"Official" Audyssey thread (FAQ in post #51779) - Page 497 - AVS Forum
First ... 495  496  497 498  499  ... Last
Receivers, Amps, and Processors > "Official" Audyssey thread (FAQ in post #51779)
pepar's Avatar pepar 10:10 AM 05-29-2009
A very good, short read on "The History and Future of Surround Sound" (distilled largely from T. Holman's book "Surround Sound, Second Edition: Up and Running") can be found here. For where we might be headed, I draw your attention to the last paragraph.

audyssey's Avatar audyssey 11:14 AM 05-29-2009
Quote:
Originally Posted by pepar View Post

I would pick a 7.1 speaker system that fit my budget and trust that the manufacturer had "balanced" the system. As far as the AVR goes, you wil need to match power output to the speaker system you select.

Chris might have some insight into a "formula" for allocating $$$ between the AVR and speaker system . .

This is good advice, but there is no formula that I have for allocating $$$
Mike_WI's Avatar Mike_WI 11:25 AM 05-29-2009
Quote:
Originally Posted by SoundChex View Post

Not so 'off the wall' as you might think! NHK Japanese tv says it plans to start broadcast-to-home transmission of Super Hi-Vision (SHV) tv "within the next 10 years", and SHV is 22.2 channel audio [with 11 front speakers...]

Cool, but do most/many Japanese homes have that much room?

Mike


Mike_WI's Avatar Mike_WI 11:29 AM 05-29-2009
Quote:
Originally Posted by pepar View Post

A very good, short read on "The History and Future of Surround Sound" (distilled largely from T. Holman's book "Surround Sound, Second Edition: Up and Running") can be found here. For where we might be headed, I draw your attention to the last paragraph.

Interesting.

Here is a different take on things...

"He also pointed out how the right and left side channels of their six-channel array were used for low-channel effects, thus introducing an early equivalent of today's .1 subwoofer channel. Holman amusingly pointed out that the .1 of 5.1 multichannel is more a marketing concept than a reality. Considering the percentage of bandwidth actually needed for low channel effects, he said that it could more accurately be described as 5.005 surround sound."

Mike
SoundChex's Avatar SoundChex 11:57 AM 05-29-2009
Quote:
Originally Posted by Mike_WI View Post


Quote:
Originally Posted by SoundChex View Post

Not so 'off the wall' as you might think! NHK Japanese tv says it plans to start broadcast-to-home transmission of Super Hi-Vision (SHV) tv "within the next 10 years", and SHV is 22.2 channel audio [with 11 front speakers...]

Cool, but do most/many Japanese homes have that much room?

Mike


(1) I believe the current development consortium includes the BBC and Italian tv also...

(2) With 22.2 discrete channels you get lots of good mixdown options for room configurations that include fewer speakers (incl. 2.0, 5.1, 7.1 backward compatibility).

(3) My home office is only 11' x 11', but I would happily(?) expand the existing 8.1 system to 18.2 (skipping the Center-Overhead, and 3 'lower layer' satellite speakers.)
pepar's Avatar pepar 12:04 PM 05-29-2009
Quote:
Originally Posted by audyssey View Post

This is good advice, but there is no formula that I have for allocating $$$

It might be moot; the equipment you want will always cost more than the equipment you can afford.
giomania's Avatar giomania 12:09 PM 05-29-2009
According to the posting dates, Chris Marten's third installation of his visit to Audyssey should be posted today. Here is what he will focus on:

In part 3 of this blog series, I'll discuss the sound of Audyssey DSX, questions that the DSX experience raised for me, and other new technology developments from Audyssey. Stay tuned for more.

Here is the link to the web page:

http://www.avguide.com/blogs/chris-martens

Mark
Sirquack's Avatar Sirquack 03:39 PM 05-29-2009
Chris any ideas on my sub problem from my above posts and distance being closer than actual. Hope my forum name is not scaring people off. Randy
pepar's Avatar pepar 09:47 PM 05-29-2009
Quote:
Originally Posted by giomania View Post

According to the posting dates, Chris Marten's third installation of his visit to Audyssey should be posted today.

