Total thumbs up OCA!!! I've learned so much by watching and working through your videos. I thought the SOTA walkthrough was good. But this was unbelievable the different that I could both hear and measure. Thank you so much.
Two quick notes: In REW export all to text, if you change setting 'Export Text Delimiter' to Tab, the numbers will go directly into Excel columns without having to monkey around with the Excel delimiter. Also, if you adjust distances and levels in the Audyssey MultiEQ app instead of in the receiver web portal, although more cumbersome, they will be saved into the .ady file and you won't lose them if you change between different calibrations. Cheers and thanks for another great tutorial!
Ernie, I'm guessing you mean make the adjustments in the the json editor? There are six decimal places for both trim and distance. For trim, did you try entering something other than half dB increments? Wondering if it will accept levels that aren't .5 or whole dB steps.
@@vincentdemarco6374 no I meant inside the audyssey muiltEQ app on your iPad or other device. When you open your calibration file, one of the submenus is (something like) Speaker Results, inside you’ll find crossovers, distances, and levels. If you input the values here they’ll be saved into your calibration .ady file
Here is my feedback having completed this process. Firstly great learning experience. Going through things a few times makes them easier to follow and dots start connecting. The graphs looking stunning and I was able to follow each step all the way. Then I gave it a listen. My first impression was that the speakers seemed to have come alive a lot more and sounded more immersive but sounded a bit harsh in some instances and the bass seemed weaker to me and also little exaggerated in the upper bass frequencies. I went back and forth with my old settings and the new ones. Since I have a minidsp and a Denon 4700 which has two speaker presets, it was a lot easier to compare. I found the bass sounding better in my old setting so I went back to it but applied some of the learnings from this tutorial to it especially the correction of the impulse responses. That made a big difference and everything sounds better than before. So I ended up with new settings using my old and new and it sounds great. What was interesting was that the FR between the latest REW and the early access differ. Not sure why but they do. Also for others the controls to see IR distances on IR tab is Ctrl + right mouse key and drag. Thanks OCA
First of all thank you and congrats on your achievement. REW has been going through an extensive development stage recently and some trace arithmetic basics have been changed although there would not be a difference in the final filters produced.
I have a Denon x3700h andI ended up doing the same as you admittedly, after praising this mothod so much. My results were similar - the mid bass was way too exaggerated and the rest of the speakers sounded like they were crossed over at 200hz, ie; no bass even where the speakers can handle 60hz or even 80hz. I applied the impulse response to all speakers and now my original Audyssey calibration sounds much better. Surprisingly though, the biggest improvement came from turning Loudness Compensation off in the now perfected original calibration. However, after finding the original calibration sounded better, I still learnt so much from this tutorial! I feel like I am now an advanced REW user! Thank you OCA! I hope you can find why the calibration (after following this tutorial) ended up sounding the way it did. I followed it exactly as instructed. I can't wait to try your next tutorial....:)
You're channel is really exceptional in regards to range and depth of the topics. But also practical and precise. Awesome work and I hope you get a much bigger audience soon. Thanks!
Hello. First of all, thank you very much for providing the calibration file. I wanted to make a comment about how to use it: I was trying to use the Audyssey mic to check the frequency response of my system playing pink noise and using REW's rta. And I believe the calibration file for the Audyssey mic you provided has the positive and negative values inverted. My system was already calibrated following the avr's built in audyssey setup procedure. First I measured with no calibration file at all, and I saw a frequency roll off starting at about 7KHz as expected according to what you showed in your other video about the hidden calibration file of the Audyssey mic. Then I loaded your provided calibration file and took another measurement. Instead of getting a flatter response, the frequency roll off starting at 7KHz was stronger. So I opened your calibration file in notepad texteditor, changed all negative values to positive and changed all positive values to negative, and saved with a new name. I loaded this new calibration file into REW, took a new measurement and got an almost flat frequency response. (The system was set to "flat" multeq profile, not "reference"). The calibration file for REW should show the mic output level at each frequency, not the required compensation. That is to say, if the mic's response is rolling off after 7khz, the values in the calibration file after 7khz should be negative. I hope my comment to be useful for other users who are in the same situation as me: not having a calibration mic and wanting to use the audyssey mic in REW. (Forgive me if I did something wrong, perhaps there is a way to load the provided file in REW so that it works as it should and I am not aware of it. I found that the file works inverting the values and I wanted to share it. Also forgive me if you or another user already said the same thing about the calibration file before).
You were right, it was having the inverse effect on the measurement. I've now updated the calibration file in all videos with the inverted one. Thanks for your attention!
Thank you OCA; great video again🙂 i want to try that; in the beginning you mentioned that also MultEQ-X user can apply this method. But how to do this, because the .mqx-files are different and not JSON formated (the Data-section is another format i think)!?
I don't use mqx and not sure what the format of an mqx file is but at the final stage of this method instead of inversion, you can apply REW'S auto EQ to the responses and transfer these filters to multeq-x.
@@ocaudiophile Okay, thank you! But then i think better be careful with statements like this: 'We don’t need ‚MultEQ-X‘, but MultEQ-X users can also easily implement the method', because as it turns out the method can not be used!
@@didi_p7756 There's nothing too wrong with that statement. The method can be implemented for mqx with REW's auto EQ instead of inversion.I have detailed videos on how to use REW's auto-EQ with these receivers.
@@ocaudiophile What do you think about the following workflow for MULTEQ-X users (assuming, as part of the measurement process, MULTEQ-X gets the speaker distances/levels correct) 1. Set REW preferences as prescribed in this video 2. Take all measurements (I.e. three measurements per speaker) with Audyssey enabled how without any Audyssey filters applied (note: MULTEQ-X allows any Audysseyfilters to be disabled via excluding’ ALL measurements in the app 3. Take RMS Average of each set of speaker measurements (I.e. RMS average of the three measurements) Question: Is Vector averaging preferred??? 4. Use REW to EQ the response of each speaker’s RMS average 5. Import REW correction/eq filters into MULTEQ-X app 6. Upload to receiver and call it a day!
Thank you so much! I can attest that this is the best my stereo speakers have sounded (and was a lot faster than a full Atmos setup). I was previously using the room corection provided by my WiiM, which was an improvement over baseline. But after disabling that and trying this, it sounds fantastic and now my movies and games can benefit as well. Well done!
Your username is absolutely Perfect Sir. Excellent instructions and great job describing what you are doing and why. Thank you for sharing your knowledge and expertise with us.
Finally! My system is a new one with you. Especially with my low end Denon 1700h without multieq xt32 the difference is huge in bass management. Thank you so much. Somebody needs to take the method and create a blog post out of it. Because a video sometimes is not the ideal medium while working on such things . Much love ❤❤❤🎉🎉🎉
i have started writing down from your video the steps to do. but still ongoing. a nice blog post on medium or somewhere with nice tables regarding settings in REW, direct download links, hint blocks and so on would make it much better to understand. it would be also possible to make first a 1:1 tutorial out of it without any deep knowledge background which can be collapsed from special blocks. what you think? @@ocaudiophile
You've my full support but some techniques have evolved and there are automation scripts in the immersive audio and especially Audyssey ART videos. You might wanna take a look at them as welll.@@Squizzy911
With the instructions of the Dolby Atmos 9.1.6 video I was able to get also a measuring of the Atmos speakers for my 7.1.4 setup on a MacBook Pro, using VLC media player on my Mac. Afterwards quit simple choose the wav file of the sweeps in REW, start measuring and play in VLC the speaker you are needed, every speakers has is own chapter, great work. Now this evening I can go further with the instructions of this video with my 36 measurements. Great to do this job with a Mac, I am not an experienced windows user.
Very strange that Mac users too much problems have with REW. And there is more an advantage for a Mac because you need no special drivers. I use Ventura or on another system Sonoma, and the latest REW from the REW-website. On every mac will plays VLC plays and you give the right files and sweeps in you 9.1.6 video, maybe you can these also add to this video. Furthermore a Mac User must setup his mouse as a secundary mouse. So you can speak more optimistic in your videos for Mac Users!!!
I spoke to John about the inversion with [A] / [B] magnitude division and it is working correctly. What threw me off was that the calculation in REW displays as the differences on a 0db line. I was expecting it to visually invert the response. However, If you export an inversion and pick an area that looks off in REW. For me its was 16K. Look at the difference in the measure response vs the target curve. Mine was -2.5db in REW. This means that the inversion should be +2.5db and in the text file it was. The biggest thing is the way REW is displaying the correction. It only displays it as the differences on the 0db SPL graph. I had to get my head around this. In the end, A / B magnitude division does work as intended, and the 0db line becomes the target curve.
OCA, thank you for this detailed info. I'm completely new at this hence have a question relating to the beginning of your video where you transfer the Default.ady file to your phone then send it to the AVR, I assume via the MultiEQ editor. I did this and my AVR asked to save it to one of the two presets (I saved to preset 1). I assume this is correct? My real question comes at 6:48 of the video when you then use some sort of in web interface protocol to interact with your AVR. What is the purpose of this? Or can this all be done by directly using the AVR menu via the remote control? Are you using this web interface to simply show your viewers the changes or is there something more behind this action? When I selected my new preset 1 setting and viewed via the AVR (using my remote) the changes you note were there except all my distances were not changed to 0.04m ... they remained at the original Multi EQ actual speaker distance measurements. I apologize for this newby question.
You can use whichever preset you like. I keep one preset for full manual calibration and one preset with Audyssey and Dynamic EQ using this method. The web interface commands can all be done from the receiver menu/remote, this is just a short cut but also a lot easier to change settings from your PC instantly. Some newer receiver models use a different setting in the ady for speaker distance. You can manually reset them in the reciever setup menu or in the online Json editor before you save the ady file.
Thank you!! With all the videos on your site which one do you recommend I use to maximize my 7.2.4 system (2021 AVR Denon X6700H)? I first experimented by adding the Harmon Curve (as outlined in one of your videos) … which achieved a much improved response. Of course I now want to see what more we can do with the system. Cheers and thanks again for all your sage advice.
FYI, I tried numerous times to download the REW V5.20.14 file, it appears to work, however it never includes the “house curve” tab under preferences? When I download the REW V5.20.13 it does include the “house curve” tab under preferences. Am I missing something? Can I use the .13 version and still successfully work thru your video(s)? Keeping in mind I’m a REW infantile. 😂
Last questions and I won't bother you anymore. 😁 I decided to try your "Supreme Audyssey Calibration" video (assuming that might be what you advise further to my question above?). I've updated all the preference tabs info. per your video including adding the mic calibration file as I'm using the original Audyssey mic. At 14:00 in the video I'm trying to run the measurement but it gets hung up at "10% Waiting on timing reference" and the graph at the bottom left corner simply keeps scrolling? When running the Check Levels I get the message "Ready to measure..." with the the message "Level Low, -61.2 dBFS, 78db SPL" . FYI, given the Level Low message I tried turning the AVR volume up and still got the message "Level Low, -41.7 dBFS, 98 dB SPL" but it did not help. I then also experimented in the Preferences area "Soundcard" Tab to see about calibrating my soundcard (not sure if I even needed to do that?) and could never get the Input and Output levels even close to matching one another as the volume was waaaay to loud at the original Sweep Level setting of -12 dBFS. I had to adjust the Sweep Level setting to -30 dBFS, in order to turn up the AVR volume enough to match the input/output reading and not blow my speakers (which level was still very loud). Lost in REW Space ... hope you had a great day today and can help.
@@ocaudiophile i have the x3700 denon BUT in the web interface i don't HAVE the setting "LOW frequency effects" !! am i doing something wrong or maybe some models don't have it ?! update : i had to play a movie with LFE to be able to SEE that setting !
i got stuck at the default.ady file transfer at 6:37. multeq app says "the specified file is not located in the devices local storage or is not a proper curve file". Windows sees it as a ady-file but nothing helps here. could you maybe help with this?
Hello. I did the JSON editing part. I opened it back in JSON to make sure it was showing the new values. Still after importing it into my AVR, I checked the levels and distances and they were not as expected. Any idea what could be wrong? Thanks!!
Then it should work fine. Some XT users encountered similar problems. Have you converted responses of ALL speakers including the subwoofer to perfect (1,0,0,0....)?
Please a little clarification for me, At around 51:00 I made copies of the sweeps, applied Align Spl (for speech), and then compared sweeps with their copy and applied Spl Offset. Do the new Spl offsets replace the AVR Speaker Levels previously set or do you add the Spl Offsets to the AVR Speaker Levels? I tried both and the results were no better than the first applied inversion (before 51:00). I've previously aligned my subs following your tutorial. Got them flat. The distance from the sub to the center speaker measures between 7-9 meters depending if you measure from the bottom or top peaks. The center speaker distance to the MLP is 1.91 meters. So my total distance is approximately 11 meters to add to the subwoofer distance in the DENON AVR!? 7.90 meters is the highest distance that can be entered. I'm confused.
I understand the confusion. Actually REW now shows the SPL offset applied to a measurement during "Align SPL" process in the Info window under "Align Offset:" so you don't need to make copies. just check these for each measurement. I explained this method in detail around 08:35 minutes into the following video: th-cam.com/video/4gVUdArG7oA/w-d-xo.html
@@ocaudiophile In the Audyssey calibration for Dolby Atmos Music video you apply Cross correlation alignment with the Vector Avg Sweeps and Actual Time(distance) Aligned measurements is that new? Should it be employed here also?
@@CarlosGomez-THX_1138 It takes the precision one step further and helps in finding the correct crossover points. As you cannot adjust speaker distances with high accuracy in the reciever, this extra step helps REW to use the actual speaker distances and calculate phase interactions with the sub better. It's up to you to try and test the results.
In REW, after applying Align SPL @1000Hz and entering the Levels in the AVR, can we then apply, in REW, Align SPL @200Hz and adjust the Levels in the AVR? or do we apply the Align SPL as depicted in this video?
Hey OCA, how have you been? I've noticed that A over B magnitude division is overcorrecting on the EQs I'm doing. Do you have any idea why? I've seen that treble sometimes needs an increase of 1.5db, but the A over B is increasing by more. Like 2db. What could be causing higher numbers than needed? Is it also taking into account reflections in the sweep? Can I remove that? Or just measure with an RTA?
I guess these deviations are caused by the complicated maths behind the division operation between two FFT's especially when it's not over a flat line but a target curve. The beginning and the end needs some mathematical gizmo to be handled. I use direct 1/A inversion lately over a flat line for that very reason.
@18:02, if the default is set to my Marantz AVR, is there any special way that Windows Media Player (or anything else) needs to be setup in order to play the height channels? Or will it play them natively?
@@ocaudiophile Thanks! How about when I select the configuration for the Marantz in Windows? 7.1 or do I choose another option? I think there's one that shows up as Atmos, but the diagram disappears when I select it.
@@ocaudiophile This is definitely still not working for me. My default is set to the Marantz AVR and stereo configuration, but the height files are playing out of fronts left and right.
Thank you very much for taking the time to do this. My receiver only has multeq (not 32 or xt) so I was intrigued by this video. Unfortunately, after completing the process, I got no frequency response below 120hz whatsoever. My sub was set at -12db to hit the target curve and even after boosting it to +12db in the audyssey settings and turning the gain on my amp up an additional 30%(maxed out) there was maybe 1-2mm of travel from my subwoofer and no bass from my bookshelves. I intend to redo the process from creating new corrections forward, potentially with another curve, but I am curious if maybe multeq does not have the resolution that multeqxt32 has and as a result, some numbers may not transfer to my avr. For instance, when adding the 241 points to the custom “target curve points”, could audyssey be ignoring 100 of them or something to this nature because it is the lesser audyssey version?
