Audyssey's Next Generation of Room Correction (MultEQ-X)

[Ruspamen]

They should match like that, yes, and we've established with our own REW measurements that this should produce the best results.

Thanks. Ultimate answer, for those who use manual adjustment without enabling audyssey, is it better to use multeq-x to find the distances and multiply them by 0.875 and insert them into the avr, or use a tape measure and then do the conversion?

 

[KMO]

never noticed any such differences as displayed on the screen vs physical tape measurements,

Well, no, you wouldn't. Are you misunderstanding the issue?

The problem is not what's displayed on the screen, or the way the microphone method computes distances, it's the way what's entered into the distance menu gets translated into actual delays.

If you have a speaker configured at 1.00m and another at 2.00m (accurate to tape measurements), the delay that gets added does not correspond to the delay you need for the 1 metre difference. You get a 3.3ms delay when you need a 2.9ms delay to have the surround arrive in sync.

Multiplying all distances by 0.875 corrects for this. Changing those speakers to 0.87m and 1.75m gives you the 2.9ms delay you need for the 1 metre difference.

 

[KMO]

Thanks. Ultimate answer, for those who use manual adjustment without enabling audyssey, is it better to use multeq-x to find the distances and multiply them by 0.875 and insert them into the avr, or use a tape measure and then do the conversion?

Now I'm a bit confused. If you're using MultEQ-X, why not get it to upload the numbers itself (doing the 0.875 for you)? What's the manual entry for?

If you just don't want to enable Audyssey room correction, then you can just turn that off separately. Having the MultEQ-X upload the filters along with distances, trims and crossovers doesn't mean you have to enable Audyssey correction mode, and you can edit all of the distances, trims and crossovers after upload.

Anyway, I believe MultEQ-X's acoustic measurement work quite well, from seeing other people's results, and I think I'd favour it if I had no other measurement solution, and I had any doubt about differences between speakers. In particular, it's the best bet you've got of getting an initial correct sub alignment. Tape measure alone won't work for your sub. I see no particular reason to doubt MultEQ-X's readouts.

I'd trust tape measure more for relative distances between speakers if they were all identical.

 

[KMO]

If the audyssey measurement is different from the actual tape, which should be taken as a reference to be multiplied by 0.875?

Impossible to say, really, and nothing to do with the 0.875.

If the Audyssey measurement is different from the tape, which to trust more?

I have seen people report occasional oddities with MultEQ-X numbers that make no sense, suggesting it can have the occasional glitch, but on the whole it seems to work. And it can figure out latencies for active speakers, which a tape measure can't.

I would say if tape and MultEQ-X are grossly different for similar speakers, it indicates a problem - aside from a constant offset. A constant offset (eg every speaker being 15cm further away in MultEQ-X) has no effect, and can be brushed aside as a latency quirk.

If they're quite similar, then it doesn't really matter. I wouldn't waste time trying to pick at it unless you have a tool like REW to actually see the real alignment results.

The overall 0.875 correction is worthwhile because it's correcting per-speaker differences of the order of up to a whole foot, which is significant in a home theatre scenario.

Nit picking about a couple of centimetres difference would be getting a bit silly, particularly as the AVR only has 3cm resolution itself.

 

[peng]

Well, no, you wouldn't. Are you misunderstanding the issue?

The problem is not what's displayed on the screen, or the way the microphone method computes distances, it's the way what's entered into the distance menu gets translated into actual delays.

If you have a speaker configured at 1.00m and another at 2.00m (accurate to tape measurements), the delay that gets added does not correspond to the delay you need for the 1 metre difference. You get a 3.3ms delay when you need a 2.9ms delay to have the surround arrive in sync.

Multiplying all distances by 0.875 corrects for this. Changing those speakers to 0.87m and 1.75m gives you the 2.9ms delay you need for the 1 metre difference.

Okay, I did misunderstand. I should have read your posts more carefully. Sorry and thank you for you patience.

