[arnyk]

  You don't need listening tests to demonstrate subsample timing resolution of 16/44.1. You can demonstrate it with measurements very easily.
 
Now, if you wanted to demonstrate the audibility of subsample timing shifts, you'd need an ABX, but if you just want to demonstrate that the format is capable of that kind of timing resolution, a simple measurement on an oscilloscope will suffice. It's shown very clearly at around 21 minutes in this video:
 
http://xiph.org/video/vid2.shtml

 
The files for doing an ABX study of this issue can be found here:
 
http://www.hydrogenaud.io/forums/index.php?showtopic=107570&view=findpost&p=899713
 
The designated sample delays are for 24/192. Since we've got positive ABX results for a 2 sample dealy, the ability of 16/44 to encode and decode subsample delays is supported.
 
People who talk about "The evidence of their ears" should download these files and FOOBAR2000 and its ABX plug-in and obtain some real and genuine "Evidence of their own ears".  No out-of-pocket cash required.

 

[FFBookman]

  Because lots of tests have shown that it's an exceedingly subtle difference, only audible with very careful, fast-switch comparisons on specially chosen "killer" samples? You're the one making the claim that goes against the preponderance of current evidence, so you have to support it or we will assume that you're talking nonsense (and it seems that in this case, that's a pretty safe assumption).

Show me an ABX test that proves anything sounds better than anything else. 

 

[RRod]

 
Show me an ABX test that proves anything sounds better than anything else. 

 
ABX is for testing if differences can be heard, not about testing preference.

 

[FFBookman]

https://youtu.be/r_wxRGiBoJg
 
discuss...

 

[FFBookman]

   
ABX is for testing if differences can be heard, not about testing preference.

differences between a memory and a real sound.
or preferences between  a memory and a real sound?
 
either way, you aren't comparing the two directly.  memory of sound and sound processing are done by two very different parts of the brain.  
 
this invalidates ABX tests for mixed music quality testing, as far as i'm concerned.

 

[RRod]

 
differences between a memory and a real sound.
or preferences between  a memory and a real sound?
 
either way, you aren't comparing the two directly.  memory of sound and sound processing are done by two very different parts of the brain.  
 
this invalidates ABX tests for mixed music quality testing, as far as i'm concerned.

 
If I walked into a room and you had a favorite movie of mine on, and you asked me if it were in 720p or 1080p, I might have a hard time telling. If, while I was watching the movie, you suddenly changed it from 720p to 1080p, I'd be able to tell (assuming the screen size is large enough). My "memory" of the movie would have nothing to do with it.

 

[maverickronin]

  differences between a memory and a real sound.
or preferences between  a memory and a real sound?
 
either way, you aren't comparing the two directly.  memory of sound and sound processing are done by two very different parts of the brain.  
 
this invalidates ABX tests for mixed music quality testing, as far as i'm concerned.

 
You might want to rethink that logic a little bit...
 
Since it's also impossible to compare 2 sounds by playing them at the the same time, that only leaves memory.  If memory is also invalid for comparisons then there's no way left for humans to compare any sounds at all.  Therefore it's impossible for you to definitively state the24/192 FLAC sounds any different than 32kbps created from a circa 1996 encoder because there is no valid way for humans to make such comparisons.

 

[FFBookman]

   
If I walked into a room and you had a favorite movie of mine on, and you asked me if it were in 720p or 1080p, I might have a hard time telling. If, while I was watching the movie, you suddenly changed it from 720p to 1080p, I'd be able to tell (assuming the screen size is large enough). My "memory" of the movie would have nothing to do with it.

Visual is different, we can flash an image in our mind and overlay that for a quick comparison.
 
I see pixels, I can roughly estimate how many there are in a small section of the video that I focus on. You change resolutions, I rescan and recount the pixels in that area and determine if resolution has gone up or down.
 
Also all of the site data comes in through the eyes. If we imagine an image overlay, it appears in our view just like the original.
 
