"Official" Audyssey thread (FAQ in post #51779) - Page 1782 - AVS Forum
Forum Jump: 
Reply
 
Thread Tools
post #53431 of 72568 Old 05-03-2012, 11:51 AM
Senior Member
 
Sam1000's Avatar
 
Join Date: Nov 2005
Posts: 485
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 16 Post(s)
Liked: 11
Quote:
Originally Posted by kbarnes701 View Post

LOL. Sam may appreciate a moment of levity amongst his current woes with Audyssey I do hope the combined experience and expertise in this great thread is able to help him solve his problem.

But what a price he paid for that 4311 eh? $800, new and boxed. If he decides to sell it and try something else, he will surely recoup every cent.

:-) I do appreciate the levity and all the help I'm getting. However, I'm a different user which paid $800 for his Denon. I paid almost double that amount for my 5508 :-(

Speakers: Pioneer/TAD S-1EX, S-7EX, Ascend HTM200
Pre/Pro: Onkyo PR-SC5508,Bel Canto Pre-6, Sherwood 972
Amp: Outlaw 770
Source:Oppo 105, Ciunas DAC, SB Touch
Sam1000 is offline  
Sponsored Links
Advertisement
 
post #53432 of 72568 Old 05-03-2012, 12:09 PM
AVS Special Member
 
mogorf's Avatar
 
Join Date: May 2009
Location: Budapest, Hungary
Posts: 4,429
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 201 Post(s)
Liked: 117
Quote:
Originally Posted by pepar View Post

Hi Jeff,
Feri asked me this on Facebook Audyssey Tech Talk.......
Best,
Chris

Cute! That only goes to show he knows all his chicken in his yard by their name, eh?

To Tandy and Sam, if I'm correct as we know both of you have a Denon-4311 AVR, so in order to follow up your queries, would it be possible for the two of you to post some more details of your systems:

1. Speaker brand and model
2. Couple of photos of your room setup
3. Speaker level trims set by Audyssey per speaker/sub
4. Distances set by Audyssey per speaker/sub

Thx.

P.S. For Sam, it's not Odyssey but Audyssey and it's not Onkyo but Aunkyo (just kidding!).
mogorf is offline  
post #53433 of 72568 Old 05-03-2012, 12:12 PM
AVS Addicted Member
 
kbarnes701's Avatar
 
Join Date: Jan 2008
Location: Main Listening Positon
Posts: 18,759
Mentioned: 6 Post(s)
Tagged: 0 Thread(s)
Quoted: 1937 Post(s)
Liked: 2001
Quote:
Originally Posted by Sam1000 View Post

:-) I do appreciate the levity and all the help I'm getting. However, I'm a different user which paid $800 for his Denon. I paid almost double that amount for my 5508 :-(

Ooops, sorry. So you are. I was so dumbstruck by the price it has affected my reason
kbarnes701 is offline  
post #53434 of 72568 Old 05-03-2012, 12:22 PM
AVS Special Member
 
AustinJerry's Avatar
 
Join Date: Jan 2003
Location: Austin, Texas
Posts: 7,567
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 763 Post(s)
Liked: 793
Quote:
Originally Posted by kbarnes701 View Post


Incidentally, I have done further tweaking of the sub distances.

Just curious, Keith, if the distances were "optimized" before, how did you get it even better?

What is the red line on the OM graph?
AustinJerry is online now  
post #53435 of 72568 Old 05-03-2012, 12:33 PM
AVS Special Member
 
mogorf's Avatar
 
Join Date: May 2009
Location: Budapest, Hungary
Posts: 4,429
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 201 Post(s)
Liked: 117
Quote:
Originally Posted by kbarnes701 View Post


Like you, I'd wish for it to ping the mains and the subs working together and then base the final decisions on those measurements, but I’d guess there is a good reason it doesn't do that. Not that I have any idea what that reason might be of course

Keith, wouldn't that mean that instead of a simple ping another full set of mic measurements around the bubble would be needed to be done for the algorithm to re-calculate the filters around the splice one by one from 5. 1 to 11.3, whatever? Just thinking out loud. Maybe a next Pro-Pro version will do.
mogorf is offline  
post #53436 of 72568 Old 05-03-2012, 12:37 PM
AVS Addicted Member
 
kbarnes701's Avatar
 
Join Date: Jan 2008
Location: Main Listening Positon
Posts: 18,759
Mentioned: 6 Post(s)
Tagged: 0 Thread(s)
Quoted: 1937 Post(s)
Liked: 2001
Quote:
Originally Posted by AustinJerry View Post

Just curious, Keith, if the distances were "optimized" before, how did you get it even better?

