Debate Thread: Scott's Hi-res Audio Test - Page 82 - AVS Forum
Forum Jump: 
 
Thread Tools
Old 07-26-2014, 09:04 AM - Thread Starter
AVS Addicted Member
 
amirm's Avatar
 
Join Date: Jan 2002
Location: Washington State
Posts: 18,375
Mentioned: 4 Post(s)
Tagged: 0 Thread(s)
Quoted: 981 Post(s)
Liked: 547
Quote:
Originally Posted by koturban View Post
It is certainly an interesting phenomena. I guess that if the media being played has no ultrasonic content, and their equipment isn't capable of ultrasonic reproduction (including transducers) then one needn't be concerned and enjoy their system. Seems like a case of over-performance creating unintended consequences.
The only interesting thing is that after years and years of challenges that no one can tell these differences apart, we all of a sudden can. All those textbook snippets that were supposed to prove we can't, didn't stop us. All the stomping the feet that the tests must have been sighted, must have lacked level matching, must have been created by people who didn't know what they were doing, did not apply here either.

We have run test after test that Arny has put forward and passed it. Passing just one of these was supposed to be an impossibility but here we are, and Arny keeps giving us more homework. Because Arny can't hear what we hear I expect these shots in the dark to keep going.

So yes, "interesting" does not even begin to describe what we managed to get done in this one thread.

Quote:
Btw, I cruised on over to the aforementioned forum and I see that someone's ego has been inflated to a point that he is considering submitting a paper to AES about his "result". Oh my.
Didn't know that it was an ego driven thing to get things published at AES. We have another reason then to dismiss tests like Meyer and Moran.
Charles R and stereoeditor like this.

Amir
Retired Technology Insider
Founder, Madrona Digital
"Insist on Quality Engineering"
amirm is offline  
Sponsored Links
Advertisement
 
Old 07-26-2014, 09:07 AM
AVS Special Member
 
krabapple's Avatar
 
Join Date: Apr 2003
Location: in a state bordered by Kentucky and Maine
Posts: 5,360
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 177 Post(s)
Liked: 229
Quote:
Originally Posted by koturban View Post
Btw, I cruised on over to the aforementioned forum and I see that someone's ego has been inflated to a point that he is considering submitting a paper to AES about his "result". Oh my.
Well, peer review of *that* should prove interesting

(Btw, AES conference papers /abstracts aren't peer reviewed; JAES papers are)

It's rather remarkable that the few published tests audibility of hirez vs standard have never found such a set of straight-up difference-positives (some of which appear to be due to IM distortion) as we have here using Arny's files. This includes Oohashi et al, who only 'detected' their 'hypersonic effect' using a setup specifically aimed at optimizing discrimination (it used *musical* signal with lots of HF content -- gamelan IIRC -- and a custom extended-bandwith playback rig), as differences in 'unconscious' (brain blood flow and EEG), or 'conscious' (preference* scale testing) response -- plain conscious difference testing a la ABX as we're doing here produced no evidence of audible difference.

It also includes, of course, Meyer and Moran 2007, as well as attempts to replicate Oohashi et al's conscious response results (no one, AFAIK, has tried to replicate their unconscious response work). Results from those were nothing like the rather 'robust' number of successes being reported here for 96/24 vs 44.1/16.

So we have a whopping big outlier here. What to make of it? I think I know what the peer reviewers will want.


Meanwhile no one at Hydrogenaudio seems to be aware of our AVSF test or the our remarkable results. I'll make them aware, but anyone who becomes interested there will certainly want access to the test files.

Last edited by krabapple; 07-26-2014 at 09:26 AM.
krabapple is offline  
Old 07-26-2014, 09:53 AM
Senior Member
 
koturban's Avatar
 
Join Date: Jan 2007
Posts: 386
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 155 Post(s)
Liked: 102
Quote:
Originally Posted by amirm View Post
The only interesting thing is that after years and years of challenges that no one can tell these differences apart, we all of a sudden can. All those textbook snippets that were supposed to prove we can't, didn't stop us. All the stomping the feet that the tests must have been sighted, must have lacked level matching, must have been created by people who didn't know what they were doing, did not apply here either.

...
The fact that one "suddenly can" when seemingly nothing has changed would set off alarm bells for anyone unencumbered by confirmation bias.

But you go right ahead and publish that paper.
tubetwister likes this.
koturban is online now  
Old 07-26-2014, 10:00 AM - Thread Starter
AVS Addicted Member
 
amirm's Avatar
 
Join Date: Jan 2002
Location: Washington State
Posts: 18,375
Mentioned: 4 Post(s)
Tagged: 0 Thread(s)
Quoted: 981 Post(s)
Liked: 547
A side note guys. Since running these tests I saw my lightning fast laptop come to a crawl. This is a core i7 laptop with 16 gigabytes of memory by the way and discrete graphics.

Anyway, I looked at perfmon and the problem was right there:



See it? Look at the memory usage of audiodg.exe. It has climbed up to over 13 gigabytes! Worse yet, it was leaking memory fast with the counter running like there was no tomorrow. So the CPU usage is also impacted.

Audiodg is the audio filter graph user process. It is triggered by running Foobar. Once there, it hits on a memory leak bug causing the runaway condition above. You would need to manually kill the process and restart foobar every time you want to use the latter to run these tests. Otherwise, in span of a couple of hours this morning it had already gobbled up more than 2 gigabytes! This is on Windows 7. Maybe fixed in Windows 8.
Chu Gai likes this.