 

[MarkI]

Thanks everybody,

This has gone a slightly different direction than I had hoped. Have any of you played around with virtually moving the speakers out/or in. To find a sweet spot. Or as D Live claims, delivers a larger sweet spot, accurate staging. I’ve heard this before, how can that make a difference. How can you know. How can you measure it. Which I find funny because when you buy speakers you listen to them. Hopefully you have a good reference material that you’re very familiar with. And you can sit back and listen to and in my case verify the changes you’ve made. Whether they’re better or worse. Over the last few months I’ve spent a lot of time with multEQ, gotten some worse results but slowly been dialing in the better ones.

the space I’m in has been blueprinted, meaning any and all measurements are almost perfect. My speaker placement is done with A laser and of course old fashion tape measures. Something I can confidently tell you I know what I’m doing with. So I have metric and regular imperial tape measures. When setting up the current system it all was carefully measured, the spacing and placement of everything. To a single point in between the sweet spot of two seats. Which is marked with blue tape. When I take my measurements with Audessy I’m using the exact same - exact first location. So I can set up the stand with mic and measure from there. This is a two person job and takes a bit of time. Ceilings are 12 feet tall. Do that a few times and you get actual real measurements within about a quarter of an inch. Maybe you’ve heard the term measure twice and cut once. John Dewey thinks what you guys are calling in anomaly which is about a 4 inch difference to the fronts is due to the size of my space. Again it’s a huge sunken living room that opens up into about 10,000 square feet. So the pressure doesn’t get built up the way it would in an enclosed space. But again who knows. It’s not worth trying to prove that the actual measurements are correct. Which they are.

And I am most certainly not trying to create a scandal about the Audessy mic and it’s measuring abilities.

If you look below the actual measurements are the actual measurements. When I go and open up multEQ-x, have it take all its measurements, you can see the measurements it comes up with below. and yes there about 10cm consistently. What it dumps into the Marantz is a fairly significant difference. Again that’s about the foot.

Actual - MultEQ-x - Marantz
F-L 3.68 - 3.85 - 3.37
F-R 3.68 - 3.85 - 3.37
C 3.47 - 3.57 - 3.12
S-L 3.10 - 3.24 - 2.83
S-R 2.21 - 2.40 - 2.10
B-L 2.46 - 2.63 - 2.30
B-R 2.46 - 2.61 - 2.28

Amos-speakers
F-L 3.34 - 3.45 - 3.02
F-R 3.34 - 3.44 - 3.01
R-L 3.22 - 3.30 - 2.89
R-R 3.22 - 3.32 - 2.91

This all started for me a few weeks back when the time thing issue was brought up. And I went about it a completely different way. Which led me to where we’re I’m at now. By virtually moving my speaker out which I believe I’m saying right is actually pulling them in. The sound-stage delivers a larger sweet spot and more accurate staging.

I didn’t stumble upon this by opening up the Marantz and moving all of the speakers 1 foot closer/or playing around in the metric side with 10cm increments. Which by the way because of this thread / form I have now I have my Marantz left in the metric setting. So again thank you. Instead I loaded my actual numbers into multEQ only to find that what landed in the Marantz was different. You guys obviously understand this computation or mathematical formula that happens. I didn’t. So I change the numbers manually, reloaded them until I came up with a number of 4.21 which then landed 3.68 in the Marantz. This is my actual measurement. Not sure what’s happening in between however again it changed my sound stage. For the better. If you’re in California in the Danville area you’re more than happy to come over, sit back and tell me what you think. By the way I have probably two dozen different MultEQ-X profiles saved that we can compare it to.

Marantz Correction added to get Actual distances
FL-3.37 +84cm = 4.21 - 24.92% - 3.68
FR-3.37 +84cm = 4.21 - 24.92% - 3.68
C—3.12 +85cm = 3.97 - 27.24%- 3.47
S-L-2.83 +52cm = 3.55 - 25.44 - 3.10
S-R-2.10 +42cm = 2.52 - 20% - 2.21
B-L-2.30 +51cm = 2.81 22.17% 2.46
B-R-2.28 +53cm = 2.81 23.24% 2.46

Amos-speakers
F-L-3.02 +80cm = 3.82 - 26.49% - 3.34
F-R-3.01 +81cm = 3.82 - 26.91% - 3.34
R-L-2.89 +79cm = 3.68 - 27.33 - 3.22
R-R-2.91 +77cm = 3.68 - 26.46% - 3.22

By doing it this way I obviously changed the delay. Which I verified using Umik-2 and REW. One of the cool features of the new app is you can right click any of the measurements and retake them. So just another thing I tried was going back in and re-measuring the first position slot of all speakers. Shifted The mic around until I got larger numbers needed to land the actual numbers back into Marantz to see if that would get the delay back to zero on starting on the left front main. Almost worked.