What is the red line on the OM graph?

The red line is the starting point, with the sub distances set by Audyssey. The blue line is after optimisation. The graph refers to measurements taken from the latest calibration run with slightly different mic positions to those used before (and which relate to the earlier graphs I posted). Sorry, AJ, I should have made that more clear.

On the previous run, I 'optimised' the distances to 12.2/12.8 Sub 1/2 respectively and the result was in the graphs posted before. I could not achieve a better result than that, as seen in the graphs.

On the latest run, with the new mic positions (heavily biased around the MLP but this time only 12 positions not 14) Audyssey set the subs to 12.8/11.2. Tweaking the distances to 12.2/12.0 gave the results in the latest graph. So the red line is with the subs set at 12.8/11.2 and the blue line with them at 12.2/12.0. Further adjustments produced no further improvement.

As soon as I get a round tuit I will make a calibration using positions very similar to your own, ignoring the other seat entirely. I am hopeful that that may smooth the curve even more, as I am thinking there may be an element of compromise at the MLP introduced by the filters created wrt to the other seat. If not, well I can go back to my current best, which is called my '1st May Calibration 12.2/12.0'. And I have remembered to make a note of the sub trims to in deference to the s/w bug we know about.
kbarnes701 is offline  
post #53437 of 72568 Old 05-03-2012, 12:45 PM
AVS Special Member
 
AustinJerry's Avatar
 
Join Date: Jan 2003
Location: Austin, Texas
Posts: 7,567
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 763 Post(s)
Liked: 793
^ Keith, I understand now, thanks. One of the downsides of the "distance tweaking" process is that the distance adjustments need to be re-checked and adjusted every time a new calibration is completed. Unfortunately, this adds quite a bit of time to an already lengthy calibration process. I guess it's the price we pay for being perfectionists.
AustinJerry is online now  
post #53438 of 72568 Old 05-03-2012, 12:46 PM
AVS Addicted Member
 
kbarnes701's Avatar
 
Join Date: Jan 2008
Location: Main Listening Positon
Posts: 18,759
Mentioned: 6 Post(s)
Tagged: 0 Thread(s)
Quoted: 1937 Post(s)
Liked: 2001
Quote:
Originally Posted by mogorf View Post

Keith, wouldn't that mean that instead of a simple ping another full set of mic measurements around the bubble would be needed to be done for the algorithm to re-calculate the filters around the splice one by one from 5. 1 to 11.3, whatever? Just thinking out loud. Maybe a next Pro-Pro version will do.

Yeah - I can never see it happening. The neat thing is that with the OM or REW mic and software, it is very easy to measure the combined output of the subs and mains and see what is really going on. In my case, this technique has made a terrific difference to my LF quality (and quantity) and, of course, it means there is no need for Audyssey to do the job (for those with independent measuring gear anyway).

My current setup is the first time I have never felt the need to raise the sub trims by 2-3 dB. I think I was doing that before to try to compensate for the phase issues I was experiencing but, at the time, didn't know I had. Since working on those, my bass is amazingly improved and, for the first time, it 'feels' right in terms of relative loudness. It would be impossible to do the 'sub distance adjustment trick' without measuring gear - you need to tweak the distances and observe the effect in real time.
kbarnes701 is offline  
post #53439 of 72568 Old 05-03-2012, 12:49 PM
AVS Addicted Member
 
kbarnes701's Avatar
 
Join Date: Jan 2008
Location: Main Listening Positon
Posts: 18,759
Mentioned: 6 Post(s)
Tagged: 0 Thread(s)
Quoted: 1937 Post(s)
Liked: 2001
Quote:
Originally Posted by AustinJerry View Post

^ Keith, I understand now, thanks. One of the downsides of the "distance tweaking" process is that the distance adjustments need to be re-checked and adjusted every time a new calibration is completed. Unfortunately, this adds quite a bit of time to an already lengthy calibration process. I guess it's the price we pay for being perfectionists.

Agreed - but the results definitely make the time spent worthwhile. I am simply astonished at the difference that small tweak has made.