Not there yet . .
JonFo's Avatar JonFo 03:35 AM 05-30-2009
Quote:
Originally Posted by SoundChex View Post

(1) I believe the current development consortium includes the BBC and Italian tv also...

(2) With 22.2 discrete channels you get lots of good mixdown options for room configurations that include fewer speakers (incl. 2.0, 5.1, 7.1 backward compatibility).

...

While I'm happier than most to cram more speakers into my dedicated HT (heck, I even pre-wired for a ceiling (overhead) channel ten years ago), I find these new n-channel discreet systems to be nuts.

Whatever happened to good old vectored audio steering systems, where the playback system could have anywhere from as few as three to as many as the preamp could sustain?

Early attempts at this was stuff like Ambisonics, but I have to assume progress has been made on these fronts in the past 20 or 30 years.

Anyone know?


BTW- can you image sitting through the Audyssey calibration for a 22 speaker array using 8 positions? That's like, 176 measurements. At an average of one per every 30 seconds (plus some time to move mics), that could take an hour and 45 minutes to complete the measurements

Chris, I think you're going to have to speed up the measurement process (DSX is already creating pressure due to this)
MACCA350's Avatar MACCA350 06:03 AM 05-30-2009
Chris, Just realised that even Denon's graphs show that Audyssey is boosting the low frequencies, at least down to the limit of their graphs.
Take a look at these snaps I took and compare the 3808's graph to the measured result(Audyssey on/off at line level). You can even see in the 3808's graph that even it is showing that Audyssey it is in fact boosting at 20Hz the limit of the graph, you could extrapolate that out to extend lower also.

This also backs up the sub measurements taken as these are almost a perfect match. This also means that the graphs that the 3808 shows are post normalisation.

Denon Left

Measured Left


Denon Right

Measured Right


Have you had a chance to look into this further?

cheers
Sirquack's Avatar Sirquack 09:02 AM 05-30-2009
I believe Chris has mentioned that these Denono graphs are a very crude representation of what Odyssey is applying to even out the FR. If you flip these graphs, you will see what your FR was before Audyssey, the opposite of what your seeing. So Audyssey is actually not boosting those low frequencies, but rather creating a dip to bring down the peak "before" Audyssey.

Unless I misunderstood Chris earlier in this thread, that was my understanding.
MACCA350's Avatar MACCA350 11:02 AM 05-30-2009
Quote:
Originally Posted by Sirquack View Post

I believe Chris has mentioned that these Denono graphs are a very crude representation of what Odyssey is applying to even out the FR. If you flip these graphs, you will see what your FR was before Audyssey, the opposite of what your seeing. So Audyssey is actually not boosting those low frequencies, but rather creating a dip to bring down the peak "before" Audyssey.

Unless I misunderstood Chris earlier in this thread, that was my understanding.

You've lost me Why would I need to flip those graphs? If you are referring to flipping just for informational purposes here's one with a previous Audyssey Auto setup run:

Post #8661

Blue = Audyssey filters line level (from previous graph)
Green = Audyssey off inroom(8 position average)
Red = Audyssey on inroom(8 position average)

Those graphs in my previous post show how Audyssey is affecting the preamp output with the mains running fullrange, the straight red and purple line is with Audyssey off, the green and blue lines are with Audyssey on.

While those Denon graphs are a crude representation they are remarkably accurate to what is happening at the actual outputs, both of which show Audyssey is boosting 20Hz and below, the sub channel you can see is being boosted all the way down below 2Hz

Post #14334

Subwoofer output with Audyssey OFF calibrated as reference
Blue=Audyssey off line level
Green=Audyssey on line level

cheers
audyssey's Avatar audyssey 11:24 AM 05-30-2009
Quote:
Originally Posted by MACCA350 View Post

Chris, Just realised that even Denon's graphs show that Audyssey is boosting the low frequencies, at least down to the limit of their graphs.

It's really important to remember that these graphs show the filter and not the acoustical response. So, a 4 dB (or 9 dB in the sub plot) boost to an acoustical response that is 20-30 dB down because of the natural speaker roll off is not really an issue.
larryy's Avatar larryy 12:51 PM 05-30-2009
MACCA350: I'm still learning about Audyssey and don't even have my Denon AVP and Audyssey gear yet, but having worked in signal processing and data analysis, I think I understand the graphs, so maybe I can help.

