"Official" Audyssey thread (FAQ in post #1) - Page 1784

I think that nobody who has done the 'sub distance adjustment trick', including Mark Seaton who pioneered it and Craig John who brought it to us here, and all the others who have found it to be very useful, have suggested that your method is required. You may be right - I don't know. What I do know is that the sound and the graph I posted speaks for itself wrt to the improvement I have obtained.

I am not sure how your method could be achieved in practice. AUIU you are suggesting taking several mic readings around the MLP to mimic those done with Audyssey's mic. And then averaging those measurements. How would you then adjust the sub distance settings in the AVP? Each time you adjust the sub distance by 0.2 of a foot, you need to see the impact of that adjustment on the graph. How would that be possible using the method you describe?

Tediously!! But we are looking for perfectionism, aren't we.

Notheless, not only for the "trick" with distance ajdustments, but I remember Chris has always enphasized that their research has clearly shown that a single measurement in space would surely not show what the ears are hearing. It was also a suggestion of Chris that if anyone would want to verify what Audyssey is doing it should be done by multiple measurement and by averaging them. I know REW can do averaging of graphs, how about OM?

Much of the time seems to be spent calculating the filters in relation to the crossovers chosen by the user. Pro ranks the XO's from 'best' to 'less best' and you can accept the recommended ones or choose from the others in the list. Only when the XOs have been selected does the calculation of the filters begin and this seems to require a heck of a lot of processing power, which is why I think they cannot do this in an AVR with the consumer version. That is speculation of course but knowing the difference in processor power between a modern laptop and an AVR it seems reasonable.


Ah yes - that would be desirable and something I have often wished for myself. The big problem with the consumer versions of Audyssey is that if you get a good calibration you are always fearful of trying to make it even better because of the inability to save the existing one and return to it if your experiment is not successful. HST, there is a way to save one calibration at least - see here:

6. Is it possible to save and recall an Audyssey MultEQ calibration? (click here)


Yes - Pro allows you to save the calibration results as a file. You can do as many different calibrations as you wish and save them all. If you want to go back to a particular calibration you can run Pro s/w and reload the calibration and then upload it to the AVR/AVP. The process of doing that takes less than 15 minutes. You can change things in the calibration before re-uploading it to the AVR if you wish (and also save those changes if you wish) so it is very flexible. I can't personally see how that could be accomplished in an AVR with its limitations. This, for me, is one of the big benefits of Pro as it allows as much experimentation as you like, with the security of knowing you can always return to an exact, known good calibration.


It would indeed be great to have that facility in XT32 etc.

This mirrors my own results almost to perfection. On my latest tweaks I have been able to see a similar sort of boost, over a limited but important part of the spectrum and this has produced the listening experience you describe. I think it is the first time I have ever been able to boost the bass output without introducing undesirable side effects. And your point about this being a 'free ride' in terms of headroom is an excellent one - yet another reason for the astonishing efficacy of this simple, post-Audyssey adjustment.


Same here. I let favoured guests use the 'money seat' and the rest, as you say, are so overwhelmed by what they hear (often hearing really good movie sound for the first time) that they are blown away anyway. For me, and you too it seems, this is the nirvana scenario. For once my downside of only having two seats works to my advantage and makes life easier wrt to calibrations.


I know I have thanked you before for bringing this technique to our attention, Craig, but I am so overwhelmed by how effective it is that I feel the need to say 'thank you' to you every time we 'meet' here. Please forgive the effusiveness It is without doubt one of the most effective 'tweaks' I have ever made.

I admire your perseverance, Feri, but asking the same question 100 times won't produce a different answer

Set up your mic and REW, take a measurement at the MLP and save it. Then start by adjusting Sub 1 in either direction, until you see the graph improve. Keep adjusting until the graph starts to deteriorate and them move back to the distance where the best result was obtained. Repeat for Sub 2. When you have improved the graph even further by adjusting Sub 2 distance, go back to Sub 1 distance to double check that you can get no further improvement. If you can, then go back once more to Sub 2 adjustment and see if you can squeeze a little more improvement out. Go back and forwards like this as much as you need to until you are happy you can make no further improvement.

That's how I do it anyway, but I am the new kid on this particular block. I am sure others with more experience will chime in if I am misleading you at all.