Amir
Retired Technology Insider
Founder, Madrona Digital
"Insist on Quality Engineering"
amirm is offline  
Old 07-26-2014, 10:11 AM - Thread Starter
AVS Addicted Member
 
amirm's Avatar
 
Join Date: Jan 2002
Location: Washington State
Posts: 18,375
Mentioned: 4 Post(s)
Tagged: 0 Thread(s)
Quoted: 981 Post(s)
Liked: 547
Quote:
Originally Posted by koturban View Post
The fact that one "suddenly can" when seemingly nothing has changed would set off alarm bells for anyone unencumbered by confirmation bias.
If you are hearing alarm bells, that is one of the reasons you can't differentiate between the files.

But sure, there are alarm bells. Those bells say that no one should take these DBT challenges seriously from you all. You won't run and report on them yourself.

Quote:
But you go right ahead and publish that paper.
I may wait a year or two to see how much suffering the thought of that paper getting published causes .

Amir
Retired Technology Insider
Founder, Madrona Digital
"Insist on Quality Engineering"
amirm is offline  
Old 07-26-2014, 10:18 AM
AVS Special Member
 
Frank Derks's Avatar
 
Join Date: Jul 2002
Location: Region A,B,C
Posts: 1,897
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 103 Post(s)
Liked: 89
Quote:
Originally Posted by amirm View Post
A side note guys. Since running these tests I saw my lightning fast laptop come to a crawl. This is a core i7 laptop with 16 gigabytes of memory by the way and discrete graphics.

Anyway, I looked at perfmon and the problem was right there:



See it? Look at the memory usage of audiodg.exe. It has climbed up to over 13 gigabytes! Worse yet, it was leaking memory fast with the counter running like there was no tomorrow. So the CPU usage is also impacted.

Audiodg is the audio filter graph user process. It is triggered by running Foobar. Once there, it hits on a memory leak bug causing the runaway condition above. You would need to manually kill the process and restart foobar every time you want to use the latter to run these tests. Otherwise, in span of a couple of hours this morning it had already gobbled up more than 2 gigabytes! This is on Windows 7. Maybe fixed in Windows 8.
Don't see this happening in my system.


Used the foobar ABX tool for a while but I see the memory usage going up and down around 14,000K.
Frank Derks is offline  
Old 07-26-2014, 11:32 AM
Senior Member
 
koturban's Avatar
 
Join Date: Jan 2007
Posts: 386
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 155 Post(s)
Liked: 102
Quote:
Originally Posted by amirm View Post
If you are hearing alarm bells, that is one of the reasons you can't differentiate between the files.

But sure, there are alarm bells. Those bells say that no one should take these DBT challenges seriously from you all. You won't run and report on them yourself.


I may wait a year or two to see how much suffering the thought of that paper getting published causes .
No need to inflict such suffering on yourself, Amir.

You know you want to.
koturban is online now  
Old 07-26-2014, 11:47 AM
AVS Addicted Member
 
Ratman's Avatar
 
Join Date: May 2002
Location: Collingswood, N.J.
Posts: 14,662
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 235 Post(s)
Liked: 315
Quote:
Originally Posted by koturban View Post
No need to inflict such suffering on yourself, Amir.

You know you want to.
LOL...
I know I don't wake up 1st thing in the morning, hop on the forum hoping to get a "one up" on Arny.
Ratman is offline  
Old 07-26-2014, 01:02 PM
AVS Special Member
 
tubetwister's Avatar
 
Join Date: Feb 2013
Location: Sacramento delta N. Cal. US Don't trust any air I can't see ☺
Posts: 3,232
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 640 Post(s)
Liked: 387
Quote:
Originally Posted by krabapple View Post
btw for newbies here, the way to quote correctly is to make sure that 'quote="whoever"' is the opening tag on *every* quoted section, even if the sections are from the same post. This is a new thing at AVSF

e.g. (I've omitting opening and closing brackets here, so the examples don't turn into quotes! But every 'quote' or /quote tag should have brackets around it)

(open)quote=R2D2(close)

blah blah section 1

(open)/quote(close)

reply to 1




(open)quote=R2D2(close)

blah blah section 2

(open)/quote(close)


reply to 2


etc
Thanks I'll have to get with that

Hires Music formats ..............."Why does it sound like a CD ?" ............. can we make it louder "?
"The wireless music box has no commercial value. Who would pay for a message sent to nobody in particular?"
- David Sarnoff's associates at RCA the 1920's -
tubetwister is offline  
Old 07-26-2014, 01:07 PM
AVS Addicted Member
 
arnyk's Avatar
 
Join Date: Oct 2002
Location: Grosse Pointe Woods, MI
Posts: 14,387
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 763 Post(s)
Liked: 1178
Quote:
Originally Posted by Ratman View Post
LOL...
I know I don't wake up 1st thing in the morning, hop on the forum hoping to get a "one up" on Arny.
I've usually watched TV for several hours before the first time I log onto AVS every day...

Charlie Rose, NASCAR Race Hub, BBC news, the previous nights CBS national news, Bloomberg Surveillance...

that gets me bored enough... ;-)
arnyk is offline  
Old 07-26-2014, 01:13 PM
isa
Member
 
Join Date: Jun 2014
Posts: 37
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 33 Post(s)
Liked: 18
Quote:
Originally Posted by amirm View Post
The only interesting thing is that after years and years of challenges that no one can tell these differences apart, we all of a sudden can.
...We have another reason then to dismiss tests like Meyer and Moran.
I'll try to be kind, but I'm rather disappointed in the quote above and similar. I agree a difference or differences has/have been detected. But we've not seen anything credible on what difference(s) has/have been detected, much less the underlying cause(s). A correlation that the differently-sounding files have different bit depth and sampling rates does not mean causation by bit depth or sampling rate by itself/themselves. Several have appealed to using basic scientific method principles to investigate all that and eliminate possible sources of error or unintended differences before leaping to conclusions, but those appeals have apparently been unpersuasive.