So here’s a question, if one were to have a clean configuration saved in the Marantz. And you went into the distant settings and changed the distances in the metric side, let’s say I pulled in everything 30 cm or whatever. Would the delay still stay at zero or close like it does when you enter a clean configuration?

Thank you all.

 

[KMO]

So I change the numbers manually, reloaded them until I came up with a number of 4.21 which then landed 3.68 in the Marantz.

Aha, that's the bit I didn't get - I didn't spot the relationship between those.

Yes, that behaviour's correct - if your real physical measurement is 4.21m, you want to see 0.875 * 4.21m = 3.68m in the AVR.

So here’s a question, if one were to have a clean configuration saved in the Marantz. And you went into the distant settings and changed the distances in the metric side, let’s say I pulled in everything 30 cm or whatever. Would the delay still stay at zero or close like it does when you enter a clean configuration?

A constant change to every speaker - such as adding or subtracting 30cm - should have absolutely no effect.

The receiver identifies the most distant speaker (FL or FR in your case), outputs sound to that with zero added delay, and adds delays to the other channels to compensate for them being closer to the listening position, based on how much closer they are.

If you only add a constant to everything, the distance from FL to C remains the same 21cm, and produces the same 21/30 = 0.7ms delay on the centre channel.

But multiplying everything by a constant scales the delays. If you entered the raw numbers 4.21m and 3.97m in, you'd have got a 24/30 = 0.8ms delay. But that's too long for a 24cm difference - sound travels faster than that - 34.3cm/ms, not 30cm/ms. The fudged distances 3.68m and 3.47m have this 30/34.3 adjustment (0.875) to produce the correct delay for 24cm: 24/34.3 = 0.7ms.

Your new measurements certainly did change the original MultEQ-X measurements, but not by that much. As I showed in my previous table, you can subtract off the average constant 13cm shift (which has no effect), and study the remaining offset changes.

Both the before and after tests did incorporate the needed 0.875 correction factor, so there should have been nothing majorly wrong to start with.

 

[MarkI]

Aha, that's the bit I didn't get - I didn't spot the relationship between those.

Yes, that behaviour's correct - if your real physical measurement is 4.21m, you want to see 0.875 * 4.21m = 3.68m in the AVR.


A constant change to every speaker - such as adding or subtracting 30cm - should have absolutely no effect.

The receiver identifies the most distant speaker (FL or FR in your case), outputs sound to that with zero added delay, and adds delays to the other channels to compensate for them being closer to the listening position, based on how much closer they are.

If you only add a constant to everything, the distance from FL to C remains the same 21cm, and produces the same 21/30 = 0.7ms delay on the centre channel.

But multiplying everything by a constant scales the delays. If you entered the raw numbers 4.21m and 3.97m in, you'd have got a 24/30 = 0.8ms delay. But that's too long for a 24cm difference - sound travels faster than that - 34.3cm/ms, not 30cm/ms. The fudged distances 3.68m and 3.47m have this 30/34.3 adjustment (0.875) to produce the correct delay for 24cm: 24/34.3 = 0.7ms.

Your new measurements certainly did change the original MultEQ-X measurements, but not by that much. As I showed in my previous table, you can subtract off the average constant 13cm shift (which has no effect), and study the remaining offset changes.

Both the before and after tests did incorporate the needed 0.875 correction factor, so there should have been nothing majorly wrong to start with.

Thanks for the response.
I really appreciate it. John’s coming over tomorrow and we’re gonna dig deeper into why the rears sound so much better and the panning from the sides to the rears also. Obviously Will be playing around in REW and try to figure out how a 1 foot shift could make such a big difference. At least what I’m hearing.