Looking at the first graph in your most recent post (the only one with all three colors), it seems clear that your room has a resonance that boots the signal in a broad range around 65 Hz (as seen by that large peak in the green/Audyssey-off curve), and Audyssey is doing a good job of neutralizing it by reducing power in that range (the dip in the blue/Audyssey-on-line-level), producing an expected flattened response curve, as desired, in that range (as seen in the red/Audyssey-on curve in that range). Going the other way there is a notch in power around 200 Hz that Audyssey gives more energy to in order to compensate. And there are various other small or large notches compensated for, including about 640 Hz and 2.8 - 4 KHz. Apologies if that's all obvious to you already.

It certainly does appear in that second graph that Audyssey is boosting your bass response in the low end, but you didn't include the Audyssey-off in-room and expected Audyssey-on in-room curves in your most recent post, and you didn't include all the curves or provide a color legend in the previous post, so it's hard to say exactly what's going on, but I suspect the red and purple lines in your earlier post correspond to what Audyssey expects you to hear as a result of their line-level modulation, and those look pretty flat. Unclear, though, as the absolute levels are so different from the later curves you show, and, again, I'm a novice to the specifics of the Audyssey package. If you show the sub-woofer graph with all three curves it would be easier to interpet.

Oh, and what does it sound like?

HTH.

- larryy
larryy's Avatar larryy 12:55 PM 05-30-2009
One thing I'm a little disturbed about with all room-EQ software is that there doesn't appear to ever be an actual, measured in-room response after the equalization is applied. It's like doing a scientific test and not bothering to measure the results. I guess they have enough experience with their system that one can have a modest degree of confidence in their projected post-EQ response curves. But I'd much rather spend a bit more time taking measurements and make it an iterative process. Measure. Make adjustment 1. Measure with adjustment 1. Make adjustment 2. Measure with adjustment 2. Make adjustment 3. That would probably be sufficient for the trickiest of environments. Is this supported with Audyssey Pro? To incorporate knowledge of previous measurements and adjustments in making subsequent adjustments? Or should we start a competitor to Audyssey and ARC?

I see that people are using various pieces of software themselves to measure the results. That's a good idea. I just think the correction software should take those measurements into account.
audyssey's Avatar audyssey 01:32 PM 05-30-2009
Quote:
Originally Posted by larryy View Post

One thing I'm a little disturbed about with all room-EQ software is that there doesn't appear to ever be an actual, measured in-room response after the equalization is applied.

Larry, that's the beauty of linear system theory. Through system identification one obtains the impulse response of the speaker-room system. That's not as straightforward as one would find in a DSP textbook as it involves combining multiple measurements and weighting them properly to account for the spatial distribution of problems and also for psychoacoustics. But, once that is done and you have the impulse response for each speaker then there is no need to measure with iterations. The final room response equals the measured response convolved (in the time domain) with the impulse response. The Pro Audyssey software shows these results, but the built in version is limited by what each manufacturer wants to show in their OSD. So far, none have chosen to show the before-after responses.
bluesky636's Avatar bluesky636 02:55 PM 05-30-2009
Quote:
Originally Posted by audyssey View Post

Larry, that's the beauty of linear system theory. Through system identification one obtains the impulse response of the speaker-room system. That's not as straightforward as one would find in a DSP textbook as it involves combining multiple measurements and weighting them properly to account for the spatial distribution of problems and also for psychoacoustics. But, once that is done and you have the impulse response for each speaker then there is no need to measure with iterations. The final room response equals the measured response convolved (in the time domain) with the impulse response. The Pro Audyssey software shows these results, but the built in version is limited by what each manufacturer wants to show in their OSD. So far, none have chosen to show the before-after responses.

