or Connect
AVS › AVS Forum › Audio › Receivers, Amps, and Processors › "Official" Audyssey thread (FAQ in post #51779)
New Posts  All Forums:Forum Nav:

"Official" Audyssey thread (FAQ in post #51779) - Page 2303

post #69061 of 70886
Quote:
Originally Posted by batpig View Post

I know you are being facetious but XT and regular MultEQ have very high resolution subwoofer filters. Only 4x less than XT32. It's the bass in the satellite channels where the most drastic difference lies.

In the midst of this lets not forget that you can still have a wonderful sounding system with lesser versions of MultEQ. Obviously if you are buying new the only reason not to get XT32 would be budgetary as Chris K notes.

Concerning the resolution. XT does ~11ms of correction with a FIR filter. This translates to ~91 Hz of the frequency response resolution (this could be "simplified" for understanding as a minimal distance between two peaks or two dips on the pre-out graph). So, if we are looking at it in the logarithmic scale there are much more resolution per octave for high frequencies than for low frequencies. This is the nature of FIR filter.

Then when we go to XT32 resolution is 32 times higher, so it is less than 3 Hz. For all channels. And the subwoofer channel of XT has 4 times less or ~11 Hz resolution.

Look at the graphs posted earlier and you will see pretty much that!
post #69062 of 70886
Quote:
Originally Posted by pbarach View Post
 
Quote:
Originally Posted by kbarnes701 View Post

When Audyssey made XT correct the high frequencies the way it does - where far less correction is needed and may actually make the result worse (or as Roger says "do harm") - and correct the low frequencies hardly at all - which is where correction is actually needed, I’d say this implementation of a room EQ system was fundamentally flawed. XT is doing the exact opposite of what a well-designed room EQ system should be doing.

I saw your graphs, and I'm trying to reconcile them with my own experience of using as AS-EQ1 (same resolution on the subwoofer channel as XT32) and then finding how much better my system sounds with Multi EQ XT turned on at the same time.

 

The AS-EQ1 is a terrific upgrade and brings XT32 functionality to the bass - I had the same unit when I had XT in my 5007. The system does indeed sound so much better when the AS-EQ1 is in the loop.

 

I assume you mean that you have tried playing content with the EQ1 in the loop but with and without XT and have not noticed any 'contamination' in the HF?  I suspect that if you now move to an XT32 unit, you would hear a marked difference in the higher frequencies.  The graphs are undeniable and show all that 'hair' in the upper frequencies and if you read posts from the time of the original posting of rickardl's graphs, there is further information there. Also, Igor has done a lot of work in this area too but I am not sure if his own results were published in this thread or in the Onkyo 818 thread - I am sure he will remember though.

post #69063 of 70886
Quote:
Originally Posted by toofast68 View Post
 
Quote:
Originally Posted by kbarnes701 View Post
 

 

Or maybe it's like XT, and doesn't EQ the sub channel very well at all. LOL.


So if have XT, and can't afford to buy a new AVR just for XT32, what does I do ?  See my setup link at the bottom - decent mid range setup, but nothing sure high end.

 

Beeman has given some excellent suggestions IMO. You could also try to find an AS-EQ1 on the secondhand market. This will bring about a superb improvement (I testify from personal experience - at the time that was the single most significant upgrade I had ever made). It effectively gives you XT32 in the bass, which is where you need it most. You could also consider adding some room treatments, if possible, and then just turning Audyssey off and relying on the EQ1 and the room treatments. Or you could use REW to analyse the room and create and upload correction filters to a MiniDSP 10x10.  Audyssey isn't the only way to deal with room-induced problems. It's perhaps one of the easiest ways, but there are many other possibilities.

 

I can see that all this talk of the flaws and deficiencies in XT can be somewhat soul-destroying for people who have XT. But surely, it is better to learn of the flaws and deficiencies and at least be an informed consumer?  For the sensible, like you, (as opposed to those in permanent denial) it causes them to question what they have and whether they can do better and I would have thought this was a good thing, not a bad thing.