I don't think number of custom target curve points is a problem for the type of MultEQ as long as the app accepts them (I think the limit is somewhere near 65,000 points). However, MultEQ uses significantly less number of FIR taps compared to XT32 while generating FIR filters which will effect LOW BASS correction very much. In other words, Audyssey may not be able to correct properly to the target curve it's pointed at especially below 100Hz.
@@ocaudiophile Quite appreciative that you took the time to reply. I hope i dont seem as though i am asking for help diagnosing my own issues here. I am simply trying to contribute issues i am having in the event someone else runs into these problems. I am now concerned that the measurements i took with REW were being influenced by the existing Audyssey profile's corrections ( i had a traditional audyssey correction applied prior to starting this guide). I will start from scratch with a factory reset and report back. Once again, thank you.
Hi OCA, In the video at 43:57, I don't see the category "customTargetCurvePoints" in my .ady file in the JSON Editor. Using a Marantz Cinema 50. So I'm not sure where I should input the REW response. Thanks in advance for your help.
Is your MultEQ Editor app up to date? Custom curve points should definetely be there. There have been people who applied this method successfully with Marantz CInema 50.
@@ocaudiophile it was some bug with the exported .ady file. I exported again and the setting was available in JSON Editor. Finally finished all the steps in this method and I can immediately notice a significant improvement in the height channels. However, the bass response seems to very low. I have double checked that LFE level is set to 0 and not at -10. Still can't figure out what's causing this. In REW, the subwoofer response looked fine.
Loudness management is on in most models by default. It could be a reason but best way is to remeasure speakers and sub with the new settings active and check in REW what's going on.
@@ocaudiophile thank you! When measuring sub impulse response, the distance adjustment from Center Channel peak impulse to Subwoofer peak impulse is showing to be 12 meters (!!). Obviously this cannot even be configured. So I'm not sure if I'm doing something wrong. Is there any other way to time align the sub?
@@Judddernaut are you using minidsp for the sub? It usually introduces such delays causing 12m distance requirements. You can align subwoofers first peak rather than the highest peak with the acoustic reference speaker instead.
amazing tutorial. many thanks! my 2 subs on mini-dsp seem to arrive much later than receiver can set (+6.8m to center). in this case would I just set to maximum available distance? or is there anything smarter I can do?
Million dollar question ;) There's not much you can do to compensate for the time delay introduced by MiniDSP in the reciever. *Play around with bass sync setting if you've a Denon (Marantz don't have it), that might add some delay compensation. *You can switch to IIR filters (biquads) in MiniDSP if you are using FIR filters (taps) which introduce delays to the system. *And last resort: You can align subwoofer's previous impulse peak instead of the highest peak to the center channel
Thank you! I've waited a long time for someone to do a tutorial manually tuning HT/atmos speakers. Bless you! .....Initially, I tested the sweep files on all my speakers and determined that two of my Atmos speaker's high frequencies dropped precipitously around 1000Hz. Thank you for the files to test them. In the Denon AVR-X4700H Low Frequency Effects are displayed when "This item can be selected when a Dolby Digital or DTS signal or DVD-Audio is played." I did not set this initially and continued with the tutorial with 7.1. Should I have played a DTS movie and then set this at the beginning? What do I set the Reference Level Offset given I did not set Low Frequency Effects? Thank you for bringing my Home Theater to life. My 7.1 sounds so good I wonder whether I needed Atmos.
I suggest you use Denon's web interface to change its settings. WHen you start palying any of the sweep files with Windows Media Player, all these settings will sho up in the web interface and you can adjust them. RLO has nothing to do with LFE. LFE -10dB offset is just a temporary setting to see the sub response where it really is relative to other speakers. RLO is a setting which adjusts the strength of DYnamic EQ.
Thank you for the excellent video. In the narration you state midrange compensation should be off, but in the JSON editor it shows midrange compensation box as checked (true). Should the box be checked or unchecked? Thank you
Uncheck it for every channel except for the subwoofer channel where it's irrelavant but may help reduce noise in the upper frequencies. The much newer Audyssey ART video is easier to implement btw.
I have taken measurements at the three points, and am now ready to time align them. At about 20:30, you mention a new REW feature called something like 'copy selections to other overlay graphs'. How are you getting that pop-up that contains that new feature?
@@ocaudiophile I must not be on the same version of REW as you, since that command is not available when I right-click on that bottom window. I am on the released Mac version of REW, version 5.20.13.
Hey man. I finally got around to giving this a shot. My results were pretty good. Only thing I changed was I made the target curve give me way more bass (about 10db to match the uncorrected curve of my speakers) and I made Audyssey not do any correction after 10khz because the boost of highs was way too fatiguing in my room. Using this method has made my home theater sound like a real theater. It doesn't sound like speakers, it sounds like sound. No it's not perfect but nothing will ever be perfect in audio. Thanks for the video!
@@ocaudiophile Yeah man you are doing some really good work. Do you take your measurements with sofa cushions on or off? I think the high end boost might be because I measured with my sofa cushions on but I saw another video from HD MOVIE SOURCE saying to measure with the cushions off.
@chadkirk925 The differences are too small to measure precisely. I'd rather put effort to place the mic tip where your ears will exactly be during measurements and leave the room exactly as it will be during film watching for better results.
This is really great. I appreciate you sharing your knowledge!! Quick question - do you recommend doing this for 2.1 channel music listening? I have a 5.1 system with Audyssey default calibration. I disabled the mid range compensation using the app because that sounds better to me. I use it for movies. But for music I switch to L/R Bypass and 2.1 stereo mode because that sounds better to me. It comes close to Pure Direct on the Marantz 5015 receiver but a bit better due to the SVS sub filling in. I would love to improve the music experience but I need to set up REW and buy the umik-1 microphone
You can use Audyssey's microphone and REW is free. I extracted the calibration file for it (check my latest video's description link) and its results are quite satisfactory and timing is better than a USB mic. With proper sub-main time alignment and graphic EQ, you can get nice stereo sound.
Thanks for the idea@@ocaudiophile. Unfortunately I cannot connect the Audyssey mic to my Macbook Pro as it does not have a mic input. Neither does my Windows Thinkpad for that matter. I will order the umik-1 in a few days. So you do recommend doing a good calibration for stereo listening
Especially if you have easy to drive mains, you could get good results. With higher end stereo speakers, these AVRs are not powerful enough in general. Good speaker placement and descent DSP always prevails though.
@@ocaudiophile I have B&W 683 towers rated for 89dB sensitivity in the front. Marantz 5015 driving them. I will get external amp later. 80% music 20% movies. I am happy with music just want to see how it can get better
First, thank you very much for this guide. I am eager to try this for myself. I have one questions: can I do this process without using DEQ? Different movies are mixed differently and for compressed audio I have to turn up the volume which would mess up DEQ, right? And dont we have to SPL all speakers to about 75db or so if not using DEQ? And can I measure and EQ two subs with REW and MiniDSP before wallthrough this guide? Or is it better to use your Sub Alignment and cross corelation video?
You should measure and create filters for your system at the volume you like to listen to it. If you take these measurements with DEQ on, DEQ will then adjust the calibration for louder or dimmer volumes on the go so it's useful for the problem you mention. It has some odd adjustments for surround and height speakers, etc. but still, it's better than nothing. If you are using multiple subs with a MiniDSP, it's better to time align the subs first using the sub alignment video and dial in the delay settings in MiniDSP.
Wow, this makes a ton of difference! The clarity in the sound is much improved, thank you! However, I think the sound after adjustment lacks a bit of 'punch' to it. Is there something I should be doing different or maybe go for another target curve? Is there one you can recommend to try in that case? Keep up the good work 🙌
Great video, while thinking this through I wonder how you adapt for custom curve in the bass region. Will you just create a target which will not be just "flat" harman curve, but will rather have a little bassboost around 20-100hz? I remember the harman curve you used in your "Audyssey hacks for epic Dolby Atmos" Video had a slightly different harman curve with a little bass boost. So what is the correct harman curve, or to rephrase it, did you edit the harman curve in the former Video more towards your taste in the bass region?
The one in the video is standard, only with extended frequency range. The bass boost will depend on the windowing and to some extent smoothing used on the response. The method in the video yields a very satisfactory bass when applied.
very impressive, thank you very much, another step up to the moon :) by the way, for the LFE cut off at 120Hz, it is not related to setting into the AVR (LPF for LFE) ? and an other thing for the filter curve generation, MultEQ-X app offer the possibity to import REW fiters, it say "EQ is limited to 10hz to 20khz, -10db to +10db, and Q 0.1 to 10" do we have to make sure that the generated filter curve is not going above that ? and thinking if it is better to create a filter curve from the EQ by using a generic filter that match this parameters ?
No, it's not effected by the LFE cut off in the receiver (that was also my first thought and I tried increasing it to 250, no change in response!) You can't replicate these inverted responses accurately with EQ but you don't need to. In REW EQ filter types, choose "Generic" as manufacturer and "Configurable PEQ" as model. Then you can set all these limits for MultEQ-X you've mentioned in the configuration and REW's auto-EQ will "match response to target" within these parameters. Then you can export them to MultEQ-X. Don't forget to keep the target level same for all speakers after initially calculating it as explained in the method.
@@ocaudiophile one little tip about REW zooming, it focus on the cross in the graph, so make sure to clic on the impulse start before cliquing on zoom+. I say that because it hapen to me and now I figured out it helps
I am afraid no. Macbooks don't support bitstream output of DD+ over HDMI, they can only decode to LPCM. I heard that some people managed to do it with some paid software but you need to check the forums. I really am not a Mac guy.
You can however switch speaker cables (not so hard if you're using banana plugs) and use one of the surround channels to play from all the Atmos speakers. The correction files will be accuarte. For timing, you will need to do a calculation based on the distance of the speaker you used
At the 23:30 point in your video, you are going to vector average front left three measurements. You also included the copy of the time-aligned measurements. Why did you include the three original measurements, if you had the time -aligned version?
All 3 measurements were already time aligned with cross correlation. The 4th is a copy of the central measurement and a copy of of a time-aligned measurement is also time-aligned. The reason I added the central measurement twice to the average is to give it more weight in the average. The reason for evarging them is to find the true speaker response in a larger area.
Hello, OCA. Thank you for all you do. I am really enjoying your channel and learning a ton. I have a question and hoping you could answer it... If my peak impulse on both of my fronts is >0 and my peak impulse for my center is
oh, btw, I have my fronts connected to a 2 channel amplifier and my Center channel and surrounds connected to a 5 channel amplifier. The Center peak peak impulse is 0. i'm just curious if that is an indication that the speaker is hooked up incorrectly.
It doesn't necessarrily mean that but if they are the same speaker brands it's quite likely. Best way is to compare Overlays/phase response of center with the rest of the speakers across the frequency bands by inverting polarity of the center speaker. You can do that from within REW without changing wires and decide which polarity is correct.
Hey, thanks for this great tutorial. I noticed that the amplitudes of FL and FR are exactly opposite to the rest of the speakers. Do I need to change the plugs on my speakers? Is this the fabled wrong phase? I use a down fire subwoofer which according to impulse response should have a distance of 11 meters and the curve looks more like a mountain without a plane like yours. Is the sub wrong or should I try it in frontfire? I can not set 11 meters distance in the receiver nor do I know later how I want to set the level with the curve of the subwoofer it looks like a mountain with a peak as I said. Thanks in advance for your help
Compare the overlays/phase graphs of front speakers inverting polarity of one of them (REW can do that with one click) and phases match better inverted, then do so physically. It's uncommon front speakers having opposite impulse peak direction but not impossible.
Thanks for your answer. I have switched the cables and now every amplitude shows in same direction. Or is it ok when some speakers have inverted amplitudes? I thought this causes bass problems
Hey, sorry to bug ya again. With the subs - am i supposed to align to the target curve while i have the -10db LFE in the AVR? So when finished, the subs will be 10db hotter than the target curve? Primary reason i ask is this is causing me to run my subs awfully hot. Granted, i am calibrating pretty hot overall, target curve is at 76.4db. But if i use the -10db, im having to keep my sub AVR setting at 0 and my three PB1000 pros are running at -6db, just to hit the harman curve. Or should I be matching at the LFE 0db. It could be just because of the type of subs i have (hindsight is 20/20, should have gone with at least the 2000 pros).
If your subwoofer measurements range is 0-24000Hz like other speakers, you can leave LFE setting at 0. -10dB setting is only useful if you're measuring subs 0-200Hz because when range is low, the receiver thinks it's an LFE only signal and boosts it by 10dB. Btw, in the latest early access version of REW, you can apply -10dB to LFE automatically during measurement, you don't need to change anything in the receiver.
@ocaudiophile Ahh. Got ha. I've been leaving the sub measurement at the same range as the speakers, no wonder I've been having to crank up the volume!
Hey OCA, trust you are well. Got a question on the „determine crossover frequencies“ part of your video. I own 2 pairs of atmos upfiring speakers and I am wondering how to find the correct crossover. Should I vector average TFL and TFR and then trace arr. (merge A to B) with the vector of FL and FR or with the sub? As far as I experienced the atmos effect is better when using a high crossover (e.g. 180). I read somewhere that some bass management systems route in a 2 step way. E.g. Front Height speakers signals below crossover 180 being routed to fronts which are crossed over at 110 hz from there to the subwoofer as usual. What do you think, thanks Sir?
Normally, they should been crossed over with the sub and the method you explained shluld work well. I haven't heard an information on two step crossover for height speakers. What receiver are you using?
Thanks. I use a Marantz 7706 processor. Have read somewhere about the 2 step crossover in an older interview with Andrew Jones since he was working with Pioneer. Besides that I found out measuring the heights as „large“ it’s bass response is kinda same than the front/rear speaker it sits on (measured as large as well -> only looking at the bass region below 200 hz or so). This tells me that the processor is allegedly doing some internal crossover to avoid overdriving the height speakers. When measureing the heights crossed over there is surprisingly no big difference between the frequency responses comparing 200/180/150, weird. As there is no big difference between 120 and 150hz I chose 150 hoping that supports the bouncy effect better.
Makes sense. But if your upfiring speakers are separable, I would suggest trying to hang them up in the front wall and use them as front height speakers. The reflection atmos effect is will always be compromised relatively.
Thanks, this yoúll need to tell my better half 😂, I guess I need to live with that relatively good result. Thanks for your support and that great stuff.
Hey OCA, I would like to ask if the „merge b to a“ method to determine the crossover for each pair of loudspeakers is still valid or if you have a newer tip than the method you used in the A R T video (finding intersection points for each loudspeaker which then is entered into the excel file)? I ask because when I use this method (vector average -> merge b to a), set the crossover accordingly (in my case 110 hz) and remeasure, the bass went up from approximately 70 hz to 100 hz showing a huge suckout at the 110 hz crossover point. Any useful tip on this one? Thanks Sir.
I believe, the very best crossover frequency is the frequency your speakers' spec -3/-6dB roll off frequency. This is usually given in the speaker manual. There's no way to phase align speakers and the sub with Audyssey (time alignment is not exactly the same thing) so it's a bit of a hit and miss really. If you want to go through the burden, measure your speaker pair with each possible crossover over frequency when Audyssey is on with REW and pick the flattest response.