That rule implies that Audyssey gets it 100% right and therefore it cannot be improved on. All the results from those who have tested it and graphed it show that Audyssey does not get it 100% right in this particular instance.

That is correct - each sub is done independently.


No. One is in front right, the other is about 1 third the way along the left wall of the room (more or less only place I can put it).


Yep - see my more detailed reply to you.

Yes, "it" being best SQ at MLP. Which is what some of us really care the most about.

BTW, I'm reasurred to see repeated and measured confirmation of the "tight mic grouping around MLP technique", which I've advocated for a long time as improving SQ at MLP.

I'm not trying to verify what Audyssey has done. I am trying to improve what it has done, and for that aim this technique works exceptionally well.

Your starting point, Feri, and forgive me for saying so, seems to be that Audyssey is 'perfect' and therefore its results can never be improved on.

My starting point, OTOH, is that Audyssey is very, very good but in some areas it fails, for whatever reason. This bass phase adjustment issue is one of those areas. I have no idea why Audyssey does not get the sub distances right for the best FR - I read all the various and learned speculations with great interest though. But what I do know, from my own tests and experiments and results is that Audyssey's results can sometimes be improved on by post-Audyssey adjustments. This has been confirmed here (and in the Pro thread) by many others who have far more knowledge and experience than I will ever have. It was my *ears* that first alerted me to a problem with my Pro calibration - you can read about it in the Pro thread if you are interested - look for my first post with a graph and start there. Briefly, my bass was lacking after my first Pro run. It didn't need measuring to discover this - it was immediately obvious. This led me on a journey of discovery which culminated in my adjusting the phase of my subs and the end result has been such a huge improvement (as Craig also mentions in his recent post here) that I can scarcely believe it.

Many people make post-Audyssey adjustments - the raising of XOs is the most common one. This, as we know, is an issue in the AVR not an issue with Audyssey as such, and this may be the case here, as Harrison has speculated. I don't really care where the issue arises - what I care about is that I can fix it and improve the SQ in my room.

It is interesting that one of the other very common post-Audyssey adjustments is to make the bass a little hot. The cry "where has my bass gone since I ran Audyssey" is heard too often to ignore. The usual response is "you are not used to flat bass - give it time". But suppose that instead it is this issue that we are discussing - that Audyssey has the splice wrong and that what is really needed for those who feel their bass is lacking is to tweak the sub distances? Since making this adjustment, this is the first time I have felt that I do not need to run my bass hot at all. It is now strong and full and tight etc with the trim settings Audyssey picked. All my other attempts to make the bass stronger (by raising the trims or fiddling with DEQ etc) have been unsatisfactory for one reason or another. This one small post-Audyssey tweak I made has solved that problem entirely - the best bass I have ever had, with no undesirable side-effects at all.

I disagree with Harrison on one point and that is when he says that we (the consumers) should raise the issue with Denon and Onkyo (if it is indeed their issue). My POV is that Audyssey should raise the issues with the AVR manufacturers. We are buying a product from Audyssey that should work with the AVR that the product is applied to and in my view it is for Audyssey to ensure that their product works properly with the AVR, not the end user. But fundamentally, Harrison is right in the sense that *someone* should raise this issue with Audyssey, encouraging them to discuss it with their AVR partners. It is such a pity that Chris had to leave this thread (although he may still lurk for all we know.... Hi Chris!! )

Yes indeed. Another deviation from Audyssey's 'rules' that works well for us. I have yet to try a run with all positions around the MLP, ignoring the other seat entirely, but I will. My current best run was with *almost8 that mic setup but not quite. I am hopeful that I can squeeze a little more performance out.

I was going to do it this weekend but it is a holiday weekend here in the UK and we have house guests, so it will have to be next week now. I will, of course, report back my findings.

Interesting point. We have indeed always been told that all settings in the AVR are ignored during autosetup. Do we actually know whether, when the subs are pinged together, the individual distance is applied? If not, this may shed further light on the distance twiddle technique, especially if those finding the tweak helpful are those with subs placed at quite different distances from MLP.

Yes it suppose it presumes measurements taken inches away at each ear would be exactly the same as that point. It further presumes you keep you head locked in space at that point for a two hour movie.