 

And if it is impossible to do anything at all, for financial or WAF or any other reason, well, at least you are informed and one day may be able to take the next step. We are almost all in this position one way or another - few, if any, of us have 'perfect' systems. I, for example, labor under the constraints of a very difficult and unsuitable room and I know that my SQ would likely take a step up if I had a bigger room - but because that is not currently possible, I have to do the best I can with what I have. That doesn’t mean I bury my head in the sand with regard to the flaws and deficiencies - it means I understand what they are, try to mitigate them in any way open to me, and hope that what I learn may one day be applied to a better room.  Meanwhile, I enjoy immensely what I have, while at the same time, acknowledging that there is better.

post #69064 of 70886
Quote:
Originally Posted by batpig View Post
 
Quote:
Originally Posted by toofast68 View Post
 
Quote:
Originally Posted by kbarnes701 View Post

 

Or maybe it's like XT, and doesn't EQ the sub channel very well at all. LOL.


So if have XT, and can't afford to buy a new AVR just for XT32, what does I do ?  See my setup link at the bottom - decent mid range setup, but nothing sure high end.

You also have the option of doing nothing and enjoying your system if you don't actually find any audible flaws. The quote above isn't really accurate either IMO, XT does a fine job with the sub; as discussed above the biggest gains are in the other channels where XT32 amps up the filter resolution by 32x (thus the name) and also distributes it's resources more effectively.

Don't let a pedantic discussion by bleeding edge critical listeners convince you that someone is wrong with your system if you are currently happy with it.

So the process is really backwards -- first YOU need to define what you think is wrong, if anything. Don't go looking for problems just because you heard about a solution.

If you are intrepid and looking for cheap improvements, nothing will yield more bang for the buck than dropping $100 on a USB mic and learning to use free REW measurement software to optimize your room acoustics.

 

I agree with this post, batpig, FWIW.  Your points are eloquently made, especially that last one about room acoustics. If someone can get their room acoustics right, then the need for Audyssey to do very much is greatly diminished anyway.

 

I'd say my room currently sounds better with Audyssey switched OFF than it did, when untreated, with Audyssey switched ON. But it sounds best of all treated and with Audyssey ON.

post #69065 of 70886
Quote:
Originally Posted by kbarnes701 View Post

I suspect that if you now move to an XT32 unit, you would hear a marked difference in the higher frequencies.  The graphs are undeniable and show all that 'hair' in the upper frequencies

When I have the bread, I'll get rid of the hair wink.gif
post #69066 of 70886
Quote:
Originally Posted by sdrucker View Post
 
Quote:
Originally Posted by batpig View Post

You also have the option of doing nothing and enjoying your system if you don't actually find any audible flaws. The quote above isn't really accurate either IMO, XT does a fine job with the sub; as discussed above the biggest gains are in the other channels where XT32 amps up the filter resolution by 32x (thus the name) and also distributes it's resources more effectively.

Don't let a pedantic discussion by bleeding edge critical listeners convince you that someone is wrong with your system if you are currently happy with it.

So the process is really backwards -- first YOU need to define what you think is wrong, if anything. Don't go looking for problems just because you heard about a solution.

If you are intrepid and looking for cheap improvements, nothing will yield more bang for the buck than dropping $100 on a USB mic and learning to use free REW measurement software to optimize your room acoustics.

True about all of that - downloading REW, picking up a USB mic, and exploring the REW guide linked to KBarnes and AustinJerry's signatures is the single most useful thing that anybody seeking to do more than "set and forget" their Audyssey setup can do, IMO. You'll learn how to improve your sub and speaker placements and compare the results of your calibration to a non-calibrated baseline. Or compare the results of different calibrations, with proper pre/post protocols and attention to detail.

But to a broader point, BP, you've hit the fault line between two extremes of AVSers that read this thread.

On one hand, there's the people that are simply asking technical questions or basic questions about their AVRs, how they handle multiple subs, mic positions etc. Most if not all of these can be covered in the Audyssey guides and your own Denon setup guides to an extent, but we all have our individual journeys we want to discuss. And many of these folks, if not the lurkers and/or former readers of the thread, can stop once they get a calibration they're happy with, in whatever flavor of Audyssey they work with. They may participate here at times once they've hit "success", but they can walk away from the table and enjoy their systems, whether it's XT or XT32, or a hybrid with other tools like an AntiMode or MiniDSP. And the simplest reflection of "walk away" isn't sexy, it's silence. And posting elsewhere more than here.

