[arnyk]

 

   
I would surprise me if any hi-res vendor would now willingly donate material to such a test, especially if results were again destined for a professional publication. Too much to lose; nothing to gain.

 

But why is a vendor required? Could they not just download commercially available hi res files, purchase some DVD A's and SACD's and do the test themselves?

 
Its been done many times with negative results.

 

[kar13]

   
 
 
 
I don't see where in the above any of the points I raised were even touched on, so the entire above makes no sense to me.  Perhaps a little fact-based discussion of the points raised rather than the wave of a hand and flat dismissal would help?
 
For example, using IEMs addresses zero of the points I raised.  They are fine, I use headphones, earphones and loudspeakers about equally often, but so what?
 
One of the points - the first point about audiophile evaluations not even being tests, can't be "taken care of" because it is an inherent failing.
 
The casual apparent  attempt at sort-of randomization shouldn't impress anybody who even took just first year statistics.

Please make an effort to understand, instead of accusing somebody with flat dismissal or wave of hand or whatnot.Or maybe I'm not understanding your rhetoric, as there is a miscommunication.
 
iem's matter, because of the passive noise reduction, which helps in concentrating on the tracks, which implies an unmeasured improvement in listening, no matter how small.
 
Even though this is going to be a regurgitation of precautions which were diligently carried out, will attempt it one last time.
If you still feel the same, please tell me the manner in which, you would perform another test to obviate the very issues you raised.I would be glad to do that.
 
(2) was taken care of by the tracks being burned in random manner, as you were unable to glean from me having mentioned it before..
 
(3)I had mentioned in my original post that level matching had been done!I don't change the loudness level in the middle of a listening session, using loudness meter and gain adjusted accordingly.
 
(4) was again taken care as I have listened to the same duration of interval especially with the 2 different versions of Take Five.Not just the different versions of the tracks, but also their bits were burned to create a wav file which was then used for listening purpose.It was also burned to a
 
Your 1st point subsumes 5th point, you made.
I have no vested interests as I don't own any expensive equipment, and no known bias/prejudice resulting from any psychological/physical reasons, afaik.If you are kind enough to point them out, i will make another A/B test to try and include it.Otherwise, you can happily assume that my results are a biased/subjective.
 
But, that was kind of my point with my original post that experience is the only test.

 

[arnyk]

 

Originally Posted by kar13 /img/forum/go_quote.gif  
iem's matter, because of the passive noise reduction, which helps in concentrating on the tracks, which implies an unmeasured improvement in listening, no matter how small.
 

 
IEMs are just a means to an end that can be achieved by a number of other means. Looks to me that by overstating thier benefits you are giving them unique properties that they don't have. Sure they matter, but they aren't the only useful tool and they don't cure the biggest problems with your evaluations.
 

[(2) was taken care of by the tracks being burned in random manner, as you were unable to glean from me having mentioned it before..

 
Wrong. I had figured out that you used CDs you burned with purportedly random tracks as your randomizing technique and have already said as much.  Enough randomized tracks for good statistics is 16 which is an awkwardly large number of random tracks to burn, but there is still this problem - maintaining their order as a secret from you, particularly if you are the one who burned them.
 
 

(3)I had mentioned in my original post that level matching had been done!I don't change the loudness level in the middle of a listening session, using loudness meter and gain adjusted accordingly.
 

Loudness meters are generally not reliable or precise enough to provide the +/- 0.1 dB level matching that is required.

 

 
(4) was again taken care as I have listened to the same duration of interval especially with the 2 different versions of Take Five.Not just the different versions of the tracks, but also their bits were burned to create a wav file which was then used for listening purpose.It was also burned to a
 

 
I doubt that you were able to maintain the ca. 1 millsecond time matching required by these means.
 

I have no vested interests as I don't own any expensive equipment, and no known bias/prejudice resulting from any psychological/physical reasons, afaik.If you are kind enough to point them out, i will make another A/B test to try and include it.Otherwise, you can happily assume that my results are a biased/subjective.

 
The above appears to be a claim that you know what your personal biases are and can control them. This is simply not possible for anybody.
 
 

 

[FFBookman]

 
I have tried and haven't found any difference between 16/44 cd and 16/44 from 24/96.
 
