Optimal Subwoofer Alignment with Cross Correlation
HTML-код
- Опубликовано: 20 ноя 2022
- Learn how to optimally measure, align and correct your subwoofers with the use of the new "cross correlation" feature in the "Room EQ Wizard". Get your BASS right!
"home theater system"
"home theater speakers"
"home theatre setup"
"audyssey multeq xt32" "audyssey einmessen" "audyssey multeq-x" "audiosciencereview" "audyssey multeq editor app" "audyssey calibration" #rew #roomcorrection #subwoofer #subwoofers #subwoofer_bass_test #hometheater #calibration #cinemasound #dolbyatmos #audiophile #hometheatre #bassboost #bass #alignment #aligning #bassmanagement #minidsp "minidsp 2x4" "svs" "rel" 
Наука
These videos are SO HELPFUL. Thank you so much and keep up the good work!
Thank you 👍
... Best channel I've come across ... Thank you for info .. and for a prompt response..
Welcome!
I am so grateful for your REW demos. So far when I applied any of your techniques, my system has improved I have a tube amplifier and solid state amplifiers so they have different sound profiles and trying to dial those in a 2.2 stereo set up requires different tuning. Thank you. Again for an easy to follow tutorial on using REW. I’m going to try this subwoofer time aligning approach now w my MINIDSP SHD.
Good to know thanks.
@24:35 align phase slopes at cursor
thank you so much for that, I do REW since a while but as I do a new DIY Sub build with 2 PR on each sub this info is pure gold if I look at the final output with having a 2x4HD
Thank you for your tutorial. I was able to align my two front subs with my two rear subs and eq flat with Target Type NONE using the Harmon curve. No polarity reversal. I then set high and low pass filters and a compressor (As Per AVSFORUM) in my miniDSP 2x4 HD. I vary how much bass I need by using a LOW_SHELF filter at 47Hz, 3-6dB gain, and a Q 0.7. Thank you!
Fantastic!
Good Job man ，THANKS ！！ so helpful ！ KEPP ！！
Thank you too!
Thanks!
🙏
Another great video. When setting the bass Managment cut-off on the target curve do you set this to best match the existing bass curve or the value for matching your other speakers e.g. 80Hz that you would set on the AVR?
match the existing curve!
Danke!
Great video! Thanks. I just got a Minidsp and Umik1 so plan to go through your tutorial to align my 2 subs. I have an Onkyo receiver with AccuEQ for room correction. Should I run that first then this tutorial or align the subs first then run AccuEQ?
I am not familiar with AccuEQ but I'd run it first and then correct the subwoofer distances it came up with accroding to my findings with this method. Btw, there's a newer tutorial for subwoofer alignment:
ruclips.net/video/DofUiHz7Fu4/видео.html
Thanks for the great tutorial. Do we need to align SPL of both subs, before doing this? One of my subs, behind the couch is closer to the MLP and therefore louder. Or shall I just set them to the same gain if I have identical subs?
İt's easier to reach a flat response when you align their SPL at the LP beforehand. Ideally, you should also EQ them individually for excessive peaks.
Hi! Thank you so much for the great video! I am very excited to do an overhaul and recalibration of my 7.2 system based on your videos yet before I enter the rabbit hole I had a few questions. Sorry it’s a longer post.
What is the best order to watch your videos to fully calibrate ones system? I suspect going though your “Optimal Subwoofer Alignment with Cross Correlation” video to time align the multiple subs is the first step. Then go though the “Dolby Atmos Height Channels Measurement & Calibration with REW” video for levels and time alignment for the other speakers. And then go though the “Supreme Digital Room Correction with REW & rePhase” and “Excess Phase Inversion - Filter 4” videos for the final calibration for the L+R speakers . Does that seem like the best order for approaching things? Any other videos that I should be aware of?
I also had a few other questions:
Is “Supreme Digital Room Correction with REW & rePhase” the process that you think results in the best results? As I have seen your older videos such as “REW (Room EQ Wizard) Top Tricks: Convolution with Inversion (no EQ filters, all FIR!)” but just making sure that if I am to pick one video to follow it is most recent Supreme Digital Room Correction.
For the final EQ process do you think your results would be comparable or better than DRC-FIR / DRC Designer or Accurate or Audiolense or would you recommend any of these? I realize that only DRC is the only free one on that list though.
I have a Marantz receiver and have never found the published Configurable PEQ values so that was much appreciated. But if I am going to run the audio through my computer should I just ignore any and all EQ that the Marantz can do? Do you think there would be any audible downside for having the manual Marantz EQ active as a base layer and then the PC based EQ for the L+R on top of that? (FYI I have a Marantz receiver with the older MultiEQ)
I have a phantom center speaker as the Marantz just routes the center signal to the L&R speakers. If I am using a PC and a PC based EQ for the L&R speaker then should I route the center signal to the L&R via Equalizer APO and then that signal would get the appropriate L&R EQ applied to it correct?
Finally, have you thought of posting your videos and process on forms such as AVS Forms or Audio Science Review? As I think they could get more exposure and use if the process was outlined in text and then your videos for the content. If and when I get around to doing this overhaul and EQ I might post my results on ASR if that is ok with you.
Thanks for any feedback or clarification as I am excited to try all your tips and tricks out!
Thank you for your kind thoughts.
I didn't launch the videos in a particular order and all of them will achieve results but in general later dated videos are more efficient mostly because REW keeps adding features and I have to develop new techniques to make use of them. You're right that subwoofer alignement is a good start if you've multiple subwoofers. For Dolby Atmos, you've decide whether you will use Audyssey or not. Audyssey has tools for low level listening but manual calibration has more realistic sound. I keep one preset for both. If you will go the Audyssey way, follow the method in Dolby ATmos Music video, it makes use of the latest REW features and gets the best out of Audyssey.
Stereo calibration is a whole different ballgame. It's much more precise than the surround calibrations but requires a PC as 65k+ tap FIR filters are generated. They have no implementation in a surround receiver. REW top tricks is a good start. It's simple, achieves great results (people's feedback) and is supported with a written guide. Supreme DRC is a step forward but easy to get wrong. DRC,Acourate and Audiolense are the only tools which are capable of achieving SOTA DRC and I learned a lot from them. But none are fully automated, all have a learning curve and as you mentioned most are not free. REW is now capable of eveything that can be done with them and more. So if you'll dedicate your time, better improve in REW as it shows you immediately the results of any operation and is a good tool to understand what you're doing and learn from your mistakes.
I stream music from a PC to a DAC and from there to my Marantz AVR at the moment (as I sold my stereo amp and didn't replace it yet) and it's quite good especially if your front speakers are easy to drive with high sensitivity. You can hear tiny differences between different corrections applied. I use Roon in the PC so I use Roon's convolution engine but Equalizer APO works just as well except for very small timing issues. But the AVR should be in pure direct mode I am afraid. You cannot use EQ Apo and equalizer in the receiver together. You don't need that for stereo listenning, FIR filters can do all the work and a lot better. You should have separate calibrations (and your sources will be diffferent anyway) for stereo listening from PC and surround content from a bluray player or a movie streaming service.
Phantom center works (even better than a low placed center speaker if your fronts are placed correctly) and in fact you can use EQ APO with 7.1 surround (not Atmos though) if you can play surround sources from your PC.
I have posted one or two links in the forums at the beginning but I don't think they got much attention. Of course feel free to post if you like.
I think I answered all your questions.
Amazing video, thank you. As opposed to using IIR EQ bands to EQ the response, could I perhaps do a target phase inversion filter and EQ the bass with FIR for better result? Thanks.
