or Connect
Head-Fi.org › Forums › Equipment Forums › Sound Science › Old timer playing around with HD music
New Posts  All Forums:Forum Nav:

Old timer playing around with HD music

post #1 of 16
Thread Starter 

I've been out of the hobby for a while - call it 7 years or so.  Went up pretty far into the high end of things for a while and certainly had a lot of fun, but then I guess I topped out and slimmed down after a while.

 

Recently, I moved back to Tokyo and it's obvious all the audio manufacturers are pushing High Rez audio at the higher end consumer level.  E.g., Sony displays highlighting HD music in selling the MDR-1R Series Mk. 2 headphone and their new walkmans/Mora online music store service/PHA-2 portable DAC/etc.  Teac doing the same with the UD-501 DAC.  Many other examples.

 

I imagine it's different for those of you living in the States, but imagine walking into your local Best Buy and seeing high resolution audio gear being advertised instead of like Bose or Beats headphones.

 

So I got curious (and I wanted a new decent portable closed headphone), so I picked up the MDR-1R Series Mk. 2 for kicks (around $300 ish, I forget exactly).  Also got a new laptop with a really noisy headphone jack, so picked up a Meridian Explorer ($300).  So certainly not high end stuff, but more expensive, high end consumer, stuff that's specifically advertised for HD music playback.  I could warm up some of the higher end stuff here, but I actually kinda like the simplicity/portability/accessibility of this rig.  Besides, music sounds really good out of it.

 

 

Ignore the Qualia 010 and Grace m902 for now.  That's me just showing off/establishing I kept some decent gear.

 

So I downloaded a few things from HD Tracks - I kinda like it, btw.  So tired or ripping my own CDs and not motivated start on my (small) pile of SACD and DVD-A.  Very convenient to just download stuff, esp since, hopefully, they represent well mastered, properly encoded, versions of the music.  Though not a fan of repurchasing stuff yet again that I have on DVD-A/SACD...

 

Anyway, for kicks I booted into Foobar's ABX comparison thing two versions I recently bought of Daft Punk's Get Lucky:

 

HD Tracks 24/88 FLAC - album cost $18, can't do a single track d/l.

iTunes in the usual 256k (I think) AAC compressed - cost like a $1.50 for the one track.

 

I did something similar like 7 years ago as I was leaving the hobby, though less scientific (not ABX) between 320k mp3s and Apple Lossless files I ripped from my own CDs through  my speaker rig.

 

And once again, 5 out of 5 times, I could distinguish that the files are different.  Real pain in the ass because the differences are subtle (to me) and they both sounded great.  I was surprised I actually got it 5 out of 5 times, it was so close, I was expecting 50/50 guessing.

 

Anyway, took a lot of effort, but I again correctly identified that the compressed, and here, low rez, version sounded better to me every time. 

 

Lol.

 

Doesn't mean I'm giving up and just buying all my music from iTunes (though damn that would make life a hell of a lot easier...)  I think there's something really anal about many of us, but me for sure, that wants the most data as possible, even if it's not useful (and sounds ever so slightly worse less good to me lol). 

 

Anyway, that was fun.  Prob. not going to do something like this again for a few years.  ABX isn't fun.  Much nicer just to enjoy the music.

 

Best,

 

-Jason


Edited by jjcha - 5/1/14 at 12:29pm
post #2 of 16

So why do you kinda like HDTracks when a consequently lower resolution file sounded better to you than HDTracks' supposedly "high resolution" file?

post #3 of 16
Thread Starter 
Quote:
Originally Posted by elmoe View Post
 

So why do you kinda like HDTracks when a consequently lower resolution file sounded better to you than HDTracks' supposedly "high resolution" file?

 

Well, there that anal part of me that wants uncompressed files, and, to a lesser degree, high rez, just "because".  And it is SO MUCH EASIER than ripping CDs, much less SACD and DVD-A.  Oh man, if I never rip another ISO file again, and I'm not booting up the PS3 to grab my pitiful pile of SACD. 

 

 

Sometimes I am annoyed that I can't just be happy buying music on iTunes.  I know it's arguably not entirely rational, given what I just said.  Though in some ways it, arguably, is.  Anyway, even admitting that, if Apple sold lossless versions of all the music on iTunes, that's an open license just to take my money. 

 

Even though I prob would like the 256 AAC files better lol...

 

People can be weird, eh?