So to be clear multEQ has 3.85 the Marantz is showing 3.37. And my actual measurement is 3.68
So what I did was in MultEQ was input 4.21 not 3.85. 4.21 ends up showing in the Marantz AV 7705 3.68 Which is my actual measurement.
Same to all speakers just different Numbers?

That’s why I was asking if anybody has played around with moving the sound bubble in or out. Sorry for the Layman's term. Just easier for me to describe it that way. And why wouldn’t you try. It can’t hurt. Especially with this new app giving us the ability to do all kinds of stuff that wasn’t available before.

Over the last week I’ve been dumping files from Rew. Meaning of use the EQ feature and then imported them as a generic file. I’ve gotten some pretty bad results and some pretty amazing results.

I have a theory that nothing can be proven but I can’t prove it.

 

[Alice of Old Vincennes]

Okay, MultEQ-X is adding an average 13cm over "actual" in the first section, possibly due to some sort of overall latency. But a constant offset like that doesn't matter, as it's only the relative distances that are significant.

Subtracting off the constant 13cm, can help visualise the changes:

Speaker	MultEQ-X	Actual	MultEQ-X - 0.13cm	Difference
L	3.85	3.68	3.72	+0.04
R	3.85	3.68	3.72	+0.04
C	3.57	3.47	3.44	-0.03
Ls	3.24	3.10	3.11	+0.01
Rs	2.40	2.21	2.27	+0.06
Lrs	2.63	2.46	2.50	+0.04
Rrs	2.61	2.46	2.48	+0.03
Ltf	3.45	3.34	3.32	-0.02
Rtf	3.44	3.34	3.32	-0.02
Ltr	3.30	3.22	3.17	-0.05
Rtr	3.32	3.22	3.19	-0.03

Those are pretty small adjustments, so find it hard to find it is that significant. Have you looked at the REW results to confirm whether your measurements are producing more-aligned results? Note that the AVR's delay function only works in 3cm steps itself - we're quite close to its resolution.

When working at this level of precision, I begin to wonder how much different speakers might matter - can group delay due to different crossover types start to show up? It might be worth a couple of centimetres equivalent. I wouldn't be totally confident about saying that a tape measurement is better than MultEQ-X's audio measurement without further testing, unless the speakers are actually identical.

The bigger "secret sauce" correction is that the numbers (actual or MultEQ-X displayed) get reduced by a factor 0.875 when entered into the AVR. If you have a real FL measurement of 3.85 or 3.68, then they need to be entered into the AVR as 3.37 or 3.22. And MultEQ-X is already doing that. That correction is bigger than the changes you made - for the most extreme speaker pair of L versus Rs, it's equivalent to an 18cm shift. Your biggest shift was 11cm between Rs and Ltr.

If millimeters make a difference your speakers suck. Anthem Room Correction uses a tape measure.

 

[Alice of Old Vincennes]

Thanks everybody,

This has gone a slightly different direction than I had hoped. Have any of you played around with virtually moving the speakers out/or in. To find a sweet spot. Or as D Live claims, delivers a larger sweet spot, accurate staging. I’ve heard this before, how can that make a difference. How can you know. How can you measure it. Which I find funny because when you buy speakers you listen to them. Hopefully you have a good reference material that you’re very familiar with. And you can sit back and listen to and in my case verify the changes you’ve made. Whether they’re better or worse. Over the last few months I’ve spent a lot of time with multEQ, gotten some worse results but slowly been dialing in the better ones.