Then there's the rest of us, which ranges from casual hobbyists trying to do more with a multipurpose room, to people that have decades of hands-on experience with audio, and/or have professional specialties that are technical, and expect a level of rigor which is publication or at least peer-review conference quality. And those that would like to emulate them. I'm sure the identities of the latter are obvious to regular readers of this thread.

For these folks on the bleeding edge or that are methodologists by inclination, there never IS a stopping point,, and a product like Audyssey is a work in progress because there's no such thing as a completed, perfect product. To these guys, once a 2nd generation product (XT32) is available on top of the original (XT and earlier versions), and the original can be demonstratively shown to produce harmful effects on a graph, the die is cast. And anything less than that is not only flawed, but to defend it, or not show publication-level rigor at an extreme, is a lie and warrants thinly disguised ridicule for those that won't take the next step - which involves spending money usually - or that hold differing opinions. And those with too much time on their hands arguing either way LOL... smile.gif

I will say that the effective peer pressure aspect of a thread like this can be dangerous. But on the other hand, as a group we couldn't make progress without a certain degree of that to push some of us to educate ourselves. But keep in mind that a lot of mistakes and missteps can be par for the course. It's up to you if you want to take that chance. However, since this is A/V Science, not a Facebook page for the fans of Audyssey smile.gif, we can't dismiss either.

And you have to admit that those XT vs. XT32 graphs CAN be convincing.

 

Another exceptionally eloquent post IMO. And again, I agree, FWIW, with all the points you make. Even if the knowledge gained from the journey hacks me off in many ways, I still prefer to be an informed traveller.

 

And I don't think anyone in 'Group A' is not served well when they come here with a setup question etc about Audyssey. I have never seen a question of that sort go unanswered here, even in the midst of one of the more esoteric discussions. Personally, I think that the presence of the two 'Groups' is what makes this thread unique and is one of the reasons why it is the biggest on AVS by far.

post #69067 of 70886
Quote:
Originally Posted by BeeMan458 View Post
 
Quote:
Originally Posted by sdrucker View Post

On one hand, there's the people that are simply asking technical questions or basic questions about their AVRs, how they handle multiple subs, mic positions etc. Most if not all of these can be covered in the Audyssey guides and your own Denon setup guides to an extent, but we all have our individual journeys we want to discuss. And many of these folks, if not the lurkers and/or former readers of the thread, can stop once they get a calibration they're happy with, in whatever flavor of Audyssey they work with. They may participate here at times once they've hit "success", but they can walk away from the table and enjoy their systems, whether it's XT or XT32, or a hybrid with other tools like an AntiMode or MiniDSP. And the simplest reflection of "walk away" isn't sexy, it's silence. And posting elsewhere more than here.

The basic group of users that you describe, needs to trot on over to the subwoofer forum. Lots of good guys and setup/subwoofer information to be had. And we're happy to help people spend their money, too. tongue.gif

 

Which forum is that Beeman? I am always interested in learning more about subs....

post #69068 of 70886
Quote:
Originally Posted by pbarach View Post
 
Quote:
Originally Posted by kbarnes701 View Post

I suspect that if you now move to an XT32 unit, you would hear a marked difference in the higher frequencies.  The graphs are undeniable and show all that 'hair' in the upper frequencies

When I have the bread, I'll get rid of the hair wink.gif

 

And that is an excellent response IMO. But at least you now know about the 'hair' and know what problems it reveals and know that you may one day do something about it. If, as some would seem to prefer, the thread never went into all of this in detail, then none of us would know that there was better around the corner. Ignorance may be bliss, but it is still ignorance :)

 

I also think that all of this discussion of the relative merits of XT and XT32 might serve those who are currently only contemplating Audyssey perhaps for the first time, or who are considering replacing their AVR perhaps for reasons other than its room EQ capabilities.  Now they are much more 'informed consumers', they may decide they do not want an XT unit at all and would rather wait until they can get their hands on an XT32 unit. They may even re-prioritize their upgrade path because of what they have learned. And there are very good XT32 units available now for less than the $1,000 mark - something which brings XT32 within reach of a much bigger potential audience. All of this seems like a good thing, to me. Far better than 'buying blind' in the belief that perhaps there isn't all that much difference between the two Audyssey flavors.