@@ocaudiophile, thanks Sir. With Java Excel one can measure LR but cannot with the other speaker pairs, would then measure each speaker with the different crossovers and then vector average the pairs provide the requested solution and is the merge b to a tool still useful? My fronts go down to 26 hz (Nubert Nuvero 140), Center goes down to 36, Surrounds to 36 as well and Atmos speakers 113 hz. So fronts will be fullrange, Center and surrounds 40 (probably 60), atmos 110? Wouldnt that be too many bass sources in the room as I have a huge null in the right front speakers response due to floor or table reflection (I think). What is your opinion on that? If you are interested I can provide my REW file (already corrected speakers) via google drive so that you can have a look, what do you think?
You are right about too many bass sources to manage especially with Audyssey's limited adjustments. I choose to select 40Hz XO for my full range speakers for that reason just to leave bass management to subs. There's really not much you can do for the dips after carefully time aligning subs and speakers. I could have a look but when it comes to crossover frequency selection there's nothing I can do better by looking at measurements. You may try measuring speakers+sub at 40Hz crossovers when sub has customtargetcurve loaded and the speakers are not and regenerate customcurvepoints for speakers. @@saschawehde8072
Thank you so much for these great tutorial videos. One thing I’m a bit confused about is that information seems to overlap in your videos and I’m not clear if there’s a particular order to watch them in. If I do the Supreme Audyssey Calibration for example, are there other tutorials I should follow next? What video would you say is best to start at and follow for Audyssey (Home Theatre focus)? Thanks again!!!
There are improvements in the techniques in the most recent videos mostly because REW has been continiously improving. I think, you can get some additional tips from "Audyssey for Dolby Atmos music" video after the "Supreme Audyssey video"
Hi, how to use these protocols and what will be the benefits. I couldn't understand the use and benefits and how to apply. I've gone through the document but couldn't understand the application.
Yet another rookie question : whats more important, setting your avr to - 24.5 and check levels are around 70-71 for C (with 0 db all across the board in the avr volumes) or, raising the level on your avr to hit 75 db in check levels?
I did my original calibration using your process and the test signals built into REW: the results were significantly better than Audyssey, and I thank you for that. Now I want to experiment with your 'Atmos Sweeps' files. When I use the signals that are built into REW, I can specify which speaker I want to use as the reference. How does REW sync the test signals when they come from your individual .mp4 files? In other words, what does REW put at t=0 for each speaker signal?
When I select 'from file' in the measurement pane, the option to select a specific channel as the timing reference goes away. Is there a way to do name a specific channel as the reference when using your files?
@@ronaldippolito6785 They all use center channel as acoustic reference but REW will adjust for that if you're using a different one as long as you don't change it during measurements including all non-"from file"measurments.
At about 26:10, you change the speed of sound in the 'preferences' pane from 343 meters per second to 300. Can you please explain what you are doing, and why that works? Thank you...
That's the constant used is the Sound United (Marantz/Denon) receivers to convert time delays to distance and the reason Audyssey will get them all wrong. At some very recent models, they might have fixed this.
@@leo180811 I am not sure. They say all models 2022 upwards were fixed with a firmware upgrade. You can compare the real distance of a speaker which you can manually measure with the Audyssey calculated distance and undrestand.
Hello and thank you for sharing your knowledge and work. One question so far: You changed low frequency effects to -10db due to Atmos db bosst. Does this apply to a standard 5.1 sytem like I have too or should I leave this option untouched ? Thanks again
Thanks again for your work! Really enjoy how my system is sounding now on my Denon x4400. If I upgrade to an x6700, can I just import my saved profile from my old x4400 to the new x6700? Or do I re-do calibration to the new receiver?
You can manually edit the ady files side by side and copy/paste objects in json online editor but 6700 is a different beast and probably has different crossover rolloffs etc. I'd re-do the measurememts.
@@ocaudiophile Thank you, I didn't see this version when I looked! Now I have that trace option. Looks like the mac version doesn't have graph metrics though, which is a bummer! Thanks for your help.
This is all really great, but I am confused about a discrepancy between the Speaker Level settings at two points in your video. At 36:58 you show values for, say, the LCR at -1.0/+2.5/-1.0, but at 46:00, after explaining to take a note of the Speaker Level and distances settings before uploading the curve-edited .ady file, you enter values of -4.0/+0.5/-4.5 for the same three speakers. How/why do those values differ when they should have been the same as the noted values prior to uploading the curve-edited .ady file indicated at 36:58?
:) Your attention to detail is amazing and you're right but in the middle of the calibration during recording the video, I figured I forgot to take a note of the levels so I used levels from another measurement. Just ignore the differences and follow the method!
@@ocaudiophile Thanks for that info - good to know, and now my brain can rest easy :) I did have another question after giving this process some thought: If one doesn't like what Dynamic EQ does to the sound in one's system, does this method work without having that enabled? I find that DEQ really bloats the bass, and makes it sound too overbearing, so I personally turn it off.
@@discostu668 I never turned DEQ on myself for the exact same reasons but this method tames all its disadvantages at the usual volume level you listen to your system and makes use of its boosts at lower volume levels. If you still don't want to use DEQ or DV then you can do a manul calibration with Graphic Equalizer. You'll find two videos on that in my channel.
@@ocaudiophile That is also very good to know. Once I get around to the small re-arrangement of my HT system, I'm definitely going to try this method out. Thankfully, I'm only running a 3.1 system, so I won't have to go through so many measurements! I'm looking forward to seeing how this method changes the sound in my space. The thing that struck me the most was the method used to determine the speaker crossovers.
I am using the latest beta release of REW (5.20.14 for mac), and have run into two problems. The first is that when I try to determine the distance from the cross-correlated channels, my sub distance is calculated as significantly further away than the actual measured distance. I should note that I ran the measurement with the sub phase angle at 270 degrees. Could that be the cause?
Subwoofer impulse peaks are very often too far to be compensated by the distance limits in the receivers and this is normal. It's due to the time delays introduced by the filters in the sub (or by MiniDSP if you're using one). You can try aligning its firts peak with the acoustic reference speaker instead.
Hey mate! Sorry, i have another question. Im following this guide to a T, using MultiEQ. Ive noticed that, Im getting a 10db difference in the LFE channel when using the REW sweep vs. The sweep you provide (this sweep is 10db hotter). I verified that I have Low Frequency Effects set to -10 in the AVR. I also verified that this only occurs with the LFE channel (the other sweeps match perfectly.) When I move the LFE effect back to 0, the LFE sweep from REW matches the provided sweep at -10 perfectly as well. Any ideas? And if i cant figure it out, which one would you recommend i use? I have peaks with the provided sweep oved 90db. When using the REW sweep I have peaks over 80. If I move the LFE Effects back to 0, the provided sweep is hitting over 100db.😊 Thanks!
ALways use REW's LFE sweep because Atmos LFE sweep has a low pass filter and introduces extra delays due to that. If you download the latest early access version of REW, there're now measurement settings specific to LFE channel so you don't have to tamper with receiver LFE volume settings.
@Obsessive Compulsive Audiophile hey - I tried using the -10lfe option in REW. I then set up LFE Effects back to 0. The results between selecting the -10db REW LFE option, and not, are nearly exactly the same. There is a slight 3db boost at about 80hz when not using the -10db option in REW, but it doesn't appear to be reducing the LFE channel by 10db. I'm wondering if it's a bug? I'm using ASIO and channel 4. Just wondering if I should stick with the AVR option if the results in REW do not appear to change much when using the new LFE option.
@@thecrlghs The volume depends on the frequency range measured. If you measure LFE from 0hz to 24000hz like other speakers, the volume will not be boosted. If you measure 0hz-200hz for example, receiver will think it's an LFE signal and will boost it. Whether you use REW or dim it in the reciever or simply measure everything 0-24000 without chnaging any volume setting is up to you. They are the same thing.
New to the channel and really like it. What’s the best calibration process if you don’t have a laptop? I have Audy MultEQ XT32 on the Denon and the MultiEQ app, but no laptop.
At 8:20, you tell us how to use the control protocols to set the maximum volume to that which you use for watching movies. I watch movies at a lower volume than you do. How did you convert your listening volume into the value you sent for maximum volume?
Once we set the max volume ( with the protocol command) that volume that we watch movies, should we make all our measurements at that volume on the preamp?
@@ronaldippolito6785 Yes, especially if you intend to use dynamic eq or dynamic volume because Audyssey will apply different levels of boost to different channels depending on the volume of the amp.
Hi OCA -Thanks for all your videos, but this one I can't wait to get started on! 2 questions that might be very stupid so please forgive me I have the Marantz SR-6012 - it does have the Audyssey app but I just can't get the web interface to work. I get Error 403 - Forbidden/Access forbidden. 1) should I have the option use the web interface at all? If not I might just have to use setup via the remote control - will work but takes much longer 2) but when I access the setup from the remote control - under audio/surround parameter I don't have "low frequency effects" at all Will that give me problems? Kind regards
I had 6011 and now have 6014 and both have Web access, maybe network control is off in your receiver menu setup. Certain setup parameters only come out when there's an audio signal being fed. Play an atmos track and try reaching that setup menu item.
When playing sweep file from media player, the response curve capture by REW is 15 db lower than the response curve captured by playing sweep directly from REW. What should be done?
Turn off loudness in the receiver. The menu comes up only during Atmos signal playback so can be hard to find. If that doesn't help then your receiver is one of the few models which applies dialnorm setting to Atmos signals and you cannot do much about it.
Hello, could you please tell me the order of videos i should follow. For example, do i first do the sub alignment qnd house curve before doing the steps in this video? Or do this then align the subs?
Did you delete a video (similar to this one) you previously had on your channel? I remember you sharing a file to input in json editor specifically for 80hz (you also had one for 120hz). That procedure worked great for me. I moved my seating & was hoping to use that video as a reference again but can't find it. I've tried this video, but every time I try to import the "dummy" ady file (after using the json editor) into MultEQ, it states that the file is corrupt.
Hi OCA. It is the last thing you remind us to do, but just wanted to double check. Do we change the Low Freq Effect back to 0db after we determine the crossover frequencies, or at some point before that?
Thanks you so much for sharing your knowledge and enthusiasm with other fans of home cinema systems. I will follow this tutorial and aply it in my Denon receiver. In a first phase I will use the original Audyssey microphone, with the calibration that you shared for REW, to evaluate the results. Later I'm thinking ordering a calibrated microphone, but I'm undecided between the UMIK-1 and the UMIK-2. If budget is not an issue, do you consider the UMIK-2 a better option? If yes, why? Thank you very much and continue the awsome work :)
For AVR calibration Umik-1 is really good enough. If you will dive into advanced digital room correction for stereo hifi in the future, I would rather go for an audio interface and a XLR analog microphone. XLR mics have better timing accuracy. Basic sets will cost less than an Umik--2.
@@miguelmedinaramos You can choose between ASIO and Java EXCL drivers in REW. For ASIO, you'll need to install ASIO4All and you don't need to bother with Windows sound settings. All settings can be done from ASIO control panel. If you use Java EXCL drivers, Windows default player should be your AVR and the AVR should be in Movie/MultCh-in mode.
Just to give feedback it was quite simple after I use another PC. My PC is outdated (hardware and software) thus the ASIO4ALL did not worked. I will update my PC and surely will work. ;)
Some of my impulse peaks are on the top and some are on the bottom. Should I align to the "highest" peak or the peak that is on the same side? Thank you
No, always align for the strongest peak regardless of its direction. Btw, compare the phase graphs in Overlays and invert the polarity of a speaker if its impulse peak is in the opposite direction of the others. You can simulate that with "invert polarity" button under SPL & Phase tab and physically switch poles of the speaker if you decide to invert for real.
@@ocaudiophile Thank you! However the issues are my subwoofers and I can't easily change the speaker wires because of how they're installed. I checked "isReversePolarity" to "true" on the .ady file and it didn't change anything.
@@ocaudiophile Thats what I tried. I changed the.ady file on the subwoofers: I changed "isReversePolarity" from "false" to "true" on the .ady file and it didn't change the direction on the impulse peak when measured. Edit: The "isReversePolarity" command doesn't seem to work for me. I tried it on my center channel speaker and the impulse was the same - however if I physically switched the speaker wires around it does change... maybe its a bug?
I didn't try it myself tbh because I also switch banana plugs behind the receiver if I need to but I would think it's there in the ady for a reason. Maybe it's only used to remind the user to change polarity if Audyssey detects phase anomaly.
I'm stuck at importing txt file into Excel "Audyssey Curve editor". Somehow I have commas instead of dots in dB values..And then when I apply "convert text to colums" with Space ticked, the excel erases commas in dB values or example : from -3,965 i get -3965, and this of course don't work later in JSON. EDIT : I found the solution - when export I needed to tick "Use computer's number format"
Hi d1234 I have the same issue as you have (or had) - but you have found a solution. Unfortunately I still have the problem. When I have the raw text in Notepad it wil say ex -3.965 but as soon as I paste it into Excel it converts to -3,965 so comma instead of dot - and I can't get JSON to accept this format. Have you succeded in that d1234? I can't for the life of me find out what the heck I'm doing wrong?! Still hopeful regards
It's related to your Windows regional settings which effect use of . or , for decimals. You can fix that either through Windows settings or by simply editing the formula in XL.
At 12:00, you describe how to set up the room curve. The settings you give will yield A Harman room curve. To generate that Harman curve, is REW changing the test tone output to be higher at 20 Hz, then decreasing the amplitude all the way out to 20,000 Hz?
I don't understand, yet. When I look at your response curves, they look like they drop about 7 dB from 20-20,000 Hz. I guessed that the sweep amplitude decreased at the slope you entered in the equalizer preference panel, and level matching is how the response fit the Harman curve. If the sweep volume is constant through that range, what drives the response to a constant negative slope?
@@ronaldippolito6785 The low frequencies are always boosted by the room walls and there's also usually additional downward slope intentionally given by the manufacturers if you check their anechoic response because we tend to hear these frequencies higher than they actually are so in fact it's flat to our ears.
Another great video, thanks again. I can't seem to get audio out of my front height speakers using any of the "Atmos sweeps" files, any ideas? I have double checked all settings for my receiver in windows, made it default device, no enhancements, etc When using the dolby-test_tones_9_1_6 file it does match correctly for all speakers so I'm a bit confused. It's a 5.2.2 system and a denon x4400H (should it be in "Multi Ch in" mode?)
Some things to try: Select "Movie/Stereo" or "Movie/Mult ch in" in the receiver, reinstall Windows Media Player, with default settings, Check sample rates to be all at 48kHz in Windows sound settings, Denon should be the default player, if all else fails try with VLC media player with Windows direct sound and HDMI bypass option active.
Hello and big thank you for the great tutorial. After following the steps I finally hear my Dolby Enabled Speakers reflecting from the sealing. The sound in movies is impressive, compared to the sound the system does with just Audyssey. I don't know what the people at Audyssey tought when inventing the correction curves for the Atmos speakers, but it is not working well. The Atmos effect with my 4 Dolby Enabled Speakers is now, implementing your method, great. I now understand why people think, the "bouncing" speakers are crap. Audyssey cuts them in such a way they really sound as crap. One thing, that makes me think that I might have made a mistake at some point through the tutorial is the fact, I still have to adjust to the high pitch of the sound. With Audyssey it is somehow warmer...And my correction curve in MultyEQ app after finishing your method looks like a zigzag with multiple zigzags... Don't know if this should be the outcome...If you can share your MultyEQ curve after implementing your method I would be glad to compare to main, just to be sure, main is OK...I cannot actually call my outcome a "curve" :) Big Thanks for your time and the comprehensive why of taking me trough the tutorial.
Don't worry, it's gotta be a zigzag. The method uses Audyssey curves to apply the actual correction. The zigzags you see are the equalization applied to each speaker. I didn't undertstand what you mean by adjusting high pitch of the sound?