I should have added above that I also intend to play around with the curve editor at some stage (thanks to Jeff's help and advice) to see if I can get an even better result. I will probably do a new calibration first using the mic positions we just discussed to see if I can get a better 'starting point'. Then tweak the curve as the icing on the cake.

The only thing that concerns me is that for two weeks now I have been spending more time measuring and tweaking than I have watching movies - but once I get it as good as I think I can get it, then that will change.
kbarnes701 is offline  
post #53440 of 72568 Old 05-03-2012, 12:53 PM
AVS Special Member
 
AustinJerry's Avatar
 
Join Date: Jan 2003
Location: Austin, Texas
Posts: 7,567
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 763 Post(s)
Liked: 793
Quote:
Originally Posted by kbarnes701 View Post


The only thing that concerns me is that for two weeks now I have been spending more time measuring and tweaking than I have watching movies - but once I get it as good as I think I can get it, then that will change.

Seems like I remember you saying this before, Keith. Face the reality--you are hooked....
AustinJerry is online now  
post #53441 of 72568 Old 05-03-2012, 12:59 PM
AVS Addicted Member
 
kbarnes701's Avatar
 
Join Date: Jan 2008
Location: Main Listening Positon
Posts: 18,759
Mentioned: 6 Post(s)
Tagged: 0 Thread(s)
Quoted: 1937 Post(s)
Liked: 2001
kbarnes701 is offline  
post #53442 of 72568 Old 05-03-2012, 01:02 PM
AVS Special Member
 
mogorf's Avatar
 
Join Date: May 2009
Location: Budapest, Hungary
Posts: 4,429
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 201 Post(s)
Liked: 117
Quote:
Originally Posted by kbarnes701 View Post

Yeah - I can never see it happening. The neat thing is that with the OM or REW mic and software, it is very easy to measure the combined output of the subs and mains and see what is really going on. In my case, this technique has made a terrific difference to my LF quality (and quantity) and, of course, it means there is no need for Audyssey to do the job (for those with independent measuring gear anyway).

So you are ahead of the next pro version even if done manually.

Quote:


It would be impossible to do the 'sub distance adjustment trick' without measuring gear - you need to tweak the distances and observe the effect in real time.

Keith, just out of curiosity when you say tweaking in real time I suppose that means only one mic position. Is that position at the MPL (I suppose)? Then if so how does that tweaking translate to the multi-point fuzzy-logic requirements of Audyssey's principles?
mogorf is offline  
post #53443 of 72568 Old 05-03-2012, 01:06 PM
AVS Special Member
 
mogorf's Avatar
 
Join Date: May 2009
Location: Budapest, Hungary
Posts: 4,429
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 201 Post(s)
Liked: 117
Quote:
Originally Posted by kbarnes701 View Post

That, Jerry, is a terrifying thought

And then you will have a next splendid two weeks to re-watch all your film archives. Got enough beer on stock in the fridge yet?
mogorf is offline  
post #53444 of 72568 Old 05-03-2012, 01:22 PM
Member
 
tandy1000rl's Avatar
 
Join Date: Oct 2005
Posts: 198
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by pepar View Post

Hi Jeff,
Feri asked me this on Facebook Audyssey Tech Talk. It could be that they have a damaged mic (humidity or static spark can kill the HF mic response resulting in a boost). Or it could be that the speakers are set to Large and so the sub is not doing anything. Without data it's hard to say.

There is nothing specific about a big room that would cause the algorithm to behave this way.

Best,
Chris

May 02, 2012 10:13 pm
User photo
pepar

Hi Chris,

After over-thinking and re-parsing Feri's question and your reply, it was thought that the question should be asked a different way.

The context for the mulligan is Pro's initial curve options based on room volume, comments regarding Audyssey Reference being a translation of X-Curve for the average residential-sized space and the user's space being considerably larger than that.

So Pro allows fine-tuning, but the absence of this in the consumer version is not a possible factor in what the user is experiencing?

Thanks,
Jeff

May 03, 2012 05:05 am
User photo
Chris Kyriakakis
Audyssey

The standard Audyssey target curve rolls off the highs so it wouldn't be causing a boost unless the starting (before) response has severely rolled off highs. That would not be due to the large room, but to the speaker or the measurement (far off axis).