But i have a question..
is the effect of different downsampling algorithms used noticeable to a discerning ear..because of dithering?

You are correct that every dithering algorithm sounds different. At least every one I've heard (about 6 total) sound different on my rig.  
 
This was tested when reviewing music I mixed at 24bit that was returned to me with various dithers applied when downsampling to 16bit. My engineer wanted me to blindly pick my favorite.
 
Most of the people arguing about 16bit being max resolution have never tested different dithers on their own mixes, so they are not intimately familiar with how it sounds.
 
I can't say I have a favorite, or can really describe the differences better than saying the size, shape, and timbre of the overall room changes. When you spend many hours mixing a piece of music at 24bit you know exactly where each instrument sits and how the various voices blend together on the EQ.
 
The dithers that have 'noise shaping' impart an artificial boosted sound to my ears. I think when targeting mp3 format the stuff is used to brighten and thin the sound a bit before lossy time.
 
That said - dither is critical. Turning it off and leaving downsample artifacts is a bad idea.
 
It seems to me that if CD's would have been 20bit/48k we wouldn't even have 'hi-res' now, because that's probably enough bandwidth for no-one to want more. I think you can push about 3000k true bitrate though that format, which is enough for most commercially recorded music.

 

[FFBookman]

btw Kar13 - be prepared for every angle of absurdity telling you that you can't hear, don't understand, and are just a flat out idiot for even thinking that we have any better digital audio than 1978.
 
they scream ABX tests but they can't point to one ABX test that shows any bit of pre-recorded music is better than any other.  the test is not used for that but that doesn't stop some from claiming everything is garbage unless it passes an ABX test.
 
most of us can't pass ABX tests consistently even with major changes in quality (like 8bit files). this is because of flaws in the test make it completely inappropriate for this use. it will only result in confusion, which is why the side of "there is no such thing as quality" use them.
 
 
Hey guess what Arnyk -- 50% of people can't reliably pick out the differences between a $10,000 mic and a $10 mic when listening to some random person talking through their computer.  Does this make the two mics the same?  Does this make Aretha Franklin want to sing through the $10 mic?  Will it sound the same when Aretha sings through it in a recording studio?  Who even cares what mic she uses besides Aretha Franklin and her producer?  Do artists you love dissapoint you with this audio snake oil of using quality instruments and microphones?
 
The point is this is an artistic discussion. If you bring horrible content or measuring equipment only to the debate you won't get far.  You've got to account for the artistry of the creation and production of the music, and let the listener make the choice. What I really hate about some on this board is how they take this stance that you are just an ignorant fool if you want better quality.

 

[arnyk]

 
I have tried and haven't found any difference between 16/44 cd and 16/44 from 24/96.

 
That is known as evidence. It is evidence that answers the question below.
 
 

 
But i have a question..
is the effect of different downsampling algorithms used noticeable to a discerning ear..because of dithering?

 
A conversion from 24/96 to 44/16 involves downsampling, digital filtering and dither.
 
If you haven't found any difference between 16/44 cd and 16/44 from 24/96 then you have combined downsampling, digital filtering and dither and found no difference.
 
Why would all three cause no audible difference, while just one of them cause an audible difference?
 
They are all different, independent and don't cancel or augment each other. 

 

[RRod]

 
most of us can't pass ABX tests consistently even with major changes in quality (like 8bit files). this is because of flaws in the test make it completely inappropriate for this use. it will only result in confusion, which is why the side of "there is no such thing as quality" use them.
 

 
I can consistently pass ABX on 8bit files, if the original file is of high enough dynamic range, which is the exactly what the bit depth affects. The second I get to some über-compressed metal track, suddenly the loss of bits becomes inaudible. For any given track in my collection, there is some bit/sample rate spec at which I can finally pass an ABX.

 

[interpolate]

I can't help thinking people are worrying about things that yet to happen or even matter.