 

When even Chris Kyriakakis, Audyssey's CTO, says he couldn’t honestly recommend any other flavor of MultEQ than XT32, other than for reasons of simple finance, then I think that really tells us all we need to know.

post #69069 of 70886
Quote:
Originally Posted by mogorf View Post

Actually it is the AVR that will set a speaker to full band if -3 dB point measured and reported by Audyssey is below 40 Hz (even if it's only 39 Hz). But still and even with a little room gain/boundary gain the eq'ing of XT down to 10 Hz is doubtlul. IMHO. On the other hand the XT32 curve shows some signs of a roll-off + the infamous normalization boost.

Well, in my setup the Audyssey Pro Kit suggests full range as well, so it is not just an AVR issue.

As for the EQ on the specific graph you mention, I don't see any sub 20Hz filtering, but just a constant gain with very small fluctuations.
post #69070 of 70886
Keith,

Maybe it's my eyes, but your red curve (Audyssey Flat XT32) seems to show quite a bit of high frequency boost, especially at the very top. The biggest difference between that and the XT curve at the top end seems to be that the boost provided by XT32 is much smoother (less "hairy"). In a way, that smoothness is a surprise to me. I would expect the higher resolution compensating curve of XT32 to be pretty hairy, since both XT and XT32 are trying to make up for room anomalies, and most unsmoothed room/speaker combinations (based on the curves I've seen) are pretty jagged, even at the top, so I'd expect the mirror image compensation curve right out of the pre-out to be jagged as well ... unless the multiple mic positions of Audyssey and their "fuzzy logic" tend to smooth this out ... but I only have two ears, and they have many more mic positions. Still confused. If I were to go over to XT32 someday, should I cluster the mics around the eight ear positions of myself and the two usual guests?
Edited by garygarrison - 1/19/14 at 6:10pm
post #69071 of 70886
xtvsxt32.jpeg 64k .jpeg file
Quote:
Originally Posted by garygarrison View Post

Keith,

Maybe it's my eyes, but your red curve (Audyssey Flat XT32) seems to show quite a bit of high frequency boost, especially at the very top. The biggest difference between that and the XT curve at the top end seems to be that the boost provided by XT32 is much smoother (less "hairy"). In a way, that smoothness is a surprise to me. I would expect the higher resolution compensating curve of XT32 to be pretty hairy, since both XT and XT32 are trying to make up for room anomalies, and most unsmoothed room/speaker combinations (based on the curves I've seen) are pretty jagged, even at the top, so I'd expect the mirror image compensation curve right out of the pre-out to be jagged as well ...
Actually, the less jagged curve with XT32 is very good! The jagged XT curve was doing more harm than good at the higher frequencies...

Quote:
Originally Posted by Roger Dressler View Post

Bingo! XT32 is much less busy >1 kHz where it is not needed and even potentially harmful, and much busier <1 kHz where it is really needed. Two huge improvements over XT.
Quote:
Originally Posted by sdurani View Post

... Low frequencies are not only where more correction is needed, but also where corrections are most effective. In fact, I know a couple of people who don't bother EQing above Schroder [frequency]. Looking at the measurements rickardl posted makes me feel that Audyssey finally (after all these years) got it right with xt32.

Edited by rickardl - 1/20/14 at 12:44am
post #69072 of 70886
Quote:
Originally Posted by garygarrison View Post

Keith,

Maybe it's my eyes, but your red curve (Audyssey Flat XT32) seems to show quite a bit of high frequency boost, especially at the very top. The biggest difference between that and the XT curve at the top end seems to be that the boost provided by XT32 is much smoother (less "hairy"). In a way, that smoothness is a surprise to me. I would expect the higher resolution compensating curve of XT32 to be pretty hairy, since both XT and XT32 are trying to make up for room anomalies, and most unsmoothed room/speaker combinations (based on the curves I've seen) are pretty jagged, even at the top, so I'd expect the mirror image compensation curve right out of the pre-out to be jagged as well ... unless the multiple mic positions of Audyssey and their "fuzzy logic" tend to smooth this out ... but I only have two ears, and they have many more mic positions. 

