Next step would be EQ in REW But then I am lost. How do I transfer the measurements to the MiniDSP? Do I add the individual delays for the subs then add the EQ file to each of the four subs in the MiniDSP?
Great tutorial although using dark mode made it a little hard to see some of the measurement lines. I did the manual method and am looking forward to trying this out to see if I can improve on things.
@@jeffmery It went great! I managed to flatten out my response noticeably and learning how to use the alignment tool helped me align my mains as well. FYI you can click the slider button, and then use the arrow keys to adjust the delay rather than clicking and dragging to fine tune things easier once you get in the neighborhood.
Excellent video and very good job explaining the process! I have a naive question; I have two SVS PB2000 PRO is there a way to align them without miniDSP before running audyssey and sub distance tweak?
Unfortunately I don't know the answer to that. You need to be able to apply delay to one of the subs to align them for the most positive summation. I don't think the SVS DSP has a delay capability. You might make some progress using the phase adjustment on the sub, but that's not the same as using delay to align the phase like we do with the alignment tool. I have never used Audyssey so I don't know what things you can set before / after calibration that will stick. I think it will vary with the AVR and even the version of MultEQ / SubEQ. I wish I had more experience there for sure.
This will depend on the type of subs you have. If they are all identical, set the gain control on their amplifiers to the same position and go. I like to use the middle position so I can turn the subs up or down if needed. If the subs are different models, but similar capabilities, then level matching at the MLP can work. If the subs have differing capabilities, you can try distortion matching. Essentially, turn up the level on the least capable sub until distortion becomes an issue. Then back it off just a bit. Then turn up the more capable sub(s) until their distortion matches that of the less capable sub(s).
@@jeffmery Thanks for the reply Jeff. My subs are all identical, but some are right in a corner and the rears are shooting into a open space, so their output under 50 hz differs quite a bit(no room gain). Basically the back pair is not contributing at all underneath 50 hz. They are also futher away. Measuring at the MPL they are about 3 db lower in output between their outputs. This is why I am wondering if the summation will then work correctly.
@@poleepwka I'd set the gain on each to the same level. I would not allow any gain in the PEQ filters for those subs. If they're ported, make sure they have protection at or below the port tune. They should if they are commercially available subs, but it's best to verify.
Jumping on the bandwagon to let you know how superb your tutorials are! I've watched tutorials from Home Theater Gurus as well as Home Theater Gamer when I first got into the miniDSP + REW game, but I recently discovered yours and I'm loving how thorough you are in your explanations, plus the fact you use a Mac & UMIK-2! I'm sending you a message on the AVSforum with some questions I have, but figured I'd also ask one here... - Should I start off my sub measurements using the LFE channel or the Center channel? I know you mentioned the LFE is 10dB louder and can cause issues when aligning to your Center/Mains.
Thanks for the feedback! I used the Home Theater Gurus video myself when I got started. Being on a Mac made some things easier (and harder). Plus, the alignment tool didn't exist when he did his excellent video. I figured I could give back a bit to the community and help out with my own versions :-). I'm happy to answer here or over on AVS. If you're a new user there, I think there's a minimum post count you have to hit to send a DM (just FYI). When aligning multiple subs, using the LFE is just fine. The only alignment we care about there is the timing amongst the subs. The volume really doesn't matter at all as long as it's sufficiently above the nose floor and not changed during the measurement process (set it and forget it until you're done!) When aligning subs with the mains (L+R) or center, I use those channels respectively. I just mute the subs when measuring the speaker, and then unmute the subs and disconnect the speakers to get the sub-only output. Then it's just alignment tool and go. It's been a while since I watched any of my own videos (always awkward to see/hear myself present). However, I *think* the SPL difference I mentioned was probably specific to the mains+subs configuration with MSO. It definitely causes problems there. I like to be consistent in my own personal processes, though, so it has become my habit to use the mains or center when aligning the subs to those respective channels.
@@jeffmery Thanks for the quick reply! Also good to know about the minimum post count on AVS. I'm getting ready to align my subs, EQ, then align to my mains before running Audyssey. I'll be using the newly released Audyssey MultEQ-X via Parallels. Have you heard about it? My Denon AVR-X6700H has 2 sub preouts, but my initial research of course lead me to the conclusion that Audyssey could not align dual subs properly, and that it is best to use a miniDSP 2x4HD + REW, then Audyssey MultEQ-EXT32 via the phone app (since it allowed for fine tuning). I had done this months back and was fairly satisfied, but now I'm learning so much more especially with these newer updates to REW & Audyssey. - Assuming you aren't familiar with the new Windows version of Audyssey MultEQ-X, would you still say the above method is better for aligning & EQ'ing dual subs rather than ditching the miniDSP method and relying solely on Audyssey MultEQ-X? *I know I hadn't mentioned MSO, I literally just downloaded it yesterday and have been studying up on how to use it after REW. For now, I'm just going to start with my MLP, then soon I might try integrating MSO for multiple seating positions. I read in one of your other comments somewhere that Audyssey does not account for phase, plus after investing in all this equipment my inclination is to continue using the miniDSP haha. - Would you mind if I document a high level process of what I've done then share it with you via email? I don't want to blow up the YT comments section if I don't have to haha. I also want to be respectful of your time, so no biggie either way! I can imagine providing what must be similar responses to multiple people would lose its luster after a while😆
@@FiletMyYawn I did see the MultEQ-X announcement, but really don't know much about its capabilities. Unfortunately I haven't owned a Denon AVR in a long time. My advice is as accurate as it can be for not having direct hands-on experience ... I would still take care of the subs first within the miniDSP first (either via REW or MSO). The important thing is to align and EQ to flat in the miniDSP. Don't put a house curve there at all. Then run the MultEQ-X process and use it to both fine tune and add your house curve. If you add the house curve prior to running room correction, the room correction software will almost certainly EQ the curve right back out to flat. I want to be clear I'm not picking on Audyssey at all. It can do a good job with 2 subs. However, you've already got the hardware and experience so why not use that to make it sound as good as you possibly can? Audyssey can get the timing pretty good between the subs, but just because the timing is right, it doesn't mean there aren't still issues (like phase cancellation) that it can't account for properly. You can send it over to jeff at bisonhta dot com. There's also a REALLY great thread over on AVS for calibrating multiple subs. It happens to be very Audyssey heavy as well. Check out the first two or three posts in this thread: www.avsforum.com/threads/guide-to-subwoofer-calibration-and-bass-preferences.2958528/ There might be WAY more than you ever wanted to know there, but it's a great resource and you can jump around between sections.
@@jeffmery Really appreciate that feedback! I just aligned and flat EQ'd my subs with REW only for my MLP at this point. Now I'm slightly confused on integrating with the mains. Should I first run Audyssey, adjust the crossover if needed/desired THEN take a measurement of both channels with REW to see if there is a dip? And if there is, do I use the alignment tool and enter the delay into the AVR Sub Distance or the miniDSP for both subs?
@@FiletMyYawn Exactly. Run Audyssey. Set crossovers. Measure to see if there's a dip. If so, alignment too and adjust the delay in the AVR for the sub channel. Don't mess with the delays in the miniDSP now that things are aligned.
Great video. You do a great job explaining things. So, I only saw this video after going through the manual process of adding manual delays, measuring, changing delay, playing with inverting polarity, etc. Then, I saw this video. Question: the alignment tool that you discuss in this video doesn't actually measure the changes, but calculates the impacts of delays based on mathematical modeling? I'm wondering what assumptions the alignment tool might have that might not be correct for certain rooms. For example, my room isn't a perfect triangle, but widens by 3' in the back half of the room.
I just answered your post on AVS, but wanted to respond here too. You are correct in that I didn't actually measure the result after the change. You should *always* do that to make sure something wasn't messed up. 99% of the time, if the result doesn't match the prediction, it's because either the mic was moved between the input and verification measurement or the input measurements were taken without a timing reference. In this instance, while it seems complicated, REW is just adding the two input measurements together while delaying the phase by some amount, say 5 ms. E.G., if the phase is 180° at 20 Hz, the phase is still 180° at 20 Hz, it just happens 5 ms later. The shape of the room doesn't matter. REW doesn't know or care how the room is shaped because it's not doing a prediction. Well it sort of is because it's doing some addition, but not prediction in the way I think you mean. Additionally, the *original* measurements were taken in the same imperfect room. That means the original measurements already take the irregular shape of the room into account. It then follows that the result of adding the traces together *also* accounts for it. This is easy stuff to test and I highly encourage you to play with it! Pretty enlightening in my opinion and a little creepy when the predicted result is so accurate. (Side note: This feature didn't exist when most of the more popular sub alignment videos were made. Those videos are still 100% correct. The alignment tool is just more accurate and faster because it eliminates the guessing and extra measurements.)
Another excellent tutorial. I am trying to figure out starting point of setting sub distance before using rew and room correction; arc in my case. I have set the sub trim in my prepro to zero and set the volume to get 85db from my subs using umik1. Can someone help on setting the sub distance before i start the calibration for sub time alignment using minidsp-hd.
Thanks for the kind words! Unfortunately, I have zero experience with ARC. It's my personal habit to set the sub distance to zero before going through the sub alignment and EQ process. Technically you don't have to do that, but I like to have a baseline where I can easily repeat things from a known starting point. Setting it to zero is just an easy thing to remember. Once the subs are aligned, the room correction software should determine the proper sub distance (delay) as part of the calibration process. After the calibration is done, it's always good to use REW to verify the results for at least the main listening position. The distance may need to be adjusted if there is a dip in the magnitude response at the crossover frequency. This is commonly called the "sub distance tweak". Again, I don't know ARC so I don't know how applicable the sub distance tweak is with that software. For Dirac with DLBC, as an example, the sub distance tweak can't be used. I hope this helps, and again, sorry about my lack of ARC knowledge.
Thx for the video. I tried your method to find the delay. However, after I applied the delay on my miniDsp and re-measured, the resulting curve is no where near the predicted (aligned sum) curve. I don’t think the predicted curve generated by the alignment curve is very reliable ...
The curve is 100% percent reliable and extremely accurate. There are a couple of issues that can lead to the real result not matching the predicted result. First, is the microphone being moved between the initial and the aligned measurements. Even a few inches can result in a drastic difference given that we are working with modal frequencies. Second, is not using a timing reference on the measurements that will be aligned. The simplest way to do this is to specify "acoustic timing reference" next to "Timing:" in the measurement window. Third is taking measurements with settings in place, and then attempting to directly use or apply the values shown in the alignment tool. This is most common when aligning multiple subs that were previously aligned with another method. Settings recommended by the alignment tool are relative to the measurements used for alignment. As an example, if there were delays or EQ in place on the "input" measurements, then alignment results will have to be manually calculated relative to the existing settings. I've done this before on accident and it's absolutely maddening to try and figure it out. When you need to align, or realign subs, you should zero out any gains, delays or EQ that is set anywhere in the signal chain. Then take measurements with an acoustic timing reference to use for alignment.
Thanks for the video, it is possibly the easiest video to understand I've seen to date. I have a curiosity to consult. Have you seen how the waterfall improves in a multisub configuration and also after applying equalization? If you checked it, it would be nice if you could make a video when you could. Do you have a link where you can see your room ?. I say this in case I had put acoustic measurements of it. Written with translator. Greetings from Spain
You're welcome! Yes, I have seen improvements in the waterfall after multiple subs are aligned and equalized. There are two major reasons for this. First, the subs are working to help each other to produce an even response. There is less fighting (cancellation) and more cooperation (reinforcement). This leads to the second reason: Because the subs are working together and producing a higher overall output, most of our EQ filters can now be trims (negative) instead of gains (positive). This helps pull energy out of the peaks, and energy out of the room, which should reduce some of the ringing seen in the waterfall. However, I have also seen the waterfall get worse. This is almost always the result of increasing gain to fix a dip in the frequency response. Small amounts of gain are okay. I generally allow +3 dB on individual PEQ filters in my personal setup. The problem with gain is that it can increase ringing. This ringing will show up in the waterfall as longer decay times around the EQ center frequency. The more gain, the more ringing. The more narrow the EQ filter (also called "high-Q" filters), the greater the ringing as well. It can take time to get the best results, but if you are able, try to use only negative gain (trim) on your EQ filters. If you need to boost some frequencies, try to use the lowest amount of gain with the widest bandwidth (lowest Q) that produces the desired results. This will have the best chance of improving the waterfall without introducing more ringing. As always - measure after making a change to verify the results. Don't get caught up chasing a perfect graph. Listen with your ears to see if you actually like the results as well :-).
Many thanks for the video. That is by far the best explanation I have been able to find. I only have a single sub, but I would like to find out how to use REW, to adjust my sub best with my front speakers. I use antimode dual core between my streamer and preamp, and a dspeaker 8033 for my subwoofer. Have you made a video on the subject, or do you know where I can find one?
I've heard good things about the DSPeaker kit, but unfortunately have zero experience with them. As a result, I've never looked for videos or other guides on them either.
@@jeffmery I did not express myself well enough :) I mean, a video showing how to use REW, to analyze whether the subwoofer is adjusted to fit with the main speakers?
@@franknielsen7597 ahhh. That’s my mistake! Sorry! You want to find a video on the “sub distance tweak”. There are tons out there. A lot are focused on Audyssey-based room correction however, the cause and fix are universal. It’s pretty easy to check your system to see if anything needs to be adjusted as well.
Align this SubGroup to the MAINS ? Is it the same process but just focusing on a narrow range over the crossover freq? (like 80-100 for a 90x-over?) You said that most videos on this topic focus on integrating subs with mains but unfortunately many of them are utilizing outdated software/methods AND drag on for longer than needed in a "diary" format. Thank you for this TO-THE-POINT video. I am running a 2x4HDw/dirac to control a non-avr 2.2 setup.
You're welcome! For a 2.2 system you'll align with your mains (L+R measured and playing together, not separately). The general recommendation for home theater is to align with the center. But even for HT, I advise folks to try both and see if they prefer one over the other. A common problem to have when integrating subs with mains is a dip in the magnitude response (frequency response) at the crossover point when the subs and mains are playing together. This is due to phase cancellation happening at, or near enough to, the crossover frequency. If you do a sweep and see that dip. Use the process below to get the measurements. You'll then align at the exact crossover frequency you're using. It is generally the same process. Sweep L+R with the subs muted. Then sweep L+R with the subs on, but the mains muted (or disconnected). Use those two measurements to align things. Then run a sweep of L+R with everything playing to verify the fix.
@@jeffmery Great. Thank you for confirming my thoughts! I see in the latest REW where you can select the exact frequency that you want to align two measured responses to. I previously just aligned the subs for smoothest freq response & max output, then ran DIRAC to EQ the overall setup. The Sound is great, but will now add aligning the sub group to mains at the crossover freq then re-run DIRAC to get additional improvement/optimization.
This seems like a great tool. Im trying to watch as many videos as possible befor I jump into this, and that might be a mistake :) but before alignment have you not gain/level mathed you subs to play the same level? and do you not link the subs in the minidsp?
Yes, those are things that should be done prior to using the alignment tool. I’m not sure what you mean by “link subs in the miniDSP”. If you mean linking the output channels together, then it’s a “maybe”. It can be a time saver but if you forget it’s on, it can certainly cause some unexpected things (e.g., wanting to change one sub, but it changes both and you forget why). I generally don’t link subs in the miniDSP unless the room tells me that I should do that. One example is a pair of subs at the front of the room. If the seating and subs are placed properly, then linking can make a lot of sense.
Yes I meant linking output channels. That’s what I see others do so wondered why you didn’t. Have you made a video where you explain how to gain/level match the subs prior to alignment? And when you are done with eq and audyssey do you then level match subs with the other speakers? Really like the way you break things down for us not used to all the settings and what seems obvious to those who are used to it
@@michaelbested123456 I don't do that because I've caused myself more problems than I've solved. It's a personal choice only. That feature can save time for sure. I don't have a video on gain or level matching. If your subs are identical, just set the gain knobs on the amps to the same position and done. I think Home Theater Guru's has a decent video on level matching. With Audyssey, EQ to flat, run Audyssey, check all the speaker levels to make sure they're correct, then add your house curve. if you add the house curve before running Audyssey, Audyssey will EQ it out again.
Thanks for the video! Quick question! If I am only measuring 2 subs , what do I do after I get the 1st positive sum ? And once I am happy with the response , how do I go about adding the numbers into my miniDSP ? I just put the delay # into one of my subs in the miniDSP correct ?
@@ants9574 I’m not sure what you mean? With 2 subs, there’s only one alignment to worry about. It’s always “N-1” on the number of alignments needed. 4 subs only needs three alignments (1+2, then 3 with the already combined 1+2, then 4 with the already combined 1+2+3).