Yes you can!
Thank you so much for making these videos! I successfully aligned my sub 1 & 2 by following the instructions. However for my sub 3 & 4, when I ran the Cross Corr Align, one of the trace’s impulse peak is flipped upside down against the other four traces. Do need to do anything to fix that before I proceed with Vector Averaging and the rest of the steps? Thanks again for doing this!
Cross correlation align may get subwoofer impulses wrong sometimes. Check group delay graphs of all 4 subs. The lines at high frequencies (ie above 1000hz) should be identical. You can measure the odd subs required delay on that screen with mouse right click method.
Great video as always. Since I don't have minidsp, you mentioned that when you calculate the delay for 2 subs then you can use these delay value in the AVR. In that case how you do that? Adjust the relevant distance difference between the subs? And after you run audyssey you make sure that this delay is kept?
I didn't use both subwoofer outputs since many years but I guess you can adjust their distances and volumes separately but can only EQ the sum of them. That's the first option you should try. The method will flatten the total response and timing will be precise.. Phase aligning subs between each other will require y-split cable and using single sub out of the receiver and continuous phase alignment knobs on subs rather than the common 0-180 switches.
@ocaudiophile Many thanks for the reply! I have a denon 4700h so I have 2 independent sub outputs. Also the SVS subs offer via their app to do fine phrase adjustments on the sub. Would that work in this case?
@@HarryM2022 In that case I would align the subs between each other first with the SVS app and then connect both subs as one to a single connection in Denon then adjust its impulse peak with the rest of the speakers as explained in the method.
Hi there, I love your contents so far! You have shown here doing 2 sub alignment. What about if I have third or fourth sub? How do I get around this?
You can still use REW Alignment tool and the same technique. Combine first two subs then add the third to the combination, so on and so forth.
Hi! Thanks for this very detailed tutorial. This is the second tutorial I watched and I hope you continue to make these. Here is my question:
I got 2 subs sitting next to my main speakers covered around 100 Hz. Do you think it is still best to use this method and align the subs together, then align them with the mains? Or align each subwoofer to its main?
Also, let's say I first aligned the subs with your method, then what is the next step to align the subs to my mains? Measuring both L+R subs at the same time, measuring both L+R mains at the same time, then time-align those two graphs? Thanks for your help 👏🏻
You should always align your subs between them and then eq them. Then time align the resulting sub (their vector average or aligned sum) with your fronts for the central listenning position. This is done by bringing their impulse peaks on top of each other by adjusting delays. Finally you have to crossover your L & R speakers at the correct frequency checking your response graphs. I will launch a video on that very soon!
@@ocaudiophile thanks for the reply. I'll check your channel.
Thank you for your detailed videos. Just one question....can this and all your other calibration videos, be done on Mac OS. Because, you mainly talk about using ASIO4all to get multichannel output...somehow, I have not had great luck with it and Windows, on the past. Mac OS supports multi channel natively, so I find that easier. Once again, Thank you for your dedication and hard work..
You can use Java EXCL drivers of REW with Mac for 7.1 channel but you will not be able to play Atmos sweeps with a Mac if you are dealing with Audyssey & Atmos. For Atmos with a Mac check my latest video:
ruclips.net/video/aZ4k04uAf2o/видео.html
You are a wizard. Thank you for making these videos.
One question I have - Do you have a similar style video on aligning main monitors to the subs? That is the last piece I'm unsure about. I wasn't sure if it's best to time align a small band-limited region near the XO or use the impulse response.
I have explained sub - main time alignment in most Audyssey videos but in general try aligning impulse peaks, if the distance required is too high, you can align first peaks as well. Crossover frequency should ideally be based on the speaker specs ie where their anechoic response starts to roll off.
@@ocaudiophile You are the best. To clarify, largest peak in impulse or onset of impulse? Thanks so, so much!
@@mrbbizzle the peaks you see in Overlays/Impulse tab.
Awesome video. Just wondering how you’d utilise this to EQ 4 subwoofers.
From what I understood you’d do 5 measurements in MLP then 4 measurements for any other positions.
You can align them in pairs and use the vector average of the pair as on e sub and align with the third and then vector average again an align with the fourth or you can use the free tool MSO
Again thanks for such great content. 6 questions are coming to my mind 😀and I would love to hear your opinion on those.
1. I wonder how you decided to put your crossover frequency to 80hz? Is that based on any measurement in your room or is that just your common recommendation?
2. Are you measuring the subs with crossover frequency in avr set to 80? Because when you set the eq target curve (around 32:30) you are using 100hz as your bass management cutoff, so why that when your actual crossover is at 80.
3. When you level match the subs with the main speakers, how much do you boost your subs compared to your 75db level for your mains? And at which frequency do you start boosting the bass?
4. What is your recommendation: I have 5.2 setup, the mains play through miniDSP DDRC-24 (with Dirac). Right now I am playing my mains through minidsp dirac simply because it lets me adjust the target curve sooo easily and the subs are played through the AVR (+ miniDSP 2x4 boxed). Would you rather do manual setup for all speakers as you show in your videos or just use dirac for the mains and do manual for the rest of the speakers.
5. Since I have miniDSP DDRC-24 (which is actually a 2x4hd) I could also run my mains + sub through the miniDSP, while not using the AVR Sub out. Do you see any advantage of that (using dirac to measure front+subs as one speaker), as opposed to using MiniDSP (with Dirac) for the fronts only and using the AVR Sub Out for my 2 subs?
6. Are you planning to produce videos on FIR filtering using 2x4hd?
I am afraid I have one more question😅
If you only eq the frequency response up until the crossover point e.g. 80hz, wouldn‘t that be bad for the LFE channel, as it will play up to 120hz?
80Hz is the THX standard for movies. It's accepted to be the frequency where human ear starts to "locate" the source of the sound and which speaker produces the bass frequencies below that doesn't really matter. So the idea is directing all the bass (which is almost always the most problematic area in a room) to a single source (the sub) which can be easily equalized for a flat and smooth response at the LP.
I don't have much experience with the Dirac in Minidsp but I'd rather keep it as simple as possible (at least to minimize distortion levels) and use just one minidsp box and the AVR as power amp.
You're right about the LFE channel but there's really very little content in general and even less conent above 80Hz!
I've been learning a lot from your videos, thanks!
One question about MSO which you mentioned briefly: would it not be able to generate a flatter overall response (at least in theory) since it generates a different set of PEQ filters for each sub? So we'd have 3 sets of filters total for 2 subs (shared filters on input and separate ones on each output).
Any chance you could compare the results of this method vs the best MSO optimization you can get? (Maybe an idea for a new video.)
Individual PEQ for both subs wouldn't have much benefit over PEQ over combined sub response especially for just 2 subs but MSO has advantages for 2+ subs and also can utilize all pass filters.
Hi there, First I would like to say Thank you for the time and effort you have made in making these tutorials. They are really well done, easy to follow with clear explanations.
I have followed “Optimal Subwoofer Alignment with Cross Correlation” step by step but do get the result I was expecting. I have been through the process twice and are not sure where I am going wrong.
My results were disappointing with a delay to the back sub. The result improved quite a bit without the delay, which is weird. Maybe I need to reconfigure the 2x4 HD as single mono input?
