AVS Forum banner
Status
Not open for further replies.

AVS/AIX High-Resolution Audio Test: Take 2

Tags
frontpage
132K views 939 replies 70 participants last post by  markrubin 
#1 · (Edited)


After correcting for a 0.2 dB difference in level, the native 24/96 and SRC versions of the test files are again ready for downloading.

As many of you know, I'm conducting an informal experiment with the AVS community to investigate whether or not true high-resolution audio files—with information beyond what a Redbook CD can accommodate—can be reliably distinguished from the same files that have been downconverted to CD specs. The parameters of the experiment can be found here.

Shortly after I made the audio files available for download, several members reported that one of each pair of files was 0.2 dB lower in level than the other. This was determined to be the result of subjecting a copy of the native 24/96 file to two sample-rate conversions—first down to 16/44.1 and then back up to 24/96—in order to hide the fact that it has a lower resolution. It turns out that the sample-rate conversion software—Sonic Solutions Sonic Process in this case—lowers the level of the file being converted by 0.1 dB to provide some headroom so the calculations will not cause any clipping. (Most sample-rate converters do this, not just Sonic Process.) Because the "hidden" 16/44.1 version went through two conversions, it was 0.2 dB lower in level than the native 24/96 version, which is close enough to the just-noticeable difference (JND) that listeners might pick the native high-res version on that basis.

A number of solutions to this problem were proposed, including a couple of rather complicated, multistep conversion procedures. We also experimented with normalizing both files as well as simply increasing the level of the converted file by 0.2 dB and decreasing the level of the native file by 0.2 dB. In addition, we considered raising the level of the file that would be converted to CD specs by 0.2 dB before doing the conversion, but there wasn't enough headroom in two of the three files to do that without clipping.

In the end, neither Mark Waldrep (aka Dr. AIX), who provided the files, nor I believe that the procedure by which the levels are matched makes much difference for this experiment. Still, I wanted to process the files as little as possible, so I rejected the idea of normalizing all of them. Because I subscribe to the "cutting is better than boosting" school of thought, I decided that the best course of action was to decrease the level of the native 24/96 files by 0.2 dB rather than boost the level of the converted files. Yes, that decreases the dynamic range of the native file, but only by 0.2 dB, which still leaves it beyond what CD can handle in each case. I have checked all three pairs of files using foobar2000's ReplayGain Scan function, and each pair is now within 0.01 dB of each other, well below the JND of perceptibility.

Of course, some AVS members will object to this approach, but I seriously doubt there would be unanimous approval of any solution within the AVS community, so I made an executive decision to do it this way. Those who believe my solution is unreasonable are free to simply not participate and to argue their position in the comments here and in the other related threads; all I ask is that you keep it respectful and not personal. Just keep in mind that this is an informal—and, I hope, fun—exercise that invites AVS members to make their own determination about whether or not the extra information in a bona fide high-res audio file makes a perceptible difference over the same content that conforms to CD specs.

Some members have also suggested that the results of the experiment should not be limited to those with audio systems capable of reproducing the extra information in the native high-res files. It's fine with me if you want to submit your determination of which files are high-res regardless of your system's capabilities, but all submissions must include a complete list of all components used in the test, including makes and models, as well as how they are connected together. I will compile the results separately for those with truly high res-capable systems and those with systems that can reproduce no more than CDs can deliver.

The new audio files are now available by clicking on the links below. As before, we've zipped each file so all browsers will initiate a download rather than trying to play them, and we've combined all six files in a single zipped package. The "A" and "B" assignments are new, so they are labeled "A2" and "B2" to make sure there is no confusion with the previous files. To make the assignments, I used the random-number generator at random.org, which is based on atmospheric noise rather than a pseudo-random algorithm—hence, it is truly random. I set the range from 1 to 100 and let it generate a random number for each pair of files—if the number was odd, the native high-res version became A2; if it was even, the native high-res version became B2.

So let's try this again, shall we? If you have already submitted your determinations, please do so again with these new files, and be sure to include a complete description of your audio system. Remember, this is not an ABX test—simply listen to each pair of files as many times as you wish, at the same volume level for both files, and see if you can tell them apart. And please don't go poking around the files with audio software—just listen. If you can, decide which one is high-res, A2 or B2, and send me your determinations by PM with the subject "AVS AIX HRA Test 2."

Thanks to all who identified the level-difference problem and offered suggestions about how to correct it.