 

It's the 'hair' in the XT trace that is the problem. It is doing way too much to the higher frequencies and subsequent listening tests by experienced members (eg Igor) have verified aurally what the graph us showing.  Don't confuse the pre-out traces with the in-room measurements. Almost all graphs you see are in-room and they will show substantial 'jaggies' at the HF end if no smoothing is used, usually from comb filtering which has been induced by the room.  If you look at the XT32 preout trace you can see that most if its efforts are concentrated at the bass end of the spectrum, as it should be, with far less correction being applied at the top end. XT does the reverse, which is why I described it (controversially it seems) as "flawed".

 

Quote:

 Still confused. If I were to go over to XT32 someday, should I cluster the mics around the eight ear positions of myself and the two usual guests?

 

Cluster around the most important areas you wish to correct for. If you are only interested in one seat, cluster around that one seat. If you are interested in several seats then do a broader mic position pattern. The mic patterns aren't really any different with XT than with XT32.

 

d)3.   Where should I position the mic for best results?

post #69073 of 70886
Quote:
Originally Posted by garygarrison View Post

Keith,

Maybe it's my eyes, but your red curve (Audyssey Flat XT32) seems to show quite a bit of high frequency boost, especially at the very top. The biggest difference between that and the XT curve at the top end seems to be that the boost provided by XT32 is much smoother (less "hairy"). In a way, that smoothness is a surprise to me. I would expect the higher resolution compensating curve of XT32 to be pretty hairy, since both XT and XT32 are trying to make up for room anomalies, and most unsmoothed room/speaker combinations (based on the curves I've seen) are pretty jagged, even at the top, so I'd expect the mirror image compensation curve right out of the pre-out to be jagged as well ... unless the multiple mic positions of Audyssey and their "fuzzy logic" tend to smooth this out ... but I only have two ears, and they have many more mic positions. Still confused. If I were to go over to XT32 someday, should I cluster the mics around the eight ear positions of myself and the two usual guests?
In order to properly grasp the concept, you need to better understand what is going on.

1) Yes, an in-room measured response at a particular position will show a lot of 'hair' in the higher frequencies. This is due to constructive and destructive interference from the reflected acoustuc waves arriving at the mic at the different frequencies. Some waves will interfere constructively (peak), some destructively (dip).

At these higher frequencies, wavelengths get very short. For a specific kHz, a peak at one measuring spot may change to a dip 1" away. Therefore, electronically correcting the peak at that one measurement location is pointless, as the results will be off at any other location.

2) The above is why it's not only pointless to apply such fine corrections in the higher frequencies, it's possible you would produce even worse results by doing so. Every flavor of Audyssey below XT32 does it though.

3) XT32 OTOH, uses a more practical and elegant approach. Instead of applying corrections to fine details (which as described above is either futile, or potentially detrimental), XT32 DOES still apply corrections above Schroeder, but in broader strokes, i.e. if the speaker's measured response shows a dip of 5db from 2kHz to 5kHz, XT32 will try to boost it to flatten the dip. Likewise, if the measured response shows a 4db rise from 8kHz to 16kHz, XT32 attempts to squash it flat (or to fit the standard Audyssey curve with roll off).

Unlike the previous iterations, XT32 does NOT attempt to correct the peak at 9,876Hz, the dip at 9,882Hz, the peak at 9,897Hz etc., which is the way it should be.

You'll also note that in the pre-out graphs of XT vs XT32, XT32 shows much finer corrections in the lowest octaves vs the much cruder broader strokes of XT in the lowest octaves (where correction is actually useful). In other words, the allocation of resources for correction in XT is completely backward to what is optimal. Where you SHOULD have the finest granularity for correction in the lowest octaves, then taper the granularity at higher frequencies, applying only general broad strokes to produce a flatter response, XT has the crudest corrections in the lowest octaves, with granularity increasing with frequency which is the opposite of optimal. As Roger's comment points out, it took them till XT32 to get this right.



