[dprimary]

Is anyone aware of a record company that sells both DSD and PCM versions of the same recording? I want to do a test and need some source material.

 

[Music Alchemist]

http://www.head-fi.org/a/list-of-lossless-and-high-res-music-flac-alac-aiff-dsd-dxd-etc-download-sites

 

[johangrb]

http://store.acousticsounds.com/superhirez

 

[Music Alchemist]

And here are tons of free high-res downloads: https://hifiduino.wordpress.com/free-hi-res-music/

 

[Yuri Korzunov]

  Is anyone aware of a record company that sells both DSD and PCM versions of the same recording? I want to do a test and need some source material.

 
Hi Dprimary,
 
For comparing need know where original and where converted.
Here we can comparing audio file converter's quality.
 
Sometime possibly meet simultaneous records in PCM and DSD.
However there the will comparing mastering art.
 
Different audio resolution format possibly can give improvement of sound if it match to DAC's "native resolution" (without changing resolution and format DSD/PCM).
 
There two ways: buy matched resolution or convert resolution at home.
 
Me seems, fully correct comparing of DSD and PCM versions of the same recording is impossible by technical reasons.
 
Best regards,
Yuri Korzunov
(AuI ConverteR 48x44 developer)

 

[dprimary]

Thanks for the help.
It looks like http://downloadsnow.net and http://www.2l.no/hires/index.html are the closest to what I am looking for. The both list the recording format. Download.net did not seem to have any recording that had been originally recorded in DSD128 all of them started as DSD64, 2l seems to mostly be recorded on DXD.

 

[Music Alchemist]

FYI, when DSD is properly converted to PCM, there's no audible difference between the two. Same goes for when you convert 24-bit to 16-bit, or even lossless to 256 kbps AAC. The real benefit of high-res audio has nothing to do with the fact that it's high-res per se, but simply because the recording, mix, master, and so on are of higher quality. Oddly enough, certain software programs (such as HQPlayer) can resample audio in real-time in a way that does make an obvious difference.

 

[Yuri Korzunov]

Hi Music Alchemist,
 
When we oversample audio file, it doesn't improve sound quality of audio file as itself.
 
However DAC do oversampling inside to own "native" (without further oversampling) sample rate.
 
Oversampling need for applying more (than analog filter) qualitative PCM  DAC's digital filter.
 
I suppose DSD DAC use digital filtration too. Digital filtration help improve sound.
 
DAC as any hardware decission is limited in computing resources.
 
Computer based inline oversampler almost don't limited in computing power - available memory, CPU speed, big CPU cache, multi-core. But time conversion of music fragment must be less its time duration.
 
Computer based offline oversampler we can consider as not limited in using computer power. Of course practically we don't ready wait infinite time while files will converted However offline converter able work long time in background, applying "heaviest" algorithms.
 
Of course, impossibly guarantee sound improving via oversampling. And oversampling must be performed on pro level.
 
It depend on full system "audio file - audio player - driver - operation system - DAC - analog part".
 
Resume:
High resolution don't improve containing in file information, however allow DAC work in more optimal mode.
 
Best regards,
Yuri Korzunov

 

[Music Alchemist]

  Hi Music Alchemist,
 
When we oversample audio file, it doesn't improve sound quality of audio file as itself.
 
However DAC do oversampling inside to own "native" (without further oversampling) sample rate.
 
Oversampling need for applying more (than analog filter) qualitative PCM  DAC's digital filter.
 
I suppose DSD DAC use digital filtration too. Digital filtration help improve sound.
 
DAC as any hardware decission is limited in computing resources.
 
Computer based inline oversampler almost don't limited in computing power - available memory, CPU speed, big CPU cache, multi-core. But time conversion of music fragment must be less its time duration.
 
Computer based offline oversampler we can consider as not limited in using computer power. Of course practically we don't ready wait infinite time while files will converted However offline converter able work long time in background, applying "heaviest" algorithms.
 