I realize Amir's specific comment on M&M above was a tongue-in-cheek remark about egos and publishing, but M&M has been invoked many times here in ways that make me cringe. My view is that any tests conducted in this thread have not yet established anything credible that confirms or refutes the conclusions in the Meyer & Moran study. Lest we forget, their actual conclusions were as follows:
Quote:
3 CONCLUSIONS
We have analyzed all of the test data by type of music
and specific program; type of high-resolution technology;
age of recording; and listener age, gender, experience, and
hearing bandwidth. None of these variables have shown
any correlation with the results, or any difference between
the answers and coin-flip results.
The previous work cited, some of it at the very beginning
of the CD era and some more recent, pointed toward
our result. With the momentum of widespread “high-rez”
anecdotes over the last decade, culminating in the Stuart
assertions, we felt the need to go further and perform a
thorough, straightforward double-blind level-matched listening
test to determine whether 16/44.1 technology would
audibly degrade the sound of the best high-resolution discs
we could find. We used a large and varied sample of
serious listeners; we conducted our tests using several different
types of high-quality playback systems and rooms;
and we took as much time as we felt necessary to establish
the transparency of the CD standard.
Now, it is very difficult to use negative results to prove
the inaudibility of any given phenomenon or process.
There is always the remote possibility that a different system
or more finely attuned pair of ears would reveal a
difference. But we have gathered enough data, using sufficiently
varied and capable systems and listeners, to state
that the burden of proof has now shifted. Further claims
that careful 16/44.1 encoding audibly degrades high resolution
signals must be supported by properly controlled
double-blind tests.
I appeal to everyone to avoid strawmans to make this thread a better use of electrons, time and disc space.

Last edited by isa; 07-26-2014 at 01:29 PM. Reason: computer crash while posting. Doh!
isa is offline  
Old 07-26-2014, 01:49 PM - Thread Starter
AVS Addicted Member
 
amirm's Avatar
 
Join Date: Jan 2002
Location: Washington State
Posts: 18,375
Mentioned: 4 Post(s)
Tagged: 0 Thread(s)
Quoted: 981 Post(s)
Liked: 547
Quote:
Originally Posted by isa View Post
I'll try to be kind, but I'm rather disappointed in the quote above and similar.
I appreciate your kindness.

Quote:
I agree a difference or differences has/have been detected. But we've not seen anything credible on what difference(s) has/have been detected, much less the underlying cause(s).
Now you know the problem with demanding forced choice, binary tests like ABX. It gives you no qualitative analysis. If you expected such, the nature of such testing is not understood.

I am glad you accept what has been accomplished. That a difference has been detected. The experimenters have proposed these tests as to show lack of differentiation between original file and the converted one to 16/44.1. As you say we have proven an audible differentiation exists. Therefore no claim of transparency can be made in these conversions.

I know this is a painful turn of events and it invalidates thousands of posts where we have said otherwise. Remember, all of those posts were based on what we had read, not what is in front of us in hard, tangible data and content. Nor have you ever been to question the people who have run these tests before.
Quote:
A correlation that the differently-sounding files have different bit depth and sampling rates does not mean causation by bit depth or sampling rate by itself/themselves.
That is all that the experimenters say they have done to give us the two versions. This is what Scott/Mark have done. And this is what Arny has done.

Now, you could doubt the veracity of their statements and using instrumentation/measurements/etc. that shows that they have done more than what they say with respect to changing the sampling rate/bit depth. But you haven't done that. You are simply practicing FUD, hoping mere words can create doubt.

Remember, I repeated Arny's test as I post last night and the simple act of reducing the bit depth and sample rate in Audition CC allowed me to hear the differences in double blind, ABX, computer controlled testing. Countless people use this tool to create music from high-res to CD. This data says conversions done in this tool do not provide transparency using Arny's master file.

Quote:
Several have appealed to using basic scientific method principles to investigate all that and eliminate possible sources of error or unintended differences before leaping to conclusions, but those appeals have apparently been unpersuasive.
Once again, you make this like I created the test and am reporting the results of the same. I did not. The test was created by others. Across multiple tests, positive outcomes have been achieved, not just by me but by others. You agree such a positive difference is heard. Yet you think there needs to be further objections?

The objections voiced are emotional and anything but "scientific." They lack understanding the very nature of the tests being asked to run. They lack protesters own data and results. They represent people who can't hear the differences and pass the same tests.

Quote:
Any tests conducted in this thread have not yet established anything credible that confirms or refutes anything in the Meyer & Moran study. Lest we forget, their actual conclusions were as follows:
Yes, let's note what they have said:

"There is always the remote possibility that a different system
or more finely attuned pair of ears would reveal a
difference. But we have gathered enough data, using sufficiently
varied and capable systems and listeners, to state
that the burden of proof has now shifted. "


Seeing how you accept that I am hearing differences that others cannot, then it reasons that I may be one of those people with "finer attuned pair of ears." As are the others who have found differences in this thread.

Remember, the very objection you voice here applies to them in a huge way. You have no measurements whatsoever that when they ran the content through A/D to D/A that they actually changed the signal. We know from other data post their testing that their material may have lacked ultrasonic content.