@@ocaudiophile If I compare the sound of the outcome now with the sound of audyssey, then I feel like the sound is much more "sharper" (almost scratching my ears form time to time), less "warm", and voices come out "thinner", thinner perhaps because of the presence of the high frequencies, which were not there before...For me now, it is as if I have different speakers, and I have to get comfortable with them first... When I look for the crossovers with your method, my crossover for the Dolby Enabled Bouncy Speakers comes out at 90 Hz. What would you recommend me to do with this relatively low frequency. Should I crossover them higher like 150 hz? Somewhere in your video you talked about high speakers and crossing them also at 80 hz, to avoid hearing the top sound out from the subwoofer on the ground. Or perhaps I misunderstood something...Of course my dolby enabled speakers are not high speakers...Perhaps thatswhy I should choose the higher crossover? Front speakers have the option for 60 hz and 90 hz Center comes out on 40 hz or 80 hz Surround at 60 hz , 90 hz or 120 hz What frequency would you choose in my place? Now I am mrunning Front @ 90 hz; Center @ 80 hz; Surround @ 90; Dolby Enabled Modules @ 150 hz Thank you in advance!
@@ocaudiophile I have only one Sub (SVS SB 2000 the older version, not the pro version) at the present moment. My room is small 6,50 m to 5,05 m to 2,47 m. My speakers are Heco The New Statement as front, Heco Celan GT 42 as center, Heco Revolution 3 as surround, Heco AM 200 as Dolby enabled speakers in a 5.1.4 configuration. All of this is connected directly to my AV-Receiver Marantz SR 8012. Appreciating your help. Thanks.
@@miroslavpetkov9895 Ok, if you don't use MiniDSP for your sub you can time align the sub properly with the receiver's distance settings. MiniDSP causes extra processing delays which makes this very difficult. When the sub is time aligned, it will be very easy to see where the crossover frequencies should be. For the treble problem you're having, suggest you follow the method in the Dolby Atmos Music video. I used Dolby's own target curve there which is much softer at the high frequencies.
Awesome! Do you recommend investing $80 for calibrated UMIK-1 microphone for setup or ok to use the mic that comes with Audyssey (I just purchased Denon x4800h receiver that have Audyssey multiq 32x with mic)?
Congrats on your purchase, that receiver has the same processor inside as the top models and is made in Japan. You can use the Audyssey mic that came with the receiver but make sure you use the calibration file with it while measuring speakers in REW. You can find the the link to the calibration file of these mics in this video: th-cam.com/video/KIozAzQH3CM/w-d-xo.html
@@leo180811 No, you can even use Audysssey mic with REW to measure your speakers. The tutorial should be showing how. If not that one, definteley the earlier ones on manual tuning.
Hello OCA. After - Pinpointing crossover frequencies for each speaker... I get following error - Error: Speaker has no bass response according to your measurement file. Please check! - program execution has stopped! Please let me know what I might be doing wrong.
Starting down the path and already have a question. JSON is new to me. I was cutting and pasting and something happened in the editor that collapsed the tree view to one location. The editor just shows an error with 16384 items. How do I go back to see the full tree and continue my work??
Fixed by switching to text mode then back to tree. Uploaded the new audy file to the file (had to change extension from json to audy) but the measurements aren't .4 and the levels are 0 like you show. Whats up??
Just click "New" in the top left menu and drop the .ady file in it again. It will be refreshed. Also check your MS Office regional settings. Some Excel versions can replace , and ; which might be causing the error.
@@ocaudiophile Appreciate that, sir. New to REW and looking forward to getting things dialed in and learning along the way. Already subscribed and appreciate the content!
You should Switch the “Network Control” setting to “Always On" in the receiver menu. You can leave the sound profile at stereo, it will change automatically.
@@ocaudiophile Thanks :) Network Control was already ON but I saw on forums that Web Control for SR7012 is no longer available (only for SR7015), I have to rollback the FW, but no idea how to hack the AVR FW for this So for stereo I leave Movie Stereo or Pure Stereo ?
@@Mitholman73 I don't change this setting at all, it automatically switches depending on the incoming signal. Stay away from pure direct modes as they bypass the subwoofer and Audyssey settings.
Just an idea. Create the PEQ in REW, import in Rephase, export impulse response of the filters back in REW, create division of the imported filter versus a flat impulse. This would get the FIR filters from the PEQ. Just tested and seems to be working very well. You can also further PEQ the result and apply another bank of filter correction in Rephase for even further refinment of the initial response. Do you see any issues with this approach?
It should work but with one inversion in REW, you already create the required PEQ settings as custom curve edit points for Audyssey. Besides in general leess is more in digital correction . Too much eq especially in high frequencies will cause boxy sound. Try the method in Audyssey for Dolby Atmos music video where there's less correction and you might prefer that one for movies as well.
@@ocaudiophile Thank you! I will. I learned so much from your videos. I looked at one of my first measurements after Audyssey automatic calibration and it's night and day versus what I got following your methods!
I am stuck at the ASIO4ALL mic driver issue beyond logic (red exclamation). Trying to resolve the past 2 days. Any help regarding this ?? Can I use the JAVA driver instead ??
I am having an issue where my front left and right mains have a huge null at 120hz when doing the vector average. REW says at ~123hz the phases of L and R are off by 180 degrees. Any ideas?
180 deg difference between phases of left and right speakers is indeed the reason for the drop in VA and it will be audible (or rather inaudible) when both speakers play that frequency, too. Probably different SBIR effects due to asymmetric speaker placement. Check for 70/140cm extra distance travelled from a speaker to LP reflected from surrounding walls that the other speaker doesn't have. Either move the speakers /LP elsewhere or direct 123Hz and below to the subwoofer with a high crossover.
@@ocaudiophile bumping up the xo is one of the things I tried. You are right, my living room has no wall to the left, so the speaker setup is asymmetrical. I also have 2 subs of differing sizes, both on the right side, in opposing corners (Front Right & Rear Right). It's the position where I get the smoothest blending - with the caveat that there is a null right at, you guessed it, 120hz lol. For trying out combinations in REW, I measured each speaker as large and I also measured the subs using the LFE channel with my LPF for LFE set to 250hz so I could get the maximum range. Here is the link to that .mdat file: drive.google.com/file/d/1Yt9Ml_EA3Op8J8suJxTuAHYPex8VvzCT/view?usp=sharing
Adjust FL -0.5dB, FR +0.5dB, SW -1dB. EQ 1/48 smoothed versions of all from 20Hz to 200Hz with 5dB boost for a Dr Toole curve with target level at 52.62dB. Export all 3 filters to the ady and cross over the mains (FL & FR) with your sub @150Hz. That seems to be the only way you can fix taht huge hole around 120Hz without moving speakers or using heavy passive absorption.
Regarding the "Speed of sound" value setting/correction. What is your reference when determining whether your AVR is using 300 or 343 m/s? Do you measure it with a tape/ laser, or do you measure with both Audyssey mic and Umik at the same position and compare?
It's caused by the AVR so it's mic independent. I measure with laser just one of my speakers and compare with Audyysry's own finding. The difference is usually too large and even with tape measure you would understand your receiver is using 300m/s. Btw, after a recent update my Marantz sr6015 now uses 343m/s. Many other models have been fixed but not all.
I'm confused as to what's going on. During my first measurement I set Denon AVR to 60 (scale 0-98) where I usually listen at. Audyssey Mic at listening position at 12 feet away or so. I got error saying that level was low and asked me to increase volume. Thought it was strange, but increased it to 70 and and that was loud and it took measurement with REW. Then, following your step by step instruction I tried same with supplied FL mp4 file. At the same AVR volume it sounded much lower and error was "The highest level in measurement is -54dBFS, for greater accuracy increase the input volume, the sweep level or AV receiver volume to raise the highest input level above -35 dBFS." Thought it was strange since my AVR was set at -10 DB (70 on the scale 0-98). Any ideas?
@@ocaudiophile This was done using ASIO drivers, and still it seems that Audyssey mic at 12 ft away at deafening 70 volume setting on AVR still comes in as low level according to REW. Mic gain is set at 100% in windows.
I've spent 2 hours today trying to figure it out. I'm gonna go deaf. Nothing is working. I keep getting errors in REW that levels are low. I boosted mic gain in Windows settings by 30 db (all the way up). AVR volume is between 65-70.....This happens either with internal REW calibration sound playback or outside calibration files. Tried both ASIO and EXCL drivers with calibration mic file you provided for Audyssey mike installed. Any ideas? @@ocaudiophile
@gr8nrg187 Did you "check levels" in preferences tab after you select input and output channels? You should be seeing output level around the top and mic level close to that and adjust volumes accordingly.
that's what I did. The output level is good near the top at -12dbfs but mic level even if I stand close to speakers is about -65 or so. Windows set mic volume to 100 and boost to the top. Is it possible my mic is defective or your custom curve for it does not work? Should I just buy UMIK-1 or just blast my speakers while setting measurements at around 75 (scale 0-98)? ? @@ocaudiophile
There was but MultEQ editor app got an update recently and now can host 4 subs and subwoofer syntax changed a lot. Your best option is to ask it in this forum: www.avsforum.com/threads/announcing-ratbuddyssey-a-tool-for-tweaking-audyssey-multeq-app-files.3006886/page-87
Thanks for that great tutorial. In a previous video you also used parametric eq filters (for yamaha), but i am using multeqx on my denon. Can i also use paraeqs with this tutorial and do i meed the hassle with python and excel, when i can use multeqx?
Denon does not have parametric EQ, it has only a graphic eq with fixed frequency bands but of course you can use that insetad of Audyssey. I have one preset with Audyssey and one preset with GEQ in my Marantz set up.
@@ocaudiophile thanks for the answer. Since multeqx support the usage of parametrics eq i will try my luck. You can also apply them without measuring your speaker wirh audyssey, thats what i want to try with rew then according to yout videos.
abit off topic but not really :) do you ever use the aligment tool? i read alot about it thru the different forums, that its the "correct" way to set your sub "distance" right etc..
I am not sure if anyone has asked this question before: with this method you calibrate with DEQ enabled at the MV level that you typically listen, but what happens to the response if you don't listen at the MV level where you calibrated everything? I know DEQ boosts its effects the lower you go in volume, but does it maintain the same level of (for a lack of a better word) neutrality throughout the response?
Total thumbs up OCA!!! I've learned so much by watching and working through your videos. I thought the SOTA walkthrough was good. But this was unbelievable the different that I could both hear and measure. Thank you so much.
Glad to hear it!
Thanks! Your tutorials and videos are a must for any HT enthusiast. You should be required reading!
Thank you!
Two quick notes: In REW export all to text, if you change setting 'Export Text Delimiter' to Tab, the numbers will go directly into Excel columns without having to monkey around with the Excel delimiter. Also, if you adjust distances and levels in the Audyssey MultiEQ app instead of in the receiver web portal, although more cumbersome, they will be saved into the .ady file and you won't lose them if you change between different calibrations. Cheers and thanks for another great tutorial!
Ernie, I'm guessing you mean make the adjustments in the the json editor? There are six decimal places for both trim and distance. For trim, did you try entering something other than half dB increments? Wondering if it will accept levels that aren't .5 or whole dB steps.
@@vincentdemarco6374 no I meant inside the audyssey muiltEQ app on your iPad or other device. When you open your calibration file, one of the submenus is (something like) Speaker Results, inside you’ll find crossovers, distances, and levels. If you input the values here they’ll be saved into your calibration .ady file
It only accepts 0.5dB trim adjustments
Thank you again for the wonderful tutorials and amazing help and support!
My pleasure!
Here is my feedback having completed this process. Firstly great learning experience. Going through things a few times makes them easier to follow and dots start connecting. The graphs looking stunning and I was able to follow each step all the way. Then I gave it a listen. My first impression was that the speakers seemed to have come alive a lot more and sounded more immersive but sounded a bit harsh in some instances and the bass seemed weaker to me and also little exaggerated in the upper bass frequencies. I went back and forth with my old settings and the new ones. Since I have a minidsp and a Denon 4700 which has two speaker presets, it was a lot easier to compare. I found the bass sounding better in my old setting so I went back to it but applied some of the learnings from this tutorial to it especially the correction of the impulse responses. That made a big difference and everything sounds better than before. So I ended up with new settings using my old and new and it sounds great.
What was interesting was that the FR between the latest REW and the early access differ. Not sure why but they do. Also for others the controls to see IR distances on IR tab is Ctrl + right mouse key and drag.
Thanks OCA
First of all thank you and congrats on your achievement. REW has been going through an extensive development stage recently and some trace arithmetic basics have been changed although there would not be a difference in the final filters produced.
I have a Denon x3700h andI ended up doing the same as you admittedly, after praising this mothod so much. My results were similar - the mid bass was way too exaggerated and the rest of the speakers sounded like they were crossed over at 200hz, ie; no bass even where the speakers can handle 60hz or even 80hz. I applied the impulse response to all speakers and now my original Audyssey calibration sounds much better. Surprisingly though, the biggest improvement came from turning Loudness Compensation off in the now perfected original calibration. However, after finding the original calibration sounded better, I still learnt so much from this tutorial! I feel like I am now an advanced REW user! Thank you OCA! I hope you can find why the calibration (after following this tutorial) ended up sounding the way it did. I followed it exactly as instructed. I can't wait to try your next tutorial....:)
You're channel is really exceptional in regards to range and depth of the topics. But also practical and precise. Awesome work and I hope you get a much bigger audience soon. Thanks!
Thanks 👍🙏
These tutorials are brilliant
Thank you.
Thank you for all the work you put into this!
WOW !!!!!!!!!!!!!!!!!!! MASSIVE DIFFERENCE !
Thanks for your work!
Welcome!
We love your stuff!
thanks!
Hello.
First of all, thank you very much for providing the calibration file.
I wanted to make a comment about how to use it:
I was trying to use the Audyssey mic to check the frequency response of my system playing pink noise and using REW's rta. And I believe the calibration file for the Audyssey mic you provided has the positive and negative values inverted.
My system was already calibrated following the avr's built in audyssey setup procedure.
First I measured with no calibration file at all, and I saw a frequency roll off starting at about 7KHz as expected according to what you showed in your other video about the hidden calibration file of the Audyssey mic. Then I loaded your provided calibration file and took another measurement. Instead of getting a flatter response, the frequency roll off starting at 7KHz was stronger. So I opened your calibration file in notepad texteditor, changed all negative values to positive and changed all positive values to negative, and saved with a new name. I loaded this new calibration file into REW, took a new measurement and got an almost flat frequency response. (The system was set to "flat" multeq profile, not "reference").
The calibration file for REW should show the mic output level at each frequency, not the required compensation. That is to say, if the mic's response is rolling off after 7khz, the values in the calibration file after 7khz should be negative.
I hope my comment to be useful for other users who are in the same situation as me: not having a calibration mic and wanting to use the audyssey mic in REW.
(Forgive me if I did something wrong, perhaps there is a way to load the provided file in REW so that it works as it should and I am not aware of it. I found that the file works inverting the values and I wanted to share it. Also forgive me if you or another user already said the same thing about the calibration file before).
I'll check that when I can (I'm away on holiday). It's remotely possible that cal file I've shared is reversed but I hope you're wrong ;)
You were right, it was having the inverse effect on the measurement. I've now updated the calibration file in all videos with the inverted one. Thanks for your attention!
Thank you OCA; great video again🙂 i want to try that; in the beginning you mentioned that also MultEQ-X user can apply this method. But how to do this, because the .mqx-files are different and not JSON formated (the Data-section is another format i think)!?
