Head-Fi.org › Forums › Equipment Forums › Sound Science › 320 kbps MP3 vs. normal audio CD listening Sound quality
New Posts  All Forums:Forum Nav:

320 kbps MP3 vs. normal audio CD listening Sound quality - Page 27

post #391 of 516
Quote:
Originally Posted by skamp View Post

Better use foobar's ABX component or any other proper ABX software, if you want to conduct a proper test.

Like chewy4 said, in order to ABX 16 bit vs. 24 bit, just take a 24 bit file, downconvert it to 16 bit and upconvert it back to 24 bit.

You want to set a number of trials beforehand (12 is good, 16 if you like) and stick to it. Hide the results until you're done. Save the log, look at the score: a statistically significant result (i.e. successful) is inferior to a 5% probality that you're just guessing.

But you cannot playback 16 bit and 24 bit in Foobar using WASAPI without changing the settings between tracks.. So you are suggesting truncating the 24 bit files to 16 bit to do the test?

post #392 of 516
Quote:
Originally Posted by jvandyk View Post

But you cannot playback 16 bit and 24 bit in Foobar using WASAPI without changing the settings between tracks.. So you are suggesting truncating the 24 bit files to 16 bit to do the test?


hmm I think he was expecting that its common sense to set output mode to 24bit... what the post saying is on how to produce the 16bit test sample from the 24bit counterpart

post #393 of 516
Quote:
Originally Posted by jvandyk View Post

But you cannot playback 16 bit and 24 bit in Foobar using WASAPI without changing the settings between tracks.. So you are suggesting truncating the 24 bit files to 16 bit to do the test?

 

He's saying compare two 24 bit files, one of which is actually a 16 bit file that has been upconverted back to 24 bit. Because they're both recognized as 24 bit (even though one only has 16 bit information in it) you won't have to change  the settings and should be able to do a proper abx test.

post #394 of 516

my mind is blown

post #395 of 516
Quote:
Originally Posted by gnarlsagan View Post

 

He's saying compare two 24 bit files, one of which is actually a 16 bit file that has been upconverted back to 24 bit. Because they're both recognized as 24 bit (even though one only has 16 bit information in it) you won't have to change  the settings and should be able to do a proper abx test.

So downsample one 24 bit file and reconvert back to 24 bit using the Foobar processing versus playing back in 16 bit and 24 bit native.  This would rely on Foobars ability to downsample and upsample..from a program that is already limited.

 

I would like to compare WASAPI (bit perfect) files played back at 16 bits and 24 bits in their native formats. No Foobar downsampling and upsampling.

 

Has anybody ever actually done this here? If so, where are the results?


Edited by jvandyk - 1/24/13 at 3:38pm
post #396 of 516
Quote:
Originally Posted by jvandyk View Post

So downsample one 24 bit file and reconvert back to 24 bit using the Foobar processing versus playing back in 16 bit and 24 bit native.  This would rely on Foobars ability to downsample and upsample..from a program that is already limited.

Adding 8 zeros is a trivial operation. This has nothing to do with down/up-sampling, the sampling rate doesn't change if you pad 16 bit to 24 bit samples.

16 bit -> 24 bit -> 16 bit is a lossless operation.

 

I do not know what you mean with limited. All I see is your limited understanding of what's going on.


Edited by xnor - 1/24/13 at 3:47pm
post #397 of 516

Here is a nice read from Robert Harley on the subject.

http://www.avguide.com/forums/blind-listening-tests-are-flawed-editorial

post #398 of 516
Quote:
Originally Posted by xnor View Post

Adding 8 zeros is an very simple operation. This has nothing to do with down/up-sampling, the sampling rate doesn't change if you pad 16 bit to 24 bit samples.

16 bit -> 24 bit -> 16 bit is a lossless operation.

 

I do not know what you mean with limited. All I see is your limited understanding of what's going on.

I guess you forgot about the first step- truncating the file. Got the second part right though!