In our situation Arny has been kind enough to post spectrums and others such as myself have confirmed that for sure the sampling rate and ultrasonic content has been changes. The credibility of what we are doing then is far higher than theirs if you are objective and data driven.

Quote:
I appeal to everyone to avoid strawmans to make this thread a better use of electrons, time and disc space.
You can appeal all you want. What matters is data not personal appeals. Unfortunately you don't have any data to counter these results. There is nothing scientific about creating doubt with words and throwing terms like "strawman" around.

Tests were created by parties with no interest in the outcome that is produced. You want to say their test fixture and content is faulty, please make your cased to them. Don't abuse the term "science" by going after listeners of the test.

You want to advance the discussion and not waste the electrons, run Arny's 32 Khz test vs 96 Khz. Focus and pay attention to what you are hearing. Remember that multiple people from different walks of life, equipment, gear, hearing ability, etc. have heard differences that are 100% reliable.

And oh, thanks again for your kindness.
Charles R likes this.

Amir
Retired Technology Insider
Founder, Madrona Digital
"Insist on Quality Engineering"
amirm is offline  
Old 07-26-2014, 02:16 PM
AVS Addicted Member
 
Chu Gai's Avatar
 
Join Date: Sep 2002
Location: NYC area
Posts: 14,968
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 319 Post(s)
Liked: 596
What differences did you hear?
tubetwister likes this.

"I've found that when you want to know the truth about someone that someone is probably the last person you should ask." - Gregory House
Chu Gai is offline  
Old 07-26-2014, 03:29 PM
AVS Special Member
 
Frank Derks's Avatar
 
Join Date: Jul 2002
Location: Region A,B,C
Posts: 1,897
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 103 Post(s)
Liked: 89
Quote:
Originally Posted by amirm View Post
On WBF Forum someone said they thought Arny's resampling of keys jingling to 44.1 was not optimal and asked me to perform another test. I did not have much time but I did use Audition CC to resample the 96 Kh file to 44.1 using TPDF dither. Here are the results:

-------

foo_abx 1.3.4 report
foobar2000 v1.3.2
2014/07/24 20:27:41

File A: C:\Users\Amir\Music\Arnys Filter Test\keys jangling amir-converted 4416 2496.wav
File B: C:\Users\Amir\Music\Arnys Filter Test\keys jangling full band 2496.wav

20:27:41 : Test started.
20:28:07 : 00/01 100.0%
20:28:25 : 00/02 100.0%
20:28:55 : 01/03 87.5%
20:29:02 : 02/04 68.8%
20:29:12 : 03/05 50.0%
20:29:20 : 04/06 34.4%
20:29:27 : 05/07 22.7%
20:29:36 : 06/08 14.5%
20:29:44 : 07/09 9.0%
20:29:55 : 08/10 5.5%
20:30:00 : 09/11 3.3%
20:30:07 : 10/12 1.9%
20:30:16 : 11/13 1.1%
20:30:22 : 12/14 0.6%
20:30:29 : 13/15 0.4%
20:30:36 : 14/16 0.2%
20:30:41 : 15/17 0.1%
20:30:53 : 16/18 0.1%
20:31:03 : 17/19 0.0%
20:31:07 : Test finished.

----------
Total: 17/19 (0.0%)

So, still audible .
Did you resample your converted file back to 2496 with or without dither?


Did you check the levels after your conversion(s)?


Did you use the key jangling part of the track or the im check and focused on the clicks?


Dogs in the vicinity?
tubetwister likes this.
Frank Derks is offline  
Old 07-26-2014, 03:58 PM
AVS Special Member
 
tubetwister's Avatar
 
Join Date: Feb 2013
Location: Sacramento delta N. Cal. US Don't trust any air I can't see ☺
Posts: 3,232
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 640 Post(s)
Liked: 387
Quote:
Originally Posted by Frank Derks View Post
Don't see this happening in my system.


Used the foobar ABX tool for a while but I see the memory usage going up and down around 14,000K.
Looks like in some cases the testing software may have some resource (maybe core scheduling ?) or memory allocation (leakage ) issues . ofc on different systems and or software that could possibly present or not?

Foobar 2000 might be ported to Linux
might work a little better for testing without windows overhead ? It's not like this is prime 95 or anything it shoulder be anyway .

EDIT ADD Looks like in Amirs case at least it *may be a Windows Audiodg.exe process memory leak
from what Google says AFAIK ,Google search results are addressed further down in the tread here.

Any modern CPU (other than low budget or a slow laptop ) and certainly a Haswell or similar i7 should be well more than adequate .

CPU should not have any issues at all with anything audio unless a software ,HDWE sub system ,sleepy hdd or OS issue.

I can run tons of stuff including studio DAW+ plugins +editor + 3-4 midi controllers and a web browser all in real time on 3 screens in the studio on a sandy bridge i5 and not break 20-25% CPU utilization if that.
OTOH one poorly written VST can bring things to a complete halt in a hurry also.


I understand the scientific question " does hires generally present an audible difference ? ) It may to some degree or may not AFAIK ?

No conclusive results as far as presenting the consumer a significant value added value on it's own merits AFAIK .
In any event that should been answered long ago anyway and there would be no need for this discussion.

The fact we are having this discussion at all this late in the game ofc presents what the likely the outcome would be to the significant added value question . No problems convincing the HT enthusiast or many consumers of the significant added value of multi channel digital sound processing because the result is readily apparent.

FWIW IMO the industry is running out of ways to differentiate product and would also like to open up new business
opportunities , I believe that is much of what the industry marketing of hires is they shouldn't have to push it
if the marketing claims lived up to the realities for the average consumer . I believe the free market and word of mouth would take care of that like it did with the iPod,iPhone etc.