Even then you have to get past the selfish disregard aspects of the experience for others in the room not being the best it can be.

The context under the "I don't care about the other seats" chatter is that generally there are no "others" in the room. But of course, that would make a comment about selfishness completely out of line.

In my case, I am single, retired, and a hermit. There are no other listeners. I lead a lonely life that is made more enjoyable only by my selfish attention to optimizing sound quality at the MLP.

In Gary's case, I'm surprised anyone would want to be in the same room with him.

Guys, correct me if I'm wrong, but does XT32 have a higher filtering resolution than XT with the Pro kit? Any Pro users who switched to XT32? I know the kit includes a higher quality mic, but then apparently XT32 has better measuring resolution, but I need confirmation on that. Is there really that much of a difference?

goneten - answered on the Pro Kit thread.

I am not sure that is accurate, guys. Here is what we said in the recently-updated Setup Guide:

Sub EQ HT includes an automated subwoofer volume calibration process, which allows you to individually level-match subwoofers connected to the Sub1 and Sub2 outputs: This automated level adjustment process occurs prior to starting the calibration measurements.

After conclusion of the automated level adjustment process, the calibration measurements begin. During the measurement process, MultEQ XT32 with Sub EQ HT measures each subwoofer output (Sub1, Sub2) independently to determine the appropriate level and delay (distance) setting. It also measures the response of both subwoofers (Sub1 + Sub2), and then utilizes this combined response when creating the equalization filters.


Mark

I was curious if the other seats would be detrimentally affected, since I assume the change is only affecting the lower frequencies? A 100Hz wave at 68 degrees fahrenheight is lambda = c / f = 343 / 100 = 3.43 meters.

Is this line of thought correct or not?

Mark

All this good discussion about adjusting the subwoofer distances to increase output at the crossover point has inspired me to update the document describing the technique hosted at 4Shared.

I found a great post by Mark Seaton about group delay, and have also added that to the document. I also thought it was a good idea to add this section, and I am looking for input:

The Basic Theory Explained

Audyssey measures the subwoofer(s) and satellite speakers independently to determine their delay from the first microphone placement position, which is usually the Main Listening Position (MLP). Audyssey does not measure the combined response of the subwoofer(s) and main speakers together. Therefore, if the frequency response at the crossover region is not optimal, Audyssey is not aware of this.

The process described below is used to determine if the frequency response at the crossover region is optimal. The process basically consists of measuringat the MPLthe combined response of the subwoofer(s) and main speakers together. Any measured dip in the frequency response is usually caused by phase cancellations, and this is corrected by adjusting the subwoofer(s) distance and re-measuring until the response improves.

Altering the distances does not alter the subwoofer correction filters calculated by Audyssey. However, there is some thought that altering the distance settings of multiple subwoofers could alter the interaction between them, as the Audyssey filters were calculated based on measuring the combined response of the subwoofers at multiple measurement locations. However, the measured results speak for themselves, and if you choose to go down the rabbit hole, please read on.

Mark

You've seen others in my room??? Odd. 99% of the time I am in there I seem to be the only one. I must look harder. Thanks for the useful tip.

Originally Posted by SoundofMind
"...Do we actually know whether, when the subs are pinged together, the individual distance is applied?..."



Mark, I was just stating that I'm not absolutely sure individual distances are applied when pinging together. They may well be. But if they're not, it adds a dimension of error we had not previously anticipated.

Note that because of the Pro kit 2 sub "bug", the 2 sub distances are set in the processor as identical even when they're not.this post> We have heard that will be corrected by a FW update.

I thought he must have been peeping through my windows and seeing people who aren't really there. I wonder if he has imaginary friends?

ROFLMAO. You just made my day with that one.

Only when reloading a former calibration though.

I will be doing just that later today and will report on the sub distances. I *hope* I don't have to re-run my two 13-point calibrations!!!

Yeah, that is interesting, and would be nice to know; we need information (good or bad) to feed our Audyssey OCD!

Mark

If you do, pay attentiong to the number of sub pings!

Mark

edited, thnx

Starting with measurement position #1, it pings sub 1, sub 2 and then sub 1 + 2 simultaneously. Subsequent measurement positions, it is also sub 1 + 2 simultaneously.

Is there disagreement on this?

Jeff