 

And really, does every reply to me have to have another abhom?

post #399 of 516
Quote:
Originally Posted by jvandyk View Post

Here is a nice read from Robert Harley on the subject.

http://www.avguide.com/forums/blind-listening-tests-are-flawed-editorial

Quite funny how examples of a few flawed tests lead this guy to concluding that all blind listening tests are flawed. By the same "logic" math must be flawed because a lot of students make calculation errors.

 

Quote:
Originally Posted by jvandyk View Post

I guess you forgot about the first step- truncating the file. Got the second part right though!

 

And really, does every reply to me have to have another abhom?

Why would you truncate anything? I thought you wanted to compare 16 to 24 bit files.

 

I guess you meant ad hom with "abhom".

post #400 of 516
Quote:
Originally Posted by jvandyk View Post

So downsample one 24 bit file and reconvert back to 24 bit using the Foobar processing versus playing back in 16 bit and 24 bit native.  This would rely on Foobars ability to downsample and upsample..from a program that is already limited.

 

I would like to compare WASAPI (bit perfect) files played back at 16 bits and 24 bits in their native formats. No Foobar downsampling and upsampling.

 

Has anybody ever actually done this here? If so, where are the results?

24bit to 16bit conversion is very simple, either direct cut-down or with dither enabled (which I think should be better).

16bit to 24bit is even more simple, no conversion is involved, just a process of filling in 0's.

 

The conversion can be done in foobar, just "right click the song > convert to..."

post #401 of 516
You aren't listening to what people are saying to you.
post #402 of 516
Quote:
Originally Posted by kn19h7 View Post

24bit to 16bit conversion is very simple, either direct cut-down or with dither enabled (which I think should be better).
16bit to 24bit is even more simple, no conversion is involved, just a process of filling in 0's.

The conversion can be done in foobar, just "right click the song > convert to..."

Yes, you suggest truncating one file and leaving the other alone with foobar. I understand that and reject that is the same as playing native 16 bit on a 16 bit dac versus native 24 bit on a 24 bit dac.
post #403 of 516
Quote:
Originally Posted by jvandyk View Post


Yes, you suggest truncating one file and leaving the other alone with foobar. I understand that and reject that is the same as playing native 16 bit on a 16 bit dac versus native 24 bit on a 24 bit dac.

Are you worried that the conversion of the 16 bit file is going to cause problems in the file? It's not, but if it did wouldn't that make it even easier to tell the difference?

post #404 of 516
Quote:
Originally Posted by xnor View Post

Quite funny how examples of a few flawed tests lead this guy to concluding that all blind listening tests are flawed. By the same "logic" math must be flawed because a lot of students.

Yes, I do think blind test listeners are more interested in math!

Edited by jvandyk - 1/29/13 at 12:11pm
post #405 of 516
Quote:
Originally Posted by chewy4 View Post

Are you worried that the conversion of the 16 bit file is going to cause problems in the file? It's not, but if it did wouldn't that make it even easier to tell the difference?

Yes. Altering 8 bits of information from the 24 bit file in order to do a so called blind test is a flawed test IMO. One has to assume all 8 bits of that truncated info is not audible for it to be a valid test. Not going to waste my time with that. In the Harley article I linked earlier, Bob Stuart from Meridian said comparing 16 bit masters to 24 bit was like turning on a light switch. Not different mastering, just different bits. Most subjective writers agree with Stuart. Of course, this thread is really about comparing low bit rate 320 to 16 bit. Again, one has to assume the roughly 75% of data removed from the 320 is not audible to honestly say they are the same.

But like I said earlier, I do think 320 sounds surprisingly good considering. 256k is clearly audibly worse than lossless in my opinion.
Edited by jvandyk - 1/29/13 at 12:23pm
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Sound Science
Head-Fi.org › Forums › Equipment Forums › Sound Science › 320 kbps MP3 vs. normal audio CD listening Sound quality