Can you please offer advise or instructions as I would love to complete this process and achieve the desired result
My System
Pure AV Power Conditioner/Cleaner PF60
Purpose built HTPC, Windows 10, Jriver 30 Media Center, Asus Essence STX 11 Sound Card
Emotiva XMC-2 Processor
Emotiva XPA Gen-3 5 Channel Amp
Onkyo A9755 2 Channel Amp
MiniDSP 2x4 HD (setup a Dual Mono for Dirac reasons with disappointing results)
Front Sub distance to LP 3.9m - Rear Sub 2.15m to LP
MiniDSP 8x8
Dual SVS PB4000 Subs
SVS Ultra Front L/R & Center
Bowers & Wilkins 703 (side channels)
Bowers & Wilkins WM6 (rears)
It's hard to detect the problem in this comments section without seeing the actual .mdat file. You might need to be choosing different dips to align the subs. Your system is capable of using even FIR filters as taps and I am sure can produce very satisfactory results. Try this: FIrst align each sub volume to be identical at the LP. If there are severe peaks in each subs response, first eq them individually to flatten those. REW has the exact filter settings for Minidsp 2x4 HD. Then experiment with the alignment tool to produce a flat combined response which also doesn't comprimise SPL at the lowest produced frequencies. You can usually achieve that with a minimal delay to one of the subs. And then EQ the combined response again with REW. FInally time align the combined sub response as one sub with the rest of the speakers by checking its impulse peak.
O.K Thanks for the suggestion.
What I have done is change the processor crossover from 80 to 200hz while completing the process
I now have a delay at 2.85ms as opposed to 33.85ms
The resulting sweep on completion now look much flatter, have ever not as flat as my result doing the process as others recommend in other RUclips clips (Home Theater Gamer)
So think I am now heading in the right direction ?
Hey! Thanks for another cool video. How would you change the process if there are more than 2 subfoowers? And how do you then align resulting sub's response with mains? I have to low-tuned tapped horns and two ported subs. Mains are synergy type horns playing from about 60hz up. Two ported subs are mainly to fill in the dips caused by the room. So tapped horns play lower and also high-passed lower, ported subs play higher and high pased higher. Would you still follow more or less the same procedure?
I'd suggest using MSO for more than 2 subs.
@@ocaudiophile Can you please make a basic video about MSO with the new updates?
Thank you for the tutorial. Did I miss it or you didn‘t provide a measurement after you applied those filters in order to see how close the predicted and actual results are?
I didn't but I did "Generate measurement from predicted" which will produce an identical impulse response to the actual measurement.
I've been watching your clips for quite a while now. Awesome insight into REW. It seems like it's capabilities are infinite 🤯.
I have a full digital setup and currently using a DEQ2496 between my source and DAC to apply some PEQ. All is presented through a 2.2 active solution, studio subs and monitors with full balanced connections.
Can you reccomend me a hardware solution that can work in the digital domain (I want to keep my DAC) in which I can apply REW generated filters? I am thinking of something that can apply eq and time corrections on defined freq ranges, so I can apply sepparate filtering for subs and monitors.
Does such a device exist? I didn't manage to find a MiniDSP solution for this, but I may be missing something.
Thank you for your imensly insightfull work! 👍🏻
I checked the Behringer unit, it can apply delays between main and aux signals. If time alignment between each sub and between the subs and the speakers can be done there then all you will need is to add convolution filter capacity to your source. If it's a PC/Mac, you can use free software tools like Equalizer APO or Camilla DSP which are perfectly capable of processing all the FIR filters you can ever create. Paid software like JRiver and Roon also have excellent DSP processing capabilities. Without a computer in the equation, you are limited to MiniDSP options which have very limited FIR filtering capabilities but you can still do basic crossover and box phase corrections.
Great video's appreciate the effort. Can you please explain or point me in the correct direction for time/phase alignment of 1 SUB and 2 Mains in a stereo setup? Using a mini DSP channel 3 output to SUB, channel 1 and 2 out both on. I don't under stand how you set the delays this way, they don't match up in REW using time alignment for just a output 3 SUB time adjustment, and if I implement delays in left and right it will get them out of sync for the stereo presentation, won't it? Thank You
In the most common and practical way, you should only time align left, right and sub to reach the LP all at the same time and crossover your mains with the sub at 80Hz or below. This is all about adjusting their time delays in Minidsp such that both speakers' and the sub's impulse peaks are on top of each other in REW/Overlays/Impulse tab. In this video, I have explained how to do that:
ruclips.net/video/J7mToOXTt-o/видео.html
This is practical because you manage your bass response with just one sub and you've a lot more freedom for its placement in the room and you can EQ the sub quite extensively and as necessary without degrading the overall sound.
But if you have large speakers and you want to make use of their woofers along with the sub for more uniform bass in the room I would suggest first to practice for the best sub location with REW's "Room Simulator". When you're done, either measure L & R speaker together or "vector average" left and right speaker measurements (almost the same thing except ofr 6dB volume difference) and use this single response for L&R and the subwoofer response as the two measurements to align with each other in the "alignement tool" as explained in this video.
@@ocaudiophile I really appreciate the thorough reply sir. I will follow your suggestions and give it a go! That is as soon as the woman will let me do some more measurements. She asked me to take a little break on what she calls the "whoop whoop" noises for a couple days. 😉 Keep up the great work, I don't know anyone else on YT that takes things to the depth you do, I'm the same way, thus the "whoop whoop" prison for a bit.
@@robertp5367 My neighbors have no idea what that sound they regularly and repeatedly hear is. They suspect it must be coming from my flat but they can not give it a meaning and luckily they would rather not question 😂
Hmm, I wonder if it would be possible to use these tools to time align and find optimal eq for lfe+main (given low enough mains), given only one real sub. Probably complex, since two or more sources would be fixed, and three eq curves to apply.
You could but you are forced to normal time align FL and FR with each other according to their distances from LP so as to keep sound stage at the centre. Then you can align FL+FR combination with the sub with the phase method simultaneously picking the correct crossover frequency.
Thank you for the tutorial. You didnt mention anything about level matching or on what volume the subs should be measured. Can you help?
Ideally, both subs should be level matched at the LP and left at that. Manipulating aligned response with individual subwoofer volumes is not a meaningful practice IMO.
Hey @ocaudiophile,
If taking repeated/Multiple sub and Main measurements for use in MSO, should I cross corelate each speaker to itself and then vector average or is there a better way to do that?
Should I make a minimum phase calibration file for my mic to use for MSO measurements as it does take phase into account? Should I use this minimum phase calibration file for all my measuremnts?
Thanks for your help.
If you are measuring at multiple mic positions, you should cross correlate each speaker/sub to the centered measurement at the MLP before vector averaging.
Since MSO can also use allpass (phase) filters, you would have some (albeit minor) benefit from using min phase version of the mic calibration file.
Coming to this video a bit late, but I was wondering why you advice against EQ'ing if you've flipped the phase of one sub to obtain the smoothest response?
This is what I've done, and I've applied EQ with, seemingly, really good results as far as frequency response and uniformity between seats is concerned.
Thanks for a very informative video and thorough walkthrough
I must have advised against EQ'ing each sub individually. You will still need to EQ for the summation due to cancellations/additions in the combined phase responses.
@@ocaudiophile, ah that makes sense.
Still, your video made me try and redo my calibration using in-phase and a higher delay, and though it did not give me quite as good consistency between seats, I found that flipping the phase was actually costing me a lot of output around 27 Hz. So I think I'll keep the slightly worse consistency in return for much less need to EQ
Imo time adjustment of the subwoofer group with the rest of the speakers is also equally important. One trick is to make sure high frequency group delays are all in sync.