Depending on how large and reverberant the room is it could certainly need more high frequency roll off to balance the response perceptually. This is something one could do with the Pro software by, for example, selecting the SPMTE 202M preset or manually adding the roll off.

However, the way it was presented to me earlier it was implied that Audyssey is actually boosting the highs because the room is so large. That is not the case unless there is something wrong with the mic.

Best,
Chris

Thanks for sharing! I do not understand his answer, though, as the logic employed seems very chicken-or-egg to me:

1. The target curve for a smaller room is "flat out to about 8 kHz, then dips to about –2 dB at 10 kHz and then a little more at 20 kHz."

Source: last post here

2. For sake of argument, lets say the recommended target curve for a larger room is flat out to 8 kHz, dips to -4 dB at 10kHz and then a little more at 20 kHz.

Source: numbers are just a guess, but representative and feasible based on visual inspection of the 3 target curves in the pro kit manual

3. In his reply above, he acknowledges a larger room "could certainly need more high frequency roll off to balance the response perceptually".

4. Thus, how does using curve 1 in a larger room not directly equal greater boost (or less cut) in the high frequencies when using Audyssey? In this example, would we not expect 2 extra dB's of 10kHz energy from the loudspeakers when using the original (small room) target curve vs using the recommended (large room) curve? It's clear that the room is not creating that extra 2 dB's...Audyssey's EQ is creating that extra energy because we are telling it to through our choice in target curve.

I agree with his first paragraph--that a room cannot boost HF on it's own accord. Likewise, I have never experienced HF boost issues in any setup with EQ bypassed. It's only when Audyssey or some other EQ is engaged does it (potentially) become an issue.

I acknowledge that my large room speakers are more polite than my small room speakers, and maybe this is playing in to the equation as well. In theory, though, if Audyssey is boosting the HF of my polite speakers to get them to an intended, optimal (i.e. non-bright) level, then it should not be offensive and bright but rather pleasing instead (like a veil has been lifted)--since they are finally at an optimal level. But the degree of EQ needed to get there is acknowledge by Chris to vary by room size/reverberation characteristics in the high frequencies. That said, is reference level minus 2 dB the correct level for 10kHz in my room or is reference level minus 4 dB right? It sure sounded a lot worse at minus 2 dB versus minus 4 dB (i.e. too bright).

Great further reading on the topic here, which I have slowly been trying to digest.

Finally, "However, the way it was presented to me earlier it was implied that Audyssey is actually boosting the highs because the room is so large. That is not the case unless there is something wrong with the mic."

Would a bad mic and/or faulty equipment not have deleterious effects regardless of the room size? I think this response misses the intended point.

"However, the way it was presented to me earlier it was implied that Audyssey is actually boosting the highs because the room is so large."
and
"There is nothing specific about a big room that would cause the algorithm to behave this way."


Well sort of. To summarize the entirety of the above:
A major goal of the Audyssey algorithm is to achieve a target frequency response. Audyssey recommends different target frequency responses based on room size to "achieve correct spectral balance". By definition, then, the algorithm intentionally creates more HF energy using curve 1 than using curve 2 or 3. Thus, using curve 1 in a large room (or a small room for that matter) will be brighter (i.e. more boost/less cut) than using curve 2 in that same room, all else being equal. How a consumer user is expected to deal with this is beyond me, since they have only 2 curves from which to work--the flat curve and the brightest "roll-off" curve. The curves intended for larger rooms, which have greater HF roll-off, are simply not there. Thus, their systems will be perceived as brighter than intended, all else equal.
tandy1000rl is offline  
post #53445 of 72568 Old 05-03-2012, 01:28 PM
AVS Special Member
 
djbluemax1's Avatar
 
Join Date: Oct 2004
Location: MI
Posts: 2,272
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 25 Post(s)
Liked: 283
Quote:
Originally Posted by kbarnes701 View Post

Now I DO have an idea what that reason might be

Thank heavens for OmniMic/REW etc then which does allow us to measure the subs and the mains playing together. It seems to be the only way to ensure that dual subs (especially) are performing at their best. I was truly shocked at my experience, described ad nauseam in the Pro thread, where my bass was emasculated so much until I discovered the phase problems and corrected them (I salute the AVS Gurus once more). I have never had bass as good as I currently do.