View attachment 207413
View attachment 207414View attachment 207415

the space I’m in has been blueprinted, meaning any and all measurements are almost perfect. My speaker placement is done with A laser and of course old fashion tape measures. Something I can confidently tell you I know what I’m doing with. So I have metric and regular imperial tape measures. When setting up the current system it all was carefully measured, the spacing and placement of everything. To a single point in between the sweet spot of two seats. Which is marked with blue tape. When I take my measurements with Audessy I’m using the exact same - exact first location. So I can set up the stand with mic and measure from there. This is a two person job and takes a bit of time. Ceilings are 12 feet tall. Do that a few times and you get actual real measurements within about a quarter of an inch. Maybe you’ve heard the term measure twice and cut once. John Dewey thinks what you guys are calling in anomaly which is about a 4 inch difference to the fronts is due to the size of my space. Again it’s a huge sunken living room that opens up into about 10,000 square feet. So the pressure doesn’t get built up the way it would in an enclosed space. But again who knows. It’s not worth trying to prove that the actual measurements are correct. Which they are.

And I am most certainly not trying to create a scandal about the Audessy mic and it’s measuring abilities.

If you look below the actual measurements are the actual measurements. When I go and open up multEQ-x, have it take all its measurements, you can see the measurements it comes up with below. and yes there about 10cm consistently. What it dumps into the Marantz is a fairly significant difference. Again that’s about the foot.

Actual - MultEQ-x - Marantz
F-L 3.68 - 3.85 - 3.37
F-R 3.68 - 3.85 - 3.37
C 3.47 - 3.57 - 3.12
S-L 3.10 - 3.24 - 2.83
S-R 2.21 - 2.40 - 2.10
B-L 2.46 - 2.63 - 2.30
B-R 2.46 - 2.61 - 2.28

Amos-speakers
F-L 3.34 - 3.45 - 3.02
F-R 3.34 - 3.44 - 3.01
R-L 3.22 - 3.30 - 2.89
R-R 3.22 - 3.32 - 2.91

This all started for me a few weeks back when the time thing issue was brought up. And I went about it a completely different way. Which led me to where we’re I’m at now. By virtually moving my speaker out which I believe I’m saying right is actually pulling them in. The sound-stage delivers a larger sweet spot and more accurate staging.

I didn’t stumble upon this by opening up the Marantz and moving all of the speakers 1 foot closer/or playing around in the metric side with 10cm increments. Which by the way because of this thread / form I have now I have my Marantz left in the metric setting. So again thank you. Instead I loaded my actual numbers into multEQ only to find that what landed in the Marantz was different. You guys obviously understand this computation or mathematical formula that happens. I didn’t. So I change the numbers manually, reloaded them until I came up with a number of 4.21 which then landed 3.68 in the Marantz. This is my actual measurement. Not sure what’s happening in between however again it changed my sound stage. For the better. If you’re in California in the Danville area you’re more than happy to come over, sit back and tell me what you think. By the way I have probably two dozen different MultEQ-X profiles saved that we can compare it to.

Marantz Correction added to get Actual distances
FL-3.37 +84cm = 4.21 - 24.92% - 3.68
FR-3.37 +84cm = 4.21 - 24.92% - 3.68
C—3.12 +85cm = 3.97 - 27.24%- 3.47
S-L-2.83 +52cm = 3.55 - 25.44 - 3.10
S-R-2.10 +42cm = 2.52 - 20% - 2.21
B-L-2.30 +51cm = 2.81 22.17% 2.46
B-R-2.28 +53cm = 2.81 23.24% 2.46

Amos-speakers
F-L-3.02 +80cm = 3.82 - 26.49% - 3.34
F-R-3.01 +81cm = 3.82 - 26.91% - 3.34
R-L-2.89 +79cm = 3.68 - 27.33 - 3.22
R-R-2.91 +77cm = 3.68 - 26.46% - 3.22

By doing it this way I obviously changed the delay. Which I verified using Umik-2 and REW. One of the cool features of the new app is you can right click any of the measurements and retake them. So just another thing I tried was going back in and re-measuring the first position slot of all speakers. Shifted The mic around until I got larger numbers needed to land the actual numbers back into Marantz to see if that would get the delay back to zero on starting on the left front main. Almost worked.

So here’s a question, if one were to have a clean configuration saved in the Marantz. And you went into the distant settings and changed the distances in the metric side, let’s say I pulled in everything 30 cm or whatever. Would the delay still stay at zero or close like it does when you enter a clean configuration?

Thank you all.

This is why I despise Audyssey.