 

Best,

 

-Jason

post #4 of 16

Well, most of my music is in FLAC format, and to my ears it sounds better than any 256 ACC file. That said yes, you're pretty weird I must admit :biggrin:

post #5 of 16
Thread Starter 

Well, taking a step back from it all, this could be the explanation.

 

I find the Meridian and the 1R mk 2 both lean on the thicker/fuller/richer side of things.  Which is fine, but I prefer a clean, detailed, usually thinner sound. 

 

I could distinguish the FLAC vs AAC because the former sounded ever so slightly, barely barely barely thicker and heavier to me.  The latter sounded better to me today.

 

Might my preferences swing the other way with a different DAC and the 010 and my old Dynamight?  May very well be that way. 

 

Which is why I guess I like having the lossless/high rez stuff.  I can always compress/resample on my own later if I want to.  Anyway, the interesting thing is how subtle all of this is on the recently acquired gear.  If this is all I had, I'd prob buy on iTunes.  But it's not all I've got, and in my experience I notice differences more on my speaker rigs.

 

Now that I think about it, the 320k mp3/ALAC test was done on thicker/warmer speakers...

 

Best,

 

-Jason


Edited by jjcha - 5/1/14 at 1:06pm
post #6 of 16
Quote:
Originally Posted by jjcha View Post
 

Well, taking a step back from it all, this could be the explanation.

 

I find the Meridian and the 1R mk 2 both lean on the thicker/fuller/richer side of things.  Which is fine, but I prefer a clean, detailed, usually thinner sound. 

 

I could distinguish the FLAC vs AAC because the former sounded ever so slightly, barely barely barely thicker and heavier to me.  The latter sounded better to me today.

 

Might my preferences swing the other way with a different DAC and the 010 and my old Dynamight?  May very well be that way. 

 

Which is why I guess I like having the lossless/high rez stuff.  I can always compress/resample on my own later if I want to.  Anyway, the interesting thing is how subtle all of this is on the recently acquired gear.  If this is all I had, I'd prob buy on iTunes.  But it's not all I've got, and in my experience I notice differences more on my speaker rigs.

 

Now that I think about it, the 320k mp3/ALAC test was done on thicker/warmer speakers...

 

Best,

 

-Jason

 

It could be that the iTunes version is just slightly louder, only .5dB would do the trick. Louder is nearly always perceived as better.
Also, in the case of RAM there are multiple mastering versions floating around. Confounding stuff.
Link (yes it's 'puteraudiophile, pinch of salt and all that)

Have you tried converting the hi-rez track to 256AAC, and then compare it to the original HDTracks version?


Edited by limpidglitch - 5/1/14 at 2:34pm
post #7 of 16

Are you sure they're from the same master and carefully level matched? Either a slightly different master or a slight level mismatch could easily account for your ability to ABX the files successfully, since a 256k AAC should be pretty near audibly transparent (though it's not impossible that you could actually be hearing compression artifacts).

post #8 of 16

If you're having that much trouble getting a consistent impression of the sound, odds are it doesn't matter.

 

To me, AAC 256 VBR is transparent to CD quality lossless and CD quality lossless is transparent to 24/96

post #9 of 16
Thread Starter 
Quote:
Originally Posted by bigshot View Post
 

If you're having that much trouble getting a consistent impression of the sound, odds are it doesn't matter.

 

To me, AAC 256 VBR is transparent to CD quality lossless and CD quality lossless is transparent to 24/96

 

Well, I think I'm in the same general ballpark with your thoughts.  While I can discern differences on a ABX testing, it's just not worth the damn effort and it doesn't really matter.

 

The question I have is while any one given impact may not be discernible, do many non discernible impacts in the aggregate start to have a discernible impact.  It's kind of the philosophy that Team Sky has in bike racing.  I've got my guess, but honestly, I don't know.

 

As for all the points raised about level matching, making sure they come from the same master, etc., I agree that those have an impact on ABX testing.  But my test wasn't to test if I can hear a difference just in the files - it was to see if I can hear a difference between one source of music for me (HD Tracks) versus another (iTunes).

 

But my gut is the files came from the same source and were level matched.  Nothing scientific other than the differences were so utterly miniscule, I'd be surprised if they came from different masterings.  But honestly, I don't know.


Edited by jjcha - 5/1/14 at 5:30pm
post #10 of 16
Quote:
Originally Posted by jjcha View Post
 

 

Well, I think I'm in the same general ballpark with your thoughts.  While I can discern differences on a ABX testing, it's just not worth the damn effort and it doesn't really matter.