Incidentally, I have done further tweaking of the sub distances. Not only have I ironed out (last week) most of that huge 27dB dip that was occurring before, but look at this latest graph and notice the way that the dB level has been boosted by a few dB (and brought closer to flat at the same time) between 20 Hz and about 60 Hz (and especially between 30 Hz-55 Hz). This has resulted in a truly terrific bass output now - but at the same tight nice and controlled and tight.



That dip at about 180Hz still irks me though...

Yes, that's why I've been telling folks, to REALLY know what you're getting (unfortunately), you need to get measuring gear and learn how to use it. Otherwise, for all those folks out there, yes, they may gotten lucky and been one of the folks that truly has an optimized FR right off the bat, but you would never know and can't be sure without measuring it.

BTW, about that 180Hz dip;
IIRC, Omnimic's default is to apply 1/6 octave smoothing to the graphs (which seems to be what your graph is using as it lacks the granularity of a graph with zero smoothing applied). It lacks sufficient resolution to see what's really going on in the bass region. If you saved the measurement, you might be able to just reload it and toggle the graph to No Smoothing so you wouldn't even need to take another measurement (not sure about Omnimic, but REW has this capability). Looking at the graph with No Smoothing may show that the the Q in fact, indicates a null.

I have nulls at my MLP at ~150Hz and 190Hz as shown in my graphs with no smoothing applied below. With 1/6 octave smoothing applied, they would look like the dips in your graph.

Quote:
Originally Posted by kbarnes701 View Post

Yeah - I can never see it happening. The neat thing is that with the OM or REW mic and software, it is very easy to measure the combined output of the subs and mains and see what is really going on. In my case, this technique has made a terrific difference to my LF quality (and quantity) and, of course, it means there is no need for Audyssey to do the job (for those with independent measuring gear anyway).

My current setup is the first time I have never felt the need to raise the sub trims by 2-3 dB. I think I was doing that before to try to compensate for the phase issues I was experiencing but, at the time, didn't know I had. Since working on those, my bass is amazingly improved and, for the first time, it 'feels' right in terms of relative loudness. It would be impossible to do the 'sub distance adjustment trick' without measuring gear - you need to tweak the distances and observe the effect in real time.

Since XT32 has the ability to ping 2 sub outputs together, I wonder why they don't have it ping the subs + mains together. I would think it wouldn't be THAT difficult for Audyssey to ping Mains + Subs, tweak distances, ping again and repeat till the FR is smooth.


Max
LL
djbluemax1 is offline  
post #53446 of 72568 Old 05-03-2012, 01:32 PM
AVS Special Member
 
mogorf's Avatar
 
Join Date: May 2009
Location: Budapest, Hungary
Posts: 4,429
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 201 Post(s)
Liked: 117
Quote:
Originally Posted by tandy1000rl View Post

Thanks for sharing! I do not understand his answer, though, as the logic employed seems very chicken-or-egg to me:

Dear Tandy, me thinks you are getting a bit lost in too many details at once. Tell me what target curve would you choose for music materials other than Flat?
mogorf is offline  
post #53447 of 72568 Old 05-03-2012, 01:44 PM
AVS Special Member
 
mogorf's Avatar
 
Join Date: May 2009
Location: Budapest, Hungary
Posts: 4,429
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 201 Post(s)
Liked: 117
Quote:
Originally Posted by djbluemax1 View Post

Since XT32 has the ability to ping 2 sub outputs together, I wonder why they don't have it ping the subs + mains together. I would think it wouldn't be THAT difficult for Audyssey to ping Mains + Subs, tweak distances, ping again and repeat till the FR is smooth.


Max

Hi Max,

I think pinging two subs still means we are in the same channel, namely the sub channel, while pinging sub + speaker would mean inter-channel pinging. BTW, why stop at sub + mains and not ping all satellite channels with the sub-channel? Algorithm, future development and marketing comes to mind. Again, just thinking out loud.
mogorf is offline  
post #53448 of 72568 Old 05-03-2012, 01:56 PM
Member
 
tandy1000rl's Avatar
 
Join Date: Oct 2005
Posts: 198
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by mogorf View Post

Dear Tandy, me thinks you are getting a bit lost in too many details at once. Tell me what target curve would you choose for music materials other than Flat?

In my large room I use High Frequency Roll Off 2 in the pro kit with the following edits for both movies and music:

Hz, Db
20, 0
2000, 0
5000, -0.25
10000, -0.75
24000, -1.5

In my small room I use High Frequency Roll Off 1 with no edits.