I don't use mqx and not sure what the format of an mqx file is but at the final stage of this method instead of inversion, you can apply REW'S auto EQ to the responses and transfer these filters to multeq-x.
@@ocaudiophile Okay, thank you! But then i think better be careful with statements like this: 'We don’t need ‚MultEQ-X‘, but MultEQ-X users can also easily implement the method', because as it turns out the method can not be used!
@@didi_p7756 There's nothing too wrong with that statement. The method can be implemented for mqx with REW's auto EQ instead of inversion.I have detailed videos on how to use REW's auto-EQ with these receivers.
@@ocaudiophile What do you think about the following workflow for MULTEQ-X users (assuming, as part of the measurement process, MULTEQ-X gets the speaker distances/levels correct)
1. Set REW preferences as prescribed in this video
2. Take all measurements (I.e. three measurements per speaker) with Audyssey enabled how without any Audyssey filters applied (note: MULTEQ-X allows any Audysseyfilters to be disabled via excluding’ ALL measurements in the app
3. Take RMS Average of each set of speaker measurements (I.e. RMS average of the three measurements) Question: Is Vector averaging preferred???
4. Use REW to EQ the response of each speaker’s RMS average
5. Import REW correction/eq filters into MULTEQ-X app
6. Upload to receiver and call it a day!
Thank you so much! I can attest that this is the best my stereo speakers have sounded (and was a lot faster than a full Atmos setup). I was previously using the room corection provided by my WiiM, which was an improvement over baseline. But after disabling that and trying this, it sounds fantastic and now my movies and games can benefit as well. Well done!
Great to hear!
Please do a moving microphone technique with Audyssey next.
Sure thing!
Your username is absolutely Perfect Sir. Excellent instructions and great job describing what you are doing and why. Thank you for sharing your knowledge and expertise with us.
You are very welcome
Even though I run Dirac, I've learned something from every one of your videos. Thank you!!
Great to hear!
Great video, loved it. Learning a lot from your videos, I appreciate all the hard work that you've put into this video.
Glad to hear it!
Tak!
First ever DKK asset I have in my life ;)
@@ocaudiophile well..we are almost neighbors so :D
@@_shogun_gaming_ 😅
Thanks!
Finally! My system is a new one with you. Especially with my low end Denon 1700h without multieq xt32 the difference is huge in bass management. Thank you so much. Somebody needs to take the method and create a blog post out of it. Because a video sometimes is not the ideal medium while working on such things . Much love ❤❤❤🎉🎉🎉
Thanks
Could you do the blog post pls @Squizzy911
Noted!
i have started writing down from your video the steps to do. but still ongoing. a nice blog post on medium or somewhere with nice tables regarding settings in REW, direct download links, hint blocks and so on would make it much better to understand. it would be also possible to make first a 1:1 tutorial out of it without any deep knowledge background which can be collapsed from special blocks. what you think?
@@ocaudiophile
You've my full support but some techniques have evolved and there are automation scripts in the immersive audio and especially Audyssey ART videos. You might wanna take a look at them as welll.@@Squizzy911
Thanks
Many thanks!
@@ocaudiophile
Only fair with a the help you have given me!
With the instructions of the Dolby Atmos 9.1.6 video I was able to get also a measuring of the Atmos speakers for my 7.1.4 setup on a MacBook Pro, using VLC media player on my Mac. Afterwards quit simple choose the wav file of the sweeps in REW, start measuring and play in VLC the speaker you are needed, every speakers has is own chapter, great work.
Now this evening I can go further with the instructions of this video with my 36 measurements. Great to do this job with a Mac, I am not an experienced windows user.
Glad to hear that. You're one of the very few who could play the Atmos sweeps with a Mac!
Very strange that Mac users too much problems have with REW. And there is more an advantage for a Mac because you need no special drivers. I use Ventura or on another system Sonoma, and the latest REW from the REW-website. On every mac will plays VLC plays and you give the right files and sweeps in you 9.1.6 video, maybe you can these also add to this video.
Furthermore a Mac User must setup his mouse as a secundary mouse. So you can speak more optimistic in your videos for Mac Users!!!
Excellent video with excellent results, I congratulate you for the great work and for the dissemination
Thank you very much!
I spoke to John about the inversion with [A] / [B] magnitude division and it is working correctly. What threw me off was that the calculation in REW displays as the differences on a 0db line. I was expecting it to visually invert the response. However, If you export an inversion and pick an area that looks off in REW. For me its was 16K. Look at the difference in the measure response vs the target curve. Mine was -2.5db in REW. This means that the inversion should be +2.5db and in the text file it was. The biggest thing is the way REW is displaying the correction. It only displays it as the differences on the 0db SPL graph. I had to get my head around this. In the end, A / B magnitude division does work as intended, and the 0db line becomes the target curve.
OCA, thank you for this detailed info. I'm completely new at this hence have a question relating to the beginning of your video where you transfer the Default.ady file to your phone then send it to the AVR, I assume via the MultiEQ editor. I did this and my AVR asked to save it to one of the two presets (I saved to preset 1). I assume this is correct? My real question comes at 6:48 of the video when you then use some sort of in web interface protocol to interact with your AVR. What is the purpose of this? Or can this all be done by directly using the AVR menu via the remote control? Are you using this web interface to simply show your viewers the changes or is there something more behind this action? When I selected my new preset 1 setting and viewed via the AVR (using my remote) the changes you note were there except all my distances were not changed to 0.04m ... they remained at the original Multi EQ actual speaker distance measurements. I apologize for this newby question.
You can use whichever preset you like. I keep one preset for full manual calibration and one preset with Audyssey and Dynamic EQ using this method. The web interface commands can all be done from the receiver menu/remote, this is just a short cut but also a lot easier to change settings from your PC instantly.
Some newer receiver models use a different setting in the ady for speaker distance. You can manually reset them in the reciever setup menu or in the online Json editor before you save the ady file.
Thank you!! With all the videos on your site which one do you recommend I use to maximize my 7.2.4 system (2021 AVR Denon X6700H)? I first experimented by adding the Harmon Curve (as outlined in one of your videos) … which achieved a much improved response.
Of course I now want to see what more we can do with the system. Cheers and thanks again for all your sage advice.
FYI, I tried numerous times to download the REW V5.20.14 file, it appears to work, however it never includes the “house curve” tab under preferences? When I download the REW V5.20.13 it does include the “house curve” tab under preferences. Am I missing something? Can I use the .13 version and still successfully work thru your video(s)? Keeping in mind I’m a REW infantile. 😂
Last questions and I won't bother you anymore. 😁 I decided to try your "Supreme Audyssey Calibration" video (assuming that might be what you advise further to my question above?). I've updated all the preference tabs info. per your video including adding the mic calibration file as I'm using the original Audyssey mic. At 14:00 in the video I'm trying to run the measurement but it gets hung up at "10% Waiting on timing reference" and the graph at the bottom left corner simply keeps scrolling? When running the Check Levels I get the message "Ready to measure..." with the the message "Level Low, -61.2 dBFS, 78db SPL" . FYI, given the Level Low message I tried turning the AVR volume up and still got the message "Level Low, -41.7 dBFS, 98 dB SPL" but it did not help. I then also experimented in the Preferences area "Soundcard" Tab to see about calibrating my soundcard (not sure if I even needed to do that?) and could never get the Input and Output levels even close to matching one another as the volume was waaaay to loud at the original Sweep Level setting of -12 dBFS. I had to adjust the Sweep Level setting to -30 dBFS, in order to turn up the AVR volume enough to match the input/output reading and not blow my speakers (which level was still very loud). Lost in REW Space ... hope you had a great day today and can help.
It's moved to EQ window.@@bradmcneely4745
This was fantastic. Learned a lot from your videos. I'm excited to try them out on my system.
Great to hear!
@@ocaudiophile i have the x3700 denon BUT in the web interface i don't HAVE the setting "LOW frequency effects" !! am i doing something wrong or maybe some models don't have it ?!
update : i had to play a movie with LFE to be able to SEE that setting !
i got stuck at the default.ady file transfer at 6:37. multeq app says "the specified file is not located in the devices local storage or is not a proper curve file". Windows sees it as a ady-file but nothing helps here. could you maybe help with this?
Android apps need to use Google drive for transfer for some reason.
@@ocaudiophile thank you so much for the quick reply! i'll try that.
Hello. I did the JSON editing part. I opened it back in JSON to make sure it was showing the new values. Still after importing it into my AVR, I checked the levels and distances and they were not as expected.
Any idea what could be wrong?
Thanks!!
Are you on MultEQ-XT or XT32?
@@ocaudiophile XT32
Then it should work fine. Some XT users encountered similar problems. Have you converted responses of ALL speakers including the subwoofer to perfect (1,0,0,0....)?
@@ocaudiophile Yes I checked a few times and all of them are showing the correct numbers including the subwoofer. Bummer :/
Danke!
Many thanks, you didn't need to!
Please a little clarification for me, At around 51:00 I made copies of the sweeps, applied Align Spl (for speech), and then compared sweeps with their copy and applied Spl Offset.
Do the new Spl offsets replace the AVR Speaker Levels previously set or do you add the Spl Offsets to the AVR Speaker Levels? I tried both and the results were no better than the first applied inversion (before 51:00).
I've previously aligned my subs following your tutorial. Got them flat. The distance from the sub to the center speaker measures between 7-9 meters depending if you measure from the bottom or top peaks. The center speaker distance to the MLP is 1.91 meters. So my total distance is approximately 11 meters to add to the subwoofer distance in the DENON AVR!? 7.90 meters is the highest distance that can be entered. I'm confused.
I understand the confusion. Actually REW now shows the SPL offset applied to a measurement during "Align SPL" process in the Info window under "Align Offset:" so you don't need to make copies. just check these for each measurement. I explained this method in detail around 08:35 minutes into the following video:
th-cam.com/video/4gVUdArG7oA/w-d-xo.html
@@ocaudiophile In the Audyssey calibration for Dolby Atmos Music video you apply Cross correlation alignment with the Vector Avg Sweeps and Actual Time(distance) Aligned measurements is that new? Should it be employed here also?
@@CarlosGomez-THX_1138 It takes the precision one step further and helps in finding the correct crossover points. As you cannot adjust speaker distances with high accuracy in the reciever, this extra step helps REW to use the actual speaker distances and calculate phase interactions with the sub better. It's up to you to try and test the results.
In REW, after applying Align SPL @1000Hz and entering the Levels in the AVR, can we then apply, in REW, Align SPL @200Hz and adjust the Levels in the AVR? or do we apply the Align SPL as depicted in this video?
@@ocaudiophile it works!
Hey OCA, how have you been? I've noticed that A over B magnitude division is overcorrecting on the EQs I'm doing. Do you have any idea why? I've seen that treble sometimes needs an increase of 1.5db, but the A over B is increasing by more. Like 2db. What could be causing higher numbers than needed? Is it also taking into account reflections in the sweep? Can I remove that? Or just measure with an RTA?
I guess these deviations are caused by the complicated maths behind the division operation between two FFT's especially when it's not over a flat line but a target curve. The beginning and the end needs some mathematical gizmo to be handled. I use direct 1/A inversion lately over a flat line for that very reason.
@18:02, if the default is set to my Marantz AVR, is there any special way that Windows Media Player (or anything else) needs to be setup in order to play the height channels? Or will it play them natively?
No change needed in WMP default settings
@@ocaudiophile Thanks! How about when I select the configuration for the Marantz in Windows? 7.1 or do I choose another option? I think there's one that shows up as Atmos, but the diagram disappears when I select it.
I leave it at stereo
@@ocaudiophile This is definitely still not working for me. My default is set to the Marantz AVR and stereo configuration, but the height files are playing out of fronts left and right.
@@vincentdemarco6374 How about when you shut down REW and play these files from WMP?
Thank you very much for taking the time to do this. My receiver only has multeq (not 32 or xt) so I was intrigued by this video. Unfortunately, after completing the process, I got no frequency response below 120hz whatsoever. My sub was set at -12db to hit the target curve and even after boosting it to +12db in the audyssey settings and turning the gain on my amp up an additional 30%(maxed out) there was maybe 1-2mm of travel from my subwoofer and no bass from my bookshelves. I intend to redo the process from creating new corrections forward, potentially with another curve, but I am curious if maybe multeq does not have the resolution that multeqxt32 has and as a result, some numbers may not transfer to my avr. For instance, when adding the 241 points to the custom “target curve points”, could audyssey be ignoring 100 of them or something to this nature because it is the lesser audyssey version?
I don't think number of custom target curve points is a problem for the type of MultEQ as long as the app accepts them (I think the limit is somewhere near 65,000 points). However, MultEQ uses significantly less number of FIR taps compared to XT32 while generating FIR filters which will effect LOW BASS correction very much. In other words, Audyssey may not be able to correct properly to the target curve it's pointed at especially below 100Hz.
@@ocaudiophile Quite appreciative that you took the time to reply. I hope i dont seem as though i am asking for help diagnosing my own issues here. I am simply trying to contribute issues i am having in the event someone else runs into these problems. I am now concerned that the measurements i took with REW were being influenced by the existing Audyssey profile's corrections ( i had a traditional audyssey correction applied prior to starting this guide). I will start from scratch with a factory reset and report back. Once again, thank you.
Even just getting the distances and sound levels corrected does a lot to improve the cohearency (intended)
Just finished the calibration. What a difference!!!! How can I offer you a coffee? Where is the donate button, I need it now 🤣
Can you explain the difference from Audyssey (refined post calibration) to this method? I am starting this process soon...
This one uses Audyssey and dynamic EQ so it has better low volume equalisation. It can also be more accurate compared to manual GEQ bands
Hi OCA,
In the video at 43:57, I don't see the category "customTargetCurvePoints" in my .ady file in the JSON Editor. Using a Marantz Cinema 50. So I'm not sure where I should input the REW response. Thanks in advance for your help.
Is your MultEQ Editor app up to date? Custom curve points should definetely be there. There have been people who applied this method successfully with Marantz CInema 50.
@@ocaudiophile it was some bug with the exported .ady file. I exported again and the setting was available in JSON Editor.
Finally finished all the steps in this method and I can immediately notice a significant improvement in the height channels. However, the bass response seems to very low. I have double checked that LFE level is set to 0 and not at -10. Still can't figure out what's causing this. In REW, the subwoofer response looked fine.
Loudness management is on in most models by default. It could be a reason but best way is to remeasure speakers and sub with the new settings active and check in REW what's going on.
@@ocaudiophile thank you! When measuring sub impulse response, the distance adjustment from Center Channel peak impulse to Subwoofer peak impulse is showing to be 12 meters (!!). Obviously this cannot even be configured. So I'm not sure if I'm doing something wrong. Is there any other way to time align the sub?
@@Judddernaut are you using minidsp for the sub? It usually introduces such delays causing 12m distance requirements. You can align subwoofers first peak rather than the highest peak with the acoustic reference speaker instead.
amazing tutorial. many thanks! my 2 subs on mini-dsp seem to arrive much later than receiver can set (+6.8m to center). in this case would I just set to maximum available distance? or is there anything smarter I can do?
Million dollar question ;) There's not much you can do to compensate for the time delay introduced by MiniDSP in the reciever.
*Play around with bass sync setting if you've a Denon (Marantz don't have it), that might add some delay compensation.
*You can switch to IIR filters (biquads) in MiniDSP if you are using FIR filters (taps) which introduce delays to the system.