@@jeffmery sorry about that. I meant can we put our positive sum alignment response from the 2 subs into both spots in the alignment tool and keep finding tune and trying to get it even more aligned ? Does that make more sense ? Sorry I'm noob to using the alignment tool haha
@@ants9574 I understand now. No, that is not possible. The combined response depends on the alignment of the two subs. If you were to try and align that, the original alignment would change and no longer be valid.
Great video! Maybe a noob question, but could this theoretically work in a car audio setting for time aligning main speakers to listening position, as well as sub to main speakers? Thanks!
I haven't personally used it in a car, but there's no reason you couldn't use it for that purpose. The sub process would be the same as in the video. Aligning speakers would probably be slightly different. I'd probably align them by looking at the differences in impulse response timing. That would be easier. Then align the subs if needed.
I wish you could move the slider .1 at a time with keyboard arrows or something * Found out you can, if you click the slider then use arrows it moves in .01 increments !
Hi, a general question, the Alignment tool can set delays with "allign phase at cursor" doesnt it matter to keep each sub in phase while using the same process as the video? does it only matter to get most positiv summation?
When aligning subs, we typically just care about having the most positive summation. We're aligning subs with one another which we will later present to an AVR/Processor for use. We want that "virtual" sub to be as strong and as even as possible. So we look at the entire frequency range. Typically, we alight the phase slopes or align at the specific frequency when we're integrating speakers that play at different frequencies (subs & mains, or mids and tweeters for people using REW to build speakers). I usually look at both the phase slopes and frequency options when I'm tweaking sub distances to see which yields the smoothest integration.
@@jeffmery thank u, was just wondering because I also watched a video of sub main integration with the tool. And someone said we can use it too to set delay right for the subs. I did it like ur video before. Now I used MSO. For seat to seat consistency. But still unsure if I made mistakes somewhere. So still playing around. If I use the alignment tool with phase at cursor I would need to delay 50ms between front and back subs Which I would need the mini dsp 2x4hd. Instead of my 10x10hd (only15ms there) All subs without delay,eq or change in gain play pretty flat 18-100hz. Well someday I will get it and don’t think about it anymore 😬
@@God_y_u If you use MSO, it will calculate delays for you. There's no reason to use the alignment tool between the subs going that route. You'd only use it to do the "sub distance tweak" to get smooth integration with the mains. Inside of MSO, you can specify the allowable range of delay as well. With just 15ms of allowable range, I would set the range in MSO to +/- 7.5ms for all delay blocks in the config. I hope this helps.
@@jeffmery yeah I have a working MSO. Setup atm. But it does make a lot of cuts in eq to balance everything. So still testing around To see what changes for better or worse. Thanks
Great tut. Q: If i want to eq two seperate listening positions with minidsp Do i align my two subs from each of those listening positions? Then what do i do with those two results?
Thank you! If you care about the response at more than one position, you probably need to look into Multi-Sub Optimizer (MSO). What we're doing with the alignment tool is setting timing between subs. Those are measurements taken at a single position. If we move the mic, the timing will be different at the new position. It is possible to get improvements in the non-main positions with the alignment tool, but it may not be to the degree you want. More advanced solutions such as MSO, Dirac with DLBC, or Multi-Sub EQ (Audyssey for up to 2 subs) are really required for this.
@@jeffmery Thks for the detailed reply For now i think ill pack the minidsp away Its too overwhelming I have to learn REW, minidsp, mso and ARC gen in order to optimise my two subs lol I think ill spend a few months studdying it all then give it a go
@@isaaccraig3666 It's definitely time consuming so I totally understand. IIRC, Anthem were working on a multi-sub optimization solution for ARC. I'm not sure if they have released this yet or what models will support the feature. this might be an easier option to consider (in the same way that DLBC is easier than both REW and MSO). Good luck and let me know if I can help out!
@@jeffmery Thks. For now i calibrated with ARC. It actually did a decent job. I used deep base boost to push a 50hz-70hz null up a few dB Atleast now i can relax and spend time studdying MSO and Minidsp , rew etc Get the knowledge down. Then in the nxt lockdown ill have time to get it right lol Thks again. And look forward to seeing more of your vids . 👌
Great video. I followed your steps but when I clicked on "Aligned Sum" to get the curve listed, I get a curve which is exactky the Aligned Sum but 6db lower all along. Any idea? Thanks
Hmmm. That's an interesting one that I have not seen before. I know with MSO, a 6 dB error can be introduced if the AVR/Processor is configured to send bass to both the main and LFE channels. This is usually called "LFE+Main" or "Double Bass" or something like that. That setting is something I would never, ever recommend the average person use in a home theater environment (or ever TBH). Some people like a ton of bass, but when you want accurate sound, that setting causes more problems than it solves. Double check that setting and make sure it's disabled. If it was turned on, you'll need to remeasure your subs and then try the alignment tool again. Let me know.
@@jeffmery Thanks. I do not have Extra Bass, but I have my front speakers connected to the subs via HF. However here I am measuring my subs response only (LFE channel). I will keep investigating. Amazing and useful video mate!!
@@fofocho71 Sorry I wasn't any help on this one. If you don't mind, please follow up when you find the fix. I'd definitely like to know if I come across someone else with a similar problem in the future.
Great video! Question, if I don’t have a miniDSP is there a way to apply the delays using the SVS app? I’m not sure if phase control is the same thing or not as delays, but the SVS app allows me to control the phase from 0 to 180 degrees, as well as the polarity.
Great question! I think I answered you over on AVSForum, but wanted to post the answer here as well. I'll update my response there to be more complete now that I'm off mobile. I've got an SB-1000 Pro in my office. There doesn't appear to be a way to set a delay using the SVS app. Phase and delay are related and one can be calculated for a given frequency. The question will be "does 180 degrees of phase give us enough adjustment at the frequency we care about?". If we use 80 Hz as our example crossover frequency, one wavelength is 12.5 ms long (1 second/80 cycles = 0.0125 seconds). That means that 180 degrees of phase will give us 6.25 ms of delay (half the wavelength = half the time). Whether or not that will be enough depends on the specifics required for your room. If it's enough, great, if not, you'd need to look at other options. You MIGHT get a crazy result from a delay-to-phase calculator like 738 degrees. That's okay. 720 degrees is 2 full cycles (360 degrees * 2), plus a little. 180 degrees gives us 1/2 a cycle to work with. If we subtract 720 from 738, we get an 18 degree difference which is well within the limits of the SVS controls. The most common option is the "sub distance tweak". You adjust the distance listed for your sub in your AVR or Processor to add or remove delay which gets the phase lined up at the crossover frequency. I hope this helps!
@@jeffmery Thanks! Very helpful between your post of the forum and this reply! I will test it out later today and report back on the forum. In terms of which frequency to enter into the calculator, I'm assuming that since I have a pretty significant null at around 63Hz (you can see my REW chart in the AVSForum), I would enter both the 63Hz into the calculator, and the ms delay that the Rew alignment tool gives me to figure out the phase?
@@BenjiGator23 I just took a look at your graph. There are 2 nulls at ~63 Hz and 105 Hz. When we use the alignment tool or adjust delay/phase, we're typically targeting a specific type of dip that is centered on the crossover frequency. What crossover are you using? It's possible you will see changes, either positive or negative in both dips by adjusting delay/phase, but probably not the amount of correction you're looking for. Another possibility is that you're hitting a room mode at 63 Hz. I may have missed it in the thread - What are your room's dimensions? Where, exactly, are you measuring the response with REW?
@@jeffmery Front Left and Front Right speakers are currently crossed over at 80Hz; Center channel at 90Hz. Here is a picture of my set from my original post in the forum (bit.ly/3jxxYKT). The main listening position (MLP) is approximately 6.5 feet from the front of the center TV console. Side wall to side wall is about 11 feet wide. As you can see in the picture, to the left of the left subwoofer is a stairway, and the there isn't really a left wall since it opens up into the the dining room. The right wall is a closed wall that goes to the rear wall about 21' back. Subwoofer cone to subwoofer cone is about 8 feet apart. The REW response was measured from the MLP about 6.5 feet away from the center of the TV, and the MLP is about about 8 feet from each sub equidistant.
@@jeffmery can you link to this discussion in avsforum, I was also interested in trying to use phase control on the SVS app, but instead to use it to align a 2nd sub with 1st sub where it wouldnt be a particular frequency I would be targeting but instead the whole frequency response.
DO you EQ each sub individually? then same with speakers? BTW thanks for this method. Nobody on youtube seems to get this. Worked great for me. My car was wanting like 18ft neg delay, but the slider showed me almost exact same curve at like 26" of delay
If you're using the REW EQ method, then all subs receive the same EQ. Technically, you EQ the single "virtual" sub then just apply that single set of PEQs to all of the individual subs on the output side of your DSP. You would then run your room correction software or manually level, align, and EQ your speakers (if that's your thing). I'm glad the tutorial was helpful for you!
@@jeffmery ah that makes sense, since all the subs are considered "one" sub. Individually they might be all over the place, but you only care about what is hitting the listener
Use L+R. Sweep that with the subs muted. Then sweep it with the mains disconnected. Use those two measurements for the alignment tool. I also recommend comparing the sound of aligning the subs with L+R against aligning with C. The general "rule" is L+R for music, and C for home theater. I've done both and have my preference for my use-case. I could see people favoring one over the other so I do think it's worth the effort.
There's a slider you can play with on the speaker itself in REW. Sliding to the right will add positive delay. I believe you can just click the slider and then use the arrow keys to tap and move the slider as well.
Hey, I'm going to try this out for the first time and I would love to know something before I start. With those basic measurements in the beginning, did you unhook/disable your mains to align the subs with each other? I'm pretty new to all this and the only vids I've seen so far are those from Home Theater Gamer and he did those measurements with the left speaker on. Thanks! :) PS: it's for a 5.1 system, but I also use it for music and other stereo stuff. I saw that you said that aligning it with the center is the best way to go for home theater, but in my case, it might be best to go L+R.
For aligning multiple subs with one another, the LFE channel can be swept independently of the other channels. It should be channel 4 on Windows and automatically labelled as LFE on MacOS. There's no need to sweep the other channels just to align multiple subs with one another. You will always need to use an acoustic timing reference, though, for the alignment tool to work properly. To do the "sub distance tweak" and align the combined subs with the center -or- L+R, then yes, you will sweep one of the channels with the mains disconnected, then with the subs disconnected and align those two curves. Negative values for alignment on the subs here are almost always okay. That just means *reduce* the distance of the sub in your AVR by the indicated amount. I'd suggest experimenting with both the center and L+R alignment. Center for HT and L+R for music are good rules of thumb, but I could see people preferring one over the other.
When taking raw meassurements, isn't it better to use a 3 point average from the MLP? I find that even a slight change in mic can drastically change the graph. What I mean is: take 3 meassurements in MLP for each sub. Average that result and use that result to tweak further. Is that a good approach?
This will not work. Averaging measurements removes the phase information from the sweep. It's the phase information that is being used to align things. And yes, a slight change in mic position can sometimes drastically change things. Phase is a function of distance which is really just a proxy for time. Without using more advanced techniques or solutions, this is something we have to live with. More "advanced" techniques or solutions would include things like implementing all-pass filter networks on the subs and mains, MSO, or DLBC.
@@jeffmery I found that I also can't use averages to create waterfall graphs. Makes sense but to bad. Need to be really carefull with mic placement. Thanks!
@@johandeen6096 I should also add that it's not like averages are useless. They can be very useful when you're doing more fine-grained work like building a target curve for your room. Do some research on the "moving microphone method" of measurements. Results are very similar to averaging but they're much more repeatable from session to session.
@@jeffmery Hmm, never thought of that. That seems like a good way to get an average you can use for other things. Will def. try that out! Quick question regarding this video, how did you get the raw meassurments of your subs? Just use an REW sweep with the sub cal on the subwoofer out (and using minidsp to switch between subs)?
The rule of thumb is align the sub(s) with the Center for home theater and the L+R for music. It's specifically L+R to indicate aligning with the combined output of both speakers swept at the same time. "Stereo Monitors" implies a near-field use-case which is essentially the same as the L+R alignment for music. This is exactly my config on my office 2.1 system.
Yes, but you would need to treat that as 2 subs, not 4. When we stack subs, the subs are so close they behave as a single sub in the room. From a technical perspective, subs "couple" at a frequency with a wavelength equal to 4x the distance between the drivers (paraphrasing to keep it simple). So if I stack a pair of 15" subs, and the drivers are, say 20" apart, center to center, that gives a wavelength of 80". The subs will couple and act as a single sub around 170 Hz or so and below. Use a Y-splitter and connect the L subs to the same channel and the R subs to the same channel. They use the alignment tool between the result 2 "meta" subs.
May I ask - the "Room Sim" feature.... when you 'simulate' your room with 4 subs in the positions you have - how close is the simulated response versus the actual response you get? I'm playing with the sim at the moment to work out where to add a secondary sub - and it's suggesting back right corner to couple with my big dual 18" sub on the front wall between my L and C channels. But I'd like to know how accurate that room sim feature is.
If your room is a rectangle, then the room sim is quite accurate. The less rectangular the room is, the less accurate the sim is. Rooms not fully enclosed, with angled walls/ceilings, L-shaped rooms, etc all make the predictions less reliable; even more so when those features are combined. My room is fully enclosed, but sort of an L shape comprised of other rectangles. In some dimensions (length and height), the sim is in the neighborhood. In the width, it's not so much.
@@jeffmery Thank you for that. My room is rectangle and fairly well sealed, double drywall/gyprock, good insulation, on a slab, solid door etc. The only 'anomaly' is that we have a rear riser of about 30cm/1ft in the back quarter of the room. The room sim graphs do tend to reflect what we hear in our room, but I have not done adequate tests yet to fully confirm. Thanks for the reply :)
In that case it should very accurate indeed! A riser like that will *technically* have an impact, but it's highly unlikely to be anything you can hear.
IIRC it's in 5.20 or newer. 5.20 was in beta for so long and so many features added, it's hard to keep track of for sure. Luckily 5.20 has been released and is out of beta. It's definitely a safe upgrade to make.
Can we use this process to align subs to mains or center channel for 80hz ? In REW under alignment tool... Let's say we import the combined subs after they are corrected in minidsp, then import center channel without subs. Now can we look at the 80hz (my selected crossover in AVR) and adjust delay slider for delay on subs to get flatest horizontal line at crossover without changing too much of rest of range? or would you use REW measurement page, measure subs, with center channel as timing ref channels, then see what rew says under sub measurement (right column in rew of measurements) what to add or remove for delay.. Thrn enter thst correction INTO AVR distance setting for sub?
Absolutely! I think the alignment tool the most fool-proof way to do it. This is most commonly referred to as the "sub distance tweak". The process is actually easier than you describe above. You don't have to use the sliders at all. Measure and import as you describe. Open the alignment tool. Make sure the Center is selected in the top box and the sub in the bottom. In the frequency response, click a point on the graph at your crossover frequency. In the alignment tool window select "Align phase slopes at cursor" and observe the change. Then select "Align phase at cursor". Compare this against the graph for the previous change. use the delay setting for whichever curve looks better. You can get a negative value here and that's okay. It just means reduce the delay in the AVR by the amount indicated. You might have to do some math to convert ms time to distance. Divide the time by 0.89 to convert the delay from ms to feet. Of course - it's totally okay to play with the sliders too, but I find the point-click method easier. Use it to at least get you in the right area and then potentially fine tune with sliding a bit.
@@jeffmery Cool and thanks..I know In past I tried one of those options but got weird values.. like 30 foot delay needed. when I just used values from REW measurements (using again center as ref timing.) (value in measurement box on left side of rew where measurements are) it soinded correct and looked way better... was like maybe 12 feet adjustment not 30ish.. What are we looking for exactly when aligning with main speakers crossover? are we trying to get a flat horizontal line over the crossover (let's say 80hz)? Or are we trying to get a positive summation overall from 80hz down to 20hz like you do with aligning subs? Thanks and sorry for silly question. I am also confused what we are trying to see when aligning crossover..
@@desertplanet3320 No question is silly. We all started somewhere. When aligning phase at the crossover point, we are looking for a smooth frequency response. Without getting too technical, there are very few room correction packages that take speaker phase into account. The two I'm aware of are MSO and Dirac Live with the DLBC add-on. I guarantee there are others out there; especially in the commercial or professional audio world. I'm just familiar with what I've personally worked with. The most common room correction packages like YPAO, Audyssey, etc definitely do not take phase into account. If we DON'T take phase into account, we can end up with frequency cancellation at the crossover. This will show up as a dip in the frequency response centered near the crossover frequency. The exact center depends on the specific phase misalignment which is why we look at the "crossover region" and not the exact point (there's more to it than that, but that's good enough for now). By adjusting the timing of the sub, we can line up the phase slopes to eliminate this cancellation and improve the sound. Human hearing is much more sensitive to frequency response than delay at these frequencies. The wavelengths are huge (12.5 ms long at 80 Hz). Conversely, as the frequency rises, we become very sensitive to delays, especially in the speech range. Does this help?