Why do you ask?
tandy1000rl is offline  
post #53449 of 72568 Old 05-03-2012, 01:59 PM
AVS Special Member
 
AustinJerry's Avatar
 
Join Date: Jan 2003
Location: Austin, Texas
Posts: 7,567
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 763 Post(s)
Liked: 793
Quote:
Originally Posted by mogorf View Post


So you are ahead of the next pro version even if done manually.

Keith, just out of curiosity when you say tweaking in real time I suppose that means only one mic position. Is that position at the MPL (I suppose)? Then if so how does that tweaking translate to the multi-point fuzzy-logic requirements of Audyssey's principles?

Not Keith, but we have both used the sub distance-tweaking approach to smooth bass response. Yes, the approach optimizes the response at one position, the MLP. I believe we are both on record as saying that optimizing for the MLP is our objective. Keith provided some measurements showing that the bass response is not as smooth at other positions.
AustinJerry is online now  
post #53450 of 72568 Old 05-03-2012, 02:21 PM
AVS Special Member
 
mogorf's Avatar
 
Join Date: May 2009
Location: Budapest, Hungary
Posts: 4,429
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 201 Post(s)
Liked: 117
Quote:
Originally Posted by AustinJerry View Post

Not Keith, but we have both used the sub distance-tweaking approach to smooth bass response. Yes, the approach optimizes the response at one position, the MLP. I believe we are both on record as saying that optimizing for the MLP is our objective. Keith provided some measurements showing that the bass response is not as smooth at other positions.

Jerry, the objective for optimizing for the MLP is OK with me, but my question to Keith was aimed at whether doing that with one mic position will yield the expected results or a (close proximity) multi-point measurement around the MLP would be required in this case by way of averaging them in OM/REW to get a much clearer picture. Eventhough OM/REW will be able to do an RMS average only, IMHO it might still be a better solution to "imitate" what the Audyssey algorithm will do than just doing a real-time tweak at only one single postion.

Am I wrong here?
mogorf is offline  
post #53451 of 72568 Old 05-03-2012, 02:23 PM
AVS Addicted Member
 
kbarnes701's Avatar
 
Join Date: Jan 2008
Location: Main Listening Positon
Posts: 18,759
Mentioned: 6 Post(s)
Tagged: 0 Thread(s)
Quoted: 1937 Post(s)
Liked: 2001
Quote:
Originally Posted by mogorf View Post

So you are ahead of the next pro version even if done manually.

It's one way of putting it.

Quote:
Originally Posted by mogorf View Post

Keith, just out of curiosity when you say tweaking in real time I suppose that means only one mic position. Is that position at the MPL (I suppose)? Then if so how does that tweaking translate to the multi-point fuzzy-logic requirements of Audyssey's principles?

I am only interested in one seat, Feri. I don't care what the bass is like at the other seat. Audyssey uses its fuzzy logic to great effect for EQ-ing a whole room, to give a good result for multiple seats, but if you are only concerned with one seat, the best results come, as my experiments show, from measuring fairly tightly around the MLP.

I posted a graph in the Pro thread that showed just how terrible my other seat was after using this technique. But the main benefit for me was a superior result at my own seat.
kbarnes701 is offline  
post #53452 of 72568 Old 05-03-2012, 02:24 PM
AVS Addicted Member
 
kbarnes701's Avatar
 
Join Date: Jan 2008
Location: Main Listening Positon
Posts: 18,759
Mentioned: 6 Post(s)
Tagged: 0 Thread(s)
Quoted: 1937 Post(s)
Liked: 2001
Quote:
Originally Posted by mogorf View Post

And then you will have a next splendid two weeks to re-watch all your film archives. Got enough beer on stock in the fridge yet?

LOL. Rewatching excerpts is what I have been doing most this last few days!
kbarnes701 is offline  
post #53453 of 72568 Old 05-03-2012, 02:29 PM
Wireless member
 
pepar's Avatar
 
Join Date: Jul 2002
Location: Quintana Roo ... in my mind
Posts: 25,073
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 130 Post(s)
Liked: 160
tandy1000rl - I had gotten the impression that your HF problem was a lot more than a few dB that could be attributed to the effect of a larger room.