Ya know, sometimes it is a bad thing to give people too much information. The natural tendency is to try to pick apart that which you do not understand. I am glad that I don't have graphs of my system to pour over trying to understand every little detail. I recognize that Chris is way smarter on this subject than I ever hope to be. All I really care about is the fact that when I am done calibrating my system with MultEQ, it sounds immensely better than it ever did before. Great job Chris.
xradman's Avatar xradman 04:06 PM 05-30-2009
I just completed a full Audyssey calibration on my HT system and noticed something peculiar. I have a Denon 3808 CI receiver with 7.1 Klipsch Reference based speaker system (RC25 for L/C/R, RB25 Side and rear surround, and RW12 subwoofer). I chose 4 different listening position for calibration. Not surprisingly, Audyssey set all of my speakers to "Small". However, cross over point for all the speakers was set to 60Hz for the subwoofer. Isn't this contradictory? How can the system set the cross over point at 60Hz for what it deems "Small" speakers?
audyssey's Avatar audyssey 04:22 PM 05-30-2009
Quote:
Originally Posted by xradman View Post

I just completed a full Audyssey calibration on my HT system and noticed something peculiar. I have a Denon 3808 CI receiver with 7.1 Klipsch Reference based speaker system (RC25 for L/C/R, RB25 Side and rear surround, and RW12 subwoofer). I chose 4 different listening position for calibration. Not surprisingly, Audyssey set all of my speakers to "Small". However, cross over point for all the speakers was set to 60Hz for the subwoofer. Isn't this contradictory? How can the system set the cross over point at 60Hz for what it deems "Small" speakers?

First of all Audyssey does not set speakers to Small or Large. Denon does based on their own rules. In all their recent products they use 40 Hz as the criterion. So, it's not all that surprising that with a 60 Hz roll-off point found for the Klipsch speaker that Denon set your speakers to Small with a 60 Hz crossover. That's what Small means: set a crossover point and redirect the bass below that point to the subwoofer.
crops85's Avatar crops85 04:54 PM 05-30-2009
Hey
I found my bass was humming so I went to the forums and ran into giomania's Audyssesy setup guide, which was really informative, so I re-ran the setup.

Now it seems that I have NO bass at all.
My "Low pass" on the sub can't be turned off (only as low as 40hz) so I turned it all the way up, and set the volume at 12 o'clock, and after doing audyssesy the subwoofer trim level was -12 db so I turn the volume down according to the guide. I kept turning the volume down and down, and finally got the trim level to settle at -3.5 db (the guide says try to get "the subwoofer trim level in the range of ±3 dB", I know it is not in that range but the volume was so low at this point, i still wanted to have some volume)

Now here is the problem. To get the sw trim level at -3.5 db, the volume knob is at 8 o'clock, which is less than a quarter (25%) on, I'd say about 15% on.

Is this correct, because I barely have any bass in my system now

Any help would be greatly appreciated.

Thanks
audyssey's Avatar audyssey 04:58 PM 05-30-2009
Quote:
Originally Posted by crops85 View Post

Is this correct, because I barely have any bass in my system now

It is correct if your speakers are set to Small. Otherwise there is no bass being sent to the subwoofer. If they are not set to Small you have to do that manually.
xradman's Avatar xradman 05:24 PM 05-30-2009
Quote:
Originally Posted by audyssey View Post

First of all Audyssey does not set speakers to Small or Large. Denon does based on their own rules. In all their recent products they use 40 Hz as the criterion. So, it's not all that surprising that with a 60 Hz roll-off point found for the Klipsch speaker that Denon set your speakers to Small with a 60 Hz crossover. That's what Small means: set a crossover point and redirect the bass below that point to the subwoofer.

Maybe I'm old school, but I would have thought "Small" speakers meant cross over point of ~120Hz or so.
audyssey's Avatar audyssey 05:30 PM 05-30-2009
Quote:
Originally Posted by xradman View Post

Maybe I'm old school, but I would have thought "Small" speakers meant cross over point of ~120Hz or so.

Nothing to do with old school or new school. Small simply means: turn on bass management and redirect the content from that speaker to the subwoofer. The traditional value for that was 80 Hz, but every manufacturer uses their own definition. More on that in my blog (link below).
xradman's Avatar xradman 06:41 PM 05-30-2009
Quote:
Originally Posted by audyssey View Post

Nothing to do with old school or new school. Small simply means: turn on bass management and redirect the content from that speaker to the subwoofer. The traditional value for that was 80 Hz, but every manufacturer uses their own definition. More on that in my blog (link below).