 

[KMO]

So to be clear multEQ has 3.85 the Marantz is showing 3.37. And my actual measurement is 3.68
So what I did was in MultEQ was input 4.21 not 3.85. 4.21 ends up showing in the Marantz AV 7705 3.68 Which is my actual measurement.

Okay, that's not what I thought you meant. (The confusion here is because you've been using "actual" to describe two different numbers).

Now you seem to be saying the real, actual tape distance is 3.68m, but you've put an inflated 4.21m into MultEQ-X to try make it send 3.68m to the AVR.

That's now undoing the correction work for the odd speed-of-sound constant.

You say you have REW, but I'm not clear what measurements you're making. If you are sending sweeps to individual channels, you should be able to see the relative timing of them.

If your real FL and SR measurements are 3.68m and 2.21m, then the real flight delay difference should be 4.3ms. ((3.68-2.21)/343).

If you set them to have the same distance (eg 2m for both) in the AVR, so no relative correction delay, REW should show FL arriving at the mic 4.3ms later than SR.

With your original MultEQ-X uploaded measurements of 3.37m and 2.10m in the AVR, the AVR would have arranged a 4.2ms delay. ((3.37-2.10)/300). This is pretty close. With that correction, REW should show FL arriving 0.1ms later than SR. The basic granularity of the adjustment is only 0.1ms (=3cm), so it's hard to do better.

With your setup of putting real measurements 3.68m and 2.21m into the AVR, the AVR will arrange a 4.9ms delay. ((3.68-2.21)/300). This is an overcompensation. REW should show FL arriving 0.6ms earlier than SR. (Equivalent to an 18cm measurement error).

If you put your real measurements 3.68m and 2.21m into MultEQ-X, MultEQ-X will set the AVR to 3.22m and 1.93m. The AVR will arrange a 4.3ms delay ((3.22-1.93)/300), which is correct. REW should show FL and SR arriving simultaneously (or within 0.1ms). And that's what we're aiming for.

You should be using REW to confirm that speakers are aligned. Easley, who did most of the investigation on this, had these before and after REW measurements:

The top "before" graph is what happens when you put real distances into the AVR, and that's what I expect you to be getting now. (A bit better, because your speaker distance spread is smaller - he gets a 800us spread, while you'd have 600us).

If you put real distances multiplied by 0.875 into the AVR (or MultEQ-X does it for you because you give it real distances), I would expect to see results like the bottom graph. (The 100us grid lines represent match the resolution of the adjustment, so bottom graph is pretty much as good as it gets - all speakers within one step).

That’s why I was asking if anybody has played around with moving the sound bubble in or out. Sorry for the Layman's term. Just easier for me to describe it that way. And why wouldn’t you try. It can’t hurt. Especially with this new app giving us the ability to do all kinds of stuff that wasn’t available before.

I believe your current setup, if I've interpreted correctly (it's possible I've still misunderstood), is worse than where you started, purely in terms of measuring what we were trying to achieve, which is perfect time-alignment of the sounds arriving at the listening position.

If you are preferring the misaligned sound arrival, then we're into psychoacoustics. There may be a reason misaligned speakers are preferred. The assumption has always been that we want the sound to arrive simultaneously from all speakers, with equal level. Hence the distance and level adjustments.

Your misalignment is systematic - it's based on distance, and it's an overcompensation. More distant speakers arrive earlier than closer ones.

Without any adjustment, closer speakers arrive earlier than distant ones, and the brain spots that, and kind of tunes out the distant ones, and steers direction judgement of images towards the closer.

Maybe with inverted distance cues (and balanced SPL), some other uncorrected-for spatial cue is overridden. The brain starts tuning in more on the more distant speakers that arrive earlier, and this compensates for something about them being more distant, on top of the SPL correction?

I've not heard of anyone really researching this. You're welcome to try.

When experimenting, the adjustment to make is the multiplicative one - multiplying all distances by a constant.

