Official High Res download not high res?
Sep 3, 2014 at 6:07 PM Thread Starter Post #1 of 19

rovo

New Head-Fier
Joined
Aug 30, 2014
Posts
42
Likes
10
Hello fellow Audiophiles,
 
Today I bought (88.1KHz/24Bit Studio Masters) Steve Hackett - Genesis Revisited II from http://hackettsongs.sandbag.uk.com/. After the 2.7GB download completed, I unpacked and opened some of the, said to be, 88.1KHz/24bit flac files with Audacity to use the plot analysis function.
 
The resulting graph seemed to show that the audio was not 88.1KHz at all but mere CD quality, 44.1KHz.
 
http://i.gyazo.com/50e51b06554c13162659df1176808032.png
 
Is there any way to test wether the audio is actually 24bit or upconverted 16bit?
 
Thanks in advance for any advice or thoughts on this.
 
RV
 
Sep 5, 2014 at 11:46 AM Post #2 of 19
Well, I can't tell if this is real hi-res nor recommending a software to test it, but I purchased 23 hi-res albums on Qobuz http://www.qobuz.com/gb-en/ that I mostly use on my Sony ZX1, till now I'm very happy with the mastering quality of each album.
 
I think that, there should be an official list regarding hi -res albums on Wikipedia or anything similar, just google around for feedbacks and you'll be fixed
smily_headphones1.gif

 
Sep 9, 2014 at 3:16 AM Post #3 of 19
I just got a reply to my inquiries:
 
/*
GENESIS REVISITED II 24 BIT STUDIO MASTER FLAC DOWNLOAD is very definitely 24 bit. It is true that the original masters are 44.1kHz but we deliver the "Studio Quality" downloads in 88.2kHz simply because that is the accepted norm.
*/
 
While it could be said that up converting 44.1KHz to 88.2KHz is misleading, they did offer a refund if the quality of the download was not to my liking.
 
Ron.
 
Sep 9, 2014 at 3:25 AM Post #4 of 19
  Well, I can't tell if this is real hi-res nor recommending a software to test it, but I purchased 23 hi-res albums on Qobuz http://www.qobuz.com/gb-en/ that I mostly use on my Sony ZX1, till now I'm very happy with the mastering quality of each album.
 
I think that, there should be an official list regarding hi -res albums on Wikipedia or anything similar, just google around for feedbacks and you'll be fixed
smily_headphones1.gif

 
Thank you. I too have bought from Qobuz on multiple occasions, and I have no complaints whatsoever about them. To my knowledge, they do not up-convert anything. Their High res catalogue may be somewhat limited compared to other HD Music web shops but at least you know the music they DO sell as high res really is high res.
 
Ron.
 
Sep 9, 2014 at 10:26 AM Post #5 of 19
   
Thank you. I too have bought from Qobuz on multiple occasions, and I have no complaints whatsoever about them. To my knowledge, they do not up-convert anything. Their High res catalogue may be somewhat limited compared to other HD Music web shops but at least you know the music they DO sell as high res really is high res.
 
Ron.


+1
 
Oct 30, 2014 at 6:23 PM Post #6 of 19
Just bought a studiomaster from Qobuz (Ben Howard's new album). When viewing the file in spectrum, this showed up. I'm not an expert on these things unfortunately, but doesn't this indicate that the file in question is just a 'normal' CD quality file?
 
 
 
 

 

 
Nov 3, 2014 at 11:42 PM Post #7 of 19
This should be shouted from the rooftops: the master is what matters!
 
The average human hearing range is only about 20 to 20,000 hertz. If you take a 24-bit file (any sample rate) and convert it to lossless 16-bit / 44.1 kHz, it's physically impossible to hear a difference. Don't worry about being able to hear things that you can't; worry about the quality of the recording and master.
 
Also, read this article.
 
Nov 4, 2014 at 12:32 PM Post #9 of 19
  The average human hearing range is only ....

 
 
Some people are so obsessed with trying to convince other people that high res music makes no sense that they make it their holy mission and even respond to postings where their input is off topic and thus unwelcome. Linking to other people's articles to try and lend their own statements some credibility is just laughable and shallow. Who is xiph anyway and why would they stray off of their core business and go through lengths to try and convince people that distributing high res music is pointless? That in itself should be enough to make you think...
 
" The Xiph.Org Foundation is a non-profit corporation dedicated to protecting the foundations of Internet multimedia from control by private interests. Our purpose is to support and develop free, open protocols and software to serve the public, developer and business markets. " 
Do you see any connection? Me neither.
 
Back to my original post.
 
If you buy something that turns out not to be as advertised then you got ripped off. In my case it was beside the point whether I could hear the difference between 88.2KHz and 44.1KHz. It was music that was being advertised and sold as 88.2KHz while in fact, it was not. I complained, and got my money back. 
 
In other words, what was sold was not to the exact specifications as what I was led to believe; I decided not to accept that. End of story.
 
