Head-Fi.org › Forums › Misc.-Category Forums › Music › Comparing 44/16 to hi-rez 96-192/24
New Posts  All Forums:Forum Nav:

Comparing 44/16 to hi-rez 96-192/24 - Page 2

post #16 of 20
Quote:
Originally Posted by Brooko View Post
 

Even when encoding, although you are using the same source file, the encoder itself will often have the transcoded file at a slightly different volume level to the original.  And as you're aware, even 0.5 dB difference is enough to tell two files apart if your hearing is reasonably acute.  So that's issue number one.  The second issue is using the iPod to A/B - it doesn't have the software to do it.  The easiest way is as per my guide - Foobar 2000 + abx comparator + built in volume matching + 15 iterations of each track tested.  It really is an eye opener.

 

For anyone wanting to know more  - I know you've seen the guide MA - link is here : http://www.head-fi.org/t/655879/setting-up-an-abx-test-simple-guide-to-ripping-tagging-transcoding

 

One of these days you'll get the chance to try it.  I can promise you that if you've set everything up correctly, it will change your mind about what is really audible ;)

hehe...I went to bookmark the link, but it had already been bookmarked. :beerchug:

 

I actually had someone create a custom command line encoder for me in order to use QuickTime AAC with dBpoweramp.

 

iPods support ReplayGain, which seems to be all you propose in order to do volume matching.

 

The Rockbox firmware presents more opportunities - though I don't know if it's customizable enough to conduct this type of test.

 

Here's a crazier question that is somewhat relevant to this thread: since AIFF files reverse the byte order, do you think it's possible for them to sound different than WAV files under certain circumstances, despite both of them being uncompressed? It's just that many people (including famous studio professionals with decades of experience) claim to hear differences between lossless formats, and if this is so, it would only be due to how different devices process the data, which is identical after decoding. I had discounted all of this for a while, but recently began comparing again...and AIFF files seem to sound slightly more harsh and less natural than WAV when played back on my iPod Classic. The most likely scenario is that it is simply my imagination playing tricks on me, but others have reported this phenomenon as well, so I feel it's worth looking further into. If only I had the equipment to measure its analog output while playing the files...

post #17 of 20

Again - in order to make any test as foolproof as possible, the first thing we have to do is take away any outside influences (sight, volume differences, switching differences etc).  Then we have to make the test repetitive.  It's no use getting it right once - we have to show we can do it statistically the majority of the time (high 90%) in order to be relevant.

 

As AIFF and WAV are both outputting the same bitperfect pcm file - I fail to see how anyone could tell the difference.  I think it is telling that whilst there are a heap of blind tests showing that differences (when tested properly) can't be discerned, but still a lot of people talk about anecdotal evidence of so-called experts being able to tell the difference - yet there are no real tests done under proper conditions that actually show they can.  That speaks volumes to me.  If the differences can be discerned - then where is the evidence?

post #18 of 20
Quote:
Originally Posted by Brooko View Post
 

Again - in order to make any test as foolproof as possible, the first thing we have to do is take away any outside influences (sight, volume differences, switching differences etc).  Then we have to make the test repetitive.  It's no use getting it right once - we have to show we can do it statistically the majority of the time (high 90%) in order to be relevant.

 

As AIFF and WAV are both outputting the same bitperfect pcm file - I fail to see how anyone could tell the difference.  I think it is telling that whilst there are a heap of blind tests showing that differences (when tested properly) can't be discerned, but still a lot of people talk about anecdotal evidence of so-called experts being able to tell the difference - yet there are no real tests done under proper conditions that actually show they can.  That speaks volumes to me.  If the differences can be discerned - then where is the evidence?

 

Good points.

 

The concept is that certain components don't process the data in the same way, thus outputting different audio. (Hypothetically, of course.)

 

What I want to do more than anything (regarding digital audio files) is record the analog output of various devices playing different file formats. That could produce somewhat conclusive data of great benefit to both skeptics and believers, since even the slightest inaudible difference should be picked up in the recorded measurements. So if there is no difference (on that particular system), those who claim to hear one can see that there actually isn't, and vice versa.


Edited by Music Alchemist - 8/13/14 at 1:44pm
post #19 of 20

Oh - I have no problem with there being differences.  We know without doubt that there are differences between aac256 and FLAC for instance.  They can easily be shown / measured.

 

My issue is - are they audible by humans? (and each individual should be investigating that themselves to determine their own threshold IMO).

 

Sadly - that is where the disagreement is.  And even sadder - the ones that claim to hear the differences (especially those that claim differences are easily discerned), are usually the ones who have never undergone a properly set-up test, and who also refuse to do so.  They're usually also the ones who'll insist they "they rely on their ears" etc.

 

In a way it's really sad.

post #20 of 20
Quote:
Originally Posted by Brooko View Post
 

Oh - I have no problem with there being differences.  We know without doubt that there are differences between aac256 and FLAC for instance.  They can easily be shown / measured.

 

My issue is - are they audible by humans? (and each individual should be investigating that themselves to determine their own threshold IMO).

 

Sadly - that is where the disagreement is.  And even sadder - the ones that claim to hear the differences (especially those that claim differences are easily discerned), are usually the ones who have never undergone a properly set-up test, and who also refuse to do so.  They're usually also the ones who'll insist they "they rely on their ears" etc.

 

In a way it's really sad.

 

You're right on the ball. Even if a given device plays different lossless formats differently, the difference may be inaudible. I would need to take that into account in the experiment. If it could be demonstrated that all the measured differences are outside the range of human hearing, or so small as to be inaudible regardless, it would be more valuable research than merely giving a yes or no. As for the threshold of distinction, that's a trickier matter.

New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Music
Head-Fi.org › Forums › Misc.-Category Forums › Music › Comparing 44/16 to hi-rez 96-192/24