*And last resort: You can align subwoofer's previous impulse peak instead of the highest peak to the center channel
Thank you! I've waited a long time for someone to do a tutorial manually tuning HT/atmos speakers. Bless you! .....Initially, I tested the sweep files on all my speakers and determined that two of my Atmos speaker's high frequencies dropped precipitously around 1000Hz. Thank you for the files to test them. In the Denon AVR-X4700H Low Frequency Effects are displayed when
"This item can be selected when a Dolby Digital or DTS signal or DVD-Audio is played." I did not set this initially and continued with the tutorial with 7.1. Should I have played a DTS movie and then set this at the beginning? What do I set the Reference Level Offset given I did not set Low Frequency Effects?
Thank you for bringing my Home Theater to life. My 7.1 sounds so good I wonder whether I needed Atmos.
I suggest you use Denon's web interface to change its settings. WHen you start palying any of the sweep files with Windows Media Player, all these settings will sho up in the web interface and you can adjust them. RLO has nothing to do with LFE. LFE -10dB offset is just a temporary setting to see the sub response where it really is relative to other speakers. RLO is a setting which adjusts the strength of DYnamic EQ.
Thanks so much for putting this together!!
My pleasure!
Thank you for the excellent video. In the narration you state midrange compensation should be off, but in the JSON editor it shows midrange compensation box as checked (true). Should the box be checked or unchecked? Thank you
Uncheck it for every channel except for the subwoofer channel where it's irrelavant but may help reduce noise in the upper frequencies. The much newer Audyssey ART video is easier to implement btw.
I have taken measurements at the three points, and am now ready to time align them.
At about 20:30, you mention a new REW feature called something like 'copy selections to other overlay graphs'. How are you getting that pop-up that contains that new feature?
Mouse right click on the bottom window
@@ocaudiophile I must not be on the same version of REW as you, since that command is not available when I right-click on that bottom window. I am on the released Mac version of REW, version 5.20.13.
I found that there is a newer version (5.20.14), early access build. I installed that one, and it has the command that you use in your video.
@@ronaldippolito6785 Google "REW early access version"
Hey man. I finally got around to giving this a shot.
My results were pretty good. Only thing I changed was I made the target curve give me way more bass (about 10db to match the uncorrected curve of my speakers) and I made Audyssey not do any correction after 10khz because the boost of highs was way too fatiguing in my room.
Using this method has made my home theater sound like a real theater. It doesn't sound like speakers, it sounds like sound. No it's not perfect but nothing will ever be perfect in audio. Thanks for the video!
Cheers!
@@ocaudiophile Yeah man you are doing some really good work. Do you take your measurements with sofa cushions on or off? I think the high end boost might be because I measured with my sofa cushions on but I saw another video from HD MOVIE SOURCE saying to measure with the cushions off.
@chadkirk925 The differences are too small to measure precisely. I'd rather put effort to place the mic tip where your ears will exactly be during measurements and leave the room exactly as it will be during film watching for better results.
@@ocaudiophile thanks. That is my thought process too.
This is really great. I appreciate you sharing your knowledge!! Quick question - do you recommend doing this for 2.1 channel music listening? I have a 5.1 system with Audyssey default calibration. I disabled the mid range compensation using the app because that sounds better to me. I use it for movies. But for music I switch to L/R Bypass and 2.1 stereo mode because that sounds better to me. It comes close to Pure Direct on the Marantz 5015 receiver but a bit better due to the SVS sub filling in. I would love to improve the music experience but I need to set up REW and buy the umik-1 microphone
You can use Audyssey's microphone and REW is free. I extracted the calibration file for it (check my latest video's description link) and its results are quite satisfactory and timing is better than a USB mic. With proper sub-main time alignment and graphic EQ, you can get nice stereo sound.
Thanks for the idea@@ocaudiophile. Unfortunately I cannot connect the Audyssey mic to my Macbook Pro as it does not have a mic input. Neither does my Windows Thinkpad for that matter. I will order the umik-1 in a few days. So you do recommend doing a good calibration for stereo listening
Especially if you have easy to drive mains, you could get good results. With higher end stereo speakers, these AVRs are not powerful enough in general. Good speaker placement and descent DSP always prevails though.
@@ocaudiophile I have B&W 683 towers rated for 89dB sensitivity in the front. Marantz 5015 driving them. I will get external amp later. 80% music 20% movies. I am happy with music just want to see how it can get better
First, thank you very much for this guide. I am eager to try this for myself.
I have one questions: can I do this process without using DEQ? Different movies are mixed differently and for compressed audio I have to turn up the volume which would mess up DEQ, right?
And dont we have to SPL all speakers to about 75db or so if not using DEQ?
And can I measure and EQ two subs with REW and MiniDSP before wallthrough this guide? Or is it better to use your Sub Alignment and cross corelation video?
You should measure and create filters for your system at the volume you like to listen to it. If you take these measurements with DEQ on, DEQ will then adjust the calibration for louder or dimmer volumes on the go so it's useful for the problem you mention. It has some odd adjustments for surround and height speakers, etc. but still, it's better than nothing.
If you are using multiple subs with a MiniDSP, it's better to time align the subs first using the sub alignment video and dial in the delay settings in MiniDSP.
Wow, this makes a ton of difference! The clarity in the sound is much improved, thank you!
However, I think the sound after adjustment lacks a bit of 'punch' to it. Is there something I should be doing different or maybe go for another target curve? Is there one you can recommend to try in that case?
Keep up the good work 🙌
Increase the subwoofer volume 1-2dB maybe?
@@ocaudiophile thank you for the quick response! I believe it is more than just the bass that is lacking. The sound feels flat sort of ...
Audyssey ART video method is more advanced.
@@ocaudiophilethank you sir, I will try that 👍 massive thanks for all your effort
Great video, while thinking this through I wonder how you adapt for custom curve in the bass region. Will you just create a target which will not be just "flat" harman curve, but will rather have a little bassboost around 20-100hz?
I remember the harman curve you used in your "Audyssey hacks for epic Dolby Atmos" Video had a slightly different harman curve with a little bass boost. So what is the correct harman curve, or to rephrase it, did you edit the harman curve in the former Video more towards your taste in the bass region?
The one in the video is standard, only with extended frequency range. The bass boost will depend on the windowing and to some extent smoothing used on the response. The method in the video yields a very satisfactory bass when applied.
So, in the preferences section of REW, keep ‘decimate IR’ ticked? I thought you had previously advised against such?
Decimate IR should ideally be unticked but it has no effect on this correction method. I just tried to keep it as simple as possible.
Got it. Thanks!
@@ocaudiophile I get it. Definitely understand your desire to only focus on the things that matter.
2nd question. Do you know what key and mouse strokes are used on a mac to determine the speaker distances on the impulse graph in overlays?
I don't have a Mac but should be ⌘ key instead of CTRL plus mouse right click?
Configure the mouse button as "Secondary Button," hold down CTRL and the button and click and drag. Intuitively obvious, no? 🙂
very impressive, thank you very much, another step up to the moon :) by the way, for the LFE cut off at 120Hz, it is not related to setting into the AVR (LPF for LFE) ? and an other thing for the filter curve generation, MultEQ-X app offer the possibity to import REW fiters, it say "EQ is limited to 10hz to 20khz, -10db to +10db, and Q 0.1 to 10" do we have to make sure that the generated filter curve is not going above that ? and thinking if it is better to create a filter curve from the EQ by using a generic filter that match this parameters ?
No, it's not effected by the LFE cut off in the receiver (that was also my first thought and I tried increasing it to 250, no change in response!)
You can't replicate these inverted responses accurately with EQ but you don't need to. In REW EQ filter types, choose "Generic" as manufacturer and "Configurable PEQ" as model. Then you can set all these limits for MultEQ-X you've mentioned in the configuration and REW's auto-EQ will "match response to target" within these parameters. Then you can export them to MultEQ-X. Don't forget to keep the target level same for all speakers after initially calculating it as explained in the method.
@@ocaudiophile one little tip about REW zooming, it focus on the cross in the graph, so make sure to clic on the impulse start before cliquing on zoom+. I say that because it hapen to me and now I figured out it helps
@@armagedon2B noted thanks!
Great video. Do you know if the dolby surround speaker files will work with a macbook?
I am afraid no. Macbooks don't support bitstream output of DD+ over HDMI, they can only decode to LPCM. I heard that some people managed to do it with some paid software but you need to check the forums. I really am not a Mac guy.
@@ocaudiophile Thanks.
Would it work if I copied files to USB drive and played through my Oppo dvd player?
@@djcradd I never tried but it should as long as it can decode DD+
You can however switch speaker cables (not so hard if you're using banana plugs) and use one of the surround channels to play from all the Atmos speakers. The correction files will be accuarte. For timing, you will need to do a calculation based on the distance of the speaker you used
At the 23:30 point in your video, you are going to vector average front left three measurements. You also included the copy of the time-aligned measurements.
Why did you include the three original measurements, if you had the time -aligned version?
All 3 measurements were already time aligned with cross correlation. The 4th is a copy of the central measurement and a copy of of a time-aligned measurement is also time-aligned. The reason I added the central measurement twice to the average is to give it more weight in the average. The reason for evarging them is to find the true speaker response in a larger area.
Hello, OCA. Thank you for all you do. I am really enjoying your channel and learning a ton. I have a question and hoping you could answer it... If my peak impulse on both of my fronts is >0 and my peak impulse for my center is
oh, btw, I have my fronts connected to a 2 channel amplifier and my Center channel and surrounds connected to a 5 channel amplifier. The Center peak peak impulse is 0. i'm just curious if that is an indication that the speaker is hooked up incorrectly.
It doesn't necessarrily mean that but if they are the same speaker brands it's quite likely. Best way is to compare Overlays/phase response of center with the rest of the speakers across the frequency bands by inverting polarity of the center speaker. You can do that from within REW without changing wires and decide which polarity is correct.
With external amps polarity switching is very common.
Hey, thanks for this great tutorial. I noticed that the amplitudes of FL and FR are exactly opposite to the rest of the speakers. Do I need to change the plugs on my speakers? Is this the fabled wrong phase?
I use a down fire subwoofer which according to impulse response should have a distance of 11 meters and the curve looks more like a mountain without a plane like yours. Is the sub wrong or should I try it in frontfire? I can not set 11 meters distance in the receiver nor do I know later how I want to set the level with the curve of the subwoofer it looks like a mountain with a peak as I said. Thanks in advance for your help
Compare the overlays/phase graphs of front speakers inverting polarity of one of them (REW can do that with one click) and phases match better inverted, then do so physically. It's uncommon front speakers having opposite impulse peak direction but not impossible.
Thanks for your answer. I have switched the cables and now every amplitude shows in same direction. Or is it ok when some speakers have inverted amplitudes? I thought this causes bass problems
Hey, sorry to bug ya again. With the subs - am i supposed to align to the target curve while i have the -10db LFE in the AVR? So when finished, the subs will be 10db hotter than the target curve?
Primary reason i ask is this is causing me to run my subs awfully hot. Granted, i am calibrating pretty hot overall, target curve is at 76.4db. But if i use the -10db, im having to keep my sub AVR setting at 0 and my three PB1000 pros are running at -6db, just to hit the harman curve. Or should I be matching at the LFE 0db.
It could be just because of the type of subs i have (hindsight is 20/20, should have gone with at least the 2000 pros).
If your subwoofer measurements range is 0-24000Hz like other speakers, you can leave LFE setting at 0. -10dB setting is only useful if you're measuring subs 0-200Hz because when range is low, the receiver thinks it's an LFE only signal and boosts it by 10dB. Btw, in the latest early access version of REW, you can apply -10dB to LFE automatically during measurement, you don't need to change anything in the receiver.
@ocaudiophile Ahh. Got ha. I've been leaving the sub measurement at the same range as the speakers, no wonder I've been having to crank up the volume!
Hey OCA, trust you are well. Got a question on the „determine crossover frequencies“ part of your video. I own 2 pairs of atmos upfiring speakers and I am wondering how to find the correct crossover. Should I vector average TFL and TFR and then trace arr. (merge A to B) with the vector of FL and FR or with the sub? As far as I experienced the atmos effect is better when using a high crossover (e.g. 180). I read somewhere that some bass management systems route in a 2 step way. E.g. Front Height speakers signals below crossover 180 being routed to fronts which are crossed over at 110 hz from there to the subwoofer as usual. What do you think, thanks Sir?
Normally, they should been crossed over with the sub and the method you explained shluld work well. I haven't heard an information on two step crossover for height speakers. What receiver are you using?
Thanks. I use a Marantz 7706 processor. Have read somewhere about the 2 step crossover in an older interview with Andrew Jones since he was working with Pioneer. Besides that I found out measuring the heights as „large“ it’s bass response is kinda same than the front/rear speaker it sits on (measured as large as well -> only looking at the bass region below 200 hz or so). This tells me that the processor is allegedly doing some internal crossover to avoid overdriving the height speakers. When measureing the heights crossed over there is surprisingly no big difference between the frequency responses comparing 200/180/150, weird. As there is no big difference between 120 and 150hz I chose 150 hoping that supports the bouncy effect better.
Makes sense. But if your upfiring speakers are separable, I would suggest trying to hang them up in the front wall and use them as front height speakers. The reflection atmos effect is will always be compromised relatively.
Thanks, this yoúll need to tell my better half 😂, I guess I need to live with that relatively good result. Thanks for your support and that great stuff.
Hey OCA, I would like to ask if the „merge b to a“ method to determine the crossover for each pair of loudspeakers is still valid or if you have a newer tip than the method you used in the A R T video (finding intersection points for each loudspeaker which then is entered into the excel file)? I ask because when I use this method (vector average -> merge b to a), set the crossover accordingly (in my case 110 hz) and remeasure, the bass went up from approximately 70 hz to 100 hz showing a huge suckout at the 110 hz crossover point. Any useful tip on this one? Thanks Sir.
I believe, the very best crossover frequency is the frequency your speakers' spec -3/-6dB roll off frequency. This is usually given in the speaker manual. There's no way to phase align speakers and the sub with Audyssey (time alignment is not exactly the same thing) so it's a bit of a hit and miss really. If you want to go through the burden, measure your speaker pair with each possible crossover over frequency when Audyssey is on with REW and pick the flattest response.
@@ocaudiophile, thanks Sir. With Java Excel one can measure LR but cannot with the other speaker pairs, would then measure each speaker with the different crossovers and then vector average the pairs provide the requested solution and is the merge b to a tool still useful? My fronts go down to 26 hz (Nubert Nuvero 140), Center goes down to 36, Surrounds to 36 as well and Atmos speakers 113 hz. So fronts will be fullrange, Center and surrounds 40 (probably 60), atmos 110? Wouldnt that be too many bass sources in the room as I have a huge null in the right front speakers response due to floor or table reflection (I think). What is your opinion on that? If you are interested I can provide my REW file (already corrected speakers) via google drive so that you can have a look, what do you think?
You are right about too many bass sources to manage especially with Audyssey's limited adjustments. I choose to select 40Hz XO for my full range speakers for that reason just to leave bass management to subs. There's really not much you can do for the dips after carefully time aligning subs and speakers. I could have a look but when it comes to crossover frequency selection there's nothing I can do better by looking at measurements. You may try measuring speakers+sub at 40Hz crossovers when sub has customtargetcurve loaded and the speakers are not and regenerate customcurvepoints for speakers. @@saschawehde8072
Thank you so much for these great tutorial videos. One thing I’m a bit confused about is that information seems to overlap in your videos and I’m not clear if there’s a particular order to watch them in. If I do the Supreme Audyssey Calibration for example, are there other tutorials I should follow next?
What video would you say is best to start at and follow for Audyssey (Home Theatre focus)?
Thanks again!!!