@@ocaudiophile, that makes sense. I'll be redoing the integration today or tomorrow, before running DIRAC sometime in the future. That was actually partly my reasoning for trying to avoid running out of phase. It seems to confuse DIRAC and makes it force the subs out of phase with the mains, so I have to go into my MiniDSP and flip the phase on the subs back
Hey OCA, I'm going to start measuring for multiple seats. How many measurements per seat would you recommend? I have 3 seats. Would you recommend the same number of measurements for a full-range speaker also?
I'd gather measurements mostly around the MLP and take fewer measurements the father away from MLP I go. Make sure you cross correlation align all measurements for one speaker to the one at MLP before you vector average them.
@@ocaudiophile thank you. I will do 👍.
This is a great video. Unclear to me though is if you align the sub levels for each sub at the MPL before taking measurements? I have 4 subwoofers that I would like to align. You would recommend MSO ?
MSO will help you get the strongest total input from the subs but if you're not going after that, time aligning each sub to MLP and applying an EQ to their sum would also work fine.
@@ocaudiophile I would start with the manual way first. What about the difference in volume levels ? My rears subs are quite a bit futher away so naturally the are a few db lower in volume. My front subs are in the front corners and my rears are firing into a open ended room. So they have vastly different response under 40 hz. Would you be doing a tutorial for MSO?
So to align those subs I was thinking the following procedure.
1. Align sub 1 + 2
2. Creat virtual sub 1 ( 1+2)
3. Add sub 3 to virtual sub 1
4. Create virtual sub 2 (1+2+3)
5. Add sub 4
Is there any issue I should be aware off?
You should eaulize the volume levels, too. Equalizing average pink noise levels (which you can check with REW before any measurement) would suffice.
That's what MSO would do and it's the better way of course.
Hi, thanks for your detailed guide... Just to be sure I am getting you guide right...After you have your aligned curve for your both subs, and you would enter the EQ Filter to your AVR manually, what EQ Filter do you use for each sub? It will be not the aligned Curve right?
You need to only apply EQ filters to the "Aligned Sum" of the subs. WHat filters you will use depends on what DSP you are using. If you have a multichannel amp, you can use its graphic equalizer for which you can custom configure REW's generic filters for best optimization. If you have Audyssey, you can manipulate its filters for custom filters. Check my other videos on relevant topics.
@@ocaudiophile Thanks for your quick feedback. But I am still wondering. I do have with my pre-amp for each LS and Subwoofer as well its own DSP individual for each channel (20 parameters for each channel). Should I let the first Sub optimized and use the aligned Summary for the second sub? Sorry for asking again.
You need to first and foremost align two subs with each other. That's finding the required time delay between them for loudest bass performance. REW's alignment tool and the method in the video is good for that. The "Aligned Sum" generated is the summation response (combined response of the two subs) and that's what you need to again time align to other speakers treating it as your single and only subwoofer. It's also this summation for which you need to create EQ filters.
@@ocaudiophile Thanks again, since you have used the center as reference if I am not mistaken, you would align the time delay of all speakers still align to the center -> your "0". In addition what I understand is, after time alignment by the "aligned curve" you keep the sub1 eq as you measured to avoid the room influence and put the new EQ with the aligned curve to sub2? Am I right, or do I still miss any logic behind?
The acoustic reference speaker used during measurements (and it should be the same speaker for all measurements) will always show its impulse peak at time=0. You need to align all speakers and the subwoofer combination to that. If you are talking about the MultEQ Editor app ady file, there's only one filter for SW1 in there. That's where you set your combined subs filters. SW2 only has relative trim (volume) and delay (distance) setting.
Do you measure the subwoofers with the main speakers on or off? If you have them on, do you turn the crossover way up so that they are not producing bass in the subwoofers range?
Main speakers off!
Could you incorporate the method at about 31min mark with your Supreme Audyssey Calibration technique in order to create an "eq'd" curve with 1 sub and no minidsp? Or am I overthinking and the Supreme video method takes care of that already? Basically create the curve in EQ like you do here then use the predicted freq and create the inverse then import that to Aud using Json.
It's indeed possible but you'll need to switch LFE setting to LFE + Main
@ocaudiophile Is there anything special that can be done with REW to align the sub with the mains? Everyone talks about Dirac Live with Bass Control. Presumably something similar could be done with REW?
I haven't listened to the latest version but from what I read they have added some sort of excess phase removal technique in the bass freq. area which will improve the sound if done correctly. Using all pass filters to change phase is a centruy old procedure and REW has all pass filters (rePhase is even better at phase manipulation). Nothing as special as they want you to believe (and they have obviuosly added some AI gibberish on top) and I am sure the difference in phase is very little to avoid ringing and pre-echo effetcs. IMO, no automation system can surpass a custom calibration simply because they need to compromise here and there to generalize the method for every room even if they get the technique right.
Is it possible to align the phase of the subs and mains via REW if my mains are not run through a MiniDSP (they are through an AVR). The reason I ask is because Ill use this to align my subs and then can align phase at the crossover point between the main and the subs
Otherwise, I’m looking at a miniDSP flex for my mains
@@joek6207It's possible to use "All Pass" filters in REW to adjust the phase response of your subs and bring it inline with your front speakers phase at the crossover frequency. REW will create the required biquad filters for MiniDSP. It's questinable if it's worth the effort though. You will only have some SPL gain at the XO frequency at best. Phase shifts below 100Hz are not detectable to human.
Admittedly, this may be a stupid question but I presume the viewers of your videos should be performing the calibration of their systems based on the chronological order of your videos? I ask because it is a bit unclear to me if there have been any revisions to conducting the ‘full’ process in respect to successive videos relative to the first video? In other words, is the document you provided in your first video all inclusive including any revisions to your methodology (i.e. continuously updated) you may have discovered during the time-frame beginning with your first video and ending with your most recent video?
Thanks again…….very impressive work
The pdf guide covers the "inversion technique" on the frequency response as well as some measurement basics and is still very valid. I do make small changes to it from time to time if I figure out any further optimizations on the method (very few really) so it's more accurate than the video as it's not possible to "edit" RUclips videos once they're launched.
That doesn't mean that correction cannot be surpassed. Inversion doesn't do much on the phase correction front for example. But all these filters can easily be combined with several others for even better sound. In fact, I have been working on such combinations for some time now and It's almost time for another full scale digital correction video launch ;)
And it's definetely not a stupid question. This is work in continous progress and quite often I find that certain suggestions in an old video are not that optimal. But there's not much I can do other than sharing the new information in another video. I also regularly use the description sections of the videos to inform of any procedural change.
@@ocaudiophile Thanks so much. I get it. May I suggest you create a step by step guide incorporating all of the techniques you discovered (and to be continuously updated)? I have never joined any sort of patreon/paid RUclips type group but in your case I would make an exception (my ask is very labor and time intensive therefore I am of the view you should compensated in some form)!
@@ts6640 A step by step guide with every optimal correction technique I have come up with so far is on the way as we speak. I don't think it will need any major further improvement anytime soon because I guess I am literally doing all that can be done digitally to a speaker response without compromising sound quality. The results are best I have ever heard in my system so far and I have tried almost all commercial solutions as well over the years. You're absolutely right that the video launch stage takes a lot of boring effort especially with my English verbal skills but I don't do this for income, I actually have a day job in a very different area :) It's the joy of sharing valuable information for audiophiles that motivates me and also lately I have more free time than usual.
@@ocaudiophile well…..I can’t thank you enough. Anxiously anticipating the forthcoming guide!
When aligning each sub, if one has a positive delay and the other has a negative delay (to get flattest response), how do you account for the negative since Marantz will not allow you to enter a negative distance?