All in one:

AVS/AIX HRA Test Files 2

Individual files:

Mosaic A2
Mosaic B2
Just My Imagination A2
Just My Imagination B2
On The Street Where You Live A2
On The Street Where You Live B2

-----------

Like AVS Forum on Facebook
Follow AVS Forum on Twitter

+1 AVS Forum on Google+
 
See less See more
1
#5 ·
I recall that in testing my own hearing of music files in the past, using automated ABX software, I could demonstrably perceive 0.2dB as a difference, but not 0.1dB. [I recall that the subjective difference I perceived was not of loudness, simply an extremely small uncharacterised "difference".] So indeed it seems quite important to have corrected the files for comparison to bring them to equal amplitude within a very close tolerance.

As for sample rate conversions causing slight amplitude differences, I have never noticed that with the open source product Audacity [excluding any possible effects from introduced dither].
 
#6 ·
I could not tell any difference, but my system is not equipped for ultrasonic playback. I will say this, somewhere between 16,000 Hz and 17,000 Hz I lose the ability to hear a sine wave. I can't see how I'd be able to hear any recorded information at frequencies higher than that. just because it was recorded with music—not when an isolated tone is already completely inaudible to me at 20,000 Hz.
 
#7 · (Edited)
As for me through my system, I could not discern any difference.

It would have been interesting to have listened to the same files through a few of the extreme high-end audio systems at the T.H.E. SHOW at Newport Beach this year. Would I have noticed any difference?? Who knows but it would have been fun.



m
 
#8 ·
As for me through my system, I could not discern any difference.

It would have been interesting to have listened to the same files through a few of the extreme high-end audio systems at the T.H.E. SHOW at Newport Beach this year. Would I have noticed any difference?? Who know but it would have been fun.

m
I can reliably hear the difference on my laptop and headphone. No exotic system is needed. I hope everyone tries the files regardless of what system they have.
 
#14 ·
Age and Gender Selective Test!

For those under 30, are of the Female Gender under 60, but definitely not for those from the Construction Industry!

(May also exclude Younger Generation who like to listen with ear buds blasting out 110dB!)
 
#16 ·
Apologies if this is the wrong thread, but is anyone using Genelec 8260A active monitors for this test? I know they're highly regarded, and the specs ( http://www.genelec.com/products/8260a/ ) state: 26 Hz - 40 kHz (-3 dB).
 
#24 ·
Apologies if this is the wrong thread, but is anyone using Genelec 8260A active monitors for this test? I know they're highly regarded, and the specs ( http://www.genelec.com/products/8260a/ ) state: 26 Hz - 40 kHz (-3 dB).
I think those would be great speakers to use for this test. I would make sure there's no bottleneck limited to 20 kHz anywhere else in the system. I encourage you to give a listen and PM me with your results, the make and model of each component in your system, and how they are connected.
 
#17 ·
Interesting test

Heard a difference in the sound-stage presentation but that's about it. Need to try with IEMs and laptop.
 
#18 ·
I could easily determine the high res version for the Mosaic and On the Street Where You Live files. Just My Imagination was a toss up though and I guessed wrong on that one.

My set up was a Mac Pro Quad-Core Tower, a MusicStreamer II DAC, a Yamaha HTR-5550 receiver and Pioneer SP-B22 speakers.
 
#20 ·
No sure what you meant by ABX scores but after I made my determination I opened the files and looked at the Spectrogram to see the difference.

That cut off at 22kHz isn't gradual either which makes me think what is noticeable is more a re-enforcment of the lower harmonics (those that are actually audible) and which is absent in the CD quality file. Like when we add a sub-woofer and experience a fuller overall sound.

High resolution files might actually justify those super tweeters that have been around for some time. I'm looking forward to the new Sony - Core Series bookshelf speakers, that have a frequency response of 53Hz - 50kHz.
 
#23 · (Edited)
On my low cost set up an old MacBook with Sony MDR-V150 headphones ($20) plugged in it sounded good as follows
Just My Imagination A2 Mosaic B2 On The Street Where You Live A2

On a higher end set up with a Presonus Inspire audio interface plugged in via Firewire 400 to the MacBook
with a pair of AKG K271 studio headphones ($289 list) it sounded as follows: the opposite
Just My Imagination B2 Mosaic A2 On The Street Where You Live B2



my guess is the higher resolution is
Just My Imagination B2 Mosaic A2 On The Street Where You Live B2


great music and a interesting test

Trent Hinkforth
or Mr Eq
 
#32 · (Edited)
Originally Posted by Mpls-Sound View Post
I could easily determine the high res version for the Mosaic and On the Street Where You Live files. Just My Imagination was a toss up though and I guessed wrong on that one.

