Head-Fi.org › Forums › Equipment Forums › Sound Science › Is there an audible difference between HD and CD quality?
New Posts  All Forums:Forum Nav:

Is there an audible difference between HD and CD quality?

post #1 of 48
Thread Starter 

HD meaning 24-bit and 48/88/96/192 kHz sampling rates

CD meaning 16-bit and 44 kHz

 

I purchased 2CELLOS's In2ition album from HDTracks.com and it's a 24/48 album.

I downloaded the same album in CD quality just for testing's sake and it has EAC accurate rip information.

 

Setup:

  • MacBook Pro
  • Windows 7 64-bit via Boot Camp
  • Foobar2000 (24-bit WASAPI output)
  • ObjectiveDAC (24/96 output set in Windows' Volume Manager, digital volume is set at maximum)
  • Objective 2
  • AKG K 701 (8-bump headband version)

 

 

ABX Foobar2000 Test Results (Click to show)

foo_abx 1.3.4 report
foobar2000 v1.2.4
2013/05/22 02:02:36

File A: D:\Users\Michael\Downloads\2Cellos - In2ition (2013) FLAC\07 - Clocks.flac
File B: D:\Users\Michael\Music\HDtracks\2CELLOS - In2ition\07 Clocks (ft. Lang Lang).m4a

02:02:36 : Test started.
02:03:58 : 01/01  50.0%
02:04:12 : 02/02  25.0%
02:04:28 : 03/03  12.5%
02:04:47 : 03/04  31.3%
02:05:22 : 03/05  50.0%
02:06:00 : 03/06  65.6%
02:06:29 : 04/07  50.0%
02:07:25 : 05/08  36.3%
02:07:27 : Test finished.

 ----------
Total: 5/8 (36.3%)

 

 

 

I only did 8 tests because I really felt that I was guessing between the two tracks and doing further tests would have been futile. I genuinely tried to hear a difference between the two files and I simply could not make a definitive difference.

 

When I get the time later, I'll downsample a few of my HD Tracks purchases to CD quality and post my logs here.

 

Can you guys hear a difference?

  • If the answer is yes, please post an ABX log in a spoiler tag
  • If the answer is no, why do audiophiles bother to purchase these expensive HD albums?

Edited by miceblue - 5/22/13 at 4:09pm
post #2 of 48

mastering has been found to vary between "CD" and "Hi res" releases of the same named material, from major labels - basically loading the die/stacking the deck as it were

 

the only "safe", "fair" way to compare is to filter/decimate/dither yourself the one hi res source down to 16/44


Edited by jcx - 5/22/13 at 8:41pm
post #3 of 48
Thread Starter 
Quote:
Originally Posted by jcx View Post

mastering has been found to vary between "CD" and "Hi res" releases of the same named material, from major labels - basically loading the deck as it were

 

the only "safe", "fair" way to compare is to filter/decimate/dither yourself the one hi res source down to 16/44

Is that so? I personally don't own a CD equivalent of an HD album so I can't confirm that, which is why I "downloaded" the CD equivalent of the 2CELLOS album and compared the two with no good results.

I do not endorse downloading music illegally. I downloaded the album only for testing purposes and I deleted the files immediately afterwards.

I do own Seal's original Seal album via CD and I did buy his Seal Best 1991-2004 HD album (24/48).

When I compared the song "Crazy" from those two albums, there was a significant difference between the two. However, they are two different albums in the first place so this test is probably not valid.

 

ABX Foobar2000 Test Results (Click to show)
foo_abx 1.3.4 report
foobar2000 v1.1.18
2013/01/25 17:18:05
 
File A: D:\Users\Michael\Music\My HDTracks Music\Seal\Best 1991 2004\1-Crazy.flac
File B: D:\Users\Michael\Music\Seal\Seal\03 Crazy.flac
 
17:18:05 : Test started.
17:18:57 : 01/01  50.0%
17:19:12 : 01/02  75.0%
17:19:26 : 02/03  50.0%
17:19:43 : 03/04  31.3%
17:19:56 : 04/05  18.8%
17:20:18 : 05/06  10.9%
17:20:32 : 06/07  6.3%
17:20:44 : 07/08  3.5%
17:20:53 : 08/09  2.0%
17:21:00 : 09/10  1.1%
17:21:04 : 10/11  0.6%
17:21:09 : 11/12  0.3%
17:21:14 : 12/13  0.2%
17:21:20 : 13/14  0.1%
17:21:27 : 14/15  0.0%
17:21:34 : 15/16  0.0%
17:21:39 : 16/17  0.0%
17:21:51 : 17/18  0.0%
17:21:56 : 18/19  0.0%
17:22:01 : 19/20  0.0%
17:22:07 : 20/21  0.0%
17:22:11 : 21/22  0.0%
17:22:17 : 22/23  0.0%
17:22:34 : 23/24  0.0%
17:22:40 : 24/25  0.0%
17:22:46 : 25/26  0.0%
17:22:55 : 26/27  0.0%
17:23:01 : Test finished.
 