Jeff
pepar is online now  
post #53454 of 72568 Old 05-03-2012, 02:29 PM
AVS Addicted Member
 
kbarnes701's Avatar
 
Join Date: Jan 2008
Location: Main Listening Positon
Posts: 18,759
Mentioned: 6 Post(s)
Tagged: 0 Thread(s)
Quoted: 1937 Post(s)
Liked: 2001
Quote:
Originally Posted by tandy1000rl View Post

Well sort of. To summarize the entirety of the above:
A major goal of the Audyssey algorithm is to achieve a target frequency response. Audyssey recommends different target frequency responses based on room size to "achieve correct spectral balance". By definition, then, the algorithm intentionally creates more HF energy using curve 1 than using curve 2 or 3. Thus, using curve 1 in a large room (or a small room for that matter) will be brighter (i.e. more boost/less cut) than using curve 2 in that same room, all else being equal. How a consumer user is expected to deal with this is beyond me, since they have only 2 curves from which to work--the flat curve and the brightest "roll-off" curve. The curves intended for larger rooms, which have greater HF roll-off, are simply not there. Thus, their systems will be perceived as brighter than intended, all else equal.

I get the impression from the rather vague reply that Chris doesn't want to answer, for commercial reasons. To give the real answer would be to admit that the consumer version of Audyssey is optimised for an 'average' room and therefore, by Audyssey's own logic with the different Pro kit curves, is not suitable for rooms that are larger than average. That admission won't help sales. Chris is in a difficult position, wearing two hats as he has to.
kbarnes701 is offline  
post #53455 of 72568 Old 05-03-2012, 02:30 PM
AVS Special Member
 
mogorf's Avatar
 
Join Date: May 2009
Location: Budapest, Hungary
Posts: 4,429
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 201 Post(s)
Liked: 117
Quote:
Originally Posted by tandy1000rl View Post

In my large room I use High Frequency Roll Off 2 in the pro kit with the following edits for both movies and music:

Hz, Db
20, 0
2000, 0
5000, -0.25
10000, -0.75
24000, -1.5

In my small room I use High Frequency Roll Off 1 with no edits.

Why do you ask?

Tandy, a bunch of warmen are here in an alert position to try to save "private Tandy". Are you willing to provide the requested infos repeatedly ask from you or not?
mogorf is offline  
post #53456 of 72568 Old 05-03-2012, 02:32 PM
Advanced Member
 
SpotcheckBilly's Avatar
 
Join Date: Jun 2010
Location: Western Colorado
Posts: 689
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 12
Quote:
Originally Posted by mogorf View Post

And then you will have a next splendid two weeks to re-watch all your film archives. Got enough beer on stock in the fridge yet?


To paraphrase a line from a favorite movie -

"(Keith) don't need no steenkin' fridge"


Cheers,
SB
SpotcheckBilly is offline  
post #53457 of 72568 Old 05-03-2012, 02:42 PM
AVS Addicted Member
 
kbarnes701's Avatar
 
Join Date: Jan 2008
Location: Main Listening Positon
Posts: 18,759
Mentioned: 6 Post(s)
Tagged: 0 Thread(s)
Quoted: 1937 Post(s)
Liked: 2001
Quote:
Originally Posted by djbluemax1 View Post

Yes, that's why I've been telling folks, to REALLY know what you're getting (unfortunately), you need to get measuring gear and learn how to use it. Otherwise, for all those folks out there, yes, they may gotten lucky and been one of the folks that truly has an optimized FR right off the bat, but you would never know and can't be sure without measuring it.

It has certainly opened my eyes (ears).

Quote:
Originally Posted by djbluemax1 View Post

BTW, about that 180Hz dip;
IIRC, Omnimic's default is to apply 1/6 octave smoothing to the graphs (which seems to be what your graph is using as it lacks the granularity of a graph with zero smoothing applied). It lacks sufficient resolution to see what's really going on in the bass region. If you saved the measurement, you might be able to just reload it and toggle the graph to No Smoothing so you wouldn't even need to take another measurement (not sure about Omnimic, but REW has this capability). Looking at the graph with No Smoothing may show that the the Q in fact, indicates a null.