Thank you for your insightful comments. I came from Yamaha receivers which used their proprietary room EQ and wasn't aware of different approach taken by Audyssey. Your explanation makes perfect sense and engineer in me says, it's the right way to go.
Djoel's Avatar Djoel 07:43 PM 05-30-2009
Just post this on the Onkyo 886 thread, I'm getting low volume during the chirps, which I know shouldn't be the case when running the Audyssey on the 806..

If anyone has a suggestion, I'll be happy to hear it as I've become quite accustomed to listening my music listening and my HT with it on..

Thanks

Djoel
MACCA350's Avatar MACCA350 08:00 PM 05-30-2009
Quote:
Originally Posted by audyssey View Post

It's really important to remember that these graphs show the filter and not the acoustical response. So, a 4 dB (or 9 dB in the sub plot) boost to an acoustical response that is 20-30 dB down because of the natural speaker roll off is not really an issue.

It is an issue if it's draining amplifier power or causing the sub to try an produce frequencies that it cannot or causing the line level signal to clip.

Either way Audyssey's normalisation function is causing frequencies below the -3db point of the speaker to be boosted, something you originally said was impossible. Are you saying that you are not interested in implimenting a fix?

Quote:
Originally Posted by larryy View Post

MACCA350: I'm still learning about Audyssey and don't even have my Denon AVP and Audyssey gear yet, but having worked in signal processing and data analysis, I think I understand the graphs, so maybe I can help.

Looking at the first graph in your most recent post (the only one with all three colors), it seems clear that your room has a resonance that boots the signal in a broad range around 65 Hz (as seen by that large peak in the green/Audyssey-off curve), and Audyssey is doing a good job of neutralizing it by reducing power in that range (the dip in the blue/Audyssey-on-line-level), producing an expected flattened response curve, as desired, in that range (as seen in the red/Audyssey-on curve in that range). Going the other way there is a notch in power around 200 Hz that Audyssey gives more energy to in order to compensate. And there are various other small or large notches compensated for, including about 640 Hz and 2.8 - 4 KHz. Apologies if that's all obvious to you already.

It certainly does appear in that second graph that Audyssey is boosting your bass response in the low end, but you didn't include the Audyssey-off in-room and expected Audyssey-on in-room curves in your most recent post, and you didn't include all the curves or provide a color legend in the previous post, so it's hard to say exactly what's going on, but I suspect the red and purple lines in your earlier post correspond to what Audyssey expects you to hear as a result of their line-level modulation, and those look pretty flat. Unclear, though, as the absolute levels are so different from the later curves you show, and, again, I'm a novice to the specifics of the Audyssey package. If you show the sub-woofer graph with all three curves it would be easier to interpet.

Oh, and what does it sound like?

HTH.

- larryy

Thanks for your interest Larry, but I do understand the room acoustic issues and that is not why I have been posting these graphs. My interest at the moment, which has been for a while now, is simply to draw attention to the boosting being applied by Audyssey below the capabilities of the speakers, and hopefully get a fix for those people where this is causing issues.

Here is one, I've posted earlier in this thread, of the acoustical output of the sub and how the line signal is affecting the sub:

Post #8675

Red = Audyssey off Line level (1/3 filtered) reference line for audyssey filters
Green = Audyssey filters line level (1/3 filtered)
Blue = Audyssey off (1/3 filtered)
Purple = Audyssey on (1/3 filtered)
Aqua = Noisefloor (1/24 capture)

Currently the only fix is to implement a subsonic filter, and for those who don't have this built in to the sub they've had to try out products such as the FMod which have been less than ideal. FYI I have no personal issue with this since my Velo DD15 has a built in adjustable subsonic along with built in protection.

Anyway it looks like we may be at a dead end having this fixed at the problems source..........something like Audyssey applying acoustically equal boost and cut filters in the first place would mean the normalisation function would not need to be applied. You'll notice that Audyssey only applied cuts filters in the sub channel, hence the need for the normalisation process to boost the result and in effect boost any frequency that had no filters applied.

Anyway, if Chris is unwilling to accept this is a real issue that should be addressed, so be it.

cheers
counsil's Avatar counsil 09:36 PM 05-30-2009
At the recommendation of a few people, I purchased the Audyssey Pro kit. When everyone left the house today I took a couple of hours to calibrate my Denon 3808 with it for the first time. I only measured 16 out of the 32 available positions.