Would be interesting to see if any papers (at this point ) get accepted that would be surprising IMO.

Hires Music formats ..............."Why does it sound like a CD ?" ............. can we make it louder "?
"The wireless music box has no commercial value. Who would pay for a message sent to nobody in particular?"
- David Sarnoff's associates at RCA the 1920's -

Last edited by tubetwister; 07-26-2014 at 07:06 PM.
tubetwister is offline  
Old 07-26-2014, 04:19 PM
AVS Special Member
 
NorthSky's Avatar
 
Join Date: Jan 2013
Location: Canada - West Island: Vancouver, South Direction: Go East
Posts: 4,950
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 2137 Post(s)
Liked: 614
Thumbs up Thx!

Quote:
Originally Posted by krabapple View Post
btw for newbies here, the way to quote correctly is to make sure that 'quote="whoever"' is the opening tag on *every* quoted section, even if the sections are from the same post. This is a new thing at AVSF

e.g. (I've omitting opening and closing brackets here, so the examples don't turn into quotes! But every 'quote' or /quote tag should have brackets around it)
(open)quote=R2D2(close)
blah blah section 1
(open)/quote(close)
Quote:
Originally Posted by krab
reply to 1
(open)quote=R2D2(close)
blah blah section 2
(open)/quote(close)
Quote:
Originally Posted by apple
reply to 2
etc
Thank you so much for that, finally!
...Because I could not find the way of doing it (different than all other forums I frequent).

TEST: It works!

Last edited by NorthSky; 07-26-2014 at 04:23 PM.
NorthSky is online now  
Old 07-26-2014, 04:56 PM
AVS Addicted Member
 
Ratman's Avatar
 
Join Date: May 2002
Location: Collingswood, N.J.
Posts: 14,662
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 235 Post(s)
Liked: 315
James Edwards: Maybe you already answered this, but, why exactly are we here?
Zed: [noticing a recruit raising his hand] Son?
Second Lieutenent Jake Jenson: Second Lieutenant, Jake Jenson. West Point. Graduate with honors. We're here because you are looking for the best of the best of the best, sir!
Zed: [throws Edwards a contemptible glance as Edwards laughs] What's so funny, Edwards?
James Edwards: Boy, Captain America over here! "Best of the best of the best, sir!" "With honors." Yeah, he's just really excited and he has no clue why we're here.

Just sayin'
Ratman is offline  
Old 07-26-2014, 05:06 PM
AVS Special Member
 
NorthSky's Avatar
 
Join Date: Jan 2013
Location: Canada - West Island: Vancouver, South Direction: Go East
Posts: 4,950
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 2137 Post(s)
Liked: 614
We're all part of an unknown, undefined, yet to be explored and undiscovered universe.
...Full of black holes plugged by white holes. ...In a constant quantum leap of faith...
NorthSky is online now  
Old 07-26-2014, 06:42 PM
AVS Special Member
 
tubetwister's Avatar
 
Join Date: Feb 2013
Location: Sacramento delta N. Cal. US Don't trust any air I can't see ☺
Posts: 3,232
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 640 Post(s)
Liked: 387
Quote:
Originally Posted by amirm View Post
A side note guys. Since running these tests I saw my lightning fast laptop come to a crawl. This is a core i7 laptop with 16 gigabytes of memory by the way and discrete graphics.

Anyway, I looked at perfmon and the problem was right there:



See it? Look at the memory usage of audiodg.exe. It has climbed up to over 13 gigabytes! Worse yet, it was leaking memory fast with the counter running like there was no tomorrow. So the CPU usage is also impacted.

Audiodg is the audio filter graph user process. It is triggered by running Foobar. Once there, it hits on a memory leak bug causing the runaway condition above. You would need to manually kill the process and restart foobar every time you want to use the latter to run these tests. Otherwise, in span of a couple of hours this morning it had already gobbled up more than 2 gigabytes! This is on Windows 7. Maybe fixed in Windows 8.
FWIW there is a Hotfix at Microsoft for this very audiodg.exe. thing(memory leak )it applies to PlaySound function don't know if it would apply here but if you haven't seen it, it maybe worth a look there may be more hotfixes at M$ also.

Lot's of stuff on Google about memory leak and high CPU use with Audiodg. might be other fixes as well.

audiodg.exe Handle leak hotfix @ m$ also

It seems my suspicions may be correct that being a software issue ofc I'm sure you immediately knew that anyway.

Maybe I've never seen this because I am using ASIO drivers and USB interface in back (studio PC) and hacked ASIO/Asus drivers and sometimes also T racks in here.

M$probably fixed prob. in the new win8x audio stack. ofc DAW (s) and Editors have more graph's etc than anyone would ever need in the studio in back here anyway M$probably fixed prob. in the new win8x audio stack.

Have a DAW and T- racks + hacked Asio/Asus sound drivers on this PC I don't use DAW or T racks much on this one although it can do everything CPU wise as the other one for the most part .

This PC spends most of it's time wasting time on the web , or performing mundane office/email tasks and sometimes playing (not usually hires music it can resolve if needed ☺also digitally if needed to an external amplified device or just hires 192KHz SC within itself ) but instead other normal res losseless or lossey music through my Senns. or amplified 2.1 speakes in here ,or playing +upscaling DVD to TV screen in here .


As you know your CPU use shouldn't be a cause for concern but that thing is consuming lot's of memory so the slowdown is certainly understandable , been down that road before over the years now and then as I'm sure you have as well . I'm surprised win update didn't address that, may have been an overlooked optional update (not hard to do ) good luck .