My subs were identical when this video was made so I gain matched them by setting their control to the same level. That works best when the subs are the same because we know they’re working equally hard. If the subs are different I’d start by level matching them at the MLP. There’s a really good method for mixed subs called “distortion matching”. There is a post by user “DaveBoswell” on the MSO thread at AVSForum that details the process. It’s more complicated than straight level matching. However, it helps to ensure that one sub isn’t completely overdriven if they have very different capabilities.
@@jeffmeryYes, my subs are identical, HSU vft2-mk5. Thanks,still in the early stages of how this whole process works, received my mini dsp/umik-1 mic 3 weeks ago and trying to set these up.
@@kenestra123 Gain and delay from the alignment need to be input manually to the correct output channels on the miniDSP. If you took the next step and EQ'd the combined response, then yes, the EQ export needs to be imported and loaded on the output side of the miniDSP.
Buddy please little help i have one sub and i cant find the best listening position i run audyssey and gives me bad sound some frequencys are low down theres any way to find best best position and take best results? With some how..🙏🙏🙏🙏🤔
I highly recommend what is called a "sub crawl". You really need a calibrated microphone and REW to get the best results. Folks will tell you that it can be done by ear, but it can't. Just trust me on that. What you will do is setup the mic in your main listening position. Measure the sub in its current position. Now, move the sub around the room to every position where you could place it, measure it in each spot as you go. You're looking for the best response you can get. Even with this work, a single sub will always have nulls (large dips in the response) that will be difficult to address. The alignment tool can help eliminate a dip that happens at the crossover frequency after running Audyssey. It's very common for Audyssey to get the distance for the sub incorrect. When the distance is wrong, there is cancellation. It just takes adjusting the distance of the sub to eliminate it. Search for "Sub Distance Tweak" to see if that helps. For a very comprehensive dive into subs with Audyssey, I highly recommend the first few posts in this thread over on AVSForum. Mike did a great job and it is very heavy on Audyssey: www.avsforum.com/threads/guide-to-subwoofer-calibration-and-bass-preferences.2958528/
If you care about more than one listening position, MSO is likely the better solution. It is possible to use spatially averaged measurements from multiple positions within REW. This would require multiple microphones to be setup as well as the Pro license for REW to allow multi-input capture. I suppose trace arithmetic could be used to get a spatially averaged measurement, but it would be very tedious. MSO would be less effort.
@@jeffmery will the mso do time align or do more a eq fix? (I think it is doing some sort of time align but im not 100% if it is any good, and ifi need to do the time alignment in rew first..)
@@niklaskarlsson236 MSO will also do time alignment. There is no need to do that in REW first. You will need to take measurements in REW to use as input into MSO, but that is it. Just measurements.
Hi Jeff - just wanted to ask - do you need to use an acoustic timing reference (e.g.left main) whilst doing measurements for the alignment tool to work...? Thank you.. best wishes..
@@jeffmery thank you Jeff. Also, I wanted to ask you, which if these steps would you recommend for sub eq: 1. Sub alignment as above, followed by individual sub eq (flat target) in output section of miniDsp for each sub and a house curve as a low shelving filter in the input section of miniDsp... 2. Individual sub eq (flat target) followed by sub alignment as above and house curve as low shelving filter in input section of miniDsp... 3. Any other combination of sequences.. 4. Or am I completely missing the point and barking up the wrong tree..... Thank you very much for your time and diligence....Harsh
@@harshjain3801 I recommend your #2 approach. Be sure to EQ the combined response, and apply the same PEQ filters to all subs. Do NOT attempt to EQ each sub individually and apply differing EQ; you will have a bad time ;-). I like to keep the house curve separate from the sub alignment and EQ as you have described above. That can be done on an input filter in the miniDSP or within your AVR's room correction software if it's supported. I've used both methods. I currently use the room correction method because it's just more convenient for me.
@@jeffmery Jeff - thank you really for your prompt and detailed reply. But then, I have this query - each sub has its own frequency response curve - would it not actually be better to eq them individually - according to its own peaks and dips, rather than combined sub eq applied to all subs - because some of the resulting filters might compete with the frequencies if individual subs....hope I am making some sense here...🙁
@@harshjain3801 Great question. This is probably one of the biggest points of confusion with subs. To put it simply, because of the frequencies they produce, subs don't behave like regular speakers. What we hear with subs is more the influence of the room and not the speaker itself. If you have identical subs in two different locations, then you should see this in their respective in-room responses. The problem with this, is that because of the differing responses, the subs will end up fighting one another, rather than helping one another when they play the same content. When we align subs to play together, we essentially create a single "virtual" sub somewhere in the room. It's this "virtual" sub (all of the aligned subs playing together) that we want to equalize and then integrate into our systems. This is something that's easy to experiment with. Align your subs and EQ the "virtual"' sub. Measure the response. Align your subs and EQ each one individually. Measure the response. You could even go further and EQ each sub individually, then align and measure. The first process will result in the best overall response. This might make a good video topic.
I *think* you misspoke regarding the +10dB gain. On the logarithmic scale, every 10dB doubles (2x) the volume or SPL, not 10x as you stated. I'm no expert and it's possible you are referring to 'gain' as something other than perceived SPL
@@jeffmery you did say power after I watched again, my bad. Either way, a 10dB increase is immense difference. Do you have experience with REW for 2channel listening setups? I suppose my setup is 2.2 -- dual mono subs next to each LR speaker. At some point i want to get a decent mic and do room measurements with software like REW, but I suppose without something like minidsp I'm regulated to make corrections or improvements via speaker placement or room treatment only. Still. I have hope that these can rack up significant gains with the right measuring equipment and application.
@@erics.4113 I haven’t specifically done a 2.2, but the physics of multiple subs and sub+mains integration are the same. The most important thing to be aware of is that the best location for you mains is almost never the best location for your subs as well. The advantage of separate subs is that we can place them in the room where they give the best overall response. That doesn’t matter if it’s 1 or many subs. Don’t be afraid to experiment with their locations. Human hearing cannot localize frequencies below 80 Hz so you’re not going to hear the bass coming from a different part of the room. In my experience, more subs can push the frequency a bit higher as well. If your mains are capable, then you should have plenty of freedom to place the subs wherever you want. If you need to cross the mains over above, say 100 Hz, you might be able to start localizing the subs. In this case, you might be forced to have the subs near the mains. Just something to be aware of. You are absolutely correct about running a mono signal (make sure it’s the same signal) to each sub. Going back to our inability to localize low frequencies, the concept of “stereo” doesn’t really exist. Running a mono signal eliminates some types of cancellation issues that we would not otherwise be able to fix very easily. I cannot recommend a good mic and REW enough. Even without a miniDSP, it’s possible to make some improvements. Probably the most common issue to fix with just measurements is the cancellation that occurs at the crossover frequency in most setups. The issue is caused by phase cancellation which boils down to a timing problem between the sub(s) and mains. You can measure, then use the alignment tool to find out the change in the sub distance (delay) to fix the problem. Instead of aligning one sub with another, you measure your L+R together, then align the sub to that measurement at the crossover frequency.
@@jeffmery I’m into car audio enthusiasts and I want to learn REW can you able to share me some videos how to set the REW, measure the system and eq the system.
@@sainath8710 Setting up REW to take measurements is exactly the same. I have a series on getting started with REW on MacOS here on the channel. Unfortunately I'm not a car audio guy so I'm not up to date on all the things available in that world. The concepts should all be the same though.
@@jeffmery what you said was right the procedure are same, only thing is in car audio we do time alignment for all speakers because we are sitting offset not in the center and the environment is different.
I don't mean this to be a crappy answer, but it really depends on the room and the number of seats you care about. You can get good results with a single sub if you only care about a single position. You could place it near-field right behind the seat and get just about a perfect response (up to the capabilities of the sub in question of course). When you put subs somewhere else, the room creates the peaks and nulls you see in the responses above. The peaks can be knocked down with simple EQ, but physics won't let us EQ a null. For nulls, the only choice is to move the subs and seating or to add more subs to help even the response.
Yes it does, but some folks don't need to use MSO to get great results. The rule of thumb is ... If you care about a single listening position, you'll get excellent results, in less time, with less effort using the REW-only method of aligning and EQ-ing subs. This video is the "alignment" part of that process. If you care about more than one listening position, like one or multiple rows of seating, then MSO will produce better results.
Did you start from a known state (i.e. everything zeroed out)? Did you use an acoustic timing reference when taking your “to-be-aligned” measurements? Did you move the mic between taking the source and aligned measurements?
Neither. I'm on Mac OS. My understanding is that Java has become the way to go on Windows. I have a Windows laptop around here somewhere. I really need to play with REW a bit more there. Sorry I'm not of more use!
I’ve experimented with order hand haven’t found much difference in the results. Just know that if you start with a random pair of subs, the best looking result requires negative delay on the bottom sub, that’s the same as adding the exact same amount of positive delay to the sub on top.
@@kystarnes8486 On the "negative gain" no videos that i'm aware of. "Top" and "bottom" sub above refer to the alignment tool window in REW. You are aligning the sub in the bottom window, say sub "B", with the sub in the top window, sub "A" by adjust the delay for sub B If I get the best result with positive delay on sub B, great. Implement it and move on. If I get the best result with negative delay, what do I do? Let's say -2.22ms of delay on sub B is best. That means that sub B must arrive 2.22ms before sub A. We can achieve the same result by implementing 2.22ms of delay on sub A. Does that help?
90 dB at 15 Hz isn't very much. There are a ton of commercially sold, home theater subs that can do that. Have a look at data-bass.com/#/systems?_k=3k5xnu Ignore anything with "DIY" in the "Manufacturer" column. Those are home built subs and you mentioned "sold for HT" so we'll keep the discussion to commercially sold brands only. Much of that data is now a few years old so it's missing data on newer, improved models (like the SVS PB-16 Ultra I mention below). The person doing the independent testing has moved on to other things. However, you'll see plenty of well known home theater subs in there like HSU, Rythmik, PSA, SVS, JL Audio, etc. The SVS PB-13 Ultra, which is several years old and has been replaced by the much larger and much louder PB-16 Ultra, hit 89.3 at 12.5 Hz and 105.1 at 16 Hz. For reference, my current subs are Rythmik FVX15 subs. They have not been tested by the site above, but their big brother, the F15HP has. The F15HP hits 98.5 dB at 12.5 Hz and 104 dB at 16 Hz in single port mode. Per Rythmik, the FVX15 is down 3 dB across the bandwidth from the F15HP. I could theoretically hit 95 dB at 12.5 Hz and 103 at 16 Hz. I do not have my gain knobs on my amps turned all the way up so there's definitely head room to be had.
@@jeffmery thanks for your response and willingness to help What equipment or devices you need and what software you was using? After setting the above items up reading the software and what to look for is when running test on the subwoofer in reading the resultsso confusing
@@alambe7275 There are a few things you'll need. To take measurements, you need a good microphone, a laptop, and measurement software. I recommend the miniDSP UMIK-1 for the microphone. It's reasonably priced and available just about everywhere in the world. For the measurement software, I recommend Room EQ Wizard (a.k.a. REW). It's freeware and very very capable. I also highly recommend a boom-style mic stand to hold the mic while taking measurements. The UMIK-1 is about $100. A cheap boom-style mic stand is about $30. REW is free, but consider making a donation to support the person that develops and maintains it. Depending on your equipment it might be possible to implement some things, such as delay, with what you have. If not, or if you want more capabilities, I highly recommend the miniDSP 2x4 HD. This device plugs into the subwoofer or LFE output of your AVR, then your subwoofers plug into the miniDSP (so it sits between the AVR and the subs). The 2x4HD gives almost total control over the subs and allows for very fine grained bass management. The miniDSP 2x4HD costs about $250 shipped directly from miniDSP.
How many subs? How are they connected? Where are you setting the delay? The hard part is that the changes are relative to the measurement conditions. An example might make it easier ... Say you're using an AVR that has independent delay settings for each sub. Let's say those delay values are 0 ms and 4 ms for sub 1 and 2. If you measure like this (meaning with the delays active), the change for sub 2 is relative to the 4 ms of delay that was present when the initial measurement was taken. In that example, if the alignment tool shows 1.7 ms for sub 2, that 1.7 ms needs to be ADDED to the existing 4 ms. The total delay becomes 5.7 ms for sub 2. I hope that all makes sense.
@@jeffmery Nah I always set AVR pure straight before any measurements so its raw UN-molested. Always in the Minidsp output stage delay. 6 subs two cabs haha AVR-MiniDSP to External AMP each individually powered. sad because I was looking forward to that 5db gain. measured all three and then as one for baseline. After "delay" i had worse response then overall then without it. After delay i sweeped all three as one and was worse then before i added delay. Makes no sense haha. Ill try again tonight
You're right that time alignment isn't always needed. If the subs and seating are properly placed in the room, then there can be a naturally even response. However, that does require the subs to be physically separated from one another to reduce modal interactions in the space. Line arraying subs doesn't work in acoustically small spaces (less than ~1,000 cu. m. or 35,000 cu. ft.). Here are some good resources that explain the issue and recommendations to address. Welti/Devantier: audioroundtable.com/misc/Welti_Multisub.pdf Geddes: www.gedlee.com/Papers/OptimalBassPlaybackinSmallRooms.pptx CEDIA CTA-RP22 (Section 7 specifically): cedia.net/advocacy/rp22 (Last one behind a soft-wall requiring an e-mail address. Sorry about that.)
I just watch this video and my head is still spinning. th-cam.com/video/DofUiHz7Fu4/w-d-xo.html In a stereo system with 3 subs how would you phase align the subs to the main speakers?
There are different schools of thought on sub alignment to main speakers. Part of the challenge in using impulse responses is that they are biased to the arrival time of the higher frequencies. REW typically sweeps from low to high so a sub can show an odd impulse response making it difficult to align. This can some times lead to incorrect alignment and poor sound. This is what OCA is discussing when he talks about "cross correlation sometimes gets it wrong". The upside of this method is that typically the arrival time of peak energy is pretty well aligned. If this were me, and I only cared about a single seating position (as OCA does), I'd start with the alignment tool. I'd look at the combined subs and the L+R speaker together and see how I can get the best summation. I tend to favor as little delay as possible to get things aligned. Experiment with inverting the subs and delay to find the lowest amount of sub delay to give a good integration at the crossover.
One of my patrons recommended this video to other patrons during our Zoom calls tonight. Excellent job. Very easy to follow and understand.
Thanks for the comment! It's easily the most viewed video I have.
We crossed paths at the Atlanta Home Theater event a few months back, but were both so busy with things I couldn't find time to chat with you. I'm sure we'll meet at a similar event in the future. 🙂
Hey Ive watched your video 100 times but when i input the delay #s into MiniDSP it is WAY off the predicted curve when i run a measurement again. Any idea?
@@kringle924 The most common reason for this is that the measurements were taken without a timing reference. For USB microphones like the miniDSP UMIK, you need to tell REW to use an "acoustic timing reference" when the measurements are taken. This is on the measurement screen in REW. You will hear a chirp from the L speaker, then the sweep of the sub, then another chirp. Always use the same reference channel for all subs. It defaults to L which is fine. Once this is enabled, it should stay enabled which is a good thing. Always use a timing reference for every measurement you take.
The second most common reason is that the delay values were not set to 0 ms on the output side of the miniDSP. You don't need to change the delay in the AVR / Processor.
Okay THIS video is probably the single most important one regarding REW! Delays are how you align you subs so that they work together well and holy shit I've learned that I've been going about this the wrong way.
I've been following Steven Smiths videos on Home Theater Gurus and happened to come across your videos. I'll now be going through all your stuff, cause this video specifically helped my wrap my head around getting everything aligned using the "Alignment Tool". Using the slider and getting a visual representation of "whats going on" made a world of difference. Thank you so much !!
Steve’s videos are great! They’re how started out myself. The Alignment Tool feature was added after he did those so I thought I’d try to give back a little 😊
@@jeffmery I mentioned you and gave you credit on his Facebook page Home Theater Gurus. Thanks again !!
Thanks so much for this fantastic tutorial. I'm wanting to get into REW this year and I'm doing as much homework as possible but I find so many people skip steps that to them are very simple but for me as someone who's not as savvy on a computer as others can leave me sort of stranded.. Really appreciate your effort to walk us through step by step. Really looking forward to seeing your videos on how to apply EQ through mini dsp and any other info you care to share. Cheers Mark.
Jeff, great video and explanation of how to use the alignment tool. Please keep up the great work and carry on with vidoes for next steps on EQ and alignment to mains etc. There is a lot of misinformation out there. Yours is one of the best I have seen. Well done sir!
Thank you for the kind feedback! More videos are definitely in the works. I hope to get some posted this weekend.