My set up was a Mac Pro Quad-Core Tower, a MusicStreamer II DAC, a Yamaha HTR-5550 receiver and Pioneer SP-B22 speakers.
Edit : Pretty good on Pioneer SP-B22 speakers lots of folks like those , have you tried phones if so how did that work out ? .
 
#33 · (Edited)
I Just played both of the Just My Imagination and also Mosaic A2 ,B2 downloaded files (exactly level matched)in a sound editor reading unzipped WAV files directly from HDD rather than importing them as a project to assure they remained unaltered.

I switched back and forth without knowing which rez was playing from the 2:00 min mark to the 3:00 min mark on each file repeatedly on this playback chain PC >Xonar STX 192 kHz amplified sound card> into Sennsheiser studio phones that can resolve 27 kHz with ofc no added EQ or effects .

FWIW or not (at this point ) IMO I could not reliably discern any difference in playback . I might try these files in the studio on the same phones on a pro external audio interface and see what shakes out .

I also tried samples on a different very flat and accurate pair of studio tracking phones same opinion , also from the 1 to 2 minute marks in addition to the 2 to 3 min marks same opinion both ways .

As a consumer I would want any differences to be readily apparent on an ordinary playback chain that is capable of resolving these resolutions .

One should keep mind ultimately consumer opinions are what matters in the marketplace not necessarily mfr marketing advertisement , commercial website or enthusiast/hobby magazine reviews .
 
#35 ·
One should keep mind ultimately consumer opinions are what matters in the marketplace not necessarily mfr marketing advertisement , commercial website or enthusiast/hobby magazine reviews .
And this consumer (i.e. me) has managed to tell the differences reliably in double blind ABX tests.

I suspect vast majority of people won't be able to tell the difference which is fine. They should not spend the money if they can't hear the difference (assuming they believe they never will). This is no different than other things in life. You may not be able to tell the difference between a $20 steak and $10 in which case, save the money and buy the $10 one. Just be careful that you don't dictate your discrimination in this regards to others.
 
#34 · (Edited)
Precise level matching is very important in any comparative analysis or differential testing of this type .

To better facilitate evaluation One could Play them in Audacity or another sound editor not WMP that will allow precise level matching ,you can usually drag and drop the unzipped downloaded WAV or any supported files straight in and read them directly from the hdd rather than importing them as a project ensuring no alteration of the original files

OTOH for testing purposes I'm not so sure you could use the Foobar 2000 ABX comparative facility in this way unless Audacity supports the plugin also. Maybe some one could enlighten us ? ( I know I'm being lazy ☺☺)

Unless those in the know feel that the playback gain settings in Foobar 2000 are sufficient ?
IMO In any event WMC or most players should probably not be used to evaluate and or test for differentials.
without level matching in the playback chain to arrive at a valid conclusion or opinion .
 
#37 · (Edited)
armin said:
And this consumer (i.e. me) has managed to tell the differences reliably in double blind ABX tests
Perhaps That means I will not be buying a new hires player and you might ? Not to worry my PC (s) and PS3' can do that just as well and the one in the studio likely much better .

armin said:
I suspect vast majority of people won't be able to tell the difference which is fine. They should not spend the money if they can't hear the difference (assuming they believe they never will). This is no different than other things in life. You may not be able to tell the difference between a $20 steak and $10 in which case, save the money and buy the $10 one. Just be careful that you don't dictate your discrimination in this regards to others.

Yes Armin you have stated that you can hear a difference on many occasions it does not prove or disprove anything to me (maybe others ?) I never disputed that but I am free to form my own suppositions and opinions and freely express both . Did I hit a sore nerve or something none my post's in this thread were directed at you
shouldn't have ?

Remarkable laptop you have there is it an HP Beats model by chance ? Seems odd my quite possibly superior (than many laptops) playback chain (s) can not resolve a difference no?..................perhaps not.

On the steak thing ........... depends on where you buy it and how it is prepared I've had bad restaurant steaks that cost well over $20.00. and very decent $9.00 BBq rib eyes prepared at home properly ( not too many at that price lately though beef has gone up a lot in case you haven't noticed or perhaps beef isn't in your regimen .)