Max
post #69074 of 70886
Quote:
Originally Posted by djbluemax1 View Post
 
Quote:
Originally Posted by garygarrison View Post

Keith,

Maybe it's my eyes, but your red curve (Audyssey Flat XT32) seems to show quite a bit of high frequency boost, especially at the very top. The biggest difference between that and the XT curve at the top end seems to be that the boost provided by XT32 is much smoother (less "hairy"). In a way, that smoothness is a surprise to me. I would expect the higher resolution compensating curve of XT32 to be pretty hairy, since both XT and XT32 are trying to make up for room anomalies, and most unsmoothed room/speaker combinations (based on the curves I've seen) are pretty jagged, even at the top, so I'd expect the mirror image compensation curve right out of the pre-out to be jagged as well ... unless the multiple mic positions of Audyssey and their "fuzzy logic" tend to smooth this out ... but I only have two ears, and they have many more mic positions. Still confused. If I were to go over to XT32 someday, should I cluster the mics around the eight ear positions of myself and the two usual guests?
In order to properly grasp the concept, you need to better understand what is going on.

1) Yes, an in-room measured response at a particular position will show a lot of 'hair' in the higher frequencies. This is due to constructive and destructive interference from the reflected acoustuc waves arriving at the mic at the different frequencies. Some waves will interfere constructively (peak), some destructively (dip).

At these higher frequencies, wavelengths get very short. For a specific kHz, a peak at one measuring spot may change to a dip 1" away. Therefore, electronically correcting the peak at that one measurement location is pointless, as the results will be off at any other location.

2) The above is why it's not only pointless to apply such fine corrections in the higher frequencies, it's possible you would produce even worse results by doing so. Every flavor of Audyssey below XT32 does it though.

3) XT32 OTOH, uses a more practical and elegant approach. Instead of applying corrections to fine details (which as described above is either futile, or potentially detrimental), XT32 DOES still apply corrections above Schroeder, but in broader strokes, i.e. if the speaker's measured response shows a dip of 5db from 2kHz to 5kHz, XT32 will try to boost it to flatten the dip. Likewise, if the measured response shows a 4db rise from 8kHz to 16kHz, XT32 attempts to squash it flat (or to fit the standard Audyssey curve with roll off).

Unlike the previous iterations, XT32 does NOT attempt to correct the peak at 9,876Hz, the dip at 9,882Hz, the peak at 9,897Hz etc., which is the way it should be.

You'll also note that in the pre-out graphs of XT vs XT32, XT32 shows much finer corrections in the lowest octaves vs the much cruder broader strokes of XT in the lowest octaves (where correction is actually useful). In other words, the allocation of resources for correction in XT is completely backward to what is optimal. Where you SHOULD have the finest granularity for correction in the lowest octaves, then taper the granularity at higher frequencies, applying only general broad strokes to produce a flatter response, XT has the crudest corrections in the lowest octaves, with granularity increasing with frequency which is the opposite of optimal. As Roger's comment points out, it took them till XT32 to get this right.



Max

 

A superb explanation there! I think I may add this as a question to the FAQ as it has come up now a lot since the phenomenon was initially discovered. Currently the FAQ explains the differences between XT and XT32 as simply differences in resolution, but as we now know, there is more to it than that, with XT and XT32 using entirely different 'philosophies'. I think it would help people make a more informed choice if the explanation you give above, supported by rickardl's graphs, were included in the FAQ. It may also mean that when it comes up again in the future, as I am sure it will, we will have a simple reference to point to for the explanation.

 

Max and rickard - with your permission I will use the text/graphs from your posts (fully credited to you both of course).

post #69075 of 70886
Quote:
Originally Posted by kbarnes701 View Post

A superb explanation there! I think I may add this as a question to the FAQ as it has come up now a lot since the phenomenon was initially discovered. Currently the FAQ explains the differences between XT and XT32 as simply differences in resolution, but as we now know, there is more to it than that, with XT and XT32 using entirely different 'philosophies'. I think it would help people make a more informed choice if the explanation you give above, supported by rickardl's graphs, were included in the FAQ. It may also mean that when it comes up again in the future, as I am sure it will, we will have a simple reference to point to for the explanation.

Max and rickard - with your permission I will use the text/graphs from your posts (fully credited to you both of course).