Actually, I think that the next step before applying any EQ would be to capture an actual measurement. Would like to have seen a comparison between the predicted curve and the measured curve, but otherwise, a great and very, very helpful tutorial, thanks.
I completely agree. I didn't really plan on doing this video so I just winged it.
ALWAYS take before measurements so you have a good baseline for comparison.
ALWAYS take after measurements to verify things were implemented correctly, and that the results match the prediction. If the mic has not been touched between the before measurement and the verification measurement, I find the REW prediction to be extremely accurate.
@@jeffmery what i dont get is, now with this prediction curve, it will measure over 95db at lowest frequency... i think those sub can't handle it.. so a low pass filter is a u must hen?
@@Arvidje Maybe or maybe not. What you are seeing is the benefit of having phase better aligned. Phase alignment between two (or more) subs is free output. When phase is perfectly aligned, we get up to 6 dB of truly free output - meaning increased SPL that doesn't cost us any increased power. However, when phase is perfectly mis-aligned, we can have up to 100% cancellation - meaning that even with unlimited power, we can't fix the problem!
Now, it's possible that with time alignment, your combined subs are now much louder than they should be. Normally, the LFE channel is +10 dB from the non-LFE channels per the specification. If it ends up super loud after alignment and EQ, then it just needs to be turned down a bit. Room correction or the controls built into your AVR are the right tools for the job.
(I think you meant high-pass filter to protect the subs below their port tuning frequency). You might need a high-pass filter if your subs don't have one. Commercial subs almost always have one built into their amplifier, but it's best to be 100% sure and check with the manufacturer. For DIY ported subs, you will need some kind of high pass filter set at or near the tuning frequency of the port. The type and slope depends on the enclosure. Assemble-yourself boxes from the likes of GSG will tell you exactly what type of filter and the frequency. You'll need to make sure you implement this as required to avoid damaging the subs.
@@jeffmery thanks for your explanation. I understand it more. But I have one more question about that low frequence. For example if u have two different subs, one sub maxed out at 30hz, the other 18hz.. how does this play out after eq? Is it best to do a high or low pass in the output of that weakest sub to play safe or doesn't it matter, it doesn't try go lower?
@@Arvidje If you have different subs then you're most likely going to be limited by the less capable sub. It would be safest to high-pass the sub that plays to 18 Hz and match it to the other sub as best as possible. Do that before EQ. Then time-align the subs using the alignment tool. Then finally EQ the combined response.
Really would have liked to see that EQ’d to a curve and flat. That is amazing response.
Have a look at this video to see results of EQ in REW. It's based on a different set of measurements, but the result is the same.
th-cam.com/video/ARztXSmoQbE/w-d-xo.html
Hey Jeff, I had just set up my Mini-DSP over the weekend and just watching/reading through all the other tutorials was daunting. This one was much more easier to consume so I'll give it a whirl this week.
Thanks for the feedback and let me know if you have any questions. It's definitely hard to find a balance between quick and consumable, while still giving enough information for people to understand what we're doing and why. Though .... some folks probably don't care about that at all.
@@jeffmery Would you recommend setting the subs distance for the 4 subs before or after Dirac calibration?
@@michaelpiper6601 If you mean the sub distance in the AVR/Processor, Dirac will calculate those distances during the calibration. In my experience, the subs are usually (but not always) set to zero and then everything else is delayed from them. This is due to the lag most subs have due to their own internal DSP in their amp (i.e. crossover, delay, phase adjustments on the actual sub).
If you mean aligning the 4 subs behind a miniDSP before presenting as a single sub to Dirac, then you should fully align and tune them in the miniDSP first. This means setting all delays, gains, and PEQs first in the miniDSP. Then measure to verify the results before running Dirac. Dirac will see a single sub instead of 4, but it will still set its own distance and gains as above.
A further complication is DLBC. If you are not licensed for DLBC, then you should probably verify the integration with the center or mains and adjust the distance accordingly after running Dirac.
If you are licensed for DLBC, it really takes care of all of that for you (at least in my experience and my theater).
I hope this helps, but let me know if it creates more questions for you!
Thanks for such an informative video! I’ve been playing around with rew the last few days and using the alignment, time and SPL features I’ve increased my output significantly across the frequency spectrum with two subs. Upon eq and uploading to my MiniDSP I have a flat response with a small -3db dip at 53hz. Keep up the good work and keep the videos coming!😁 I’m subscribing now.
Glad it was helpful!
This is the best video I've seen on this topic. Thank you.
Three adittional tipps here:
-You should also conciser trying negative delays. It might be that the woofer you are trying to align is already too late so you cant fix it with adding delay. You will instead need to "subtract delay" (which actually means adding delay to all the woofers from the previous measurements)
-You can klick the slider and then move it with the arrow keys on your keyboard, which makes small adjustments way more convenient.
-After aligning all woofers into one "virtual" woofer and integrating it with the mains, you should definitely check out the Group delay plot i REW to see if your subs are playing "in time" with your mains. In most cases they will be A LOT too late, even if the frequency response looks good in the crossover region.
All good points. I was trying to keep things as simple as possible for folks. I would also consider group delay a much more advanced topic than simple alignment. It's also not something that's easy to correct for the vast majority of people or the equipment they have. I know Trinnov's Optimizer will correct for group delay, but I'm not sure even Dirac with multi-sub DLBC will. I think I have an open weekend coming up so I might be able to experiment with this on my system.
The upside is that at these frequencies, human hearing is much more sensitive to the magnitude response than the timing. It's why we can get away with sometimes significant delays and not notice. The best analogy I've seen is that bass frequencies are like a train. When does the signal arrive? When the first car crosses the road? The last? somewhere in the middle?
This ^^ is more for less experienced folks that might read this, Ferdinand, and not directed at you in particular. 99.9% of folks out there have never heard of group delay :-). Thank you for your comment!
@@jeffmery I fall in this category since I'm a REW Beginner. Your videos are helping me a lot understanding and learning different REW methods. Thank you!
Great job buddy 👍 Jeff (AKA Fattire @AVS)
I was going suggest making a video about the Alignment Tool but it seem your all over it 😀
I think this will help a ton of people out as video tutorials make the process seem so much easier. I can see myself linking this video a lot 👍
I found it strange that you didn't need any inversion for the back subs, this is the same when aligning my two sub 1 front 1 back no inversion needed and little to no delay.
I would for future videos turn off the the dark mode in REW, it comes on as default for newer versions and I find myself that it is harder to see the different sub FR's
Also I found when fine tuning the delay I click the button once that moves the delay and then use the right and left on the keyboard to visually get what I want.
Again Great Job
Jim
Great feedback. Thank you! I call them "rear" subs for simplicity, but there not really behind the MLP. Our listening area is a large 3-4 person couch. The "rear" subs are on the left and right sides, directly in-line with the long axis of the couch. My room is fully enclosed, but it's an odd shape, and is split use (2/3 theater and 1/3 home office). The office part is at the rear and limits my sub placement options.
Thank you! 4 subs means 4 times the power and 4 times the cone area. 1sub 90dB will redult in 2 subs 96dB and 4 subs 102dB then. 12dB extra in a perfect world. You could have added the 4 without any delay and compare to your optimzed version to show the difference between simply 4 and 4 optimized with time alignment aswell.
Headroom is a wonderful thing isn't it?! ;-). That's an excellent point on a before/after comparison. However, here's how the sausage is actually made in my room with my gear ...
Using the REW methods, I get the best results in my room *without any* individual sub delay. Call me lucky or spoiled or whatever. Adding any delay to any of my individual subs results in a worse combined response. A lot of it is due to placement of the subs and listening position in my multi-purpose room. So had I done a before/after with actual results, it would have looked worse! LOL! That is absolutely not the case for 99.9% of the people that need to do this. I wanted to be sure to show reality as opposed to the unusual case of my specific room. It was also not the case when I initially started out, so I know the process works very well and others would benefit from learning it.
That being said, REW didn't give the best total/overall response for all listening positions. That was achieved with MSO resulting in a basically perfect flat response (before house curve) from 15 - 150 Hz. The MSO method, however, *did* apply different delays to the individual subs. Obviously that's a totally different method from pure REW. I thought it important to call out that detail and be fully transparent.
Amazing video explained plainly and clearly. Thank you
Awesome video. Always a pleasure to see a demo of a system from someone that clearly understands it.
I haven't gotten into REW yet but I appreciate the primer even though this was a more advanced video I'm a genius so... 😉
Great job! I found using arrow keys vs. mouse was easier to see small delay adjustments.
Great video! +10 dB is usually perceived as a doubling of loudness, which is still really impressive considering you're just time aligning the subs. It beats doubling your wattage to get a 3 db increase
Thanks for the kind words! Free head room is definitely the best head room 🙂!
@@jeffmery Hi, Denon AVR user here. Just to confirm. You sweep your subs using LFE channel at 85dB each sub(30-40dB above noise floor method) then if the subs combined response is to loud when level matching you can use the gain controls in the miniDSP to lower the individual and combined subs volume. Our goal here is to get the most headroom possible right?
@@martinmares8998 I wouldn't adjust the levels of the subs until after they are aligned and EQ'd together. Each room correction tool is slightly different so it's best to adjust the levels as part of the room correction process. Sometimes that just means turning the subs down within the correction software. Sometimes the gain on the subs needs to be adjusted.
If you need to adjust the gain on the subs, they all need to be adjusted identically either up or down as required by your room correction software.
@@jeffmery I will follow these steps when I time Align and EQ my subs this week. Thank you!
@@jeffmery I just completed my subs time alignment and EQ'd them together using the time alignment tool as shown in your video and I was impressed how accurate this tool is not to mention all the time you save doing it this way. Do you only focus on positive summation of subs or do you check phase also? Great video and clear easy to follow instructions. Thank you!
Wow just what I have been looking for years!!!
Glad it was helpful!
You are an angel from heaven!!! Yes!!!!!!!!!!!!! Such a time saver
Waiting for my MiniDSP HD. Should be here in a couple days. Had the regular miniDSP and it didn’t do much for me. Stoked to try this tool
The most important differences, in my opinion, between the 2x4 and 2x4 HD are 1) 7.5 ms vs 80 ms available delay 2) 6 PEQ filters vs 10 PEQ filters on each channel 3) 0.9V vs 2.0V output voltage and 4) 0.9/2.0V vs 2.0V/4.0V input voltage options.
None of these alone are really enough to justify 2x the cost. However, when put together, the HD is a much more flexible device. The output voltage is probably the most important of those. Again ... just in my personal opinion.
This is good stuff. I've played around with the alignment tool, but maybe I didn't realize the power of it. Always thought minidsp was the easiest way...
miniDSP is probably the highest "value" way to achieve the most flexibility, but it's not the only way to implement things things :-). Just tweaking the distance in the AVR/Processor can provide huge benefits in the crossover region. Some AVRs/Processors and subs allow for manual EQ as well. My old Yamaha RX-A1080 allowed for 4 bands of manual PEQ on the sub. SVS "Pro" model subs have a 3 or 4 band PEQ built in to the DSP on the plate amp as well.
A calibrated mic, REW, and a miniDSP are mentioned together so often that a lot of folks think you have to do all or nothing which definitely isn't the case.
Great video. How about a video about the so called "sub distance tweak" as well, since that would be a logical next step after getting the subs aligned to one another.
I thought about this, but there are a bunch of "sub distance tweak" videos already available. That being said, it is on my list to do one "in my style" if that makes sense. There are a lot of people that appreciate the context of what is being done vs just following instructions blindly. This is still still on my list to do, but finding the time is unfortunately hard 😞
@@jeffmery there are a lot of mediocre ones out there that ramble around the topic. If you don't have time to make one (understandable) but know of one you believe is good, perhaps link to that?
Excellent video. Helped me time align much better than the measure tape results
Glad to hear it! The tape measure method just flat out doesn't work for subs. Below the Schroeder frequency (aka the transition frequency; typically somewhere between 200 - 500 Hz depending on the specific room), we're not hearing the direct sound from the speakers. We're hearing the resonances of the room they are in.
Sub phase is a function of time from the individual driver (and port if ported) as well. The alignment tool takes phase into account when summing the responses giving MUCH better results than manual methods.
@@jeffmery Do you have any advice on best alignment of multiple subs with multiple seats taken into consideration? You recommend aligning for a single seat and then move on to EQ?
EDIT: My bad. I see in the description you mention MSO. I'm back to trying to fully grok how to use that tool effectively :)
@@joker927 No worries! Yes, MSO is probably the best option there. I also think that Dirac's DLBC does a really, really good job and is about as easy as it gets for multi sub integration. However, it's only available on a Dirac-based processor and even then, there may be an up-charge for using it depending on the specific processor.
Using MSO with your DSP of choice is a much more "universal" option. A miniDSP 2x4 HD is probably the most common and recommended device, but the filters from MSO can be used with any DSP. You would need to set the filter limits in MSO to match what the DSP is capable of, but that's about it. I recently helped someone implement MSO using the Trinnov-native filters. Those have slightly different min/max values for gain, Q, etc. We just set those limits in MSO and were good to go.
Thanks for the great walkthrough. I just subscribed to the channel. Hopefully on your channel I can find other videos of you teaching us how to use this EQ and other adjustments that you mentioned at the end of the video. If you do have those videos, you might want to add links to them in the description that you wrote above for this video. FYI, In my case I used MSO and a miniDSP 2x4HD since I have 4 subs and 2 rows of seating. With MSO I only used the sub alignment feature to get better bass across all seats, and now I will run MCACC PRO on my Pioneer Elite SC-99 to get the subs and mains aligned. My goal is to then use the time alignment tool in REW after the MCACC PRO auto calibration, to tweak the time alignment of the subs and mains again. Hopefully you have a video showing this process as well.
Thanks for the feedback! Finally got time to work on videos and will be posting new ones very soon.
The alignment process is exactly the same. Instead of aligning the measurements of 2 subs, you align the measurements of the combined subs (post MCACC) and then either the Center or L+R channels (again, post MCACC).
The center or L+R measurement goes in the top box, and the sub measurement goes in the bottom (since we are aligning the subs with the mains).
If you get a positive delay number on the subs, you need to add distance to the subs in the AVR. If you get a negative delay number on the subs, you need to reduce sub distance in the AVR.
I would be interested in seeing what the results are if you change the order of sub introductions. In other words, with the subs labeled ABCD, if you added them in that order one time and then reversed to DCBA, would you still get the same answers? Have you ever done this? Maybe a dumb question but just a thought.
This is not terribly applicable for my sub, a Velodyne servo 15, as the gain runs typically about one micron off the pot stop, and add about a 1/4" circumferential movement and windows/dishes/etc are rattling.
My worst problem with this sub is that it produces standing wave resonance nodes in the room no matter where you point or place it, as it runs right down into the 32' organ pipe range very nicely. I definitely need some proper acoustical treatments, meaning some properly designed and placed bass traps for at least the first cut. Thx for the vid.
I have looked at changing the order as you describe. But to be perfectly honest, it's been so long I don't remember. The fact that I don't probably says there wasn't much of a difference. It's not much effort to try it out. I"ll see if I can find some time this weekend.
My previous sub was a Velodyne CHT-15 which I believe was one of their early servo subs. Bought it in like 2001/2002? Good sub and my brother is using it now. It definitely had plenty of power, but wasn't tuned terribly low; maybe 28 Hz?
Bass traps should help with that at least a little if you can get enough of them into the right places in the room. We also shouldn't expect too much out of a single sub when it comes to an even response. Unless it's (very) near field, physics just won't let us get even sound from one sub in a small room (to be clear, I'm talking acoustically small which is basically every room in every house ever built; smaller than an actual movie theater).
Subscribed and liked. My dude, this his how you do a tutorial!
Great video.. should have a run a sweep in the end thought ;-) Not one of these summed bass responses is an actual measurement??
I should have to illustrate how accurate the alignment tool is. I recommend trying it yourself. The measurement will directly overlay the prediction (assuming the microphone wasn't touched or moved between the initial and then verification measurements).
Thank you very much - very useful video - best on the topic. Please do follow up with the sub eq and subs to main integration videos - really looking forward to those. Saw your MSO tutorial as well, but since I have only 2 subs and one MLP, this method should suffice, in my room. Best wishes..
Will do. Life getting in the way at the moment but I still plan to do the video on subs+mains.
You just taught me sooooooo much! I can't thank you enough, you just got a new sub also !
JoeNTell turned me on to you. Subscribed. Thanks for this great tutorial.
Awesome! Thank you!
Hello Jeif
Great video easy to follow.
"I’m not using MiniDSP because I have a processor with independent subwoofer control. Should I input the time delay result directly into the processor’s subwoofer delay settings?"
Correct. Delay directly into the processor.
Wonderful explanation! Maybe I missed it but is it needed to use "Acoustic Timing Reference" or "Loopback Timing Reference" for this method?