Quote:
@ M$
You run a program that calls the PlaySound function to play a sound. However, a memory leak occurs in the Audiodg.exe process. To resolve this issue, you have to restart the Windows Audio service (Audiosrv). In this situation, other programs that depend on the service will be affected.

If you check the memory usage of the Audiodg.exe process, you find that the memory leak consumes several hundreds of bytes every time that the PlaySound function is called.

Note This issue occurs in the following operating systems:
Windows 7
Windows Server 2008 R2
Windows Vista
Windows Server 2008

Hires Music formats ..............."Why does it sound like a CD ?" ............. can we make it louder "?
"The wireless music box has no commercial value. Who would pay for a message sent to nobody in particular?"
- David Sarnoff's associates at RCA the 1920's -

Last edited by tubetwister; 07-27-2014 at 04:19 AM.
tubetwister is offline  
Old 07-26-2014, 10:10 PM
AVS Special Member
 
m. zillch's Avatar
 
Join Date: Oct 2006
Posts: 4,380
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 357 Post(s)
Liked: 289
Quote:
Originally Posted by arnyk View Post
The test is composed of a nice quiet little 4 KHz tone to train listeners what to listen for.

It is followed by twin tones 1 second duration each, with frequencies of 22+27 KHz, 32+36 Kz, and 38+42 Khz.

These are naturally missing from the downsampled versions of the file.

When listened to, all of the tones but the initial 4 KHz training tone should be inaudible, which means that unless the full bandwidth and downsampled versions sound the same, there is audible nonlinear distortion in the monitoring path..
If I understand you correctly, since I only hear the initial 4kHz training tone at 12 seconds in or so, and no subsequent 4kHz/5kHz tones after that [although I do hear a couple of prominent, transient ticks(clicks) in both files] my particular playback system is immune to the potential IM problem some others are facing. Is this correct?

Last edited by m. zillch; 07-26-2014 at 10:15 PM.
m. zillch is offline  
Old 07-27-2014, 03:10 AM
AVS Special Member
 
tubetwister's Avatar
 
Join Date: Feb 2013
Location: Sacramento delta N. Cal. US Don't trust any air I can't see ☺
Posts: 3,232
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 640 Post(s)
Liked: 387
Quote:
Originally Posted by krabapple View Post
you're killing me.

Quote:
Originally Posted by krabapple
btw for newbies here, the way to quote correctly is to make sure that 'quote="whoever"' is the opening tag on *every* quoted section, even if the sections are from the same post. This is a new thing at AVSF

e.g. (I've omitting opening and closing brackets here, so the examples don't turn into quotes! But every 'quote' or /quote tag should have brackets around it)

(open)quote=R2D2(close)

blah blah section 1

(open)/quote(close)

reply to 1




(open)quote=R2D2(close)

blah blah section 2

(open)/quote(close)


reply to 2


etc

Many thanks for the correct way to post multiple quotes here. I should have known that or figured it out
since the recent forum code /formatting changes . I must have been having quite a few senior moment (s) (or dozing off) lately ( no not nodding off that's different and not always legal sometimes ☺ ) not to figure out something as simple and obvious as this not to mention it is much easier than the way I was not doing it correctly before

Everyone should be relieved to know now that Armin and I *may have* come to some sort of a mutual
accommodation,or more likely just got bored to death with each others often lengthy, repetitive (and poorly formatted in my case) posts and replies to each other.☺I'm looking forward to not having to serial quote as much going forward even though I know how to format multiple quotes correctly now

Now if I can only learn to use the sometimes correct spell checker here more consistently ☺



Maybe thread discussion can resume proving (or disproving ) hires hypothesis now

PS: krabapple thanks again.

Hires Music formats ..............."Why does it sound like a CD ?" ............. can we make it louder "?
"The wireless music box has no commercial value. Who would pay for a message sent to nobody in particular?"
- David Sarnoff's associates at RCA the 1920's -

Last edited by tubetwister; 07-27-2014 at 07:48 AM.
tubetwister is offline  
Old 07-27-2014, 06:06 AM
AVS Addicted Member
 
arnyk's Avatar
 
Join Date: Oct 2002
Location: Grosse Pointe Woods, MI
Posts: 14,387
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 763 Post(s)
Liked: 1178
Quote:
Originally Posted by m. zillch View Post
If I understand you correctly, since I only hear the initial 4kHz training tone at 12 seconds in or so, and no subsequent 4kHz/5kHz tones after that [although I do hear a couple of prominent, transient ticks(clicks) in both files] my particular playback system is immune to the potential IM problem some others are facing. Is this correct?
If you hear the 4 KHz tone and the clicks identically in both of the test files, and no tones or other noises between the clicks, your monitoring system is clean at that particular listening level. If you adjust the volume, you should rerun the test.

The following commonly observed results indicate IM failure:

Steady or squiggly sounding tones other than the first steady 4 KHz tone. A raised noise floor. Any audible difference between the two test files.