Set one sub's delay to 0 and increase the other so as to keep the original difference between them.
My UMIK and minidsp arrived today. If during first tests inverting the polarity shows improvement, do you keep that checked and then run the phase slops are cursers or what determines if you keep the reverse polarity?
You can do that or just re-measure after actually changing poles
@@ocaudiophile Okay. Hopefully my final question before I jump in tonight. Once I follow all these steps and save the coefficient filters. Where do I import them for my mini-DSP? I would assume the output for each sub?
@@johnwicker7783 depends on the plug in and the minidsp model you'll be using. There's usualy one set of common filters for all outputs and another set for each output.
Hallo,
super Video, danke 👍
Vor dem Ganzen kommt ja noch die Positionierung der Subwoofer.
Ich habe einen offenen wohn-/Essbereich.
Das Wohnzimmerkino hat rechts die großen Doppeltüren zur Terrasse und nach links ist es komplett offen Richtung Küche.
Würdest du sagen, dass bei zwei Subs die Aufstellung einer vorne und einer hinten (beide nicht in der Ecke, aber genau gegenüber) am besten ist?
Oder beide vorne, einer links und der andere rechts (dieser wäre aber dann in der Ecke).
Danke dir
Viele Grüße und mach weiter so
Es gibt keine Formel, die ich für die Doppel -Subwoofer -Platzierung in einem seltsam geformten Raum kenne. Ich würde den ersten Subwoofer in die Ecke platzieren, um einen stärkeren Bass aufrechtzuerhalten und dann die Platzierung für den zweiten Subwoofer mit der "Crawling" -Methode zu bestimmen.
Mit dem ersten Subwoofer in der Ecke und dem zweiten Subwoofer an Ihrer LP (Ohrhöhe) müssen Sie durch den Raumboden kriechen, um den Ort zu finden, an dem der gesamte Bass für Sie am besten klingt. Sie können einen basslastigen Song spielen, den Sie gut kennen oder REWs "Noise/Pink Random/Sub Cal" spielen.
Do i need to reset the sub distances and or sub volume offset on my avr (sr6015) to zero prior running rew measurements?
All will be reset when the perfect speaker ady file is transferred to the receiver
How do you input ‘negative’ delays into MiniDSP(-5.86 ms)? I believe you could merely add 5.86ms to sub one but I have 4 subwoofers therefore doing so would not work for all four?
I guess you cannot add negative delays in Minidsp or any other DSP but the positive delays you add will cause Minidsp to send the signal to the opposite port earlier. It depends on your configuration and plug in whether or not it will effect all subs.
If you’re doing this for 4 subs what’s the order to adjust peaks for the phase alignment. Would you do all 4 at the same time? Pick 1 sub to do all the other 3 against that one? Or start with 2, then add one, then add another?
You can only align two at a time and one or both can be combinations you checked before so order doesn't matter. You can do 1-2, 3-4, 12-34. You'll probably find the fattest response. You can also use MSO
@@ocaudiophile Thank you. I currently have my subs tuned with MSO, but wanted to try this method just to see. I’ll tune 2 subs. Then combine their response to tune the third. Then combine again to tune in the 4th. Resulting in 1 virtual sub.
@Austin Smith how did your measurements turn out? I'm in the same situation. Tried MSO want to try this.
I have a question and read all the comments but it didn't clarify. When u cross align each individual sub first, i assume u get a delay value from each out of it? U have to put it in the dsp for each output first? Then u time align both sub, u also get also delay, then u calculate that at the end on 1 sub ouput in the dsp?
There's a newer video on that subject:
ruclips.net/video/DofUiHz7Fu4/видео.html
And yes you have to dial in the required relative delay (distance) between the subs in wherever you are combining them (ie MiniDSP) and then you can treat this as one sub and adjust its delay (distance) with the rest of the speakers.
Hey .. are your subs level matched to 75db prior to taking these sweeps .. EG volume level on receiver such that -12 sub calibration noise reflects ~ 75db on spl meter.
There's also "Gain" adjustment in the tool which is for relative dB levels of the two subs.
What version has "cross correlation tool" in it? I just upgraded to the latest, followed your instructions and when I reached that part that button isn't in the control section...
The website is under planned maintenance this weekend so I couldn't check. SHould be there on Monday.
So from the video, you have a miniDSP and mentioned you can apply this using your AVR. How would all the corrections from REW make it into a file that can be accepted and utilized by the AVR (in my case, Denon X8500HA)?
Pls see "Supreme Audyssey calibration" video
@@ocaudiophile Related, so I'll put it here - In attempting to combine these two videos in my head and method - at the EQ step in this video instead of selecting MiniDSP, I left it at generic/generic to input to Audyssey - then I believe I did "Export filter settings as formatted text" which gave me 5 filter points with Number/Enabled/Control Type/Frequency/Gain/... fields - then I fired up Ratbuddyssey and copied the five Frequency/Gain pairs into the "Target Curve Points" section for SW1 (as SW2 doesn't have any data or even a speaker type selected, not sure what's up with that?). Am I doing this right?
When I move on to the remainder of the cal with Supreme Audyssey calibration video, I think skip any EQing on the sub and only do the time alignment I'm guessing?
Thanks! And thank you for the super awesome videos!!
@@Hyacin75 There's only one response data and one EQ for all subs in Audyssey and is kept with SW1. That should be the EQ for the "sum" of all subs together. The relative time delays and volume levels should be entered under delayAdjustment and trimAdjustment properties of the other subs the ady file. The process is quite complicated but I've fully solved the logic. I will make a tutorial video on that when I finish the automation script.
I have an SR6015. 2 Subs. No Mini DSP. Do I need to reset the Subs distances to zero and then apply the Delay to one of the subs (ex Sub1 0M, & Sub2 + 2meters), or do I leave my sub distances as calculated by Audyssey and just add Delay to one sub, will leaving the other sub as calculated by Audyssey?
The latter is the right thing to do but Audyssey applied EQ to the sum of both subs will no longer be correct. Audyssey is quite problematic with two subs, applies strange subwoofer boosts (which it doesn't do with one sub) to different channels depending on volume levels.
You should measure your system with Audyssey settings and then decide what to do.
Please tell me what is the procedure and how to do initial level matching subs?
You can use "Check level" button in REW's "Measure" window while the mic is at the listenning position. It sends a pink noise and returns the volume level in dB of whichever speaker (or sub) you're about to measure.
I measured my left and right subwoofer and tried the align phase slopes at cursor trick, and it didn't work for me. I didn't measure 3 different positions, just the MLP for left and right sub. Am I missing something? It actually doesn't move at all for me. I'd love to know what I'm doing wrong. Is yours moving because you have more data and you've cross corr aligned and vector averaged? When I click phase slopes at cursor it looks as though it trying to phase align, because the different was only 0.06 delay. I really want to know what I'm doing wrong, I hope you can help.
The newer versions of REW have added functions in the Alignment tool (ie it can now be impulse or phase oriented). I've no way of simulating the old REW version used in this video but it should apply a delay and change the "aligned sum" response at the cursor if you select two different measurements, switch to "phase alignment" and click either "Align phase slopes at cursor" or "Align phase at cursor"!
I might send you an unlisted video showing it. I wonder if the length has anything to do with it? Because I have 2 subwoofer saved mdat files, one can use it and one cannot. The one that cannot was using 256k length.@@ocaudiophile
@hdmoviesource maybe send me mdat file? You've my email.