As a general rule, you should always use a timing reference. Generally speaking the acoustic timing reference is the easiest to use and has the fewest caveats as opposed to the loopback reference.
I am not an expert in the nuances between the two methods, though. Based on what I’ve read while learning and experimenting with MSO as well as several online exchanges with Andy, the creator and maintainer of MSO, the acoustic timing is definitely the path of least resistance.
@@jeffmery I have been using a 1/4" cable on my scarlet Solo as loopback reference with success. A bit tricky tho to set the right output inputs in your preferences.
@@kewlbug Good to know! That's one of the reasons I recommend the acoustic reference over the loopback. The loopback requires additional hardware (like the mic preamp you're using). Folks coming from a recording/studio background will have that equipment, but the vast majority of folks coming from a home theater background/perspective won't (USB mic and go!). Figuring out the right output/input is also tricky as you mention. Those inputs could also vary depending on the specific mic preamp in use.
All that being said, there may be technical advantages of a loopback over an acoustic reference. I haven't dug into it to verify that one way or the other, or under what conditions any advantages may apply.
Is this a pro upgrade? The alignment tool is not showing as an option for me under controls. Version 5.20.13
Update got it figured out and now the subs are working pretty good together besides one small area. Between 80 and 100hz it dips down even below some of just the single responses from a raw measurement and is the only area that isn’t close to the summed measurement from your video. Is there something I can do before moving on to eq.
Following your video I definitely found the delay to give the best summed responses just hoping to fix this small area before I move to eq.
Glad you figured it out! The only thing the Pro upgrade adds is the ability to measure from more than 1 microphone at the same time. The most common use-case is taking spatially averaged measurements.
This is an advanced topic, and I don't have any tutorials on it. However, assuming you're using a miniDSP, an all-pass filter on one of the subs might be able to resolve that for you. It's too complicated to go into here. This video from Meyer Sound does a good job explaining all-pass filters and some ideas on how to implement them: th-cam.com/video/_KAa39OcMoo/w-d-xo.html
@@jeffmery awesome! Thank you. So the raw measurement with alignment is definitely much better across the board and similar to the to the last summed line you get. But it’s definitely way better compared to just the single measurements I took. Before delay or anything.
Jerry great video
What about using this approach to combine subwoofer and bass to help determine a better crossover frequency between the two drivers ?
Thanks
I assume you mean the crossover between the subs and non-sub speakers? It's actually the exact same process and is commonly referred to as the "sub distance tweak". You just need to get the correct set of measurements. For home theater, subs are usually aligned with the C channel.
1. Set your crossover for C
2. Measure C from 20Hz-20kHz
3. Mute the subs, measure C from 20Hz-20kHz
4. Unmute the subs, temporarily disconnect the C speaker, measure C from 20Hz-20kHz
You now have a "before" from step 2, a C-only measurement from step 3, and a sub-only measurement from step 4.
In the alignment tool, you want to use the step 3 measurement in the top part of the alignment tool and the step 4 measurement in the bottom part. This will help to time align the subs to the C channel at the selected crossover frequency.
Let me know if I missed the mark and you were actually referring to something else on the better crossover frequency.
Hi Jerry thanks for quick response
I’ve built a fully active 3-way with sub. I’ve questioned whether I have selected the correct xover frequencies, ( sub to woofer, woofer to Mid, mid to tweeter). I’m thinking what I’ve learning from your video if I can start with woofer and align with mid ? Use the technique to see if I slid the freq up/ dwn if there is a sweet spot vs mfg xover freq?
@@zeekbruno4869 Ahh yes, I completely misunderstood. Yes, I think the alignment tool can be used for that use-case. But it doesn't have the ability to leave the timing fixed and just slide the crossover frequency around to see what works. I'm sure you know this, but crossover design gets complicated. It's about the frequency as well as the slopes on both sides, then gain, delay, inversion, and even AP filters can play a role.
I honestly haven't done any driver-to-driver crossover work outside of implementing manufacturer supplied specifications for their active speakers. There are probably tricks in REW to make things go faster, but I just don't know them.
Jeff
Thanks I’ll try it and keep in touch on my experiences
Stay well and keep forging new
Paths
Bruno
Excellent video, mate. I have been pondering if incorporating a MiniDSP while already having Dirac Live (and four independent sub outs on my AVR) is redundant. But if I’m understanding this process correctly, MiniDSP plays a role in establishing the most neutral base for Dirac to do its thing.
Thoughts?
Thanks for the feedback! With Dirac it depends on the specific AVR and what Dirac features it supports. Not all AVRs/Processors support all levels of Dirac so that's something you want to find out. There's Dirac Full-Range, DLBC (with single and multi-sub options), and ART. A miniDSP can help Dirac specifically with the "Full-Range" version. This is the entry level Dirac license that doesn't do anything special with subs or crossovers. In this case, the miniDSP would be used to pre-align and EQ the subs as one before running Dirac.
Things get a bit more interesting with DLBC (Dirac Live Bass Control). DLBC does two things: automatically aligns and integrates multiple subs together, aligns the non-sub speakers with the subs at the crossover. DLBC can be purchased in a single-sub or multi-sub license. With multi-sub DLBC, there's really no reason to have a miniDSP in the signal chain. It does a great job on its own. With the single-sub license, a miniDSP can, and probably should, be used to present a single big sub to Dirac. In this case you're really getting just the benefit of integration at the crossovers with non-sub speakers (still very good). Depending on your level of comfort and how much you like to tinker, DLBC is the closest thing to a multi-sub "easy button" around right now. It's not perfect and takes some tinkering, but it's less complicated and less costly than doing that plus a miniDSP.
ART is the top of the line room correction from Dirac. It's only available on Storm processors right now but others have announced that they'll add it in the future. With ART, you absolutely do not want a miniDSP between the subs and AVR. ART needs direct control of all subs and speakers to work properly.
I hope this helps, but if not, keep asking questions and I'll help out 🙂
The first time I aligned my subs was with the phase on my 2nd PB-1000Pro and REW sweeps, and it worked great. This time I used this method and got worse results. The actual response compared to the simulated one is pretty off for me. I tried adding each individually, and summing the front and rears then summing. I'm going to work on it again now.
The three biggest reasons for the prediction not matching the results are:
1) The mic was moved and is in a different position
2) An acoustic timing reference wasn't used
3) The measurements used for alignment were taken in what I'll call a "non-zero" or "non-reference" state.
On that last one, we need to turn off all room correction, dynamic volume, and dynamic EQ in the AVR/processor. Gains, delays, inversion, and PEQs need to be zeroed out both the AVR/processor and the miniDSP (if one is in use). Finally, the settings on the sub itself need to be at a baseline of zero phase, no inversion, crossover bypassed or set to the highest value possible, any PEQ bypassed, and gains set to a known level. I use the midpoint of my gain knob, but some subs like your SVS, have digital gain with a default of -10 dB; that's fine.
The phase setting on the SVS pro models is nice (I use the SB-1000 Pro in my desktop 2.1 setup). Single degree adjustment is awesome granularity if you don't have a miniDSP to provide extremely precise control over delay.
@@jeffmery 1. Mic never moved.
2. Not sure how you create a timing source with REW sweeps.
3. All DSP on all 4 PB-1000Pros, MiniDSP 2x4 HD, and JBL SDP-55 turned off. I know to do all that.
I ended up just doing manual sweeps and aligning myself. Got a MUCH better and flatter response.
@@andrewskaterrr Good deal. The acoustic timing reference is on the measurement screen. On the right side near the top, there's a "Timing" field. This defaults to "No Timing Reference". Change this to "Use acoustic timing reference". When you do this, an additional field will pop up below "Output".
"Output" is where you select the channel that you're going to measure. The new field below this is "Ref output". This is the channel that will be used to create the acoustic timing reference. It doesn't matter what channel you use. However, it's important that you use the same channel for all measurements that you're taking in that session. It's my personal habit to always use the left main, but that's just habit. That's all there is to it for enabling the acoustic timing reference. These settings should persist so you really only need to do this setup once.
Now, when you take a measurement, you'll hear a short chirp out of the "Ref output" channel, and then the measurement sweep will run on the selected "Output" channel.
You should always use an acoustic timing reference for every single measurement. Features like the alignment tool require them to work.
@@jeffmery thanks. I wondered if this existed but Ive never heard/saw anyone mention/show it in any of their videos on REW or MiniDSP, which blows my mind. i was actually trying to think of a way to do it myself, thank god I don’t have to lol. I googled it earlier and saw it can use the loop back input? I’m going to try when I get home in a bit.
@@andrewskaterrr 99% of home theater hobbyists, including me, use USB mics. The acoustic reference is the only option when using a USB mic. Loopback references have some specific hardware requirements due to the synchronization required between the sending and receiving sides of the actual loopback. The link below has a good explanation from Andyc56, the creator and maintainer of MSO: audiosciencereview.com/forum/index.php?threads/rew-loopback-timing-with-a-usb-mic.14273/post-438115
I can't speak to why so many videos don't discuss the use of an acoustic timing reference. Some features, like the alignment tool, are relatively new features released at the same time or shortly after some of the best, and most often referenced videos, were created. I've watched others, though, where it's clear the person is chasing views and parroting things without fully understanding the "why" behind the "what" they're doing.
There's no malice in what they do. And some viewers just want to get "better" results as fast as possible. Some people don't care about the "why". That's all cool. I want to understand what's going on and get the best results I can, not just better results. I'm still learning a TON from a lot of people too.
Just this week, someone pointed me to a video on room calibration that completely blew my mind and changed the way I think about things. I'm still trying to process everything talked about there (it was a 3-hour video!). Always be learning!!
Thanks for the video! Only problem is every time I use the adjustment tool and then measure to verify, they aren't similar. I've tried multiple times after rewatching your video & I get the same inconsistent results. Any ideas why?
Are you taking your measurements using an acoustic timing reference in REW? That's a hard requirement to use the alignment tool. If there's no timing reference then the alignment will never be correct or match the predicted response.
@@jeffmery thanks. I was using no timing reference.
@@doublet147 Easy mistake. I've done it many times by accident myself! I personally haven't run across any circumstances where I *don't* want to use a timing reference. Using one for all measurements doesn't hurt anything at all, and can keep you from having to try and duplicate measurements if you find you need one. It's a good habit to get into for sure.
Good luck, and let me know how it turns out!
@@jeffmery Worked like a charm. So much faster than manual adjusting. Thanks!
@@doublet147 Awesome!
Hi Jeff!
Thank you for your video, very educational!
I just have a question… I hope you can help me with this.
When and what sub I have to put the delays? Every time after align 2 subs? Or at the end?
Thank you!
If you're doing more than 2 subs, it has to be done in steps. We'll use sub A, B, C, and D as an example. Each time you align things, only one gets delay added.
So if we start with aligning A+B, only B will get delay. We enter delay for B, then take a verification measurement. The verification should almost perfectly match the predicted alignment if the mic hasn't moved.
Now we align the new A+B with C. Only C will get delay. Enter that for sub C then take a verification measurement to ensure it matches the the prediction.
Finally, we align A+B+C with D. Only D will get delay. Enter that for sub D and take a final verification measurement.
Awesome job! Do you see a need for this if someone only has two subs, and no MiniDSP? Thanks!
Yes it's absolutely needed, but the ability to implement will be questionable. That will depend on the specific capabilities of the gear in use. To implement this, you need to be able to set different delays for each sub. One needs to stay at zero, and the other needs to be set to the value from the alignment tool. You'll have to figure out if your AVR or processor has that capability.
There's also the issue of the granularity of that setting. Many AVRs only allow 0.5 ms granularity where the miniDSP allows 0.01 ms granularity. This doesn't mean you can't do it, it just means that the actual results might not be quite what REW predicts. It should still allow for improvement provided the subs have independent delay as indicated above.
Fantastic info! Thank for this!
Good day. I just stumbled upon this video. I've watched your MSO videos for my subs but is using REW's alignment tool better than MSO. I ask because I used MSO and had great a response, but I changed out one of my subs, so I have to start over. I aligned the manual way through the MiniDSP, but I like the REW alignment tool to better see what each delay does. Should I use MSO after alignment?
This depends on your goals. If you only care about a single listening position, then the REW-only method using the alignment tool is great. It takes less time/effort and gives great results. If you care about more than one listening position, like an entire row or multiple rows, then MSO is probably the better choice.
This all depends on the room and the placement of subs / seating in order to mitigate the room's modes. There aren't any absolute answers.
I would not try to align the subs manually and then use MSO. A manual alignment for a single listening position won't usually be the best alignment for multiple listening positions.
I like to print the aligned sum for a few different options to be able to compare them. (e.g. my best results normal and best results of inverted). Also, you dont even need to have the original traces "on". Makes it a bit cleaner. I don't mind a little bit of smoothing, (at least to get a better idea of the summation) as I'm just looking for the best "average" kinda result.
Sort of related, I've started using the VAR smoothing more with eqs, and putting most of my eq bands in the lower regions. (I've read we are more sensitive to EQ issues in the lower freq. ) So far the results are great. I've discovered you will also get a lot different eq results depending on the smoothing you have.
Variable smoothing is what should be used when you're going to apply EQ to the curve that you're looking at. That's actually what it was designed for :-). I use it almost exclusively, but will sometimes flip to no smoothing (subs) or a specific smoothing (like 1/6) depending on what I'm looking at.
@@jeffmery Yes I've recently started to ditch my YPAO, looking at the YPAO result, it does some weird stuff. It might be good for most people, but we can get a lot better with REW.
@@kewlbug I don't remember the specific AVR you have. Up until the newest flagship AVRs from Yamaha, YPAO was terrible at optimizing subs. My previous AVR was an RX-A1080 in a 7.4 system. The subs were optimized with MSO and then presented to the AVR as a single sub. I always had to tweak the distance to eliminate the dip at the crossover point. There's also no way to limit the curtains or do house curves in the older versions of YPAO.
The current flagship AVRs have a new version that adds this missing functionality. I haven't, and might not, get the chance to play with them myself. I've always liked the sound of Yamaha AVRs. I still have a 1080 running the living room system and patio on Zone 2. They've just needed help with the subs.
Don't forget that YPAO volume and Dynamic EQ (think those are the right names, not home and going from memory), will change the contours of the frequency response. You might or might not like what they do. I always turned them off completely when measuring/calibrating. Then I'd turn them on after I was happy with things.
@@jeffmery Its a 2070, so yeah the older softwares. It is nice to jump on the PC in my office, and use the web interface to enter in the PEQ numbers. Not great though with only 7 bands per channel and 4 bands for each sub. The bands aren't fully adjustable, but you can move them around a little bit trying to squash those peaks.
I guess it should be noted that one can achieve this without a miniDSP as well, if one has subs with a variable phase knob, which effectively allows one to choose 0 to 14 ms of delay for that sub (0 to 180 degrees on the phase knob).
Correct if you need 14ms or less delay. There are problems with phase knobs too. The main issue is the lack of control since most knobs are analogue. Very few sub manufacturers include digital DSP on the sub amps that provide granular control.
The second issue is that phase is highly dependent on frequency. How many degrees of delay are how many ms? At 100 Hz, 180° of phase is 5.64 ms. at 50 Hz, it's double that, or 11.28 ms, at 25 Hz, it's double again at 11.56 ms.
I don't want to give the impression that the phase knob is useless. That's absolutely not the case. It just takes more work than DSP with extremely granular delay. That could be a miniDSP, other external DSP, or within the distance/delay capabilities of the AVR/Processor in question.
Amazing 🤩 so useful and well presented.
Next step would be EQ in REW
But then I am lost. How do I transfer the measurements to the MiniDSP?
Do I add the individual delays for the subs then add the EQ file to each of the four subs in the MiniDSP?
Have a look at this: th-cam.com/video/ARztXSmoQbE/w-d-xo.html
Great tutorial although using dark mode made it a little hard to see some of the measurement lines. I did the manual method and am looking forward to trying this out to see if I can improve on things.
Thanks for the feedback! I was told the same thing by someone else and use "light mode" in my newer videos. Let me know how it goes for you.
Dark mode - all day
@@kewlbug For personal use totally agree. On a video that people will have to pause a few times to follow along maybe not so much 😃
@@jeffmery It went great! I managed to flatten out my response noticeably and learning how to use the alignment tool helped me align my mains as well.
FYI you can click the slider button, and then use the arrow keys to adjust the delay rather than clicking and dragging to fine tune things easier once you get in the neighborhood.
@@HisshouBuraiKen Awesome! Thanks for letting me know about the arrow keys as well :-)
Excellent video and very good job explaining the process! I have a naive question; I have two SVS PB2000 PRO is there a way to align them without miniDSP before running audyssey and sub distance tweak?
Unfortunately I don't know the answer to that. You need to be able to apply delay to one of the subs to align them for the most positive summation. I don't think the SVS DSP has a delay capability. You might make some progress using the phase adjustment on the sub, but that's not the same as using delay to align the phase like we do with the alignment tool.