In the latest version of the test files, which can be used for both ABXing keys jangling and the IM test, the frequencies of the first set of test tones was changed to 23 abd 27 hz, raising all of the test tones to above the Nyquist frequency for 44.1 KHz.

https://www.dropbox.com/s/ylrjezd7vc...st%20tones.zip

Last edited by arnyk; 07-27-2014 at 08:38 AM.
arnyk is offline  
Old 07-27-2014, 07:11 AM
Senior Writer @ AVS
 
imagic's Avatar
 
Join Date: Dec 2005
Location: Philadelphia, PA
Posts: 6,438
Mentioned: 31 Post(s)
Tagged: 0 Thread(s)
Quoted: 1761 Post(s)
Liked: 3089
Quote:
Originally Posted by m. zillch View Post
If I understand you correctly, since I only hear the initial 4kHz training tone at 12 seconds in or so, and no subsequent 4kHz/5kHz tones after that [although I do hear a couple of prominent, transient ticks(clicks) in both files] my particular playback system is immune to the potential IM problem some others are facing. Is this correct?
That's true, but make sure you really don't hear anything. For me, the subsequent tones were very faint.

Find out more about Mark Henninger at www.imagicdigital.com
imagic is offline  
Old 07-27-2014, 08:28 AM
AVS Special Member
 
tubetwister's Avatar
 
Join Date: Feb 2013
Location: Sacramento delta N. Cal. US Don't trust any air I can't see ☺
Posts: 3,232
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 640 Post(s)
Liked: 387
Quote:
Originally Posted by arnyk View Post
If you hear the 4 KHz tone and the clicks identically in the test files, your monitoring system is clean at that particular listening level. If you adjust the volume, you should rerun the test.

The following commonly observed results indicate IM failure:

Steady or squiggly sounding tones other than the first steady 4 KHz tone. A raised noise floor. Any audible difference between the two test files.

In the latest version of the test files, which can be used for both ABXing keys jangling and the IM test, the frequencies of the first set of test tones was changed to 23 abd 27 hz, raising all of the test tones to above the Nyquist frequency for 44.1 KHz.

https://www.dropbox.com/s/ylrjezd7vc...st%20tones.zip

Arny thanks for the additional clarification (s)

I can hear the 4 K test tones and 2 clicks very clearly
at soft and loud levels on daily driver PC with Xonar STX 192kHz Sound card w/ internal headphone amp on my Senns studio headphones very clearly .

EDIT : ADD I can hear maybe 2 faint transient clicks ^^^^while the keys jangle , the test tone clearly and 2 subsequent clicks clearly on both files . (hopefully this is OK as far as the IM thing goes ? )

2 files (at this point) appear to sound mostly identical but I don't want to state that as a definite opinion just quite yet or even today. I believe it will take much more testing on my part to form a legitimate opinion .


Result being I can have more time to train myself and possibly contribute to the discussion with some results now and then rather than just opinions and serial quotes and rebuttal . (don't fall outta your chairs guys bear with me no expert experimenter here )

Reason being when the studio PC and interface + other stuff is in use
(usually daily all day ) I can't be in there experimenting. I'm just the controlling silent partner and IT + electronics repair person in the family business.

TBH what got me thinking maybe the daily driver could do this easily was some folks using laptops and Armins recent posts regarding the difficulties he encountered . ( I thought maybe my hacked ASIO/ Asus SC drivers could be of help here) I can always bring in any additional software I would need from out back anyway .

I always have this persistent minds image of the more scientific types here using facilities that would be similar to a commercial or academic ................. or maybe the Frankenstein or Tesla laboratory's instead of a desktop PC and phones................................ Although that may not be to much of a stretch now and again!


Hopefully the hacked ASIO/ Asus drivers on here won't cause any conflicts or memory leaks in the win 7 audio stack in fact I'm anticipating just the opposite.


Will also download latest file .

Hires Music formats ..............."Why does it sound like a CD ?" ............. can we make it louder "?
"The wireless music box has no commercial value. Who would pay for a message sent to nobody in particular?"
- David Sarnoff's associates at RCA the 1920's -

Last edited by tubetwister; 07-27-2014 at 09:06 AM.
tubetwister is offline  
Old 07-27-2014, 08:38 AM
AVS Addicted Member
 
arnyk's Avatar
 
Join Date: Oct 2002
Location: Grosse Pointe Woods, MI
Posts: 14,387
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 763 Post(s)
Liked: 1178
Quote:
Originally Posted by imagic View Post
That's true, but make sure you really don't hear anything. For me, the subsequent tones were very faint.

Faint is not good. Not at all is good.
arnyk is offline  
Old 07-27-2014, 08:42 AM
AVS Addicted Member
 
arnyk's Avatar
 
Join Date: Oct 2002
Location: Grosse Pointe Woods, MI
Posts: 14,387
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 763 Post(s)
Liked: 1178
Quote:
Originally Posted by tubetwister View Post
M$probably fixed prob. in the new win8x audio stack. ofc DAW (s) and Editors have more graph's etc than anyone would ever need in the studio in back here anyway M$probably fixed prob. in the new win8x audio stack.
My Windows 8.1 Task manager shows neither hide nor hair of anything even vaguely resembling AudioDG. when running FOOBAR2000 or other audio tools that I use.
arnyk is offline  
Old 07-27-2014, 08:55 AM
Senior Member
 
stereoeditor's Avatar
 
Join Date: Feb 2010
Posts: 343
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 140 Post(s)
Liked: 39
Quote:
Originally Posted by arnyk View Post
If you hear the 4 KHz tone and the clicks identically in both of the test files, and no tones or other noises between the clicks, your monitoring system is clean at that particular listening level.
I posted to What's Best Forum spectral analyses of 3 USB-input D/A headphone amplifiers, a Meridian Prime ($2000, AC powered), Meridian Explorer ($299, USB bus-powered), and AudioQuest Dragonfly ($149, bus-powered), while they handled a 96kHz file with 39kHz and 41kHz tones, each at -6dB, ie, the combined waveform peaking at 0dBFS.