Nov 4, 2014 at 2:03 PM Post #10 of 19
Come to think of it, I don't routinely check my music in Audacity. I was somewhat disappointed with the sound of the album which made me check out some of the songs in Audacity. I had expected greater depth to the sound. I guess my hearing/brain registered the absence of certain euphonies that can be present in true high res.
 
Granted, I will probably never find out how those songs sound in true 88.2KHz, so there is no comparison possible.
 
Nov 4, 2014 at 2:33 PM Post #11 of 19
My point is that you literally can't hear anything above those frequencies. Some high-res albums are worth paying for, but only because they used a different master that may sound better, not because their sample size and sample rate are higher (which, as documented, does not yield any audible benefits).
 
Nov 4, 2014 at 3:26 PM Post #13 of 19
  Wow, my hint wasn't even subtle but you still didn't get it. Stick to your beliefs, I don't care.

 
It's not merely a belief; it is an extensively documented scientific fact.
 
I understand that you wanted to focus on how they upsampled the audio and you feel swindled, but that isn't the real problem; the problem is that they don't usually tell you which master was used. CDs are often mastered differently than the studio master. If they took the CD audio and tried selling it as a high-res download, implying that it sounds different from the CD when it actually doesn't, that is unethical...but in the same way, it is also unethical to tell your customers that an album sounds better for reasons that are impossible.
 
Nov 4, 2014 at 8:06 PM Post #15 of 19
Hello fellow Audiophiles,

Today I bought (88.1KHz/24Bit Studio Masters) Steve Hackett - Genesis Revisited II from http://hackettsongs.sandbag.uk.com/. After the 2.7GB download completed, I unpacked and opened some of the, said to be, 88.1KHz/24bit flac files with Audacity to use the plot analysis function.

The resulting graph seemed to show that the audio was not 88.1KHz at all but mere CD quality, 44.1KHz.

http://i.gyazo.com/50e51b06554c13162659df1176808032.png

Is there any way to test wether the audio is actually 24bit or upconverted 16bit?

Thanks in advance for any advice or thoughts on this.

RV

On a technical note, I don't think there's any way to identify whether or not the bit-depth is up-converted from 16-bit audio. Bit-depth just affects the noise floor.
Even if you took a 24-bit audio file, converted it to 16-bit, and re-converted it back to 24-bit, I'm pretty sure you wouldn't be able to hear a difference.

I used the SoX terminal program to do exactly this and here are the respective spectrograms:
Original 24-bit file (purchased from here: http://www.hdtracks.com/miscellaneous/the-ultimate-headphone-demonstration-disc-160143)
Code:
/Users/miceblue/sox-14.4.1/sox /Users/miceblue/sox-14.4.1/2-48\ Drum\ Kit\:\ Dynamic\ Range\ Uncompressed\ \:\ Test24.aif -n spectrogram -o spectrum242.png


Down-converted 16-bit file
Code:
/Users/miceblue/sox-14.4.1/sox /Users/miceblue/sox-14.4.1/2-48\ Drum\ Kit\:\ Dynamic\ Range\ Uncompressed\ \:\ Test.aif -b 16 -o 2-48\ Drum\ Kit\:\ Dynamic\ Range\ Uncompressed\ \:\ Test16.aif
/Users/miceblue/sox-14.4.1/sox /Users/miceblue/sox-14.4.1/2-48\ Drum\ Kit\:\ Dynamic\ Range\ Uncompressed\ \:\ Test16.aif -n spectrogram -o spectrum16.png


Up-converted 24-bit file from the down-converted 16-bit file
Code:
/Users/miceblue/sox-14.4.1/sox /Users/miceblue/sox-14.4.1/2-48\ Drum\ Kit\:\ Dynamic\ Range\ Uncompressed\ \:\ Test16.aif -b 24 -o 2-48\ Drum\ Kit\:\ Dynamic\ Range\ Uncompressed\ \:\ Test24.aif
/Users/miceblue/sox-14.4.1/sox /Users/miceblue/sox-14.4.1/2-48\ Drum\ Kit\:\ Dynamic\ Range\ Uncompressed\ \:\ Test24.aif -n spectrogram -o spectrum242.png


And here's the extreme example with a down-converted 8-bit file from the original 24-bit file
Code:
/Users/miceblue/sox-14.4.1/sox /Users/miceblue/sox-14.4.1/2-48\ Drum\ Kit\:\ Dynamic\ Range\ Uncompressed\ \:\ Test.aif -b 8 -o 2-48\ Drum\ Kit\:\ Dynamic\ Range\ Uncompressed\ \:\ Test8.aif
/Users/miceblue/sox-14.4.1/sox /Users/miceblue/sox-14.4.1/2-48\ Drum\ Kit\:\ Dynamic\ Range\ Uncompressed\ \:\ Test8.aif -n spectrogram -o spectrum8.png



As you can see, it's pretty much impossible to analyze with a spectrogram whether or not your music is a 16-bit file up-converted to 24-bit. From this, we can only conclude that the album may or may not be up-converted from 16-bit.