I have never used Audyssey so I don't know what things you can set before / after calibration that will stick. I think it will vary with the AVR and even the version of MultEQ / SubEQ.
I wish I had more experience there for sure.
@@jeffmery Many thanks for your reply!
This is a really great video. Unclear to me though is if you align the sub levels for each sub at the MPL before taking measuremenrs?
This will depend on the type of subs you have. If they are all identical, set the gain control on their amplifiers to the same position and go. I like to use the middle position so I can turn the subs up or down if needed.
If the subs are different models, but similar capabilities, then level matching at the MLP can work.
If the subs have differing capabilities, you can try distortion matching. Essentially, turn up the level on the least capable sub until distortion becomes an issue. Then back it off just a bit. Then turn up the more capable sub(s) until their distortion matches that of the less capable sub(s).
@@jeffmery Thanks for the reply Jeff. My subs are all identical, but some are right in a corner and the rears are shooting into a open space, so their output under 50 hz differs quite a bit(no room gain). Basically the back pair is not contributing at all underneath 50 hz. They are also futher away. Measuring at the MPL they are about 3 db lower in output between their outputs. This is why I am wondering if the summation will then work correctly.
@@poleepwka I'd set the gain on each to the same level. I would not allow any gain in the PEQ filters for those subs. If they're ported, make sure they have protection at or below the port tune. They should if they are commercially available subs, but it's best to verify.
Jumping on the bandwagon to let you know how superb your tutorials are! I've watched tutorials from Home Theater Gurus as well as Home Theater Gamer when I first got into the miniDSP + REW game, but I recently discovered yours and I'm loving how thorough you are in your explanations, plus the fact you use a Mac & UMIK-2! I'm sending you a message on the AVSforum with some questions I have, but figured I'd also ask one here...
- Should I start off my sub measurements using the LFE channel or the Center channel? I know you mentioned the LFE is 10dB louder and can cause issues when aligning to your Center/Mains.
Thanks for the feedback! I used the Home Theater Gurus video myself when I got started. Being on a Mac made some things easier (and harder). Plus, the alignment tool didn't exist when he did his excellent video. I figured I could give back a bit to the community and help out with my own versions :-).
I'm happy to answer here or over on AVS. If you're a new user there, I think there's a minimum post count you have to hit to send a DM (just FYI).
When aligning multiple subs, using the LFE is just fine. The only alignment we care about there is the timing amongst the subs. The volume really doesn't matter at all as long as it's sufficiently above the nose floor and not changed during the measurement process (set it and forget it until you're done!)
When aligning subs with the mains (L+R) or center, I use those channels respectively. I just mute the subs when measuring the speaker, and then unmute the subs and disconnect the speakers to get the sub-only output. Then it's just alignment tool and go.
It's been a while since I watched any of my own videos (always awkward to see/hear myself present). However, I *think* the SPL difference I mentioned was probably specific to the mains+subs configuration with MSO. It definitely causes problems there. I like to be consistent in my own personal processes, though, so it has become my habit to use the mains or center when aligning the subs to those respective channels.
@@jeffmery Thanks for the quick reply! Also good to know about the minimum post count on AVS.
I'm getting ready to align my subs, EQ, then align to my mains before running Audyssey. I'll be using the newly released Audyssey MultEQ-X via Parallels. Have you heard about it?
My Denon AVR-X6700H has 2 sub preouts, but my initial research of course lead me to the conclusion that Audyssey could not align dual subs properly, and that it is best to use a miniDSP 2x4HD + REW, then Audyssey MultEQ-EXT32 via the phone app (since it allowed for fine tuning). I had done this months back and was fairly satisfied, but now I'm learning so much more especially with these newer updates to REW & Audyssey.
- Assuming you aren't familiar with the new Windows version of Audyssey MultEQ-X, would you still say the above method is better for aligning & EQ'ing dual subs rather than ditching the miniDSP method and relying solely on Audyssey MultEQ-X? *I know I hadn't mentioned MSO, I literally just downloaded it yesterday and have been studying up on how to use it after REW. For now, I'm just going to start with my MLP, then soon I might try integrating MSO for multiple seating positions.
I read in one of your other comments somewhere that Audyssey does not account for phase, plus after investing in all this equipment my inclination is to continue using the miniDSP haha.
- Would you mind if I document a high level process of what I've done then share it with you via email? I don't want to blow up the YT comments section if I don't have to haha. I also want to be respectful of your time, so no biggie either way! I can imagine providing what must be similar responses to multiple people would lose its luster after a while😆
@@FiletMyYawn I did see the MultEQ-X announcement, but really don't know much about its capabilities. Unfortunately I haven't owned a Denon AVR in a long time. My advice is as accurate as it can be for not having direct hands-on experience ...
I would still take care of the subs first within the miniDSP first (either via REW or MSO). The important thing is to align and EQ to flat in the miniDSP. Don't put a house curve there at all. Then run the MultEQ-X process and use it to both fine tune and add your house curve. If you add the house curve prior to running room correction, the room correction software will almost certainly EQ the curve right back out to flat.
I want to be clear I'm not picking on Audyssey at all. It can do a good job with 2 subs. However, you've already got the hardware and experience so why not use that to make it sound as good as you possibly can? Audyssey can get the timing pretty good between the subs, but just because the timing is right, it doesn't mean there aren't still issues (like phase cancellation) that it can't account for properly.
You can send it over to jeff at bisonhta dot com. There's also a REALLY great thread over on AVS for calibrating multiple subs. It happens to be very Audyssey heavy as well. Check out the first two or three posts in this thread: www.avsforum.com/threads/guide-to-subwoofer-calibration-and-bass-preferences.2958528/
There might be WAY more than you ever wanted to know there, but it's a great resource and you can jump around between sections.
@@jeffmery Really appreciate that feedback! I just aligned and flat EQ'd my subs with REW only for my MLP at this point. Now I'm slightly confused on integrating with the mains.
Should I first run Audyssey, adjust the crossover if needed/desired THEN take a measurement of both channels with REW to see if there is a dip? And if there is, do I use the alignment tool and enter the delay into the AVR Sub Distance or the miniDSP for both subs?
@@FiletMyYawn Exactly. Run Audyssey. Set crossovers. Measure to see if there's a dip. If so, alignment too and adjust the delay in the AVR for the sub channel. Don't mess with the delays in the miniDSP now that things are aligned.
Great video. You do a great job explaining things. So, I only saw this video after going through the manual process of adding manual delays, measuring, changing delay, playing with inverting polarity, etc. Then, I saw this video. Question: the alignment tool that you discuss in this video doesn't actually measure the changes, but calculates the impacts of delays based on mathematical modeling? I'm wondering what assumptions the alignment tool might have that might not be correct for certain rooms. For example, my room isn't a perfect triangle, but widens by 3' in the back half of the room.
I just answered your post on AVS, but wanted to respond here too. You are correct in that I didn't actually measure the result after the change. You should *always* do that to make sure something wasn't messed up. 99% of the time, if the result doesn't match the prediction, it's because either the mic was moved between the input and verification measurement or the input measurements were taken without a timing reference.
In this instance, while it seems complicated, REW is just adding the two input measurements together while delaying the phase by some amount, say 5 ms. E.G., if the phase is 180° at 20 Hz, the phase is still 180° at 20 Hz, it just happens 5 ms later.
The shape of the room doesn't matter. REW doesn't know or care how the room is shaped because it's not doing a prediction. Well it sort of is because it's doing some addition, but not prediction in the way I think you mean. Additionally, the *original* measurements were taken in the same imperfect room. That means the original measurements already take the irregular shape of the room into account. It then follows that the result of adding the traces together *also* accounts for it.
This is easy stuff to test and I highly encourage you to play with it! Pretty enlightening in my opinion and a little creepy when the predicted result is so accurate.
(Side note: This feature didn't exist when most of the more popular sub alignment videos were made. Those videos are still 100% correct. The alignment tool is just more accurate and faster because it eliminates the guessing and extra measurements.)
@@jeffmery Thanks! I will try the Alignment Tool and see how it compares. Keep up the great videos! They do a great job explaining things!
Another excellent tutorial. I am trying to figure out starting point of setting sub distance before using rew and room correction; arc in my case. I have set the sub trim in my prepro to zero and set the volume to get 85db from my subs using umik1. Can someone help on setting the sub distance before i start the calibration for sub time alignment using minidsp-hd.
Thanks for the kind words! Unfortunately, I have zero experience with ARC.
It's my personal habit to set the sub distance to zero before going through the sub alignment and EQ process. Technically you don't have to do that, but I like to have a baseline where I can easily repeat things from a known starting point. Setting it to zero is just an easy thing to remember.
Once the subs are aligned, the room correction software should determine the proper sub distance (delay) as part of the calibration process. After the calibration is done, it's always good to use REW to verify the results for at least the main listening position. The distance may need to be adjusted if there is a dip in the magnitude response at the crossover frequency. This is commonly called the "sub distance tweak".
Again, I don't know ARC so I don't know how applicable the sub distance tweak is with that software. For Dirac with DLBC, as an example, the sub distance tweak can't be used.
I hope this helps, and again, sorry about my lack of ARC knowledge.
Thx for the video. I tried your method to find the delay. However, after I applied the delay on my miniDsp and re-measured, the resulting curve is no where near the predicted (aligned sum) curve. I don’t think the predicted curve generated by the alignment curve is very reliable ...
The curve is 100% percent reliable and extremely accurate. There are a couple of issues that can lead to the real result not matching the predicted result.
First, is the microphone being moved between the initial and the aligned measurements. Even a few inches can result in a drastic difference given that we are working with modal frequencies.
Second, is not using a timing reference on the measurements that will be aligned. The simplest way to do this is to specify "acoustic timing reference" next to "Timing:" in the measurement window.
Third is taking measurements with settings in place, and then attempting to directly use or apply the values shown in the alignment tool. This is most common when aligning multiple subs that were previously aligned with another method. Settings recommended by the alignment tool are relative to the measurements used for alignment. As an example, if there were delays or EQ in place on the "input" measurements, then alignment results will have to be manually calculated relative to the existing settings. I've done this before on accident and it's absolutely maddening to try and figure it out.
When you need to align, or realign subs, you should zero out any gains, delays or EQ that is set anywhere in the signal chain. Then take measurements with an acoustic timing reference to use for alignment.
Thanks for the video, it is possibly the easiest video to understand I've seen to date.
I have a curiosity to consult.
Have you seen how the waterfall improves in a multisub configuration and also after applying equalization?
If you checked it, it would be nice if you could make a video when you could.
Do you have a link where you can see your room ?.
I say this in case I had put acoustic measurements of it.
Written with translator.
Greetings from Spain
You're welcome! Yes, I have seen improvements in the waterfall after multiple subs are aligned and equalized. There are two major reasons for this. First, the subs are working to help each other to produce an even response. There is less fighting (cancellation) and more cooperation (reinforcement). This leads to the second reason: Because the subs are working together and producing a higher overall output, most of our EQ filters can now be trims (negative) instead of gains (positive). This helps pull energy out of the peaks, and energy out of the room, which should reduce some of the ringing seen in the waterfall.
However, I have also seen the waterfall get worse. This is almost always the result of increasing gain to fix a dip in the frequency response.
Small amounts of gain are okay. I generally allow +3 dB on individual PEQ filters in my personal setup. The problem with gain is that it can increase ringing. This ringing will show up in the waterfall as longer decay times around the EQ center frequency. The more gain, the more ringing. The more narrow the EQ filter (also called "high-Q" filters), the greater the ringing as well.
It can take time to get the best results, but if you are able, try to use only negative gain (trim) on your EQ filters. If you need to boost some frequencies, try to use the lowest amount of gain with the widest bandwidth (lowest Q) that produces the desired results. This will have the best chance of improving the waterfall without introducing more ringing.
As always - measure after making a change to verify the results. Don't get caught up chasing a perfect graph. Listen with your ears to see if you actually like the results as well :-).
Many thanks for the video. That is by far the best explanation I have been able to find. I only have a single sub, but I would like to find out how to use REW, to adjust my sub best with my front speakers. I use antimode dual core between my streamer and preamp, and a dspeaker 8033 for my subwoofer. Have you made a video on the subject, or do you know where I can find one?
I've heard good things about the DSPeaker kit, but unfortunately have zero experience with them. As a result, I've never looked for videos or other guides on them either.
@@jeffmery I did not express myself well enough :) I mean, a video showing how to use REW, to analyze whether the subwoofer is adjusted to fit with the main speakers?
@@franknielsen7597 ahhh. That’s my mistake! Sorry! You want to find a video on the “sub distance tweak”. There are tons out there. A lot are focused on Audyssey-based room correction however, the cause and fix are universal. It’s pretty easy to check your system to see if anything needs to be adjusted as well.
Align this SubGroup to the MAINS ? Is it the same process but just focusing on a narrow range over the crossover freq? (like 80-100 for a 90x-over?)
You said that most videos on this topic focus on integrating subs with mains but unfortunately many of them are utilizing outdated software/methods AND drag on for longer than needed in a "diary" format. Thank you for this TO-THE-POINT video.
I am running a 2x4HDw/dirac to control a non-avr 2.2 setup.
You're welcome!
For a 2.2 system you'll align with your mains (L+R measured and playing together, not separately). The general recommendation for home theater is to align with the center. But even for HT, I advise folks to try both and see if they prefer one over the other.
A common problem to have when integrating subs with mains is a dip in the magnitude response (frequency response) at the crossover point when the subs and mains are playing together. This is due to phase cancellation happening at, or near enough to, the crossover frequency. If you do a sweep and see that dip. Use the process below to get the measurements. You'll then align at the exact crossover frequency you're using.
It is generally the same process. Sweep L+R with the subs muted. Then sweep L+R with the subs on, but the mains muted (or disconnected). Use those two measurements to align things. Then run a sweep of L+R with everything playing to verify the fix.
@@jeffmery Great. Thank you for confirming my thoughts!
I see in the latest REW where you can select the exact frequency that you want to align two measured responses to. I previously just aligned the subs for smoothest freq response & max output, then ran DIRAC to EQ the overall setup. The Sound is great, but will now add aligning the sub group to mains at the crossover freq then re-run DIRAC to get additional improvement/optimization.
This seems like a great tool. Im trying to watch as many videos as possible befor I jump into this, and that might be a mistake :) but before alignment have you not gain/level mathed you subs to play the same level? and do you not link the subs in the minidsp?
Yes, those are things that should be done prior to using the alignment tool. I’m not sure what you mean by “link subs in the miniDSP”. If you mean linking the output channels together, then it’s a “maybe”. It can be a time saver but if you forget it’s on, it can certainly cause some unexpected things (e.g., wanting to change one sub, but it changes both and you forget why). I generally don’t link subs in the miniDSP unless the room tells me that I should do that. One example is a pair of subs at the front of the room. If the seating and subs are placed properly, then linking can make a lot of sense.
Yes I meant linking output channels. That’s what I see others do so wondered why you didn’t. Have you made a video where you explain how to gain/level match the subs prior to alignment? And when you are done with eq and audyssey do you then level match subs with the other speakers? Really like the way you break things down for us not used to all the settings and what seems obvious to those who are used to it
@@michaelbested123456 I don't do that because I've caused myself more problems than I've solved. It's a personal choice only. That feature can save time for sure.
I don't have a video on gain or level matching. If your subs are identical, just set the gain knobs on the amps to the same position and done. I think Home Theater Guru's has a decent video on level matching.
With Audyssey, EQ to flat, run Audyssey, check all the speaker levels to make sure they're correct, then add your house curve. if you add the house curve before running Audyssey, Audyssey will EQ it out again.
@@jeffmery thank you for replying
Thanks for the video!
Quick question! If I am only measuring 2 subs , what do I do after I get the 1st positive sum ?
And once I am happy with the response , how do I go about adding the numbers into my miniDSP ? I just put the delay # into one of my subs in the miniDSP correct ?
With just 2 subs, you’re done after the first alignment.
Yes, just one sub will need delay. The other will have 0.0 delay.
@@jeffmery can I keep finding tune with the first alignment?
@@ants9574 I’m not sure what you mean? With 2 subs, there’s only one alignment to worry about. It’s always “N-1” on the number of alignments needed. 4 subs only needs three alignments (1+2, then 3 with the already combined 1+2, then 4 with the already combined 1+2+3).
@@jeffmery sorry about that. I meant can we put our positive sum alignment response from the 2 subs into both spots in the alignment tool and keep finding tune and trying to get it even more aligned ? Does that make more sense ? Sorry I'm noob to using the alignment tool haha
@@ants9574 I understand now. No, that is not possible. The combined response depends on the alignment of the two subs. If you were to try and align that, the original alignment would change and no longer be valid.
Thank you, great tutorial.
You are welcome!