Unless for some reason their phases are already aligned at that cursor frequency, I'd always expect a movement regardless of the measurement length.@@hdmoviesource
I've sent you an email.@@ocaudiophile
Why cross corrilate align when you have all 4 measurements? Can you just average the measurements? Or would that be different? I'm considering measuring my subs with way. I normally only measure in one position. I have 3 seats, are more than 3 measurements helpful? I definitely want to try it this way one day.
That's how Audyssey averages measurements taken at different locations but it's not the most accurate way. Cross corr align removes distance/volume/delay differences between measurements and their vector average (not dB average like Audyssey does) reflects the true frequency response of a speaker in a larger area. If you vector average left and right speakers without time aligning them first, you will see a sharp roll off which will not reflect neither of the speakers' response. Cross corr align is similar to that time alignment.
@@ocaudiophile okay, thank you. Will take a look into this.
When ou use cross correlation you have 1o as reference... should you have it selected in the group to corelate with it or not? I would have thought it should not be selected (so 1f,1r,1L no 1o).
All responses to be cross correlated need to be ticked in ALLSPL Window and among them, the one that you want to align all others for should be selected on the left measurements panel.
@@ocaudiophile THank you kind sir for taking time to reply :). I'm working my way through this now. I used REW years ago when I used manual dials for filters and decided to get myself a minidsp now.
You say MSO does not produce results that are as good as your technique (and I am sure you are right) but your comment does spark my curiosity as to why? Is it because MSO relies too heavily on EQ whereas your technique solves many off the problems using phase alignment (therefore reducing any reliance on EQ)?
Secondly, I take it you do not subscribe to the theory one should not use ‘Q’s in excess of 6.117 (one popular school of thought is ‘Q’s in excess of 6.117 may produce excessive “ringing”)?
@@ts6640 I have only two subs and I use them only in the home theatre system. My stereo system has no subs. MSO is useful (to me at least) when you either have more than two subs or need to synch your subs with the mains (L & R) in a stereo system. With just 2 responses to optimize, I can do better (or at least equal) manually and much quicker. When I used MSO, it got stuck for hours and eventually it always suggested absurd volume alignments like kill one sub with -20dB, etc. Maybe, I used wrong settings. It's very possible since I didn't spend too much time on it. I fully support any attempt to freely share a tool for audiophiles so I am defo not against MSO. A lot of people love it and get good results.
With a phase equalizer, any Q above 1 combined with any phase shift above 45 degrees will cause ringing below 500Hz. Q values 5 and above will cause degradation in sound with IIR filters but no ringing. I don't use EQ filters for digital correction for sometime now. Check my REW Top Tricks: Inversion video.
Hi another question, unfortunately even with your guide, I do not get the predicted curve from REW if I use my both subs the same EQ. Anything you can advise to improve or to obtain the right corrected curve? I have to say that my living room is open, means that to my right site, I can go the dining area... Thanks in advance
How many different mic positions did you use while measuring your subs? If you measure at one point after, it will be q bit different.
@@ocaudiophile Actually I did both. And at both measurements the predicted curve looked quite different to the real curve after adjustment.
Just took measurements of 4 subwoofers, but didn't choose ''use acoustic timing reference'' during measuring. I guess that will be a problem?
Yes you'll have no timing difference information between the subs.
Nice tutorial, thanks! Now I'm bit confused if this tutorial help audyssey or is used to manual correct the subwoofers curve. Can I use this method ?:
1) Align Subwoofers and EQ them with 0.7 slope like in this video, then go with Audyssey measurements and leave it to calculate total time align
2) Calibrate the mic of Audyssey with rew and Umik (modding of the audyssey file imported from receiver)
3) send new file with the "target mic trick" to the receiver
4) Correct in Audyssey the wrong distance measured with 0,875 factor
It's correct? Also: DEQ on or off? I hate the loud surround speakers sound at low volume, how we correct this one? Thanks for your work
How do you EQ your subs?
@@ocaudiophile minidsp 2x4hd
@@gtalexandro Gr8! You can perfectly time align your subs with each other in Minidsp and EQ them. I am sure you do already but use only one of the sub outputs of your AVR to Minidsp. Steps 1-2-3 are all fine. I don't think Audyssey distance problem is due to speed of sound being taken as 300 m/s because when I measure Audyssey calibrated speaker timings with REW, I find them off both back and forth and some quite accurate, too. There's a weirdness in the steps of the distance inputs in the AVR. The distance doesn't change at all for some inputs and then suddenly jump. I can only adjust them perfectly by trial and error continuously re-measuring the timing with REW. So I would rather measure each speaker with REW with Audyssey on and amend speaker distance settings accordingly. Btw, Audyssey almost never gets sub channel timing relative to other speakers right so you should pay specific attention to that. Lastly, I never use DEQ, I even made a video on that, its effect on the curve doesn't make any sense.
@@ocaudiophile thank you!!! I will double check everything with rew after audyssey maked the corrections. Someone say DEQ it's important to have great bass response at low volume, I don't know if our correction with your subwoofer target curve will be overwrited with the default audyssey curve. I will measure before and after, and take gain adjustament if needed. Thank you again!!
@@ocaudiophile ps. how we can measure atmos speaker with rew? If I select the atmos channel the sound input go to the surround channel instead
Hey there... cool vid btw.. dumb question time :)... re cca ... in terms of the effect on the measurements .. now if i cca , and yes the peaks etc align or dont cca my vector average trace result looks exactly the same ...So what exactly is happening here .. 🤔 tad confused
The continuously changing black line you see in the alignment tool is actually the "vector average" of measurement 1 and measurement 2 and it is very different to their VA before the alignment.
@@ocaudiophile Thank you for a quick response.. I guess ill rewatch video in first instance.. Quickly though if there were say 3 Subwoofers instead of 2 would you follow process and get sub 1 and 2 sorted out , then repeat everything again but treat Sub 1 as the virtual sub derived previously then include sub 2 as the 3rd subwoofer to be integrated.
@@marksharp6647 Yes, exactly!
Hi just on this comment re cca again - sry :) .. Now i did 6 measurements for each sub .. If i CCA (yes overlay graph appears to align the peaks) , or dont CCA and VA both scenarios . .When in Aligment the delay it works out is exactly the same .. and the post Align result is identical in both scenarios , so im struggling to understand the direct effect it has feeding into the alignment tool result.. :) ..
@@marksharp6647 You don't CCA different subs with each other only different measurements of the same sub and then a vector average of the aligned same subs will give you an accurate measurement of that sub. This is the one you need to use for aligning this sub with the other one in alignment tool.
Hi . I am following your videos and it's really helpful. But one thing is I am unable to connect my configurations for Atmos speakers. I have downloaded the sweep but I can't Play that particular Atmos speaker . It keep it playing at the right speaker every time. Kindly help me how to configure this on my laptop. Thank you.
There can be a variety of reasons. Make sure your receiver is the default player in Windows sound settings and configured with correct sampling rate (48kHz), receiver is in movie/stereo sound mode, windows media player is the default player and has default settings. Try shutting down REW and playing the atmos sweeps with WMP and check all settings until you hear the atmos speakers.
Btw, you cannot play them with a Macbook.
@@ocaudiophile hi.. Noted. I will follow your advice and check back. Thank you for fast reply. Will update. Tq
@@ocaudiophile I am using Dell Laptop.
@@ocaudiophile hi . I followed your advice . and it's working 100% Now. I am so happy now. Thank you for the wonderful videos.
hi OCA. I am not getting cross cor align option in my REW. am i missing something. have REW 5.20.13 installed.
www.avnirvana.com/threads/v5-20-14-early-access-build.11154/
My new subwoofers come with 7 PEQ + delay in its appp. How do I configure REW to output settings for this?