As an addendum prior to adding it to the FAQ I would heartfully recommend to workout a kind of test procedure to verfy what is seen on the graps by our ears. Igor has already come up with a solution which I tried based on his guidance, but unfortunaly this topic was not followed up in more details by other members here. "Hearing is believing" that comes to mind.

smile.gif
post #69076 of 70886
Quote:
Originally Posted by kbarnes701 View Post

Max and rickard - with your permission I will use the text/graphs from your posts (fully credited to you both of course).
sure, np. And please look at my orignal posts for all graphs I included to see which to use:
http://www.avsforum.com/t/795421/official-audyssey-thread-faq-in-post-51779/42360#post_20807989
http://www.avsforum.com/t/795421/official-audyssey-thread-faq-in-post-51779/42360#post_20812908
http://www.avsforum.com/t/795421/official-audyssey-thread-faq-in-post-51779/42360#post_20812944
post #69077 of 70886
Keith Barnes wrote: "
It's the 'hair' in the XT trace that is the problem. It is doing way too much to the higher frequencies and subsequent listening tests by experienced members (eg Igor) have verified aurally what the graph us showing. Don't confuse the pre-out traces with the in-room measurements. Almost all graphs you see are in-room and they will show substantial 'jaggies' at the HF end if no smoothing is used, usually from comb filtering which has been induced by the room. If you look at the XT32 preout trace you can see that most if its efforts are concentrated at the bass end of the spectrum, as it should be, with far less correction being applied at the top end. XT does the reverse, which is why I described it (controversially it seems) as "flawed."

So if it's true that XT can do more harm than good in frequencies above Schroeder, then I would be best off using my SVS-AS-EQ1 unit (which does the same thing as XT32 for lower frequencies) and keeping Audyssey XT shut off??
post #69078 of 70886
Quote:
Originally Posted by pbarach View Post

would be best off using my SVS-AS-EQ1 unit (which does the same thing as XT32 for lower frequencies) and keeping Audyssey XT shut off??

That would be nonsense.
post #69079 of 70886
Quote:
Originally Posted by Gary J View Post

That would be nonsense.

OK, why would it be nonsense if, as some have asserted here, XT does more harm than good with the high frequencies?
post #69080 of 70886
Whether or not disabling Audyssey is "best" depends on your room's acoustics and your speakers. The audible improvements provided by any grade of Audyssey usually outweigh their individual defects. In most cases, substantial work has to be applied to improving a room's acoustical response (with absorbers and diffusers) and in upgrading speakers in order to match what Audyssey can do.
post #69081 of 70886
Quote:
Originally Posted by pbarach View Post

So if it's true that XT can do more harm than good in frequencies above Schroeder, then I would be best off using my SVS-AS-EQ1 unit (which does the same thing as XT32 for lower frequencies) and keeping Audyssey XT shut off??

Why don't you also ask the guys here to recommend a true listening test in order to verify what the graphs show for XT. Just like I did a few posts above. smile.gif And then you can decide for yourself what to do next. smile.gif
post #69082 of 70886
Quote:
Originally Posted by pbarach View Post

OK, why would it be nonsense if, as some have asserted here, XT does more harm than good with the high frequencies?

The body of knowledge here over the YEARS trumps "some have asserted".
post #69083 of 70886
Strange, this is the reply from Onkyo concerning my question about Sub EQ HT:

Alleen de topmodellen TX-NR5010 en PR-SC5509 ondersteunen Sub EQ HT.

In English: Only the top models TX-NR5010 and PR-SC5509 support Sub EQ HT.

Think Onkyo is not very sure about on what model Sub EQ HT is active........ :-(
post #69084 of 70886
Quote:
Originally Posted by pbarach View Post

OK, why would it be nonsense if, as some have asserted here, XT does more harm than good with the high frequencies?

"XT does more harm than good" is becoming a mountain very quickly in this thread, when it is nothing more than an ant hill. confused.gif

Try this, calibrate your system with Audyssey the best way you can, based on some of the helpful calibration tips in this thread. Then, sit back and listen. Try DEQ on and off. Try different DEQ RLO settings. Maybe experiment with lowering or raising the surround speaker levels a db or two. Once you get Audyssey XT dialed in to what YOU like, then turn it off. If you like that better, then run with it off. If you don't, then run with it on............. It's simple and free.
post #69085 of 70886
Quote:
Originally Posted by pbarach View Post

So if it's true that XT can do more harm than good in frequencies above Schroeder, then I would be best off using my SVS-AS-EQ1 unit (which does the same thing as XT32 for lower frequencies) and keeping Audyssey XT shut off??