See http://www.whatsbestforum.com/showth...l=1#post278647 and http://www.whatsbestforum.com/showth...l=1#post278655 .

The Prime handled this very demanding signal without any audio-band intermodulation products. There were some low-level IM products in the audioband with the two bus-powered headphone amps, which were starting to clip. However, reducing the signal's peak level virtually eliminated the audio-band IM products. I think it safe to conclude that as long as your playback system is not being driven into clipping or has pathologically poor performance (like the so-called NOS DACs), playing high-sample-rate files will not result in audible audio-band distortion, especially as real-life music signals or even Arny's keys will not have ultrasonic content approaching 0dBFS.

John Atkinson
Editor, Stereophile
stereoeditor is offline  
Old 07-27-2014, 09:07 AM
Senior Writer @ AVS
 
imagic's Avatar
 
Join Date: Dec 2005
Location: Philadelphia, PA
Posts: 6,438
Mentioned: 31 Post(s)
Tagged: 0 Thread(s)
Quoted: 1761 Post(s)
Liked: 3089
Quote:
Originally Posted by stereoeditor View Post
I posted to What's Best Forum spectral analyses of 3 USB-input D/A headphone amplifiers, a Meridian Prime ($2000, AC powered), Meridian Explorer ($299, USB bus-powered), and AudioQuest Dragonfly ($149, bus-powered), while they handled a 96kHz file with 39kHz and 41kHz tones, each at -6dB, ie, the combined waveform peaking at 0dBFS.

See http://www.whatsbestforum.com/showth...l=1#post278647 and http://www.whatsbestforum.com/showth...l=1#post278655 .

The Prime handled this very demanding signal without any audio-band intermodulation products. There were some low-level IM products in the audioband with the two bus-powered headphone amps, which were starting to clip. However, reducing the signal's peak level virtually eliminated the audio-band IM products. I think it safe to conclude that as long as your playback system is not being driven into clipping or has pathologically poor performance (like the so-called NOS DACs), playing high-sample-rate files will not result in audible audio-band distortion, especially as real-life music signals or even Arny's keys will not have ultrasonic content approaching 0dBFS.

John Atkinson
Editor, Stereophile
That's precisely what I'm left wondering, is the bit of distortion I heard relevant to the test itself, or is it because my amp was driven to clipping. Considering that Arny's file caused my AVR to shut down twice in a row when I played the system at a level where I could hear the distortion, I'm left wondering if I'm reading too much into it.

Find out more about Mark Henninger at www.imagicdigital.com

Last edited by imagic; 07-27-2014 at 09:14 AM.
imagic is offline  
Old 07-27-2014, 09:12 AM
Senior Writer @ AVS
 
imagic's Avatar
 
Join Date: Dec 2005
Location: Philadelphia, PA
Posts: 6,438
Mentioned: 31 Post(s)
Tagged: 0 Thread(s)
Quoted: 1761 Post(s)
Liked: 3089
Quote:
Originally Posted by arnyk View Post
Faint is not good. Not at all is good.
Yeah, but I'm really wondering about its relevance to the actual tests, considering what it takes for me to hear those faint tones, i.e. turn everything up to dangerously high levels, that resulted in my AVR shutting down—which never happened before for any reason other than my cat stepping on the power switch.

Sure, I'd love it if my system was dead silent playing those tones at beyond reference, but is it relevant to real-life testing? Again, I heard zero IM distortion when I isolated and played back just the key jangling ultrasonic content instead of those tweeter-busting tones. To me, that's an indication that it was not relevant to the test. It's almost like this IM distortion argument is the Yang to the HDMI jitter Ying.

Find out more about Mark Henninger at www.imagicdigital.com
imagic is offline  
Old 07-27-2014, 09:13 AM
AVS Special Member
 
tubetwister's Avatar
 
Join Date: Feb 2013
Location: Sacramento delta N. Cal. US Don't trust any air I can't see ☺
Posts: 3,232
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 640 Post(s)
Liked: 387
Quote:
Originally Posted by arnyk View Post
My Windows 8.1 Task manager shows neither hide nor hair of anything even vaguely resembling AudioDG. when running FOOBAR2000 or other audio tools that I use.
Must have been one of the win 8x improvements to the new Audio stack that being tossing Audiodg.exe out the window (sarcasm intended ) if you will .

Being a windows 7 holdout I might take a look in m$ procman or proc exp both a little (lot) better than taskman
for peeking under the hood .

I don't hear any faint tones at all ... nada just the regular tone nice and clear but also the 1 or 2 transient click (s) faintly during the keys jangling and ofc the other clicks at the end clearly also dead silence between the prominent clicks at end of file also.

TBH I think this test is well suited to analytical or revealing studio phones rather than speakers
unless maybe some reasonably flat decent studio monitors . ( I can try that as well ) I still think decent phones offer the better choice here .

OTOH I can see Imagic point of using speakers at somewhat elevated levels for real world relevancy also it does present a valid question. I might try with some mid fields
and a few 100 (s) wpc trust me that system won't clip at least at any level you would ever want to be in the same room(s) with anyway or even maybe next door if I fire up the Vintage EV Sentry III's ☺.which by their nature are very detailed and revealing.

Hires Music formats ..............."Why does it sound like a CD ?" ............. can we make it louder "?
"The wireless music box has no commercial value. Who would pay for a message sent to nobody in particular?"
- David Sarnoff's associates at RCA the 1920's -

Last edited by tubetwister; 07-29-2014 at 04:08 AM.
tubetwister is offline  
 
Thread Tools


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