If entering directly into the AVR (which is easier), the cases are:

real distance * 1.something: more overcompensation
real distance: overcompensation (distant speakers earlier)
real distance * 0.875: perfect compensation (speakers simultaneous)
real distance * smaller fraction: undercompensation (closer speakers earlier)
all distances equal: no compensation (natural flight delay)

I'm not convinced by the reports so far of people playing with the 0.875 constant, so comparing the second and third cases. I've seen people say it's better, but then there were also people doing early tests who made a change in the opposite direction (when there was confusion about which way to adjust), thus to the first case, who said that was better!

I suspect people are just hearing what they want to hear - I have no idea whether there's anything's detectable in a blind test. I just know this improves the measurements, and achieves what we were trying to achieve - perfect time-alignment.

Even if something in spatial cues might be helped by misalignment, proper alignment should benefit a whole bunch of things, helping sub integration and avoiding weird frequency response behaviour on stereo images between distance-mismatched pairs.

 

[MarkI]

Okay, that's not what I thought you meant. (The confusion here is because you've been using "actual" to describe two different numbers).

Now you seem to be saying the real, actual tape distance is 3.68m, but you've put an inflated 4.21m into MultEQ-X to try make it send 3.68m to the AVR.

That's now undoing the correction work for the odd speed-of-sound constant.

You say you have REW, but I'm not clear what measurements you're making. If you are sending sweeps to individual channels, you should be able to see the relative timing of them.

If your real FL and SR measurements are 3.68m and 2.21m, then the real flight delay difference should be 4.3ms. ((3.68-2.21)/343).

If you set them to have the same distance (eg 2m for both) in the AVR, so no relative correction delay, REW should show FL arriving at the mic 4.3ms later than SR.

With your original MultEQ-X uploaded measurements of 3.37m and 2.10m in the AVR, the AVR would have arranged a 4.2ms delay. ((3.37-2.10)/300). This is pretty close. With that correction, REW should show FL arriving 0.1ms later than SR. The basic granularity of the adjustment is only 0.1ms (=3cm), so it's hard to do better.

With your setup of putting real measurements 3.68m and 2.21m into the AVR, the AVR will arrange a 4.9ms delay. ((3.68-2.21)/300). This is an overcompensation. REW should show FL arriving 0.6ms earlier than SR. (Equivalent to an 18cm measurement error).

If you put your real measurements 3.68m and 2.21m into MultEQ-X, MultEQ-X will set the AVR to 3.22m and 1.93m. The AVR will arrange a 4.3ms delay ((3.22-1.93)/300), which is correct. REW should show FL and SR arriving simultaneously (or within 0.1ms). And that's what we're aiming for.

You should be using REW to confirm that speakers are aligned. Easley, who did most of the investigation on this, had these before and after REW measurements:

The top "before" graph is what happens when you put real distances into the AVR, and that's what I expect you to be getting now. (A bit better, because your speaker distance spread is smaller - he gets a 800us spread, while you'd have 600us).

If you put real distances multiplied by 0.875 into the AVR (or MultEQ-X does it for you because you give it real distances), I would expect to see results like the bottom graph. (The 100us grid lines represent match the resolution of the adjustment, so bottom graph is pretty much as good as it gets - all speakers within one step).


I believe your current setup, if I've interpreted correctly (it's possible I've still misunderstood), is worse than where you started, purely in terms of measuring what we were trying to achieve, which is perfect time-alignment of the sounds arriving at the listening position.

If you are preferring the misaligned sound arrival, then we're into psychoacoustics. There may be a reason misaligned speakers are preferred. The assumption has always been that we want the sound to arrive simultaneously from all speakers, with equal level. Hence the distance and level adjustments.

Your misalignment is systematic - it's based on distance, and it's an overcompensation. More distant speakers arrive earlier than closer ones.

Without any adjustment, closer speakers arrive earlier than distant ones, and the brain spots that, and kind of tunes out the distant ones, and steers direction judgement of images towards the closer.

Maybe with inverted distance cues (and balanced SPL), some other uncorrected-for spatial cue is overridden. The brain starts tuning in more on the more distant speakers that arrive earlier, and this compensates for something about them being more distant, on top of the SPL correction?

I've not heard of anyone really researching this. You're welcome to try.

When experimenting, the adjustment to make is the multiplicative one - multiplying all distances by a constant.