Great video! Maybe a noob question, but could this theoretically work in a car audio setting for time aligning main speakers to listening position, as well as sub to main speakers? Thanks!
I haven't personally used it in a car, but there's no reason you couldn't use it for that purpose. The sub process would be the same as in the video. Aligning speakers would probably be slightly different.
I'd probably align them by looking at the differences in impulse response timing. That would be easier. Then align the subs if needed.
@@jeffmery Awesome - thanks for taking the time to respond!
I wish you could move the slider .1 at a time with keyboard arrows or something
* Found out you can, if you click the slider then use arrows it moves in .01 increments !
Hi, a general question, the Alignment tool can set delays with "allign phase at cursor" doesnt it matter to keep each sub in phase while using the same process as the video? does it only matter to get most positiv summation?
When aligning subs, we typically just care about having the most positive summation. We're aligning subs with one another which we will later present to an AVR/Processor for use. We want that "virtual" sub to be as strong and as even as possible. So we look at the entire frequency range.
Typically, we alight the phase slopes or align at the specific frequency when we're integrating speakers that play at different frequencies (subs & mains, or mids and tweeters for people using REW to build speakers). I usually look at both the phase slopes and frequency options when I'm tweaking sub distances to see which yields the smoothest integration.
@@jeffmery thank u, was just wondering because I also watched a video of sub main integration with the tool. And someone said we can use it too to set delay right for the subs. I did it like ur video before. Now I used MSO. For seat to seat consistency.
But still unsure if I made mistakes somewhere.
So still playing around.
If I use the alignment tool with phase at cursor I would need to delay 50ms between front and back subs Which I would need the mini dsp 2x4hd. Instead of my 10x10hd
(only15ms there)
All subs without delay,eq or change in gain play pretty flat 18-100hz.
Well someday I will get it and don’t think about it anymore 😬
@@God_y_u If you use MSO, it will calculate delays for you. There's no reason to use the alignment tool between the subs going that route. You'd only use it to do the "sub distance tweak" to get smooth integration with the mains.
Inside of MSO, you can specify the allowable range of delay as well. With just 15ms of allowable range, I would set the range in MSO to +/- 7.5ms for all delay blocks in the config.
I hope this helps.
@@jeffmery yeah I have a working MSO. Setup atm. But it does make a lot of cuts in eq to balance everything. So still testing around To see what changes for better or worse.
Thanks
Great tut.
Q: If i want to eq two seperate listening positions with minidsp
Do i align my two subs from each of those listening positions?
Then what do i do with those two results?
Thank you! If you care about the response at more than one position, you probably need to look into Multi-Sub Optimizer (MSO). What we're doing with the alignment tool is setting timing between subs. Those are measurements taken at a single position. If we move the mic, the timing will be different at the new position.
It is possible to get improvements in the non-main positions with the alignment tool, but it may not be to the degree you want. More advanced solutions such as MSO, Dirac with DLBC, or Multi-Sub EQ (Audyssey for up to 2 subs) are really required for this.
@@jeffmery Thks for the detailed reply
For now i think ill pack the minidsp away
Its too overwhelming
I have to learn REW, minidsp, mso and ARC gen in order to optimise my two subs lol
I think ill spend a few months studdying it all then give it a go
@@isaaccraig3666 It's definitely time consuming so I totally understand. IIRC, Anthem were working on a multi-sub optimization solution for ARC. I'm not sure if they have released this yet or what models will support the feature. this might be an easier option to consider (in the same way that DLBC is easier than both REW and MSO). Good luck and let me know if I can help out!
@@jeffmery Thks.
For now i calibrated with ARC.
It actually did a decent job.
I used deep base boost to push a 50hz-70hz null up a few dB
Atleast now i can relax and spend time studdying MSO and Minidsp , rew etc
Get the knowledge down. Then in the nxt lockdown ill have time to get it right lol
Thks again. And look forward to seeing more of your vids . 👌
@@isaaccraig3666 Sure thing. Always good to just spend time enjoying it too!
Thanks for your post!
You're welcome!
Great video. I followed your steps but when I clicked on "Aligned Sum" to get the curve listed, I get a curve which is exactky the Aligned Sum but 6db lower all along. Any idea? Thanks
Hmmm. That's an interesting one that I have not seen before. I know with MSO, a 6 dB error can be introduced if the AVR/Processor is configured to send bass to both the main and LFE channels. This is usually called "LFE+Main" or "Double Bass" or something like that. That setting is something I would never, ever recommend the average person use in a home theater environment (or ever TBH). Some people like a ton of bass, but when you want accurate sound, that setting causes more problems than it solves. Double check that setting and make sure it's disabled.
If it was turned on, you'll need to remeasure your subs and then try the alignment tool again. Let me know.
@@jeffmery Thanks. I do not have Extra Bass, but I have my front speakers connected to the subs via HF. However here I am measuring my subs response only (LFE channel). I will keep investigating. Amazing and useful video mate!!
@@fofocho71 Sorry I wasn't any help on this one. If you don't mind, please follow up when you find the fix. I'd definitely like to know if I come across someone else with a similar problem in the future.
Great video! Question, if I don’t have a miniDSP is there a way to apply the delays using the SVS app? I’m not sure if phase control is the same thing or not as delays, but the SVS app allows me to control the phase from 0 to 180 degrees, as well as the polarity.
Great question! I think I answered you over on AVSForum, but wanted to post the answer here as well. I'll update my response there to be more complete now that I'm off mobile.
I've got an SB-1000 Pro in my office. There doesn't appear to be a way to set a delay using the SVS app. Phase and delay are related and one can be calculated for a given frequency. The question will be "does 180 degrees of phase give us enough adjustment at the frequency we care about?".
If we use 80 Hz as our example crossover frequency, one wavelength is 12.5 ms long (1 second/80 cycles = 0.0125 seconds). That means that 180 degrees of phase will give us 6.25 ms of delay (half the wavelength = half the time). Whether or not that will be enough depends on the specifics required for your room. If it's enough, great, if not, you'd need to look at other options.
You MIGHT get a crazy result from a delay-to-phase calculator like 738 degrees. That's okay. 720 degrees is 2 full cycles (360 degrees * 2), plus a little. 180 degrees gives us 1/2 a cycle to work with. If we subtract 720 from 738, we get an 18 degree difference which is well within the limits of the SVS controls.
The most common option is the "sub distance tweak". You adjust the distance listed for your sub in your AVR or Processor to add or remove delay which gets the phase lined up at the crossover frequency. I hope this helps!
@@jeffmery Thanks! Very helpful between your post of the forum and this reply! I will test it out later today and report back on the forum. In terms of which frequency to enter into the calculator, I'm assuming that since I have a pretty significant null at around 63Hz (you can see my REW chart in the AVSForum), I would enter both the 63Hz into the calculator, and the ms delay that the Rew alignment tool gives me to figure out the phase?
@@BenjiGator23 I just took a look at your graph. There are 2 nulls at ~63 Hz and 105 Hz. When we use the alignment tool or adjust delay/phase, we're typically targeting a specific type of dip that is centered on the crossover frequency. What crossover are you using? It's possible you will see changes, either positive or negative in both dips by adjusting delay/phase, but probably not the amount of correction you're looking for.
Another possibility is that you're hitting a room mode at 63 Hz. I may have missed it in the thread - What are your room's dimensions? Where, exactly, are you measuring the response with REW?
@@jeffmery Front Left and Front Right speakers are currently crossed over at 80Hz; Center channel at 90Hz. Here is a picture of my set from my original post in the forum (bit.ly/3jxxYKT). The main listening position (MLP) is approximately 6.5 feet from the front of the center TV console. Side wall to side wall is about 11 feet wide. As you can see in the picture, to the left of the left subwoofer is a stairway, and the there isn't really a left wall since it opens up into the the dining room. The right wall is a closed wall that goes to the rear wall about 21' back. Subwoofer cone to subwoofer cone is about 8 feet apart.
The REW response was measured from the MLP about 6.5 feet away from the center of the TV, and the MLP is about about 8 feet from each sub equidistant.
@@jeffmery can you link to this discussion in avsforum, I was also interested in trying to use phase control on the SVS app, but instead to use it to align a 2nd sub with 1st sub where it wouldnt be a particular frequency I would be targeting but instead the whole frequency response.
Thanks for the video!
Hi, does the alignment order matter? If you have 3 subs should you try all combinations?
I haven't found it to matter in my experience.
DO you EQ each sub individually? then same with speakers?
BTW thanks for this method. Nobody on youtube seems to get this. Worked great for me. My car was wanting like 18ft neg delay, but the slider showed me almost exact same curve at like 26" of delay
If you're using the REW EQ method, then all subs receive the same EQ. Technically, you EQ the single "virtual" sub then just apply that single set of PEQs to all of the individual subs on the output side of your DSP. You would then run your room correction software or manually level, align, and EQ your speakers (if that's your thing).
I'm glad the tutorial was helpful for you!
@@jeffmery ah that makes sense, since all the subs are considered "one" sub. Individually they might be all over the place, but you only care about what is hitting the listener
For this procedure, do you take measurement of your subs with the mains disconnected? Also, for the output in REW do you choose L (or R) or L+R?
Use L+R. Sweep that with the subs muted. Then sweep it with the mains disconnected. Use those two measurements for the alignment tool. I also recommend comparing the sound of aligning the subs with L+R against aligning with C.
The general "rule" is L+R for music, and C for home theater. I've done both and have my preference for my use-case. I could see people favoring one over the other so I do think it's worth the effort.
Excellent!
Quick questions, what buttons are you using to move the time alignment? So you can see the effect that each delay has?
There's a slider you can play with on the speaker itself in REW. Sliding to the right will add positive delay. I believe you can just click the slider and then use the arrow keys to tap and move the slider as well.
@@jeffmery Okay, I've never used this tool before, I'd liek to see how accurate it is, compared to what I have now.
Thanks Super helpful video.
You're welcome!
Hey, I'm going to try this out for the first time and I would love to know something before I start. With those basic measurements in the beginning, did you unhook/disable your mains to align the subs with each other? I'm pretty new to all this and the only vids I've seen so far are those from Home Theater Gamer and he did those measurements with the left speaker on.
Thanks! :)
PS: it's for a 5.1 system, but I also use it for music and other stereo stuff. I saw that you said that aligning it with the center is the best way to go for home theater, but in my case, it might be best to go L+R.
For aligning multiple subs with one another, the LFE channel can be swept independently of the other channels. It should be channel 4 on Windows and automatically labelled as LFE on MacOS. There's no need to sweep the other channels just to align multiple subs with one another. You will always need to use an acoustic timing reference, though, for the alignment tool to work properly.
To do the "sub distance tweak" and align the combined subs with the center -or- L+R, then yes, you will sweep one of the channels with the mains disconnected, then with the subs disconnected and align those two curves. Negative values for alignment on the subs here are almost always okay. That just means *reduce* the distance of the sub in your AVR by the indicated amount.
I'd suggest experimenting with both the center and L+R alignment. Center for HT and L+R for music are good rules of thumb, but I could see people preferring one over the other.
When taking raw meassurements, isn't it better to use a 3 point average from the MLP? I find that even a slight change in mic can drastically change the graph. What I mean is: take 3 meassurements in MLP for each sub. Average that result and use that result to tweak further. Is that a good approach?
This will not work. Averaging measurements removes the phase information from the sweep. It's the phase information that is being used to align things.
And yes, a slight change in mic position can sometimes drastically change things. Phase is a function of distance which is really just a proxy for time. Without using more advanced techniques or solutions, this is something we have to live with.
More "advanced" techniques or solutions would include things like implementing all-pass filter networks on the subs and mains, MSO, or DLBC.
@@jeffmery I found that I also can't use averages to create waterfall graphs. Makes sense but to bad. Need to be really carefull with mic placement. Thanks!
@@johandeen6096 Correct!
@@johandeen6096 I should also add that it's not like averages are useless. They can be very useful when you're doing more fine-grained work like building a target curve for your room. Do some research on the "moving microphone method" of measurements. Results are very similar to averaging but they're much more repeatable from session to session.
@@jeffmery Hmm, never thought of that. That seems like a good way to get an average you can use for other things. Will def. try that out! Quick question regarding this video, how did you get the raw meassurments of your subs? Just use an REW sweep with the sub cal on the subwoofer out (and using minidsp to switch between subs)?
Thanks, is it recommended to align 1 sub with your stereo monitors ?
The rule of thumb is align the sub(s) with the Center for home theater and the L+R for music. It's specifically L+R to indicate aligning with the combined output of both speakers swept at the same time.
"Stereo Monitors" implies a near-field use-case which is essentially the same as the L+R alignment for music. This is exactly my config on my office 2.1 system.
Would this work if you have a stack of Cerwin Vega Earthquakes 4 each per side in a 2x2 config?
Yes, but you would need to treat that as 2 subs, not 4. When we stack subs, the subs are so close they behave as a single sub in the room.
From a technical perspective, subs "couple" at a frequency with a wavelength equal to 4x the distance between the drivers (paraphrasing to keep it simple). So if I stack a pair of 15" subs, and the drivers are, say 20" apart, center to center, that gives a wavelength of 80". The subs will couple and act as a single sub around 170 Hz or so and below.
Use a Y-splitter and connect the L subs to the same channel and the R subs to the same channel. They use the alignment tool between the result 2 "meta" subs.
May I ask - the "Room Sim" feature.... when you 'simulate' your room with 4 subs in the positions you have - how close is the simulated response versus the actual response you get? I'm playing with the sim at the moment to work out where to add a secondary sub - and it's suggesting back right corner to couple with my big dual 18" sub on the front wall between my L and C channels. But I'd like to know how accurate that room sim feature is.
If your room is a rectangle, then the room sim is quite accurate. The less rectangular the room is, the less accurate the sim is. Rooms not fully enclosed, with angled walls/ceilings, L-shaped rooms, etc all make the predictions less reliable; even more so when those features are combined.
My room is fully enclosed, but sort of an L shape comprised of other rectangles. In some dimensions (length and height), the sim is in the neighborhood. In the width, it's not so much.
@@jeffmery Thank you for that. My room is rectangle and fairly well sealed, double drywall/gyprock, good insulation, on a slab, solid door etc. The only 'anomaly' is that we have a rear riser of about 30cm/1ft in the back quarter of the room. The room sim graphs do tend to reflect what we hear in our room, but I have not done adequate tests yet to fully confirm. Thanks for the reply :)
In that case it should very accurate indeed! A riser like that will *technically* have an impact, but it's highly unlikely to be anything you can hear.
@@jeffmery cheers mate. Thanks!
Say where is the alignment tool hiding in newer versions of REW? Google has failed me, has it been removed?
IIRC it's in 5.20 or newer. 5.20 was in beta for so long and so many features added, it's hard to keep track of for sure. Luckily 5.20 has been released and is out of beta. It's definitely a safe upgrade to make.
@@jeffmery gotcha thank!
Can we use this process to align subs to mains or center channel for 80hz ?
In REW under alignment tool...
Let's say we import the combined subs after they are corrected in minidsp, then import center channel without subs.
Now can we look at the 80hz (my selected crossover in AVR) and adjust delay slider for delay on subs to get flatest horizontal line at crossover without changing too much of rest of range?
or would you use REW measurement page, measure subs, with center channel as timing ref channels, then see what rew says under sub measurement (right column in rew of measurements) what to add or remove for delay.. Thrn enter thst correction INTO AVR distance setting for sub?
Absolutely! I think the alignment tool the most fool-proof way to do it. This is most commonly referred to as the "sub distance tweak". The process is actually easier than you describe above. You don't have to use the sliders at all.
Measure and import as you describe. Open the alignment tool. Make sure the Center is selected in the top box and the sub in the bottom. In the frequency response, click a point on the graph at your crossover frequency. In the alignment tool window select "Align phase slopes at cursor" and observe the change. Then select "Align phase at cursor". Compare this against the graph for the previous change. use the delay setting for whichever curve looks better.
You can get a negative value here and that's okay. It just means reduce the delay in the AVR by the amount indicated. You might have to do some math to convert ms time to distance. Divide the time by 0.89 to convert the delay from ms to feet.
Of course - it's totally okay to play with the sliders too, but I find the point-click method easier. Use it to at least get you in the right area and then potentially fine tune with sliding a bit.
@@jeffmery Cool and thanks..I know In past I tried one of those options but got weird values.. like 30 foot delay needed.
when I just used values from REW measurements (using again center as ref timing.) (value in measurement box on left side of rew where measurements are) it soinded correct and looked way better... was like maybe 12 feet adjustment not 30ish..
What are we looking for exactly when aligning with main speakers crossover?
are we trying to get a flat horizontal line over the crossover (let's say 80hz)?
Or are we trying to get a positive summation overall from 80hz down to 20hz like you do with aligning subs?
Thanks and sorry for silly question.
I am also confused what we are trying to see when aligning crossover..