 ---------- 
Total: 26/27 (0.0%)

 


Edited by miceblue - 5/22/13 at 8:37pm
post #4 of 48

With the exception of possible mastering differences I wouldn't say there is a real outstanding difference, generally I notice the HD tracks are quieter but that could have a lot to do with mastering since technically you have more headroom which allows for a lower noise floor.

 

I trust the Redbook standard to give me quality listening, I do purchase the occasional HD album for kicks but I don't seriously expect it to be any better than a Redbook album.

post #5 of 48
Thread Starter 

Yeah...I've spent a good chunk of money (for me at least) on HD albums. Next time there's a sale on HDTracks.com, maybe I should seriously consider purchasing the CD equivalent of the album instead of the HD ones...

 

HD album websites have some pretty good marketing strategies.  rolleyes.gif

post #6 of 48

I've written emails to several HD audio vendors asking specifically how the HD versions were created.  The answer was fairly uniform: we take them how we get therm from the record companies. In other words, they could be created any number of ways, from just up-sampling the digital CD master, or a remaster of some sort, to a re-mix of the multitrack master, then remastering to 24/96,, to ideally going back to an original 24/96 master and copying that.  For analog originals, they may be up-sampled 16/44.1 masters, remasters, or hopefully fresh 24/96 masters from the original analog.  However, the chances of the latter are VERY small, as is the 24/96 or higher original material to release file path, and mostly you could be getting a remaster of a 16/44.1 and then up-sampled version.  It would make some sense that a record company do something in the creation of their 24/96 release files to make them sound different from the CD, or we'd all be screaming "rip off!".  Of course, we sort of are anyway.

 

There are a couple of boutique labels that originate in 24/96 or higher, stay that way through production, then release in 24/96 or higher, but they are pretty far out of the mainstream and have their own unique catalogs. 

 

Comparing the CD to an HD file probably isn't a meaningful way to evaluate HD, any more than comparing an LP to a CD version.  There's just too much unknown going on between the two versions.

post #7 of 48
Thread Starter 
Quote:
Originally Posted by jaddie View Post

I've written emails to several HD audio vendors asking specifically how the HD versions were created.  The answer was fairly uniform: we take them how we get therm from the record companies. In other words, they could be created any number of ways, from just up-sampling the digital CD master, or a remaster of some sort, to a re-mix of the multitrack master, then remastering to 24/96,, to ideally going back to an original 24/96 master and copying that.  For analog originals, they may be up-sampled 16/44.1 masters, remasters, or hopefully fresh 24/96 masters from the original analog.  However, the chances of the latter are VERY small, as is the 24/96 or higher original material to release file path, and mostly you could be getting a remaster of a 16/44.1 and then up-sampled version.  It would make some sense that a record company do something in the creation of their 24/96 release files to make them sound different from the CD, or we'd all be screaming "rip off!".  Of course, we sort of are anyway.

 

There are a couple of boutique labels that originate in 24/96 or higher, stay that way through production, then release in 24/96 or higher, but they are pretty far out of the mainstream and have their own unique catalogs. 

 

Comparing the CD to an HD file probably isn't a meaningful way to evaluate HD, any more than comparing an LP to a CD version.  There's just too much unknown going on between the two versions.

Wow, that was pretty insightful; thanks for sharing that!

 

So if I were to purchase a so-called HD album in the future, how would I know if it's a genuine 24/96 master and not one that's been up-converted or re-mastered?

One of the websites I buy music from actually does distinguish between true HD masters and up-converted ones, but they are not mainstream stuff as you mentioned.

 

And you're right about comparing CD masters to HD masters. I had originally meant to ask if there was a difference between 16/44 and 24/__, but I guess me including the CD master album kind of threw that off.


Edited by miceblue - 5/22/13 at 10:18pm
post #8 of 48

Well there are some ways to spot some of the faux hi-rez recordings.  One is a sharp cutoff like one you get in 44.1khz material.  Look at the spectrogram view in Audacity for instance.  It is fairly obvious.  More than a few clearly were 44.1 somewhere along the way even if 192khz now. 

 

Then the issue about remastering is something companies aren't at all forthcoming with.  There have been a few that were remastered and had clipping in Hi-rez versions that weren't in the CD releases.  A few others clearly have some different EQ applied, though whether the old CD version was altered or the new one is something you can never know. 

 

There is a sub-forum on another site specifically for discussing such issues.  It looks like somewhere between 20% and 40% of hirez material isn't.  Though of course anyone finding this is more likely to report it so those percentages may not be a true representation of most hirez.  What is clear is that some considerable number are low rez in disguise.  What is a bit odd, is while companies universally report they simply have to take what masters were sent to them, they too could spend 5 minutes with a sound editor and find out the truth.  Yet they don't bother.  Maybe they make money and don't care, maybe something else.  Clearly in a some cases even if low rez sourced, the remastering is better than the original.  Just a bit of a let down to have it upsampled and sold as hirez when it really is just remastered. 