Select "Generic" & "Configurable PEQ" from REW EQ window and key in the Q and max/min dB limits of your sub PEQ. Once correctly configured REW Auto-EQ will generate the best filter settings within these parameters for the target curve.
The cross correlation button (in controls) is not there. Is this the version of REW I have? Is it somewhere else?
Yes you need an early access version of REW
@@ocaudiophile so everyone in the comments that has used this procedure (cross align) has an early access version of REW? I have to wait for a new version to release in order for me to use cross correlation?
@@doublet147 google "v5-20-14-early-access-build"
Hi once i input the sum aligned eq file into the miniDSP and i input the delay. How do i verify that it is actually aligned? When i run a sweep i dont get the same Avg response because i took the sweeps originally in 5 different spots . 360 circle around MLP, including MLP.
İf you cross correlation aligned all measurements to MLP, a single measurement at MLP should confirm your aligned results.
Is there anyway you can review my final EQ graph this is my first time and I am trying to figure out if I was able to do it all correctly. If you have an email, I would really appreciate it. I will also pay for your evaluation if you want please.
@kringle924 Share your mdat file and I'll have a look over the weekend
@@ocaudiophilei believe the “comments with links” setting on your account is disabled, as youtube wont allow me to post it.
All that can be enabled is enabled in this channel but YT will only allow you to post google drive and imgur links!@@kringle924
This and your other videos are missing some visuals of what it looks like, where the subs are placed, how you're actually doing the measurements, etc.
I mean, there's a lot of good info in your videos in general, but it all seems way too complicated for me to do myself. Part of this is because I've never used REW, I absolutely don't wanna install Java, and the ASIO stuff has never worked well even in WinXP.
I really would like to see a full "Here's what you need to do to set this up with physical speakers are a mic. Here's what important in running this program." etc. I don't wanna have to keep your video open to understand it; I want you to explain it in a way where I can do it myself.
Am I making sense or is this too vague?
This early REW video also comes with a pdf guide with full details on what you are asking for:
ruclips.net/video/5YcH7j2-L1Y/видео.html
What are your thoughts on MSO?
I used it more than once. The user interface is not the best and it took too long to calculate for not so optimal results. It was probably my settings. But I think beyond 2 subs (3,4) or if you intend to LFE+Main your fronts and center it's a life saver and best of all it's free.
@@ocaudiophile thank you sir.
If it would make sense. A guide from you would be much appreciated! Check out Jeff Meiry on RUclips, he is the only one on RUclips with proper MSO content and explanations.
@@BuffSquadBigBenni Thanks. I posted a new video yesterday.
Hey OCA, trust you wouldn’t mind a question on this? Assuming, the closest speaker is at 1.45 m, the max distance one can enter to the sub distance is + 6 m, so 7.45m. In this setting I have appr. 13cm missing according REW (should be 7.58 m). Have you got a workaround on hand so that this can be solved electronically (maybe it is not necessary as 13cm difference is inaudible?) or do I need to move the subs 13 cm closer to the LP physically? Thank you Sir.
I would leave the setting 13cm too short. It wouldn't make an audible difference with bass wavelengths.
Thank you Sir.
@@ocaudiophile, by the way, it is the first tiny peak which is ~ 13cm apart.
That's traumatizing!
@@ocaudiophileWhat do you mean by that? Thanks Sir.
What's "Cross Correlation" mean? I'm assuming it means having the subs work well when both are playing at the same time.
It's a time alignment technique.
Also, how would I do 3 subs? Just do one pair and then the 3rd against they aligned and EQ’d pair?
Yes, you can apply the same Alignment tool to 12 & 3 and so on but you have to give aligned combo a volume offset of 6dB before aligning them to the third one. You can do that in REW: Controls / Measurement Actions / SPL Offset / -6
@@ocaudiophile if I use aligned sum won't that add the 6dB? ALSO, a HUGE THANK YOU! I love your videos
Also, do I only apply an EQ to 1&2 or do I wait until all are aligned and then global EQ 1,2,3.
@@joek6207 EVery doubling of the volume is a 6dB increment. So when you double the number of subs, you increase their total volume by 6 dB, when you increase the number of subs from 2 to 3 it should add 4.5 dB to this total and gets complicated. Just measure each sub separately at the beginning at the same amp volume and then use the aligned sum of 12 with 3 as calcualted by REW. The new aligned sum will be accourate.
@@joek6207 Global EQ to the final aligned sum only.
Another question, what do you think of the approach for aligning subs with mains as per the following link: ruclips.net/video/9HqmkziKU48/видео.html
I have received very good results but was curious if you are of the mind a better approach exists? Admittedly, this approach is very easy to use and therefore I wonder if any other approach is even worth it (i.e. law of dismissing returns)?
Practical indeed. He's in fact aligning for the correct phase at the cross over dip frequency. It's easy and would work. If you want my obssessive compulsive comments though: The SPL volume is never stable with subwoofers, keeps moving back and forth 2-3 dB continously, there could be a louder response at some distant volume you would be omitting, highest SPL response at XO frequency is not necessarily the best setting for a sub (it's among the best though) and lastly once you set up REW to check for SPL and send a fixed frequency, play with distance, etc. you have already done all preparations for measuring the actual response so why not just do the whole sweep and then REW has many tools to find the perfect alignment.
I don’t understand how you send the time align correction or eq correction to a system?
Depends on the system
What happens if you EQ them first?
When you EQ each sub first and then combine, the result would still require more EQ as the summation of their phases will interact differently and you want to keep total EQ applied to a minimum. If you cannot get satisfactory results with this method no matter what, then you could try that but very often, it's not a necessary step.
I dont see cross correlation align in the control panel. Is that a pro feature?
Google REW early access version...
I have it now. Thanks!
Is the correlation align only on the newest version? I don't see it on 5.20.13
www.avnirvana.com/threads/v5-20-14-early-access-build.11154/
Thank you. Finally ran through all the steps. It was interesting after time alignment any align phase slopes at cursor made things worse but reverse polarity on one made it fantastic and removed any dips
@@johnwicker7783 Cool :)
@@ocaudiophile I had to start over as that was not actually the case. I've followed your video multiple time and I get much better readings when running my subs without any time alinement or EQ. As soon as I connect the mini DSP and run a measurement I get way more inconsistencies with major peaks and dips.. I wish I could contact you directly to see if maybe I am missing something.
@@johnwicker7783Did you reset Minidsp settings? At default settings, it's pretty harmless to sub response, only adds delay.
I do not understand why you are generating just 1 common EQ for both subs. To me it seems more optimal to make a dedicated EQ for each sub so that you're not cutting the wrong frequencies. Can you please explain why you propose to do it like this?
2nd question @16:45 you perform a vector average of all left subs measurements, and in the vector average there are suddenly two sharp dips that were not in the individual measurements, you also previously mentioned you were not very interested in them because they were very local. Can you please elaborate on this a bit more?
The individual EQ filters will behave very differently when you change the relative distance between the subs to optimize their combined response and you will need to apply more EQ filters at the same bands maybe even boost a frequency you have just dimmed which will be cutting the wrong frequencies. It's generally better for the overall sound to apply as little equalization as possible.
These sharp dips are due to phase cancellations which occur only at a very small area in space and even a little movement of your head will usually change their shape. Also human ear is far more sensitive to detect peaks compared to dips unless they exist through very large regions.
It's usually a lot easier to simply try a digital correction and test it with your ears rather than overthinking about its merits.