[rule]
Regarding the sample rate, it's also impossible to tell whether or not your music is up-sampled. Just because an audio file has signals that only extend to 22.05 kHz when the file is in a 88.2 kHz sampling rate format, it DOES NOT mean it's up-sampled. For all you know, the microphones used might only capture signals up to 22 kHz, but the analog-to-digital converter was sampling music at 88.2 kHz. Also, just because the music has a frequency cut-off at 22.05 kHz DOES NOT mean it's "CD-quality." Audio for CDs are usually an entirely different master from the so-called "studio master." It's a pretty easy comparison to do if you have the CD master of the album.

Here's an example from my music library.
16/44.1 accurate CD rip of Seal's "Crazy" song from his 1991 album
Code:
/Users/miceblue/sox-14.4.1/sox /Users/miceblue/sox-14.4.1/Crazy1.flac -n spectrogram -o spectrum44.png

^ so far so good

24/88.2 version of Seal's "Crazy" song from HD Tracks (http://www.hdtracks.com/best-1991-2004)
Code:
/Users/miceblue/sox-14.4.1/sox /Users/miceblue/sox-14.4.1/Crazy2.flac -n spectrogram -o spectrum88.png

^ ZOMFG omg-zors it MUST be an up-sampled CD track; holy shizznuts ball sack I just wasted my precious money!

Here's that same 24/88.2 file down-sampled to 16/44.1
Code:
/Users/miceblue/sox-14.4.1/sox /Users/miceblue/sox-14.4.1/Crazy2.flac -b 16 -r 44100 -o Crazy3.flac
/Users/miceblue/sox-14.4.1/sox /Users/miceblue/sox-14.4.1/Crazy3.flac -n spectrogram -o spectrum442.png


Here's the CD-rip file up-sampled to 24/88.2
Code:
/Users/miceblue/sox-14.4.1/sox /Users/miceblue/sox-14.4.1/Crazy1.flac -b 24 -r 88200 -o Crazy4.flac
/Users/miceblue/sox-14.4.1/sox /Users/miceblue/sox-14.4.1/Crazy4.flac -n spectrogram -o spectrum82.png



Maybe it's an up-sampled CD track, maybe it's not. My ears say the two files are not the same.
Here, you can test them for yourself. I already added ReplayGain values to make the ABX test more fair
https://dl.dropboxusercontent.com/u/2816447/Crazy1.flac
https://dl.dropboxusercontent.com/u/2816447/Crazy2.flac
https://dl.dropboxusercontent.com/u/2816447/Crazy3.flac
https://dl.dropboxusercontent.com/u/2816447/Crazy4.flac
Code:
foo_abx 1.3.4 report
foobar2000 v1.3.3
2014/11/04 17:01:59

File A: Z:\Users\miceblue\Dropbox\Public\Crazy1.flac
File B: Z:\Users\miceblue\Dropbox\Public\Crazy2.flac

17:01:59 : Test started.
17:02:08 : 01/01 50.0%
17:02:17 : 02/02 25.0%
17:02:20 : 03/03 12.5%
17:02:21 : 04/04 6.3%
17:02:23 : 05/05 3.1%
17:02:25 : 06/06 1.6%
17:02:27 : 07/07 0.8%
17:02:29 : 08/08 0.4%
17:02:31 : 09/09 0.2%
17:02:33 : 10/10 0.1%
17:02:35 : 11/11 0.0%
17:02:37 : 12/12 0.0%
17:02:39 : 13/13 0.0%
17:02:41 : 14/14 0.0%
17:02:43 : 15/15 0.0%
17:02:45 : 16/16 0.0%
17:02:59 : Test finished.

 ---------- 
Total: 16/16 (0.0%)
Code:
foo_abx 1.3.4 report
foobar2000 v1.3.3
2014/11/04 17:16:27

File A: Z:\Users\miceblue\Dropbox\Public\Crazy2.flac
File B: Z:\Users\miceblue\Dropbox\Public\Crazy4.flac

17:16:27 : Test started.
17:16:37 : 01/01 50.0%
17:16:40 : 02/02 25.0%
17:16:42 : 03/03 12.5%
17:16:43 : 04/04 6.3%
17:16:45 : 05/05 3.1%
17:16:47 : 06/06 1.6%
17:16:49 : 07/07 0.8%
17:16:51 : 08/08 0.4%
17:16:53 : 09/09 0.2%
17:16:55 : 10/10 0.1%
17:16:57 : 11/11 0.0%
17:16:59 : 12/12 0.0%
17:17:01 : 13/13 0.0%
17:17:03 : 14/14 0.0%
17:17:05 : 15/15 0.0%
17:17:07 : 16/16 0.0%
17:17:09 : 17/17 0.0%
17:17:12 : Test finished.

 ---------- 
Total: 17/17 (0.0%)


[rule]
So, in conclusion, there's no real way to know whether or not your files are up-sampled from the CD unless you own the CD itself. What you're buying, what is often advertised as "studio quality" or "studio master," is usually a separate master from the original CD master.

And finally, a nice article that I like:
http://www.innerfidelity.com/content/its-masters-damit
IT'S THE MASTERS DAMIT!!
 

Users who are viewing this thread

Back
Top