If entering directly into the AVR (which is easier), the cases are:

real distance * 1.something: more overcompensation
real distance: overcompensation (distant speakers earlier)
real distance * 0.875: perfect compensation (speakers simultaneous)
real distance * smaller fraction: undercompensation (closer speakers earlier)
all distances equal: no compensation (natural flight delay)

I'm not convinced by the reports so far of people playing with the 0.875 constant, so comparing the second and third cases. I've seen people say it's better, but then there were also people doing early tests who made a change in the opposite direction (when there was confusion about which way to adjust), thus to the first case, who said that was better!

I suspect people are just hearing what they want to hear - I have no idea whether there's anything's detectable in a blind test. I just know this improves the measurements, and achieves what we were trying to achieve - perfect time-alignment.

Even if something in spatial cues might be helped by misalignment, proper alignment should benefit a whole bunch of things, helping sub integration and avoiding weird frequency response behaviour on stereo images between distance-mismatched pairs.

Thank you,

This is quite helpful. John is coming over tomorrow. we will be looking into this.

 

[Chromatischism]

This is why I despise Audyssey.

You need to read the last page of the thread. Audyssey is working properly.

If using MultEQ-X, it is giving accurate distance measurements. Do not change the distances in the AVR as they are already offset to compensate for the inaccurate calculations in the AVR.

 

[KMO]

I do wonder if there are any other AVRs with the same "feature".

Going back, this bit of 1ft=1ms=30cm approximation is a legacy thing dating back maybe 20 years in D+M receivers.

It's quite possible others took the same shortcut for the same reason, way back on on a crude integer microcontroller, reckoning it's "good enough", and not touched it since.

Has anyone checked Anthem's?

 

[Chromatischism]

I only have Denon here, so am not of much use. Interesting thought, though.

 

[KMO]

Another thing that just occurred to me. 1ft=1ms=30cm WAS good enough, when that was the resolution.

Older Marantz (and Denon?) receivers only offered 1ms/1ft resolution on the adjustment. At that level, the bad "before" graph there is as good as you'd get anyway, due to the resolution. 800us spread is as good as you'd get.

If you zoomed out by a factor of ten, an totally non-distance-corrected-graph would look like the bad "before" and the corrected-with-bad-speed-of-sound estimate would be a nice tight "after".

The problem is that they increased the adjustment resolution by a factor of 10 to 0.1ms, but never got a factor of ten real improvement, because the error from the speed-of-sound now dominated.

 

[andyc56]

I do wonder if there are any other AVRs with the same "feature".

Going back, this bit of 1ft=1ms=30cm approximation is a legacy thing dating back maybe 20 years in D+M receivers.

It's quite possible others took the same shortcut for the same reason, way back on on a crude integer microcontroller, reckoning it's "good enough", and not touched it since.

Has anyone checked Anthem's?

I've got an Outlaw 976 (made by ATI I think) that I could do some detailed measurements on when I'm done with the Denon 3312.

 

[HarmonicTHD]

You need to read the last page of the thread. Audyssey is working properly.

If using MultEQ-X, it is giving accurate distance measurements. Do not change the distances in the AVR as they are already offset to compensate for the inaccurate calculations in the AVR.

Just to confirm. Getting a bit confused here.

If using the MultiEQX Windows software and the standard Audyssey mic for the sweeps one gets a bit different distances in MultiEQX display, than the actual (eg measured by Laser distance meter) measured distances.
BUT, when uploading the filters (without any prior manual manipulation) to the AVR (x3700 in my case), all is good, although the AVR speaker distance menu again shows different distance value.

In conclusion. Use the MultiEQX software and don’t worry about any of the distances.

Correct?

thx.

BTW. There was a firmware update today. Did that do anything in that regard.

 

[Chromatischism]

In conclusion. Use the MultiEQX software and don’t worry about any of the distances.

Right. Unless something went wrong and the distance of a speaker in MEQX is not accurate.

 

[HarmonicTHD]

Right. Unless something went wrong the distance of a speaker in MEQX is not accurate.

Thx. Appreciate it.