There are improvements in the techniques in the most recent videos mostly because REW has been continiously improving. I think, you can get some additional tips from "Audyssey for Dolby Atmos music" video after the "Supreme Audyssey video"
Hi, how to use these protocols and what will be the benefits. I couldn't understand the use and benefits and how to apply. I've gone through the document but couldn't understand the application.
Better try Audyssey ART video method, it has automation scripts which accelerates the manual processes
Yet another rookie question : whats more important, setting your avr to - 24.5 and check levels are around 70-71 for C (with 0 db all across the board in the avr volumes) or, raising the level on your avr to hit 75 db in check levels?
There shouldn't be any. A step is skipped in the video by mistake. It's explained in the description area.
@@ocaudiophile i switched my avr to 0-98 db scale, and at volume 55 its pretty close to 75 actually ;)
I did my original calibration using your process and the test signals built into REW: the results were significantly better than Audyssey, and I thank you for that. Now I want to experiment with your 'Atmos Sweeps' files. When I use the signals that are built into REW, I can specify which speaker I want to use as the reference. How does REW sync the test signals when they come from your individual .mp4 files? In other words, what does REW put at t=0 for each speaker signal?
Atmos sweeps are identical to REW sweeps in terms of timing. You can adjust speaker distances same as others
When I select 'from file' in the measurement pane, the option to select a specific channel as the timing reference goes away.
Is there a way to do name a specific channel as the reference when using your files?
@@ronaldippolito6785 They all use center channel as acoustic reference but REW will adjust for that if you're using a different one as long as you don't change it during measurements including all non-"from file"measurments.
At about 26:10, you change the speed of sound in the 'preferences' pane from 343 meters per second to 300.
Can you please explain what you are doing, and why that works? Thank you...
That's the constant used is the Sound United (Marantz/Denon) receivers to convert time delays to distance and the reason Audyssey will get them all wrong. At some very recent models, they might have fixed this.
@@ocaudiophile Is the 1700h very recent for you?
@@leo180811 I am not sure. They say all models 2022 upwards were fixed with a firmware upgrade. You can compare the real distance of a speaker which you can manually measure with the Audyssey calculated distance and undrestand.
Hello and thank you for sharing your knowledge and work. One question so far: You changed low frequency effects to -10db due to Atmos db bosst. Does this apply to a standard 5.1 sytem like I have too or should I leave this option untouched ? Thanks again
Yes and yes but best way to find out to listen the result yourself
Thanks again for your work! Really enjoy how my system is sounding now on my Denon x4400.
If I upgrade to an x6700, can I just import my saved profile from my old x4400 to the new x6700? Or do I re-do calibration to the new receiver?
You can manually edit the ady files side by side and copy/paste objects in json online editor but 6700 is a different beast and probably has different crossover rolloffs etc. I'd re-do the measurememts.
Thank you for your videos!! Is it really necessary to take 3 measurements and average them, or can I just use one measurements from the mlp?
One will be enough and even better if you are only listening at one position.
@@ocaudiophile Thank you! Im on a mac and I don't have the [A] / [B] option for some reason... only A/B. What do you recommend I do? Thanks again!
@@theweazel roomeqwizard.com/installers/REW_macos_5_20_14ea55.dmg
@@ocaudiophile Thank you, I didn't see this version when I looked! Now I have that trace option. Looks like the mac version doesn't have graph metrics though, which is a bummer! Thanks for your help.
@@theweazel It must have it, just with different buttons. Check the REW manual.
This is all really great, but I am confused about a discrepancy between the Speaker Level settings at two points in your video. At 36:58 you show values for, say, the LCR at -1.0/+2.5/-1.0, but at 46:00, after explaining to take a note of the Speaker Level and distances settings before uploading the curve-edited .ady file, you enter values of -4.0/+0.5/-4.5 for the same three speakers. How/why do those values differ when they should have been the same as the noted values prior to uploading the curve-edited .ady file indicated at 36:58?
:) Your attention to detail is amazing and you're right but in the middle of the calibration during recording the video, I figured I forgot to take a note of the levels so I used levels from another measurement. Just ignore the differences and follow the method!
@@ocaudiophile Thanks for that info - good to know, and now my brain can rest easy :) I did have another question after giving this process some thought: If one doesn't like what Dynamic EQ does to the sound in one's system, does this method work without having that enabled? I find that DEQ really bloats the bass, and makes it sound too overbearing, so I personally turn it off.
@@discostu668 I never turned DEQ on myself for the exact same reasons but this method tames all its disadvantages at the usual volume level you listen to your system and makes use of its boosts at lower volume levels. If you still don't want to use DEQ or DV then you can do a manul calibration with Graphic Equalizer. You'll find two videos on that in my channel.
@@ocaudiophile That is also very good to know. Once I get around to the small re-arrangement of my HT system, I'm definitely going to try this method out. Thankfully, I'm only running a 3.1 system, so I won't have to go through so many measurements! I'm looking forward to seeing how this method changes the sound in my space. The thing that struck me the most was the method used to determine the speaker crossovers.
I am using the latest beta release of REW (5.20.14 for mac), and have run into two problems. The first is that when I try to determine the distance from the cross-correlated channels, my sub distance is calculated as significantly further away than the actual measured distance. I should note that I ran the measurement with the sub phase angle at 270 degrees. Could that be the cause?
Subwoofer impulse peaks are very often too far to be compensated by the distance limits in the receivers and this is normal. It's due to the time delays introduced by the filters in the sub (or by MiniDSP if you're using one). You can try aligning its firts peak with the acoustic reference speaker instead.
Hey mate! Sorry, i have another question.
Im following this guide to a T, using MultiEQ.
Ive noticed that, Im getting a 10db difference in the LFE channel when using the REW sweep vs. The sweep you provide (this sweep is 10db hotter). I verified that I have Low Frequency Effects set to -10 in the AVR. I also verified that this only occurs with the LFE channel (the other sweeps match perfectly.) When I move the LFE effect back to 0, the LFE sweep from REW matches the provided sweep at -10 perfectly as well.
Any ideas? And if i cant figure it out, which one would you recommend i use? I have peaks with the provided sweep oved 90db. When using the REW sweep I have peaks over 80. If I move the LFE Effects back to 0, the provided sweep is hitting over 100db.😊
Thanks!
ALways use REW's LFE sweep because Atmos LFE sweep has a low pass filter and introduces extra delays due to that. If you download the latest early access version of REW, there're now measurement settings specific to LFE channel so you don't have to tamper with receiver LFE volume settings.
@Obsessive Compulsive Audiophile Awesome! Thanks!
@Obsessive Compulsive Audiophile hey - I tried using the -10lfe option in REW. I then set up LFE Effects back to 0. The results between selecting the -10db REW LFE option, and not, are nearly exactly the same. There is a slight 3db boost at about 80hz when not using the -10db option in REW, but it doesn't appear to be reducing the LFE channel by 10db. I'm wondering if it's a bug? I'm using ASIO and channel 4.
Just wondering if I should stick with the AVR option if the results in REW do not appear to change much when using the new LFE option.
@@thecrlghs The volume depends on the frequency range measured. If you measure LFE from 0hz to 24000hz like other speakers, the volume will not be boosted. If you measure 0hz-200hz for example, receiver will think it's an LFE signal and will boost it. Whether you use REW or dim it in the reciever or simply measure everything 0-24000 without chnaging any volume setting is up to you. They are the same thing.
Ahh. I think I follow! 😂
New to the channel and really like it.
What’s the best calibration process if you don’t have a laptop? I have Audy MultEQ XT32 on the Denon and the MultiEQ app, but no laptop.
It doesn't need to be a laptop but you need a computer to manipulate the ady file produced by MultEQ Editor
@@ocaudiophile Thank you. We need a cloud based version! Keep up the good work. Subscribed!
Would love to see a video on using RTA!
Coming soon!
At 8:20, you tell us how to use the control protocols to set the maximum volume to that which you use for watching movies.
I watch movies at a lower volume than you do. How did you convert your listening volume into the value you sent for maximum volume?
İ use 55 which is equal to -24dBfs, just replace 55 with a lower number
Once we set the max volume ( with the protocol command) that volume that we watch movies, should we make all our measurements at that volume on the preamp?
@@ronaldippolito6785 Yes, especially if you intend to use dynamic eq or dynamic volume because Audyssey will apply different levels of boost to different channels depending on the volume of the amp.
Hi OCA -Thanks for all your videos, but this one I can't wait to get started on!
2 questions that might be very stupid so please forgive me
I have the Marantz SR-6012 - it does have the Audyssey app but I just can't get the web interface to work. I get Error 403 - Forbidden/Access forbidden.
1) should I have the option use the web interface at all?
If not I might just have to use setup via the remote control - will work but takes much longer
2) but when I access the setup from the remote control - under audio/surround parameter I don't have "low frequency effects" at all
Will that give me problems?
Kind regards
I had 6011 and now have 6014 and both have Web access, maybe network control is off in your receiver menu setup.
Certain setup parameters only come out when there's an audio signal being fed. Play an atmos track and try reaching that setup menu item.
When playing sweep file from media player, the response curve capture by REW is 15 db lower than the response curve captured by playing sweep directly from REW. What should be done?
Turn off loudness in the receiver. The menu comes up only during Atmos signal playback so can be hard to find. If that doesn't help then your receiver is one of the few models which applies dialnorm setting to Atmos signals and you cannot do much about it.
Hello, could you please tell me the order of videos i should follow. For example, do i first do the sub alignment qnd house curve before doing the steps in this video? Or do this then align the subs?
Align the subs first and treat them as one and correct their summed response.
Did you delete a video (similar to this one) you previously had on your channel? I remember you sharing a file to input in json editor specifically for 80hz (you also had one for 120hz). That procedure worked great for me. I moved my seating & was hoping to use that video as a reference again but can't find it. I've tried this video, but every time I try to import the "dummy" ady file (after using the json editor) into MultEQ, it states that the file is corrupt.
th-cam.com/video/QWFyIDTe1fw/w-d-xo.html
@ocaudiophile Thank you for the response back. The link you gave sent me to a video with a private setting. So I can't view it.
Video related material:
Harman target curve:
drive.google.com/file/d/1N9C7...
Online MultEQ measurement file (.ady) editor:
jsoneditoronline.org/
Harman target curve for 120 Hz crossover (as created in the video) to directly paste into json:
[
"{20, 6.382}",
"{25, 6.592}",
"{31.5, 6.892}",
"{40, 7.152}",
"{50, 7.393}",
"{63, 7.557}",
"{80, 7.153}",
"{100, 6.13}",
"{125, 4.625}",
"{160, 3.032}",
"{200, 1.946}",
"{250, 1.19}",
"{315, 0.611}",
"{400, 0.267}",
"{500, -0.137}",
"{630, -0.327}",
"{800, -0.554}",
"{1000, -0.774}",
"{1250, -0.942}",
"{1600, -1.197}",
"{2000, -1.359}",
"{2500, -1.501}",
"{3150, -1.658}",
"{4000, -1.813}",
"{5000, -1.911}",
"{6300, -1.966}",
"{8000, -1.905}",
"{10000, -1.64}",
"{12500, -0.962}",
"{16000, 1.071}",
"{20000, 8.299}"
]
Harman target curve for 80 Hz crossover to use with MultEQ Editor app:
[
"{20, 6.657}",
"{25, 6.999}",
"{31.5, 7.482}",
"{40, 7.965}",
"{50, 8.378}",
"{63, 8.538}",
"{80, 7.752}",
"{100, 6.065}",
"{125, 3.931}",
"{160, 2.025}",
"{200, 0.984}",
"{250, 0.415}",
"{315, 0.049}",
"{400, -0.113}",
"{500, -0.392}",
"{630, -0.493}",
"{800, -0.659}",
"{1000, -0.842}",
"{1250, -0.985}",
"{1600, -1.224}",
"{2000, -1.376}",
"{2500, -1.512}",
"{3150, -1.665}",
"{4000, -1.817}",
"{5000, -1.914}",
"{6300, -1.968}",
"{8000, -1.906}",
"{10000, -1.64}",
"{12500, -0.962}",
"{16000, 1.071}",
"{20000, 8.299}"
]
@@doublet147
@@ocaudiophile Thank you!
Buddy if i dont want to have dynamic eq on and put a really house curve what i have to do???
Use Audyssey's own measurements then. The technique is explained in Audyssey ART video.
@@ocaudiophile thank you that im gonna try😇
@@ocaudiophile but can i ask you why you dont give a house curve with dynamic eq off buddy and you have to do all this process???
Check equal loudness contours!@@kuriakos36
Hi OCA. It is the last thing you remind us to do, but just wanted to double check. Do we change the Low Freq Effect back to 0db after we determine the crossover frequencies, or at some point before that?
That will not have any effect on the sound for sub and speaker crossover. LFE volume only changes special effects recorded for the sub channel only.
Thanks you so much for sharing your knowledge and enthusiasm with other fans of home cinema systems. I will follow this tutorial and aply it in my Denon receiver. In a first phase I will use the original Audyssey microphone, with the calibration that you shared for REW, to evaluate the results. Later I'm thinking ordering a calibrated microphone, but I'm undecided between the UMIK-1 and the UMIK-2. If budget is not an issue, do you consider the UMIK-2 a better option? If yes, why? Thank you very much and continue the awsome work :)
For AVR calibration Umik-1 is really good enough. If you will dive into advanced digital room correction for stereo hifi in the future, I would rather go for an audio interface and a XLR analog microphone. XLR mics have better timing accuracy. Basic sets will cost less than an Umik--2.
Much appreciated for your answer. I'm only in the Home Theater so, taking into account your comments, the UMIK-1 is good enough for my best use. 😉
I followed the first steps but I got stuck in the PC multichannel output. I was not able to do it in my NUC PC with a Realtek audio drive. Any tips?
@@miguelmedinaramos You can choose between ASIO and Java EXCL drivers in REW. For ASIO, you'll need to install ASIO4All and you don't need to bother with Windows sound settings. All settings can be done from ASIO control panel. If you use Java EXCL drivers, Windows default player should be your AVR and the AVR should be in Movie/MultCh-in mode.
Just to give feedback it was quite simple after I use another PC. My PC is outdated (hardware and software) thus the ASIO4ALL did not worked. I will update my PC and surely will work. ;)
New to REW,
Say you ran Audyssey and now starting to use REW and going to follow this video. What do you set the crossovers at to start?
All speakers are set to "Large" when you send a perfect speaker response .ady file to the AVR so no crossover at the start.
Some of my impulse peaks are on the top and some are on the bottom. Should I align to the "highest" peak or the peak that is on the same side? Thank you
No, always align for the strongest peak regardless of its direction. Btw, compare the phase graphs in Overlays and invert the polarity of a speaker if its impulse peak is in the opposite direction of the others. You can simulate that with "invert polarity" button under SPL & Phase tab and physically switch poles of the speaker if you decide to invert for real.
@@ocaudiophile Thank you! However the issues are my subwoofers and I can't easily change the speaker wires because of how they're installed. I checked "isReversePolarity" to "true" on the .ady file and it didn't change anything.
You can also change their polarity in json editor in the.ady file
@@ocaudiophile Thats what I tried. I changed the.ady file on the subwoofers: I changed "isReversePolarity" from "false" to "true" on the .ady file and it didn't change the direction on the impulse peak when measured.
Edit: The "isReversePolarity" command doesn't seem to work for me. I tried it on my center channel speaker and the impulse was the same - however if I physically switched the speaker wires around it does change... maybe its a bug?
I didn't try it myself tbh because I also switch banana plugs behind the receiver if I need to but I would think it's there in the ady for a reason. Maybe it's only used to remind the user to change polarity if Audyssey detects phase anomaly.