I believe there is also a difference in perceiving something is different by differential testing with extraordinary scientific methods or otherwise just listening is much different than 'proving that difference in the marketplace ' as small as it may be is as an added value to the ordinary average consumer or enthusiast . Hires may be a fine niche market for those few that imagine or maybe hear *some differences whatever they may be* . Markets are full of similar things nothing new.


Lastly on the " dictate your discrimination in this regards to others." You might want to to look up the definition of "dictate" as opposed to expressing an opinion in free discussion .

No offense but I believe I do not need your advice as to what or how to post or not here at AVS they have crawl bots and moderators for that thank you .

Unless you bought the Forum and now dictate forum rules here or otherwise have repealed our first amendment freedom of speech protections we are both free to continue to express our opinions ,suppositions and unproven hypothesis (s) ,facts , or whatever 'test results' we care to as fact here likewise ....................... maybe nobody told you that ...........?

No statute laws,common law torts or any other laws broken when expressing opinion or hypothesis (s) or supposition .
Is a J.S. among your many remarkable accomplishments perhaps ? I studied some of that in college as an undergrad.

We should remember this is not your forum (or mine ) and we don't make the rules here but rather we should abide by them like everyone else.

Arnin, Maybe this discussion you felt the need to initiate should be in the appropriate hires debate thread rather than this one that is only soliciting opinions either way (like mine ) and no test results (either way ) not argumentative debate
You could have stopped at the end of first sentence or just let your opinions be known as did I ,you know that would IMO have been more appropriate in this thread.

Let's not turn this thread into and argumentative C./F. like the hiresdebate thread becomes every now and then
(if you know what I mean ) but continue to freely express our opinions as I have going forward as is the stated intent of this thread.

further might I suggest if you feel the need to reply to this post let's first quote it to the appropriate hires debate thread and if you feel the need we can resume this discussion there although I don't see the need . thank you.

@ Scott Wilkinson ,Sorry for my part that being limited to this post . .

PS Armin : I'm sure many us will be looking forward to seeing some new JAES or AES hires papers . Please let us know if you hear of any .
 
#38 ·
Dropped the on the street where you live unaltered WAV files into Audacity instead of importing them to avoid alteration
hoping that some small amplitude variations might be a tell for any differences or maybe which was which. I couldn't view any difference with a cursory look . I suspect that may only prove they are expertly level matched though and nothing beyond that .

 
#39 · (Edited)
From my Youtube video's notes:

I thought it might be useful to some of our AVSforum readers to share with them the same visual aid that some of us already benefit from, namely an ultra high frequency spectrum analyzer, to help hone in on the specific parts of the hi resolution test tracks which might make identifying them easier, so as to then later focus on that specific part of the track, in comparison to the lo-res version.

There are several very important things to know before you use my video as a visual aid:

#1. It is impossible to cheat in an ABX test by using this (particular) analyzer, even if you download it yourself, because the option to use it is blocked during actual foobar ABX testing, so I'm not introducing a new "cheat method".

#2. This is for demonstration and educational use only. Always use proper safety precautions and don't listen to the files at loud levels.

#3 The audio portion of the video is low quality and compressed by Youtube, so it is therefore NOT hi-res. To use this video properly you must synchronize its playback to your external hi resolution audio (HRA) playback system. Barring that, you can simply denote the time stamp (displayed on the bottom of the video image itself, not the outer Youtube window) to then later focus on that particular song passage after watching the video.

The video is made of three, hi resolution only tracks, played sequentially. The first and last track are the same: AVSforum member ArnyK's "keys jangling full band 2496 test tones f3.wav" [The f3 part denotes it is his newest, third attempt which makes some slight alterations, and fixes some minor bugs.] He provides a download link to both it and a low res version to compare it to, at the bottom of post #2498 , or so, in the thread : http://www.avsforum.com/forum/91-au...cott-s-hi-res-audio-test-84.html#post26085394

The 2nd high resolution track, sandwiched between identical tracks #1 and #3, is the pristinely recorded song "Mosaic", kindly provided by Mark Waldrep, founder and chief engineer of AIX Records. The track selected is the current front runner for "most favorite" by people who have submitted their test results to Scott Wilkinson's thread "AVS/AIX Hi-Resolution Audio Test: Take 2", where the track is offered for download in the opening post, both in standard and Hi-Res forms [but with only letter codes so as to not divulge which is which, for testing purposes]. This and his other tracks were selected both for their extended dynamic range and extremely high frequency content, which exceeds the capabilities of the CD format.