@@ocaudiophile I meant that you can apply individual filters after having optimized the placement. It could be that they individually might require less EQ than combined. So the steps would be
(1) take the combined, aligned subwoofer response
(2) make the EQ filter (steps 1 and 2 are the same as in your video)
(3) plot the expected outcome
(4) subtract the new expected graph from the time aligned graph so you get the "delta".
(5) make EQ filters based on the differences between the "delta" and the individual subs
Or is my proposed way an example of overthinking?
This requires additional steps but it also makes sure that you're cutting the frequencies at the right sub so it could be that less EQ is applied this way.
@@l.s.1709 There's no right or wrong sub, those peaks and dips are in fact room modes and related to room dimensions. Both subs are usually identical in their sound production. When you're phase matching them for the flattest response, you're creating a new sub somewhere in the room which has different peaks and dips to either sub. Btw, I am not saying it's wrong to individually calibrate them first , it's just not a step which will audibly improve the result.
@@ocaudiophile ok understood. One last question: how to combine this with a VSA (like a recent video from yourself)? First phase/time alignment, then VBA, then EQ? Or VBA last?
@@l.s.1709 Apply VBA before anything else. I suggest you follow the method in this video for optimal VBA:
ruclips.net/video/q9aJqQpNyLY/видео.html
Merhaba! Ben halen Amerika’da yerleşiğim ve yaklaşık bir yıl önce 7.1.4 düzeninde bir home theater oluşturdum (tek sub sinyali fakat 4 büyük subwoofer ile). Sub alignment konusundaki bilgi ve tecrübeniz gerçekten etkileyici. Sistemimi akustik olarak olabilecek en iyi noktaya getirmeyi arzu ediyorum. Remote calibration servisi veriyor musunuz? Veriyorsanız kesinlikle ilgileniyorum (Teamviewer üzerinden). Sistemimdeki sub’ları şu anda MiniDsp 2x4 HD ile yönetiyorum. Umik-1 mikrofonum var. A/V processor’üm NAD M17 v2 (Dirac). Selamlar.
MSO diye free bir yazılım var, 4 sub optimizasyonu için onu indirip kullanmanızi öneririm
@@ocaudiophile Evet. Biliyorum. İndirdim bir süre evvel. Fena sonuç vermedi. Bir miktar daha üzerinde çalışmam lazım gibi.
4 sub'i birbirlerine gore ayarlamak cok kolay degil ama mumkun. REW ile olcup, mdat dosyasini buradan paylasin, bir bakayim. YT bir tek google drive link'lerini silmiyor buradan.
@@ocaudiophile Ilginiz icin cok tesekkurler. Uc farkli seating position icin her dort sub'in ayri ayri olcumlerinin linkini gonderiyorum Google Drive uzerinden. Position 1 Main Listening Positiondir. Position 2 onun solunda oturan kisi icin, position 3 de saginda oturan kisi icin. Sub'larin konumlari ideal degil. Sub 2 ve 3 projeksyon ekraninin arkasinda, yan duvarlar arasindaki mesafenin 25% ve 75% konumlarinda. Sub 1 ve 4 tek sira uzunca koltugun 2 metre onunde sol ve sag tarafta. Yaklasik olarak ekran ile arka duvar arasind ortada diyebilirim. Ancak oda dikdortgen degil ve kapatilamayan acik alanlari var (tam izole degil). MSO ile yaptigim denemelerde tam tatminkar sonuc elde edemedim. Belki gain ve delay setting'lerine belli bir ust limit koymam lazim. Bilemiyorum. Amacim sadece iyi bir frequency response'tan ote iyi bir impulse response alabilmek. Yani transient'larin temiz ve "impactful" olmasini saglayabilmek. Verebileceginiz tavsiyeler icin simdiden tesekkurler. (Butun sub olcumleri Dirac apply edilmeden, temiz olcumlerdir)
drive.google.com/drive/folders/1J31EGJCWeKTxwnqFsCUtr_3AozQibJ6v?usp=sharing
MiniDSP'de 4-way advanced plugin (48kHz) kullandiginizi dusunuyorum dolayisiyla 6 tane ortak EQ band ve 4 kanalin herbirine de 6'sar tane daha biquad filtre girilebiliyor olmali.
Filtreler bu sekilde import edebileceginiz REW/txt formatta, girilmesi gereken gain ve delay ayarlari da ayrica bir txt dosyasi icinde:
drive.google.com/file/d/1FsXidsC7Cpvt2Vhd2NVjbtQVSMI8MpQH/view?usp=sharing
Ben daha once hic 4 sub kullanmadim, test etmeden uzaktan duzeltmekte cok kolay degil ancak bu ayarlarla 13Hz-120Hz arasi fast & smooth bir bass olacagini umuyorum.
What to choose when "cross corr align" is removed in the menu in newest REW version?
time align but check in the overlays graph that impulse peaks gather on time 0. Or just use single measurement point for now, the soon to be released REW will have CC. I also added a link for the version I use in the comments under the video.
Hi OCA, just FYI a great video on this topic here: ruclips.net/video/BfhkssMB4SM/видео.html
REW has now an alignment feature on impulse response which is great! ( tested on my subs yesterday, got finally that good punch kick on good 'old funk tracks :p, and with VBA it's magic) Cheers
When making the measurements, are we measuring at full range or just 0-200hz?
Are we also measuring with a - 10db offset?
If you measure 0-200 then you need to apply -10dB offset to LFE because the receiver will boost that signal thinking its LFE signal. If you measure 0-24000, you don't need to do that. It will not be boosted.
@@ocaudiophile thank you so much for the extremely helpful information! You are friggn badass!
@@christiandickey In the latest early access version of REW, you can apply -10dB offset to LFE during measurement so you don't have to do it in the receiver settings.
@@ocaudiophile very good information! I have watched and re-watched your videos dozens of times and have found them to be the absolute best period! You have a god given talent and I'm grateful that you are sharing 🙏👍💯
Thank you for your kind words. Sharing is caring ;)
Hi OCA,
Sent a comment yesterday, but couldn’t find it your site, must have gotten lost in cyberspace. So here I will try again?
I have a late year 2020 Denon AVR X6700H (7.2.4) which is working great, and I am currently using Audyessy along with the Audyssey EQ Phone Editor App. I do have a UMIK 1 but do not have a MiniDSP 2x4.
Because I’m new to all of this and haven’t had a lot of time yet to get acclimatized, my questions below may be very obtuse for you experienced users, but here they are:
1) To avoid the cost of purchasing a MiniDSP, when using a UMIK 1 and REW to take measurements of any of my speakers, once the resultant filters are created in REW, can these results simply get exported back into an .ady file? And would that .ady file be the “perfect speaker” .ady file your show in one of your other videos?
2) I would like to use your “Multiple Subwoofer Alignment with REW” (the “Video”) to better align/marry my two SVS PB3000 subs with my mains. My mains are Polk Rti A9’s (30 hz +/- 3db). If I use the UMIK 1 and REW to measure each of the subs as is shown in the Video, must I have a MiniDSP to incorporate these results back into my system? Or, as asked in my question 1 above, can I add them into the 2 detected sub channels (channels 11 & 12) that show up in the original perfect speaker .ady file? Adding them via export and Json Online?
3) In the Video it appears you ended up with 6 equalization points/filters; how and at what place do I add these particular 6 filters back into the .ady file?
Sorry for all the newbie questions.
Brad
ruclips.net/video/LwORN-tSPjk/видео.html
Thanks!
Cheers!
@@ocaudiophile thanks for all you do. This is the most detailed alignment video I've ever seen.