@@desertplanet3320 No question is silly. We all started somewhere. When aligning phase at the crossover point, we are looking for a smooth frequency response.
Without getting too technical, there are very few room correction packages that take speaker phase into account. The two I'm aware of are MSO and Dirac Live with the DLBC add-on. I guarantee there are others out there; especially in the commercial or professional audio world. I'm just familiar with what I've personally worked with. The most common room correction packages like YPAO, Audyssey, etc definitely do not take phase into account.
If we DON'T take phase into account, we can end up with frequency cancellation at the crossover. This will show up as a dip in the frequency response centered near the crossover frequency. The exact center depends on the specific phase misalignment which is why we look at the "crossover region" and not the exact point (there's more to it than that, but that's good enough for now). By adjusting the timing of the sub, we can line up the phase slopes to eliminate this cancellation and improve the sound.
Human hearing is much more sensitive to frequency response than delay at these frequencies. The wavelengths are huge (12.5 ms long at 80 Hz). Conversely, as the frequency rises, we become very sensitive to delays, especially in the speech range.
Does this help?
Did you level match your 2 subs with REW first before using the alignment tool?
My subs were identical when this video was made so I gain matched them by setting their control to the same level. That works best when the subs are the same because we know they’re working equally hard.
If the subs are different I’d start by level matching them at the MLP. There’s a really good method for mixed subs called “distortion matching”. There is a post by user “DaveBoswell” on the MSO thread at AVSForum that details the process. It’s more complicated than straight level matching. However, it helps to ensure that one sub isn’t completely overdriven if they have very different capabilities.
@@jeffmeryYes, my subs are identical, HSU vft2-mk5. Thanks,still in the early stages of how this whole process works, received my mini dsp/umik-1 mic 3 weeks ago and trying to set these up.
Just one more question if I could,after you have your aligned sub file(2-subs only)do you then export that file to the mini dsp?
@@kenestra123 Gain and delay from the alignment need to be input manually to the correct output channels on the miniDSP. If you took the next step and EQ'd the combined response, then yes, the EQ export needs to be imported and loaded on the output side of the miniDSP.
Buddy please little help i have one sub and i cant find the best listening position i run audyssey and gives me bad sound some frequencys are low down theres any way to find best best position and take best results? With some how..🙏🙏🙏🙏🤔
I highly recommend what is called a "sub crawl". You really need a calibrated microphone and REW to get the best results. Folks will tell you that it can be done by ear, but it can't. Just trust me on that. What you will do is setup the mic in your main listening position. Measure the sub in its current position. Now, move the sub around the room to every position where you could place it, measure it in each spot as you go. You're looking for the best response you can get. Even with this work, a single sub will always have nulls (large dips in the response) that will be difficult to address.
The alignment tool can help eliminate a dip that happens at the crossover frequency after running Audyssey. It's very common for Audyssey to get the distance for the sub incorrect. When the distance is wrong, there is cancellation. It just takes adjusting the distance of the sub to eliminate it. Search for "Sub Distance Tweak" to see if that helps.
For a very comprehensive dive into subs with Audyssey, I highly recommend the first few posts in this thread over on AVSForum. Mike did a great job and it is very heavy on Audyssey: www.avsforum.com/threads/guide-to-subwoofer-calibration-and-bass-preferences.2958528/
Hi
Is there a SUM of a multi point measurements for this or is mso the best solution?
If you care about more than one listening position, MSO is likely the better solution. It is possible to use spatially averaged measurements from multiple positions within REW. This would require multiple microphones to be setup as well as the Pro license for REW to allow multi-input capture.
I suppose trace arithmetic could be used to get a spatially averaged measurement, but it would be very tedious. MSO would be less effort.
@@jeffmery will the mso do time align or do more a eq fix? (I think it is doing some sort of time align but im not 100% if it is any good, and ifi need to do the time alignment in rew first..)
@@niklaskarlsson236 MSO will also do time alignment. There is no need to do that in REW first. You will need to take measurements in REW to use as input into MSO, but that is it. Just measurements.
Hi Jeff - just wanted to ask - do you need to use an acoustic timing reference (e.g.left main) whilst doing measurements for the alignment tool to work...? Thank you.. best wishes..
Good question. Yes! You need a timing reference. When in doubt, always use a timing reference.
@@jeffmery thank you Jeff. Also, I wanted to ask you, which if these steps would you recommend for sub eq:
1. Sub alignment as above, followed by individual sub eq (flat target) in output section of miniDsp for each sub and a house curve as a low shelving filter in the input section of miniDsp...
2. Individual sub eq (flat target) followed by sub alignment as above and house curve as low shelving filter in input section of miniDsp...
3. Any other combination of sequences..
4. Or am I completely missing the point and barking up the wrong tree.....
Thank you very much for your time and diligence....Harsh
@@harshjain3801 I recommend your #2 approach. Be sure to EQ the combined response, and apply the same PEQ filters to all subs. Do NOT attempt to EQ each sub individually and apply differing EQ; you will have a bad time ;-).
I like to keep the house curve separate from the sub alignment and EQ as you have described above. That can be done on an input filter in the miniDSP or within your AVR's room correction software if it's supported. I've used both methods. I currently use the room correction method because it's just more convenient for me.
@@jeffmery Jeff - thank you really for your prompt and detailed reply. But then, I have this query - each sub has its own frequency response curve - would it not actually be better to eq them individually - according to its own peaks and dips, rather than combined sub eq applied to all subs - because some of the resulting filters might compete with the frequencies if individual subs....hope I am making some sense here...🙁
@@harshjain3801 Great question. This is probably one of the biggest points of confusion with subs. To put it simply, because of the frequencies they produce, subs don't behave like regular speakers. What we hear with subs is more the influence of the room and not the speaker itself. If you have identical subs in two different locations, then you should see this in their respective in-room responses.
The problem with this, is that because of the differing responses, the subs will end up fighting one another, rather than helping one another when they play the same content. When we align subs to play together, we essentially create a single "virtual" sub somewhere in the room.
It's this "virtual" sub (all of the aligned subs playing together) that we want to equalize and then integrate into our systems.
This is something that's easy to experiment with. Align your subs and EQ the "virtual"' sub. Measure the response.
Align your subs and EQ each one individually. Measure the response.
You could even go further and EQ each sub individually, then align and measure.
The first process will result in the best overall response.
This might make a good video topic.
Thanks great info
I *think* you misspoke regarding the +10dB gain. On the logarithmic scale, every 10dB doubles (2x) the volume or SPL, not 10x as you stated.
I'm no expert and it's possible you are referring to 'gain' as something other than perceived SPL
Good catch - I probably meant “+10 dB is 10x the power”, which is close enough to the actual number to be a good rule of thumb.
@@jeffmery you did say power after I watched again, my bad. Either way, a 10dB increase is immense difference.
Do you have experience with REW for 2channel listening setups?
I suppose my setup is 2.2 -- dual mono subs next to each LR speaker.
At some point i want to get a decent mic and do room measurements with software like REW, but I suppose without something like minidsp I'm regulated to make corrections or improvements via speaker placement or room treatment only. Still. I have hope that these can rack up significant gains with the right measuring equipment and application.
@@erics.4113 I haven’t specifically done a 2.2, but the physics of multiple subs and sub+mains integration are the same. The most important thing to be aware of is that the best location for you mains is almost never the best location for your subs as well. The advantage of separate subs is that we can place them in the room where they give the best overall response. That doesn’t matter if it’s 1 or many subs. Don’t be afraid to experiment with their locations. Human hearing cannot localize frequencies below 80 Hz so you’re not going to hear the bass coming from a different part of the room. In my experience, more subs can push the frequency a bit higher as well.
If your mains are capable, then you should have plenty of freedom to place the subs wherever you want. If you need to cross the mains over above, say 100 Hz, you might be able to start localizing the subs. In this case, you might be forced to have the subs near the mains. Just something to be aware of.
You are absolutely correct about running a mono signal (make sure it’s the same signal) to each sub. Going back to our inability to localize low frequencies, the concept of “stereo” doesn’t really exist. Running a mono signal eliminates some types of cancellation issues that we would not otherwise be able to fix very easily.
I cannot recommend a good mic and REW enough. Even without a miniDSP, it’s possible to make some improvements. Probably the most common issue to fix with just measurements is the cancellation that occurs at the crossover frequency in most setups. The issue is caused by phase cancellation which boils down to a timing problem between the sub(s) and mains. You can measure, then use the alignment tool to find out the change in the sub distance (delay) to fix the problem. Instead of aligning one sub with another, you measure your L+R together, then align the sub to that measurement at the crossover frequency.
Hi, how do you got this black background? How to get it?
Under "Preferences" -> "View"
@@jeffmery Thanks for the information 😊
@@jeffmery I’m into car audio enthusiasts and I want to learn REW can you able to share me some videos how to set the REW, measure the system and eq the system.
@@sainath8710 Setting up REW to take measurements is exactly the same. I have a series on getting started with REW on MacOS here on the channel. Unfortunately I'm not a car audio guy so I'm not up to date on all the things available in that world. The concepts should all be the same though.
@@jeffmery what you said was right the procedure are same, only thing is in car audio we do time alignment for all speakers because we are sitting offset not in the center and the environment is different.
hows the setup with only one sub?
I don't mean this to be a crappy answer, but it really depends on the room and the number of seats you care about. You can get good results with a single sub if you only care about a single position. You could place it near-field right behind the seat and get just about a perfect response (up to the capabilities of the sub in question of course).
When you put subs somewhere else, the room creates the peaks and nulls you see in the responses above. The peaks can be knocked down with simple EQ, but physics won't let us EQ a null. For nulls, the only choice is to move the subs and seating or to add more subs to help even the response.
Doesn’t MSO do this?
Yes it does, but some folks don't need to use MSO to get great results. The rule of thumb is ...
If you care about a single listening position, you'll get excellent results, in less time, with less effort using the REW-only method of aligning and EQ-ing subs. This video is the "alignment" part of that process.
If you care about more than one listening position, like one or multiple rows of seating, then MSO will produce better results.
It didn’t look like the alignment tool trace when I applied the delays to minidsp for me a ran a real sweep. Oh well
Did you start from a known state (i.e. everything zeroed out)? Did you use an acoustic timing reference when taking your “to-be-aligned” measurements? Did you move the mic between taking the source and aligned measurements?
Are you using Java or ASIO?
Neither. I'm on Mac OS. My understanding is that Java has become the way to go on Windows. I have a Windows laptop around here somewhere. I really need to play with REW a bit more there. Sorry I'm not of more use!
Do you start with the farthest sub first?
I’ve experimented with order hand haven’t found much difference in the results. Just know that if you start with a random pair of subs, the best looking result requires negative delay on the bottom sub, that’s the same as adding the exact same amount of positive delay to the sub on top.
@@jeffmerycan you explain more on this? Or is there a video on this I can watch?
@@kystarnes8486 On the "negative gain" no videos that i'm aware of. "Top" and "bottom" sub above refer to the alignment tool window in REW. You are aligning the sub in the bottom window, say sub "B", with the sub in the top window, sub "A" by adjust the delay for sub B
If I get the best result with positive delay on sub B, great. Implement it and move on. If I get the best result with negative delay, what do I do?
Let's say -2.22ms of delay on sub B is best. That means that sub B must arrive 2.22ms before sub A. We can achieve the same result by implementing 2.22ms of delay on sub A.
Does that help?
Something is off with REW measurements - look at SPL charts, nearly 90 SPL at 15 Hz?? No sub can do that, at least not those sold for HT.
90 dB at 15 Hz isn't very much. There are a ton of commercially sold, home theater subs that can do that. Have a look at data-bass.com/#/systems?_k=3k5xnu
Ignore anything with "DIY" in the "Manufacturer" column. Those are home built subs and you mentioned "sold for HT" so we'll keep the discussion to commercially sold brands only. Much of that data is now a few years old so it's missing data on newer, improved models (like the SVS PB-16 Ultra I mention below). The person doing the independent testing has moved on to other things. However, you'll see plenty of well known home theater subs in there like HSU, Rythmik, PSA, SVS, JL Audio, etc.
The SVS PB-13 Ultra, which is several years old and has been replaced by the much larger and much louder PB-16 Ultra, hit 89.3 at 12.5 Hz and 105.1 at 16 Hz.
For reference, my current subs are Rythmik FVX15 subs. They have not been tested by the site above, but their big brother, the F15HP has. The F15HP hits 98.5 dB at 12.5 Hz and 104 dB at 16 Hz in single port mode. Per Rythmik, the FVX15 is down 3 dB across the bandwidth from the F15HP. I could theoretically hit 95 dB at 12.5 Hz and 103 at 16 Hz. I do not have my gain knobs on my amps turned all the way up so there's definitely head room to be had.
Definitely confusing, I would love to calibrate my subs but this is really confusing every time I watch a video on this and I never understand it
Are there particular pieces that you don't understand? Perhaps start with questions that you have at the beginning and we can work our way through.
@@jeffmery thanks for your response and willingness to help
What equipment or devices you need and what software you was using?
After setting the above items up reading the software and what to look for is when running test on the subwoofer in reading the resultsso confusing
@@alambe7275 There are a few things you'll need. To take measurements, you need a good microphone, a laptop, and measurement software. I recommend the miniDSP UMIK-1 for the microphone. It's reasonably priced and available just about everywhere in the world. For the measurement software, I recommend Room EQ Wizard (a.k.a. REW). It's freeware and very very capable. I also highly recommend a boom-style mic stand to hold the mic while taking measurements.
The UMIK-1 is about $100. A cheap boom-style mic stand is about $30. REW is free, but consider making a donation to support the person that develops and maintains it.
Depending on your equipment it might be possible to implement some things, such as delay, with what you have. If not, or if you want more capabilities, I highly recommend the miniDSP 2x4 HD. This device plugs into the subwoofer or LFE output of your AVR, then your subwoofers plug into the miniDSP (so it sits between the AVR and the subs). The 2x4HD gives almost total control over the subs and allows for very fine grained bass management.
The miniDSP 2x4HD costs about $250 shipped directly from miniDSP.
HUH- odd I did everything to the T showed 5DB gain 5hz-10HZ and 3-5Db gain the rest, implement and it was worse lol fml.
How many subs? How are they connected? Where are you setting the delay? The hard part is that the changes are relative to the measurement conditions.
An example might make it easier ...
Say you're using an AVR that has independent delay settings for each sub. Let's say those delay values are 0 ms and 4 ms for sub 1 and 2. If you measure like this (meaning with the delays active), the change for sub 2 is relative to the 4 ms of delay that was present when the initial measurement was taken.
In that example, if the alignment tool shows 1.7 ms for sub 2, that 1.7 ms needs to be ADDED to the existing 4 ms. The total delay becomes 5.7 ms for sub 2.
I hope that all makes sense.
@@jeffmery Nah I always set AVR pure straight before any measurements so its raw UN-molested. Always in the Minidsp output stage delay. 6 subs two cabs haha AVR-MiniDSP to External AMP each individually powered. sad because I was looking forward to that 5db gain. measured all three and then as one for baseline. After "delay" i had worse response then overall then without it. After delay i sweeped all three as one and was worse then before i added delay. Makes no sense haha. Ill try again tonight
Align your subs by 'line arraying' them... you dont need to separate your subwoofers for coverage or room correction. jesus,
You're right that time alignment isn't always needed. If the subs and seating are properly placed in the room, then there can be a naturally even response. However, that does require the subs to be physically separated from one another to reduce modal interactions in the space. Line arraying subs doesn't work in acoustically small spaces (less than ~1,000 cu. m. or 35,000 cu. ft.). Here are some good resources that explain the issue and recommendations to address.
Welti/Devantier: audioroundtable.com/misc/Welti_Multisub.pdf
Geddes: www.gedlee.com/Papers/OptimalBassPlaybackinSmallRooms.pptx
CEDIA CTA-RP22 (Section 7 specifically): cedia.net/advocacy/rp22
(Last one behind a soft-wall requiring an e-mail address. Sorry about that.)
I just watch this video and my head is still spinning. th-cam.com/video/DofUiHz7Fu4/w-d-xo.html
In a stereo system with 3 subs how would you phase align the subs to the main speakers?
There are different schools of thought on sub alignment to main speakers. Part of the challenge in using impulse responses is that they are biased to the arrival time of the higher frequencies. REW typically sweeps from low to high so a sub can show an odd impulse response making it difficult to align. This can some times lead to incorrect alignment and poor sound. This is what OCA is discussing when he talks about "cross correlation sometimes gets it wrong". The upside of this method is that typically the arrival time of peak energy is pretty well aligned.
If this were me, and I only cared about a single seating position (as OCA does), I'd start with the alignment tool. I'd look at the combined subs and the L+R speaker together and see how I can get the best summation. I tend to favor as little delay as possible to get things aligned. Experiment with inverting the subs and delay to find the lowest amount of sub delay to give a good integration at the crossover.