 
But our ears do not work that way, they cannot overlay a previous sound image over an existing one. 
 
The sound exists only in our imagination. We have to reconstruct what/who/where the sound was made and we do it instantly.
 
Real sound is vibration and enters our body from all angles, and every hair follicle has a mechanoreceptor that transmits vibration data to the brain. As your head and body move the sound is adjusted and analyzed. There's also far more emotion generated by the music than the flickering screen.
 
 
 
Finally -- if you knew the TV test was coming and you reviewed the movie in 720p and 1080p before the test looking for tells, you would be able to go right to that section and determine the resolution.
 
Same way you can find crash cymbal decays, phase shifts, hi hat detail, and string noise in 24bit audio.

 

[RRod]

So you accept ABX as a valid means of testing differentiation of screen resolutions?

 

[icebear]

 
Visual is different, we can flash an image in our mind and overlay that for a quick comparison.
 
I see pixels, I can roughly estimate how many there are in a small section of the video that I focus on. You change resolutions, I rescan and recount the pixels in that area and determine if resolution has gone up or down.
....

 
ROFL,
are there any instructions how to learn this feat

.  Rainman can learn something from you for sure. Can you also do the dumped on the floor match trick - I mean these are like a 100 count max. but screen pixels will for sure be a 100 times more.
This is great stuff !!!

 

[FFBookman]

  So you accept ABX as a valid means of testing differentiation of screen resolutions?

I believe that it will be more accurate when done visually, because we can overlay and/or store off to the side of our visual space a rough approximation of the image while looking at the current image.
 
Audio just doesn't work like that, you can't compare back, there's no visual cues to temporarily store.

 

[FFBookman]

   
ROFL,
are there any instructions how to learn this feat

.  Rainman can learn something from you for sure. Can you also do the dumped on the floor match trick - I mean these are like a 100 count max. but screen pixels will for sure be a 100 times more.
This is great stuff !!!

Maybe my explanation was funny to you but it's very obvious to me, not funny at all.
 
If an image is on a screen I can quickly search for lines, shading, and anti-aliasing to determine the granularity of the image. I can take an educated guess based on known resolutions before the switch is even made.
 
Then if you switch the resolutions I can stare at the exact same spot or find a similar visual pattern and look at jaggies again. If there is about the same as before, resolution stayed the same. More jaggies, less resolution. Very simple, we all do this everyday.
 
Visually we can keep a ghost/memory of the detail and compare it intimately with a new image. Sound cannot be stored and recalled in the same way, and it cannot be ghosted/or freeze-framed in our mind.
 
Sorry I don't have the psychosomatic wherewithal to explain further. Laugh your way to further reading.

 

[FFBookman]

   
ROFL,
are there any instructions how to learn this feat

.  Rainman can learn something from you for sure. Can you also do the dumped on the floor match trick - I mean these are like a 100 count max. but screen pixels will for sure be a 100 times more.
This is great stuff !!!

You don't need rainman to determine if there's 256 matches or 5600 matches on the floor.

 

[icebear]

Throwing movies doesn't work either

 

[bfreedma]

Quote:

Originally Posted by FFBookman


Maybe my explanation was funny to you but it's very obvious to me, not funny at all.



If an image is on a screen I can quickly search for lines, shading, and anti-aliasing to determine the granularity of the image. I can take an educated guess based on known resolutions before the switch is even made.



Then if you switch the resolutions I can stare at the exact same spot or find a similar visual pattern and look at jaggies again. If there is about the same as before, resolution stayed the same. More jaggies, less resolution. Very simple, we all do this everyday.



Visually we can keep a ghost/memory of the detail and compare it intimately with a new image. Sound cannot be stored and recalled in the same way, and it cannot be ghosted/or freeze-framed in our mind.



Sorry I don't have the psychosomatic wherewithal to explain further. Laugh your way to further reading.







Can you cite your source for the highlighted quote? I see no reason for the differentiation.