I'm stuck at importing txt file into Excel "Audyssey Curve editor". Somehow I have commas instead of dots in dB values..And then when I apply "convert text to colums" with Space ticked, the excel erases commas in dB values or example : from -3,965 i get -3965, and this of course don't work later in JSON. EDIT : I found the solution - when export I needed to tick "Use computer's number format"
Hi d1234
I have the same issue as you have (or had) - but you have found a solution. Unfortunately I still have the problem.
When I have the raw text in Notepad it wil say ex -3.965 but as soon as I paste it into Excel it converts to -3,965
so comma instead of dot - and I can't get JSON to accept this format. Have you succeded in that d1234?
I can't for the life of me find out what the heck I'm doing wrong?!
Still hopeful regards
It's related to your Windows regional settings which effect use of . or , for decimals. You can fix that either through Windows settings or by simply editing the formula in XL.
At 12:00, you describe how to set up the room curve. The settings you give will yield A Harman room curve.
To generate that Harman curve, is REW changing the test tone output to be higher at 20 Hz, then decreasing the amplitude all the way out to 20,000 Hz?
No, REW's sweep volume is fixed throughout the frequency band. Most speakers naturally have this slope towards the high frequencies
I don't understand, yet. When I look at your response curves, they look like they drop about 7 dB from 20-20,000 Hz. I guessed that the sweep amplitude decreased at the slope you entered in the equalizer preference panel, and level matching is how the response fit the Harman curve.
If the sweep volume is constant through that range, what drives the response to a constant negative slope?
@@ronaldippolito6785 The low frequencies are always boosted by the room walls and there's also usually additional downward slope intentionally given by the manufacturers if you check their anechoic response because we tend to hear these frequencies higher than they actually are so in fact it's flat to our ears.
Another great video, thanks again.
I can't seem to get audio out of my front height speakers using any of the "Atmos sweeps" files, any ideas?
I have double checked all settings for my receiver in windows, made it default device, no enhancements, etc
When using the dolby-test_tones_9_1_6 file it does match correctly for all speakers so I'm a bit confused.
It's a 5.2.2 system and a denon x4400H (should it be in "Multi Ch in" mode?)
Some things to try: Select "Movie/Stereo" or "Movie/Mult ch in" in the receiver, reinstall Windows Media Player, with default settings, Check sample rates to be all at 48kHz in Windows sound settings, Denon should be the default player, if all else fails try with VLC media player with Windows direct sound and HDMI bypass option active.
Hello and big thank you for the great tutorial.
After following the steps I finally hear my Dolby Enabled Speakers reflecting from the sealing. The sound in movies is impressive, compared to the sound the system does with just Audyssey. I don't know what the people at Audyssey tought when inventing the correction curves for the Atmos speakers, but it is not working well. The Atmos effect with my 4 Dolby Enabled Speakers is now, implementing your method, great. I now understand why people think, the "bouncing" speakers are crap. Audyssey cuts them in such a way they really sound as crap.
One thing, that makes me think that I might have made a mistake at some point through the tutorial is the fact, I still have to adjust to the high pitch of the sound. With Audyssey it is somehow warmer...And my correction curve in MultyEQ app after finishing your method looks like a zigzag with multiple zigzags... Don't know if this should be the outcome...If you can share your MultyEQ curve after implementing your method I would be glad to compare to main, just to be sure, main is OK...I cannot actually call my outcome a "curve" :)
Big Thanks for your time and the comprehensive why of taking me trough the tutorial.
Don't worry, it's gotta be a zigzag. The method uses Audyssey curves to apply the actual correction. The zigzags you see are the equalization applied to each speaker. I didn't undertstand what you mean by adjusting high pitch of the sound?
@@ocaudiophile If I compare the sound of the outcome now with the sound of audyssey, then I feel like the sound is much more "sharper" (almost scratching my ears form time to time), less "warm", and voices come out "thinner", thinner perhaps because of the presence of the high frequencies, which were not there before...For me now, it is as if I have different speakers, and I have to get comfortable with them first...
When I look for the crossovers with your method, my crossover for the Dolby Enabled Bouncy Speakers comes out at 90 Hz. What would you recommend me to do with this relatively low frequency. Should I crossover them higher like 150 hz? Somewhere in your video you talked about high speakers and crossing them also at 80 hz, to avoid hearing the top sound out from the subwoofer on the ground. Or perhaps I misunderstood something...Of course my dolby enabled speakers are not high speakers...Perhaps thatswhy I should choose the higher crossover?
Front speakers have the option for 60 hz and 90 hz
Center comes out on 40 hz or 80 hz
Surround at 60 hz , 90 hz or 120 hz
What frequency would you choose in my place?
Now I am mrunning Front @ 90 hz; Center @ 80 hz; Surround @ 90; Dolby Enabled Modules @ 150 hz
Thank you in advance!
@@miroslavpetkov9895 Do you have one or two subs and how are they connected(ie MiniDSP or directly to the receiver)?
@@ocaudiophile I have only one Sub (SVS SB 2000 the older version, not the pro version) at the present moment. My room is small 6,50 m to 5,05 m to 2,47 m. My speakers are Heco The New Statement as front, Heco Celan GT 42 as center, Heco Revolution 3 as surround, Heco AM 200 as Dolby enabled speakers in a 5.1.4 configuration. All of this is connected directly to my AV-Receiver Marantz SR 8012.
Appreciating your help.
Thanks.
@@miroslavpetkov9895 Ok, if you don't use MiniDSP for your sub you can time align the sub properly with the receiver's distance settings. MiniDSP causes extra processing delays which makes this very difficult. When the sub is time aligned, it will be very easy to see where the crossover frequencies should be. For the treble problem you're having, suggest you follow the method in the Dolby Atmos Music video. I used Dolby's own target curve there which is much softer at the high frequencies.
Awesome! Do you recommend investing $80 for calibrated UMIK-1 microphone for setup or ok to use the mic that comes with Audyssey (I just purchased Denon x4800h receiver that have Audyssey multiq 32x with mic)?
Congrats on your purchase, that receiver has the same processor inside as the top models and is made in Japan.
You can use the Audyssey mic that came with the receiver but make sure you use the calibration file with it while measuring speakers in REW. You can find the the link to the calibration file of these mics in this video:
th-cam.com/video/KIozAzQH3CM/w-d-xo.html
@@ocaudiophile Hey OCA, no Calibration of my soundcard necessary?
@@leo180811 No, you can even use Audysssey mic with REW to measure your speakers. The tutorial should be showing how. If not that one, definteley the earlier ones on manual tuning.
Hello OCA. After - Pinpointing crossover frequencies for each speaker... I get following error - Error: Speaker has no bass response according to your measurement file. Please check! - program execution has stopped! Please let me know what I might be doing wrong.
Check your original ady file for absurd trim adjustment settings for subwoofer unless you've open baffle speakers with limited bass response.
Starting down the path and already have a question. JSON is new to me. I was cutting and pasting and something happened in the editor that collapsed the tree view to one location. The editor just shows an error with 16384 items. How do I go back to see the full tree and continue my work??
Fixed by switching to text mode then back to tree. Uploaded the new audy file to the file (had to change extension from json to audy) but the measurements aren't .4 and the levels are 0 like you show. Whats up??
Just click "New" in the top left menu and drop the .ady file in it again. It will be refreshed. Also check your MS Office regional settings. Some Excel versions can replace , and ; which might be causing the error.
@@ocaudiophile Appreciate that, sir. New to REW and looking forward to getting things dialed in and learning along the way. Already subscribed and appreciate the content!
Omg an all in one Audyssey cal tuto
You should Switch the “Network Control” setting to “Always On" in the receiver menu. You can leave the sound profile at stereo, it will change automatically.
@@ocaudiophile Thanks :) Network Control was already ON but I saw on forums that Web Control for SR7012 is no longer available (only for SR7015), I have to rollback the FW, but no idea how to hack the AVR FW for this
So for stereo I leave Movie Stereo or Pure Stereo ?
@@Mitholman73 I don't change this setting at all, it automatically switches depending on the incoming signal. Stay away from pure direct modes as they bypass the subwoofer and Audyssey settings.
how do i know which channel number in json editor corresponds to which speaker when im importing custom points
for instance i have 12 channels, 0 through 11
i take it 0 is front left and 1 is center based on what you did in the video but what about the rest
In CommandID there's the channel name like FL for front left speaker
@@ocaudiophile thanks must have just missed it
After all transfer files to avr can we leave reference level offset in 10db or 0db ?
0dB
@@ocaudiophile pls do tutorial how to align subwoofers with minidsp
@@avsolutions212 One way is to use REW Alignment tool as shown in my Subwoofer Alignment video and then dial in the time delays to MiniDSP.
Just an idea. Create the PEQ in REW, import in Rephase, export impulse response of the filters back in REW, create division of the imported filter versus a flat impulse. This would get the FIR filters from the PEQ. Just tested and seems to be working very well. You can also further PEQ the result and apply another bank of filter correction in Rephase for even further refinment of the initial response. Do you see any issues with this approach?
It should work but with one inversion in REW, you already create the required PEQ settings as custom curve edit points for Audyssey. Besides in general leess is more in digital correction . Too much eq especially in high frequencies will cause boxy sound. Try the method in Audyssey for Dolby Atmos music video where there's less correction and you might prefer that one for movies as well.
@@ocaudiophile Thank you! I will. I learned so much from your videos. I looked at one of my first measurements after Audyssey automatic calibration and it's night and day versus what I got following your methods!
@@horatiud.5976 Good to hear that.
I am stuck at the ASIO4ALL mic driver issue beyond logic (red exclamation). Trying to resolve the past 2 days. Any help regarding this ??
Can I use the JAVA driver instead ??
Yes but use Java EXCLusive drivers. They are as good as ASIO drivers and can do multichannel.
@@ocaudiophile Thanks for the prompt response 🙏
I am having an issue where my front left and right mains have a huge null at 120hz when doing the vector average. REW says at ~123hz the phases of L and R are off by 180 degrees. Any ideas?
180 deg difference between phases of left and right speakers is indeed the reason for the drop in VA and it will be audible (or rather inaudible) when both speakers play that frequency, too. Probably different SBIR effects due to asymmetric speaker placement. Check for 70/140cm extra distance travelled from a speaker to LP reflected from surrounding walls that the other speaker doesn't have. Either move the speakers /LP elsewhere or direct 123Hz and below to the subwoofer with a high crossover.
@@ocaudiophile bumping up the xo is one of the things I tried. You are right, my living room has no wall to the left, so the speaker setup is asymmetrical. I also have 2 subs of differing sizes, both on the right side, in opposing corners (Front Right & Rear Right). It's the position where I get the smoothest blending - with the caveat that there is a null right at, you guessed it, 120hz lol.
For trying out combinations in REW, I measured each speaker as large and I also measured the subs using the LFE channel with my LPF for LFE set to 250hz so I could get the maximum range. Here is the link to that .mdat file: drive.google.com/file/d/1Yt9Ml_EA3Op8J8suJxTuAHYPex8VvzCT/view?usp=sharing
Adjust FL -0.5dB, FR +0.5dB, SW -1dB. EQ 1/48 smoothed versions of all from 20Hz to 200Hz with 5dB boost for a Dr Toole curve with target level at 52.62dB. Export all 3 filters to the ady and cross over the mains (FL & FR) with your sub @150Hz. That seems to be the only way you can fix taht huge hole around 120Hz without moving speakers or using heavy passive absorption.
Regarding the "Speed of sound" value setting/correction. What is your reference when determining whether your AVR is using 300 or 343 m/s? Do you measure it with a tape/ laser, or do you measure with both Audyssey mic and Umik at the same position and compare?
It's caused by the AVR so it's mic independent. I measure with laser just one of my speakers and compare with Audyysry's own finding. The difference is usually too large and even with tape measure you would understand your receiver is using 300m/s. Btw, after a recent update my Marantz sr6015 now uses 343m/s. Many other models have been fixed but not all.
@@ocaudiophile Thanks! My Denon X4500H also received a couple of updates over summer, fingers crossed.
I'm confused as to what's going on. During my first measurement I set Denon AVR to 60 (scale 0-98) where I usually listen at. Audyssey Mic at listening position at 12 feet away or so. I got error saying that level was low and asked me to increase volume. Thought it was strange, but increased it to 70 and and that was loud and it took measurement with REW. Then, following your step by step instruction I tried same with supplied FL mp4 file. At the same AVR volume it sounded much lower and error was "The highest level in measurement is -54dBFS, for greater accuracy increase the input volume, the sweep level or AV receiver volume to raise the highest input level above -35 dBFS." Thought it was strange since my AVR was set at -10 DB (70 on the scale 0-98). Any ideas?
Use either java EXCL or ASIO drivers otherwise you've to adjust mix gain properly from windows sound settings
@@ocaudiophile This was done using ASIO drivers, and still it seems that Audyssey mic at 12 ft away at deafening 70 volume setting on AVR still comes in as low level according to REW. Mic gain is set at 100% in windows.
I've spent 2 hours today trying to figure it out. I'm gonna go deaf. Nothing is working. I keep getting errors in REW that levels are low. I boosted mic gain in Windows settings by 30 db (all the way up). AVR volume is between 65-70.....This happens either with internal REW calibration sound playback or outside calibration files. Tried both ASIO and EXCL drivers with calibration mic file you provided for Audyssey mike installed. Any ideas? @@ocaudiophile
@gr8nrg187 Did you "check levels" in preferences tab after you select input and output channels? You should be seeing output level around the top and mic level close to that and adjust volumes accordingly.
that's what I did. The output level is good near the top at -12dbfs but mic level even if I stand close to speakers is about -65 or so. Windows set mic volume to 100 and boost to the top. Is it possible my mic is defective or your custom curve for it does not work? Should I just buy UMIK-1 or just blast my speakers while setting measurements at around 75 (scale 0-98)? ? @@ocaudiophile
Is there a possibility with an existing measurement to activate the 2nd subwoofer channel in the .ady file?
There was but MultEQ editor app got an update recently and now can host 4 subs and subwoofer syntax changed a lot. Your best option is to ask it in this forum:
www.avsforum.com/threads/announcing-ratbuddyssey-a-tool-for-tweaking-audyssey-multeq-app-files.3006886/page-87
Thanks for that great tutorial. In a previous video you also used parametric eq filters (for yamaha), but i am using multeqx on my denon. Can i also use paraeqs with this tutorial and do i meed the hassle with python and excel, when i can use multeqx?
Denon does not have parametric EQ, it has only a graphic eq with fixed frequency bands but of course you can use that insetad of Audyssey. I have one preset with Audyssey and one preset with GEQ in my Marantz set up.
@@ocaudiophile thanks for the answer. Since multeqx support the usage of parametrics eq i will try my luck. You can also apply them without measuring your speaker wirh audyssey, thats what i want to try with rew then according to yout videos.
abit off topic but not really :) do you ever use the aligment tool? i read alot about it thru the different forums, that its the "correct" way to set your sub "distance" right etc..
I've a video on subwoofer alignment which uses that
I am not sure if anyone has asked this question before: with this method you calibrate with DEQ enabled at the MV level that you typically listen, but what happens to the response if you don't listen at the MV level where you calibrated everything? I know DEQ boosts its effects the lower you go in volume, but does it maintain the same level of (for a lack of a better word) neutrality throughout the response?
İt might not be optimal but deq defo helps at the lower volumes
Hello oca how can i make full range calibration my speakers with evo ??? Please little help
Use the method explained in Audyssey ART video. It's the best way to do full range EQ.