The graph above was 1/24th smoothing. I always use 1/24th. I did save one with 1/6th smoothing though and it is shown below:

kbarnes701 is offline  
post #53458 of 72568 Old 05-03-2012, 02:56 PM
Member
 
tandy1000rl's Avatar
 
Join Date: Oct 2005
Posts: 198
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by mogorf View Post

Cute! That only goes to show he knows all his chicken in his yard by their name, eh?

To Tandy and Sam, if I'm correct as we know both of you have a Denon-4311 AVR, so in order to follow up your queries, would it be possible for the two of you to post some more details of your systems:

1. Speaker brand and model
2. Couple of photos of your room setup
3. Speaker level trims set by Audyssey per speaker/sub
4. Distances set by Audyssey per speaker/sub

Thx.

P.S. For Sam, it's not Odyssey but Audyssey and it's not Onkyo but Aunkyo (just kidding!).

Let's just start with main L+R speakers and subs:

1. Krell Resolution 3's on metal stands filled with lead shot / two Sunfire True Subwoofer Signature EQs (stacked)
2. room layout attached (not to scale)
3. main L/R = +0.5 db, subs = -8 db
4. main L/R = 13 feet, subs 26.6 feet (not corrected for the bug in pro kit where each sub distance gets lost upon reloading a calibration)

Room layout and speaker locations cannot change per wife. Main L/R used to be set up using the Cardas Golden Ratio method 9' in front of french doors along back wall and MLP was at entrance of kitchen. That setup was glorious!! Current corner method...not nearly as good for stereo music.
LL
LL
tandy1000rl is offline  
post #53459 of 72568 Old 05-03-2012, 02:58 PM
Member
 
tandy1000rl's Avatar
 
Join Date: Oct 2005
Posts: 198
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by mogorf View Post

Tandy, a bunch of warmen are here in an alert position to try to save "private Tandy". Are you willing to provide the requested infos repeatedly ask from you or not?

;-) Just posted!
tandy1000rl is offline  
post #53460 of 72568 Old 05-03-2012, 03:04 PM
AVS Addicted Member
 
kbarnes701's Avatar
 
Join Date: Jan 2008
Location: Main Listening Positon
Posts: 18,759
Mentioned: 6 Post(s)
Tagged: 0 Thread(s)
Quoted: 1937 Post(s)
Liked: 2001
Quote:
Originally Posted by mogorf View Post

Jerry, the objective for optimizing for the MLP is OK with me, but my question to Keith was aimed at whether doing that with one mic position will yield the expected results or a (close proximity) multi-point measurement around the MLP would be required in this case by way of averaging them in OM/REW to get a much clearer picture. Eventhough OM/REW will be able to do an RMS average only, IMHO it might still be a better solution to "imitate" what the Audyssey algorithm will do than just doing a real-time tweak at only one single postion.

Am I wrong here?

I think you are missing my point. I have done the Audyssey (Pro) calibration and been given the best that Audyssey can do in my room. That is Stage 1.

My independent measurements show that Audyssey has done a good job but it is far from perfect. I posted graphs in the Pro thread to show this. I will not re-post them here as it is OT for the Audyssey thread.

Then I moved on to Stage 2. This stage involves taking independent measurements from the one and only position that matters to me: the MLP. I do not need to take several measurements all around the MLP, as I did with Audyssey, because when listening I do not move around in my seat by as much as 2 feet at a time. My independent measurements show the problems, at the MLP, that Audyssey has failed to correct. But this doesn't matter because I can do this 'fine tuning' stage myself by using the 'sub distance adjustment' method of phase realignment. I can do this in real time and observe the changes to the FR. I posted a graph above which shows the improvement made most recently to my FR. In the Pro thread I posted graphs that showed the initial enormous dip in FR that Audyssey failed to correct. This was a phase issue in Sub 2, which I corrected myself.

The result is this: Audyssey does a good job but leaves room for improvement. You can only see this room for improvement if you have independent measuring gear. Once you have seen (and/or heard - it was my ears that first alerted me to a problem and the measurements confirmed what I was hearing) the problem, it can be corrected as I have described.

The graph most recently posted above shows a much improved LF response and the final tweaking of the sub distance shows even more. This is confirmed by my listening tests. What more is there to say?
kbarnes701 is offline  
Reply Receivers, Amps, and Processors

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
Gear in this thread - Kht1005 by PriceGrabber.com

Thread Tools
Show Printable Version Show Printable Version
Email this Page Email this Page


Forum Jump: 

Posting Rules  
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off