 

The question I have is while any one given impact may not be discernible, do many non discernible impacts in the aggregate start to have a discernible impact.  It's kind of the philosophy that Team Sky has in bike racing.  I've got my guess, but honestly, I don't know.

 

As for all the points raised about level matching, making sure they come from the same master, etc., I agree that those have an impact on ABX testing.  But my test wasn't to test if I can hear a difference just in the files - it was to see if I can hear a difference between one source of music for me (HD Tracks) versus another (iTunes).

 

But my gut is the files came from the same source and were level matched.  Nothing scientific other than the differences were so utterly miniscule, I'd be surprised if they came from different masterings.  But honestly, I don't know.

 

Easiest way to check would be to take the HD Tracks Hi-Res, transcode to aac256, then run an abx between the two files.  If you're using Foobar, just apply replaygain - and that solves your volume issue.  If you can't discern any difference between the two containers coming from the same source file - then you know that either:

 

  1. Your original abx had two files with volume differences
  2. Your original abx had two files from different masters 

 

You likely already know how to set everything up - but in case anyone else wondering, here's a short guide I wrote a while ago (http://www.head-fi.org/t/655879/setting-up-an-abx-test-simple-guide-to-ripping-tagging-transcoding)

post #11 of 16
Quote:
Originally Posted by jjcha View Post
 

The question I have is while any one given impact may not be discernible, do many non discernible impacts in the aggregate start to have a discernible impact.

 

I took a song and ripped and encoded to AAC 256 VBR ten times... tenth generation. No discernible difference. So the answer is no.

post #12 of 16
Having dabbed a little into Japanese HD tracks as well - I wouldn't trust iTunes stores (or any other online stores for that matter) downloads to be from the same master as the HD track downloads - my bet is that the non-HD files are the same as the CD masters where all the horrible dynamic compression and gain boost has already been applied whereas the HD downloads are actually one step before that. I actually have some iTunes tracks and the exact equivlaent CDs at home, I might go home and check my hypothesis.

FWIW I've tried doing a proper level-matched ABX of some HD tracks against the same CD versions (with all the horrible dynamic compression and gain boost) - I got 15 out of 15 without any effort (I'm able to pick out the tracks within 10 seconds of playing them). And I'm not happy about the results - because I know that it's not that I have golden ears, but rather they screw it up THAT BAD.
post #13 of 16
Thread Starter 
Quote:
Originally Posted by bigshot View Post
 

 

I took a song and ripped and encoded to AAC 256 VBR ten times... tenth generation. No discernible difference. So the answer is no.

 

Not what I'm talking about.

 

For example, I find differences in competently made electronics (e.g., many portable DAC/amps versus the headphone out of an iPhone), to be pretty marginal.  I hear small differences, but I don't have that much confidence I'd be able to tell under a proper ABX.  Maybe I can, but even if I did, it'd be similarly a pain in the butt to discern subtle differences.  But then combine that with encoding differences, and who knows what other changes, and if the differences point in the same direction, my bet is you get something where it's a little less subtle/easier to hear the differences.

 

Not that I really have that much interest in personally testing that.


Edited by jjcha - 5/1/14 at 8:06pm
post #14 of 16

AAC is a codec that is designed to be audibly transparent at higher bit rates. By 256, I find it to be completely transparent in direct line level matched comparisons. It also doesn't appear to be additive in transcoding. Once you have run the music through the codec once, you can reencode it at the same bit rate or higher as many times as you might need to without impacting sound quality.

 

If you don't think that you can discern a difference in controlled testing, if there is a difference, it isn't audible. And ultimately, we listen with our ears. Anything beyond human hearing is only there for dogs and bats to enjoy.

post #15 of 16
Thread Starter 
Quote:
Originally Posted by nanaholic View Post

I wouldn't trust iTunes stores (or any other online stores for that matter) downloads to be from the same master as the HD track downloads - my bet is that the non-HD files are the same as the CD masters where all the horrible dynamic compression and gain boost has already been applied whereas the HD downloads are actually one step before that. I actually have some iTunes tracks and the exact equivlaent CDs at home, I might go home and check my hypothesis.

 

I think this is a good point.  I have a sample size of just 1, and I'm guessing given it's a recently released track under the "Mastered for iTunes" where they also knew they'd be releasing in 24/88 high rez, they used the same master.  But who knows on all the other stuff out there.

New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Sound Science
Head-Fi.org › Forums › Equipment Forums › Sound Science › Old timer playing around with HD music