The "Mosaic" cut needs no introduction. It is simply one of the best, most finely crafted audio recordings I have ever heard, even through my rather modest audio system. Although both the dynamic range and high frequency extension is clearly apparent, it is done without exaggeration and is completely natural.

The jangling keys sound, on the other hand, can be a little annoying [even though I'm sure Arny's recording is spot on accurate with his use of high-end, B&K 4006 mics] and are known to have more acoustical power in the ultrasonic range than any conventional musical instrument, so as you'll see, the bouncing bar graphs show a very strong signal up there. After 12 seconds of this there is a "training tone" at 4 kHz which prompts the listener as to what to listen for should their system be plagued by IM distortion in the next, brief test tone section. Hearing uniform clicks after this tone is expected, however if you hear faint 4kHz tones after this first one, or other noises (besides the clicks common to both the normal and hi-res versions), you unfortunately have an IM problem.

The last 12 seconds of Arny's track demonstrates what the ultrasonic frequencies sound like when played in isolation from the lower, CD quality only range (which might otherwise mask the ultrasonics from our perception). He has kindly filtered away for us the lower frequencies so we can attempt to listen to just these pure, ultrasonic only sounds as a learning tool.

[I recommend double clicking the video, once it starts playing, to see it full screen. Your browser must allow pop-ups for that to work, I believe.]

 
#43 ·
... The last 12 seconds of Arny's track demonstrates what the ultrasonic frequencies sound like when played in isolation from the lower, CD quality only range (which might otherwise mask the ultrasonics from our perception). He has kindly filtered away for us the lower frequencies so we can attempt to listen to just these pure, ultrasonic only sounds as a learning tool.

_________________

[I recommend double clicking the video, once it starts playing, to see it full screen. Your browser must allow pop-ups for that to work, I believe.]

https://www.youtube.com/watch?v=mOeCt8EL-9E
That is a great/peaceful tune ♪♫

 
#40 ·
m. zillch,

Nice job I will have to play it with the orig downloaded WAV files as you kindly suggested.

......... What if all the music files here are same same res and we are getting hosed ?:D.......... just kidding .
 
#42 ·
I love the idea of the test, but i would like to warm everyone.
Make sure you don't play anything too loud and don't try to hard to distinguish the differences.
I spent a lot of time on testing and trying to catch the differences between different audio files.
few months ago I could hear up to 18400hz which is pretty good as all my friends were around 14000-15000hz.
Above 16000hz it was very difficult to feel any difference. The details are so detailed that you really need to be good to tell the difference.
I could hear supersonic sound generated by audio generator but in music it was hard to hear any differences above 16000Hz.
Long story short, after all the testing i ended up with ringing in the ears.
Now i live with constant freq at around 17000-18000hz.
Trust me, you dont want this. It will ruin your life.
Please be cautious when doing all there tests.
Never go above 75-80DB to hear the difference.
If you cannot hear it at that level there is none or you just cannot perceive it.

I listened to your songs. Sometimes i feel difference sometimes i dont,
I am not sure yet.
 
#45 · (Edited)
I love the idea of the test, but i would like to warm everyone.
Make sure you don't play anything too loud and don't try to hard to distinguish the differences.


Long story short, after all the testing i ended up with ringing in the ears.
Now i live with constant freq at around 17000-18000hz.
Trust me, you dont want this. It will ruin your life.
Please be cautious when doing all there tests.
Never go above 75-80DB to hear the difference.
If you cannot hear it at that level there is none or you just cannot perceive it.
Agreed; safety first.
Ringing in the ears, tinnitus, is sometimes temporary, sometimes not. Let's hope yours is temporary.


http://www.webmd.com/a-to-z-guides/understanding-tinnitus-basics
 
#44 · (Edited)
^Hearing that part my dog stands up, cocks her head [cleverly, she knows she has better HF sensitivity on-axis] and goes into some sort of mesmerized trance. Me? I don't hear zip then, which I'm confident is true of anyone older than their teens.



Ten points to the first person who can name not just this c.1900 dog, but also his puppy, first introduced 91 years later in TV ads. No cheating.


Wikipedia:
"For sound localization in the median plane (elevation of the sound) also two detectors can be used, which are positioned at different heights. In animals, however, rough elevation information is gained simply by tilting the head, provided that the sound lasts long enough to complete the movement. This explains the innate behavior of[vague] cocking the head to one side when trying to localize a sound precisely. To get instantaneous localization in more than two dimensions from time-difference or amplitude-difference cues requires more than two detectors"
 
Status
Not open for further replies.
You have insufficient privileges to reply here.
Top