 

[castleofargh]

when the problem is about an relatively effective test contradicting another relatively effective test, it's ok to argue and often hard to know who's right. but when it's a guy with his delusions vs what most human testing is telling us(even our own controlled testing). it's not a matter of opinions anymore.
 
 
bookman makes it clear that he knows his stuff and that we're all just too ignorant to recognize the obvious. if only we had some other guys working in audio who could also tell us all about the obvious differences of dither and all things occurring below -90db, or how ABX is limited. if only hydrogen audio existed, if only Ethan Winer had made a video an some conferences talking about his own experiments on the matter and explained it all in great details, if only we had maybe someone on this very topic who is at the origin of ABX, to tell us about the pros and cons of such a test.
 
too bad none of those guys exist or talk on the net, so obviously we should keep on listening to bookman telling his stories without any form of evidence of all the times where his ego TKOed reality in a humiliating way. or tell Arny about how abx isn't objective enough but his ears are.

 

[Ruben123]

So what exactly is the problem with abx? That is doesn't give the results some want to see? If you can hear 6 different sorts of downscaled dither music samples... Why can't you hear that blind? And IF you can hear them blind, then that's sort of the same as abx. Mr Bookman.

 

[dazzerfong]

So what exactly is the problem with abx? That is doesn't give the results some want to see? If you can hear 6 different sorts of downscaled dither music samples... Why can't you hear that blind? And IF you can hear them blind, then that's sort of the same as abx. Mr Bookman.

What's wrong with ABX is that it doesn't fit with his agenda.​ Ergo, in his mind, it's pointless.

 

[StanD]

I see you guys have thrown the book at bookman. 

 

 

[castleofargh]

  I see you guys have thrown the book at bookman. 

 

I've tried the book, the baby, the bathwater, the bathtub, I even tried being nice once(I didn't like it), nothing seems to work. no way to get any evidence, no way to have him be as critical and skeptic of himself as he can be of abx, mp3 or everything else..

 

[Gr8Desire]

 
You are correct that every dithering algorithm sounds different. At least every one I've heard (about 6 total) sound different on my rig.  
 
This was tested when reviewing music I mixed at 24bit that was returned to me with various dithers applied when downsampling to 16bit. My engineer wanted me to blindly pick my favorite.
 
Most of the people arguing about 16bit being max resolution have never tested different dithers on their own mixes, so they are not intimately familiar with how it sounds.
 
I can't say I have a favorite, or can really describe the differences better than saying the size, shape, and timbre of the overall room changes. When you spend many hours mixing a piece of music at 24bit you know exactly where each instrument sits and how the various voices blend together on the EQ.
 
The dithers that have 'noise shaping' impart an artificial boosted sound to my ears. I think when targeting mp3 format the stuff is used to brighten and thin the sound a bit before lossy time.
 
That said - dither is critical. Turning it off and leaving downsample artifacts is a bad idea.
 
It seems to me that if CD's would have been 20bit/48k we wouldn't even have 'hi-res' now, because that's probably enough bandwidth for no-one to want more. I think you can push about 3000k true bitrate though that format, which is enough for most commercially recorded music.

 
Every dithering algorithm may indeed be different. Applying the same dithering algorithm to the same source twice will generally give different results.  
 
Why? Dithering i used to insert random noise in the absence of content.  
 
Contrary to what several people have suggested: Changing bit depth does not induce dithering. Whether up sampling or downsampling, you simply interpolate or remove content to get a clean result.
 
Changing sample rates will often require dithering to estimate missing data points in the new time base. Convert to an integral sampling rate, and you will not get dithering. 44/88/176K are all compatible sample rates that will not induce dithering when up or downsampled. In all cases, interpolation or truncation is used to do a conversion. 48/96/384K are also compatible sample rate groups. You simply should not mix groups. Playing 44K content at 384K may seem like a good idea. It's not. Playing 96K content at 44K is also bad. 176K content played at 44K will sound perfect.  Same goes for 48K content played at 96K.

Not rocket science.  
 

 

[RRod]

   
Every dithering algorithm may indeed be different. Applying the same dithering algorithm to the same source twice will generally give different results.  
 
Why? Dithering i used to insert random noise in the absence of content.  
 
Contrary to what several people have suggested: Changing bit depth does not induce dithering. Whether up sampling or downsampling, you simply interpolate or remove content to get a clean result.
 

 
What do you mean "induce" dithering. You either add dither or you don't. The point of it is to randomize the errors due to quantization, substituting broadband noise for the truncation distortion due to the rounding. Quantization intimately relates to the bit depth, not the sample rate, as the number of bits determines the numerical set onto which we must map the samples.