Here is a sample of what the regular Audyssey MultEQ XT did for my system. I am not sure when I ran these particular REW graphs. I have run MultEQ XT dozens of times in the past few weeks!

Attachment 144040
Blue - No EQ, Purple - Audy On, Yellow - Dyn EQ On

Here is what Audyssey Pro did for my system this afternoon (I am using REW just so you can compare to above)...

Attachment 144041
Purple - No EQ, Red - Audy On, Pink - Dyn EQ On


I have a lot of things running through my mind right now that I want to write about, but here are a few things I want to mention in this post. First off, my system sounds much better after running Audyssey Pro. I know this is subjective and broad. But there you have it. Secondly, Audyssey Pro easily detected the high pass rolloff of my dual SVS PB13 Ultras . This is something I have really been working hard to get MultEQ XT to do.

BTW, my subs were in the 15Hz tune, with the room comp feature (HPF) bypassed in both of the calibrations above.

Lastly, I wanted to mention something very weird. MultEQ XT set my sub trim level to -1.5. Audyssey Pro set it to -10 ! The gain knobs on my subs have not been touched in weeks. They are set to the 8:00 line. Anyway, one would think that I lost a lot of bass. Not so. My subs produce the same level of bass at the exact same master volume. But let me tell you this. My subs don't work nearly as hard to achieve the same SPL dB levels. I guess all that boosting in the teen Hz area was taking a lot of amp juice? What do you guys think about the huge decrease in the sub trim level?

I tested all my demo material at 0 master volume (reference). My system encountered no troubles what so ever playing at this volume. I tested with KFP, WOTW, Incredible Hulk, Pulse, and Flight of the Phoenix.

Audyssey Pro so far seems to be working wonders for my setup.
LL
LL
pepar's Avatar pepar 09:44 PM 05-30-2009
council, this is only 200hz and under. Is that all you measured?
allredp's Avatar allredp 09:56 PM 05-30-2009
Quote:
Originally Posted by counsil View Post

At the recommendation of a few people, I purchased the Audyssey Pro kit. When everyone left the house today I took a couple of hours to calibrate my Denon 3808 with it for the first time. I only measured 16 out of the 32 available positions.

I have a lot of things running through my mind right now that I want to write about, but here are a few things I want to mention in this post. First off, my system sounds much better after running Audyssey Pro. I know this is subjective and broad. But there you have it. Secondly, Audyssey Pro easily detected the high pass rolloff of my dual SVS PB13 Ultras . This is something I have really been working hard to get MultEQ XT to do.

BTW, my subs were in the 15Hz tune, with the room comp feature (HPF) bypassed in both of the calibrations above.

Lastly, I wanted to mention something very weird. MultEQ XT set my sub trim level to -1.5. Audyssey Pro set it to -10 ! The gain knobs on my subs have not been touched in weeks. They are set to the 8:00 line. Anyway, one would think that I lost a lot of bass. Not so. My subs produce the same level of bass at the exact same master volume. But let me tell you this. My subs don't work nearly as hard to achieve the same SPL dB levels. I guess all that boosting in the teen Hz area was taking a lot of amp juice? What do you guys think about the huge decrease in the sub trim level?

I tested all my demo material at 0 master volume (reference). My system encountered no troubles what so ever playing at this volume. I tested with KFP, WOTW, Incredible Hulk, Pulse, and Flight of the Phoenix.

Audyssey Pro so far seems to be working wonders for my setup.

Wow!!! That's very exciting news as I'm in the same boat, though I only have 1 Ultra that is clearly having the same issue.

PM, or post here how much the Pro kit cost? That may be just what I'm needing...

Great news for you, Counsil!!!
Tags: Audyssey , Receivers Amplifiers , Kef Kht1005 2se 5 1 Subwoofer Satellite System With C4 Subwoofer Gloss White , 5 6 7 1 7 2 Or 8 1 8 2 One Or Two Subwoofer Compatible 16 Banana Post 2 Rca Speaker Wall Plate For H
First ... 495  496  497 498  499  ... Last

Up
Mobile  Desktop