I would never do anything based on Internet theory without first validating the theory subjectively with YOUR ears in YOUR room. Your ears should be the final judge. Don't throw the baby out with the bathwater because someone posted a graph or two on an Internet forum.

Play some music with lots of high freq energy, especially material with lots of air, string plucks, cymbals, etc. Then A/B your test material with Audyssey on vs off and see if you can hear any negative impact on the high freqs, and if you can does that distortion outweigh the other benefits? This test really needs to be blind to remove expectation bias so you may need an assistant (if you have a D/M product it's easy to do yourself with the Direct/Stereo mode toggle button).

The only thing that matters is if YOU think that XT is making things better or worse. The danger of some of this drum beating is that people seek out problems where there are none, and begin to assume at its likely or worse certain that XT is doing "more harm than good". We have a theory backed with graphs about why it MIGHT do more harm than good at the high frequencies, but you need to validate this with listening tests before taking any action.
post #69086 of 70886
Quote:
Originally Posted by pbarach View Post

Quote:
Originally Posted by Gary J View Post

That would be nonsense.

OK, why would it be nonsense if, as some have asserted here, XT does more harm than good with the high frequencies?

It would be nonsense to do something based on an internet theory without first validating the prediction yourself. Although more dickishly made, I agree with Gary's point here.

Edit: and D Bone above as well.
post #69087 of 70886
Quote:
Originally Posted by batpig View Post

It would be nonsense to do something based on an internet theory without first validating the prediction yourself. Although more dickishly made, I agree with Gary's point here.

Edit: and D Bone above as well.

+1. Oh, and the listening tests suggested by many should be done by everyone interested and should be reported back here, (please),...err,...before even thinking of amending the FAQ. smile.gif
post #69088 of 70886
Quote:
Originally Posted by mogorf View Post
+1. Oh, and the listening tests suggested by many should be done by everyone interested and should be reported back here, (please),...err,...before even thinking of amending the FAQ. smile.gif

 

Not necessarily.

 

In my opinion it would be fair to add the evidence (graphs and explanation of the behavior) to the FAQ, but I strongly suggest leaving out the "flawed, inadequate, wrong, unlistenable, useless, whatever" wordings that has been used by several people in this thread.

 

The FAQ should in my opinion be fact based and leave out any personal opinions.

post #69089 of 70886
Quote:
Originally Posted by jjazdk View Post

Not necessarily.

In my opinion it would be fair to add the evidence (graphs and explanation of the behavior) to the FAQ, but I strongly suggest leaving out the "flawed, inadequate, wrong, unlistenable, useless, whatever" wordings that has been used by several people in this thread.

The FAQ should in my opinion be fact based and leave out any personal opinions.

I disagree completely that it even needs to be in the FAQ. Audyssey makes several versions of Audyssey, with each level supposedly better than the previous, and the cost goes up with each level.......That should be in the FAQ as a new user may not know there are different levels of Audyssey, and that gives them the info they need to purchase the level of Audyssey that fits within their budget. However, including fancy graphs with the general labeling that, "anything below XT32 does more harm than good" is in my opinion, irresponsible and flat out wrong.

BMW sells many levels of cars. Does driving a cheaper 3 series "do more harm than good" when compared to the more expensive 7 series? rolleyes.gif
post #69090 of 70886
Quote:
Originally Posted by D Bone View Post

I disagree completely that it even needs to be in the FAQ. Audyssey makes several versions of Audyssey, with each level supposedly better than the previous, and the cost goes up with each level.......That should be in the FAQ as a new user may not know there are different levels of Audyssey, and that gives them the info they need to purchase the level of Audyssey that fits within their budget. However, including fancy graphs with the general labeling that, "anything below XT32 does more harm than good" is in my opinion, irresponsible and flat out wrong.

BMW sells many levels of cars. Does driving a cheaper 3 series "do more harm than good" when compared to the more expensive 7 series? rolleyes.gif

Did you actually read my post before you answered???

I explicitly stated that such personal statements should NOT be included in the FAQ, and that only facts should be included.

Now why is it that you do not want factual and objective differences between various Audyssey versions to be in the FAQ?
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Receivers, Amps, and Processors
AVS › AVS Forum › Audio › Receivers, Amps, and Processors › "Official" Audyssey thread (FAQ in post #51779)