 

I am not very convinced that well done 44.1 khz lacks anything vs higher sample rates.  Any improved re-mastering has value.  So after trying a few hirez downloads I always choose 48/24 if there is a choice.  It is cheaper, and should be fine. 

post #9 of 48
Quote:
Originally Posted by miceblue View Post

Wow, that was pretty insightful; thanks for sharing that!

So if I were to purchase a so-called HD album in the future, how would I know if it's a genuine 24/96 master and not one that's been up-converted or re-mastered?
One of the websites I buy music from actually does distinguish between true HD masters and up-converted ones, but they are not mainstream stuff as you mentioned.

And you're right about comparing CD masters to HD masters. I had originally meant to ask if there was a difference between 16/44 and 24/__, but I guess me including the CD master album kind of threw that off.
Unless the vendor tells you specifically, your only option is to run the spectrum test ESLdude mentioned, hopefully on a free sample before you buy. That won't tell you if the mastering is better or not, just clue you as to the possible 16/44.1 heritage.
post #10 of 48
Thread Starter 

So, I looked at the spectrograms of my two 24/192 albums. I'm not really sure how to interpret these since I'm used to 16/44 analysis.

 

Album 1:

 

 

 

Album 2:

 

 

 

 

 

 

And a couple of 24/96 binaural albums

 

Album 3:

 

Album 4:

 

 

 

In terms of sound, and what we humans can hear, we can usually only hear up to 20 kHz with a very good set of ears. From sine wave tests on Audacity, my range of hearing only goes up to < 17 kHz.

What audible advantage does an HD recording have over its CD-quality 16/44 counterpart? It looks like Album 4 has some information at around 30 kHz, but that's way out of my hearing range.

post #11 of 48
Quote:
Originally Posted by jaddie View Post

There are a couple of boutique labels that originate in 24/96 or higher, stay that way through production, then release in 24/96 or higher, but they are pretty far out of the mainstream and have their own unique catalogs.

It's pretty normal for classical labels these days, who often offer Hybrid SACD and/or "hi-res" downloads.
post #12 of 48

Pentatone puts out hybrid CD/SACD disks. They sound good, but the one I tested sounded exactly the same on both layers after level matching.

 

The quality of the aesthetic choices made in engineering is MUCH more important than the numbers on a spec sheet. Both CDs and SACDs are capable of "perfect sound".

post #13 of 48
Quote:
Originally Posted by miceblue View Post

What audible advantage does an HD recording have over its CD-quality 16/44 counterpart? It looks like Album 4 has some information at around 30 kHz, but that's way out of my hearing range.

Higher sampling rates generally have lower levels of jitter.

 

CD playback requires the use of a brick wall filter to cut off frequencies above 22khz - this filter causes ringing.  If you look at a picture of a typical CD impulse, the signal will show ringing before and after the pulse.  Higher sampling rates can be implemented with less harsh filters and thus generate less ringing.

 

Some DACs use different reconstruction filters that eliminate the pre impulse ringing, but these will generate artifacts that are in the audible frequency range.  Higher sampling rates with these filters can move these artifacts above the audible range.

 

Higher bit rates will theoretically increase the dynamic range.

 

Now whether any of these effects are audible?  Some say they are.  Some don't.

post #14 of 48

A number of ABX test files have already been posted on this forum that allow people to find out whether they can hear any of the following:

- higher than 16 bit resolution

- higher than 44.1 kHz sample rate

- differences between 44.1 kHz reconstruction filters (ringing etc.)

So far, I do not recall seeing confirmed positive results. The artifacts of a reasonably good 44.1 kHz reconstruction filter are already above the audible range for people who cannot hear frequencies much higher than 20 kHz.


Edited by stv014 - 5/24/13 at 2:29am
post #15 of 48
Quote:
Originally Posted by bigshot View Post

Pentatone puts out hybrid CD/SACD disks. They sound good, but the one I tested sounded exactly the same on both layers after level matching.

The quality of the aesthetic choices made in engineering is MUCH more important than the numbers on a spec sheet. Both CDs and SACDs are capable of "perfect sound".

The SACD layer is used for multi-channel. It's pointless to perform a level matched stereo comparison and then say the SACD and CD layers are the same. They are not.

The "Penta" in Pentatone is a reference/allusion to five channels (as in Pentagon, pentagram and so on)..

http://www.pentatonemusic.com/index1.htm
Quote:

PentaTone Music B.V. is founded in 2001 by three former executives of Philips Classics (one of the classical music labels of Universal) together with Polyhymnia International (the former recording center of Philips Classics).

The founders of PentaTone believe that 5-channel surround sound will gradually replace today’s stereo sound, because surround sound provides an extra dimension to the musical experience. Hearing is believing. Not only the artists and the audiophiles are convinced.
Everybody who attends a demonstration confirms that the experience is more natural and superior compared to stereo sound.

Harmonia Mundi do the same (offer Hybrid SACD with stereo CD layer and multi-channel SACD layer), as do other classical labels.
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Sound Science
Head-Fi.org › Forums › Equipment Forums › Sound Science › Is there an audible difference between HD and CD quality?