Of course, impossibly guarantee sound improving via oversampling. And oversampling must be performed on pro level.
 
It depend on full system "audio file - audio player - driver - operation system - DAC - analog part".
 
Resume:
High resolution don't improve containing in file information, however allow DAC work in more optimal mode.
 
Best regards,
Yuri Korzunov

 
Yup! I thought it was very interesting how no matter you convert the files yourself, it doesn't change the sound, but HQPlayer sounds dramatically better (at least on my system) due to the processing it does.

 

[dprimary]

   
Hi Dprimary,
 
For comparing need know where original and where converted.
Here we can comparing audio file converter's quality.
 
Sometime possibly meet simultaneous records in PCM and DSD.
However there the will comparing mastering art.
 
Different audio resolution format possibly can give improvement of sound if it match to DAC's "native resolution" (without changing resolution and format DSD/PCM).
 
There two ways: buy matched resolution or convert resolution at home.
 
Me seems, fully correct comparing of DSD and PCM versions of the same recording is impossible by technical reasons.
 
Best regards,
Yuri Korzunov
(AuI ConverteR 48x44 developer)

That is my plan, I have DAC's of the same model. Through those I will play the two files and listen to them through a switch box.
 
Since I have not messed with DSD in the past I was hoping to find recordings that have been released in both formats already. If I can hear any difference then I will start looking into converting them myself.
 
Playing through two DAC's is not ideal, hopefully any production variations are inaudible. It does add more variables though. Then there is the question of software and what processing it may do to the two formats.

 

[Music Alchemist]

  That is my plan, I have DAC's of the same model. Through those I will play the two files and listen to them through a switch box.
 
Since I have not messed with DSD in the past I was hoping to find recordings that have been released in both formats already. If I can hear any difference then I will start looking into converting them myself.
 
Playing through two DAC's is not ideal, hopefully any production variations are inaudible. It does add more variables though. Then there is the question of software and what processing it may do to the two formats.

 
There's a little problem in your strategy: even if you buy/download DSD and PCM versions of an album from the same source, that does not guarantee they were derived from the same master or were even converted properly. It's a little tricky to convert DSD to PCM, but when you do it right, the differences are measurably inaudible. The only potential difference in a proper test would be if your gear cannot properly play everything. To do the type of test you want to do, it would be best to just get your hands on some DSD, then convert it yourself -- but only after learning how to do it right. But I guess if you're gonna first see if you can hear a casual difference in the first place, without jumping to conclusions, that's fine too.

 

[Yuri Korzunov]

 
That is my plan, I have DAC's of the same model. Through those I will play the two files and listen to them through a switch box.
 
Since I have not messed with DSD in the past I was hoping to find recordings that have been released in both formats already. If I can hear any difference then I will start looking into converting them myself.
 
Playing through two DAC's is not ideal, hopefully any production variations are inaudible. It does add more variables though. Then there is the question of software and what processing it may do to the two formats.

Two DAC have different sound way. Here possibly said about comparin of two systems. System as solid thing "file format - player - driver - DAC".
 
Even if playback 2 PCM with different sample rates it pass different filters or same filters in different mode.
 
However here talking about final result. Thus we can consider better system. Based on PCM or based on DSD.

 

[stv014]

The DAC difference issue (which should not be audible with well designed devices, but it is always best to minimize the sources of potential problems, including artifacts like clicking when switching sample rates or formats) can be avoided by comparing the DSD original to a version that went through a DSD->PCM->DSD conversion loop. This also makes sure there is no mastering difference. Although the extra conversion may in theory degrade the sound quality to some - ideally inaudible - extent, if the goal of the ABX test is to find out whether there is an inherent audible advantage of DSD as a distribution format over PCM, and no difference is detected in this setup, then the double conversion does not make the negative result less valid.
 

It's a little tricky to convert DSD to PCM, but when you do it right, the differences are measurably inaudible.

 
Converting DSD to PCM is not really much trickier than any other sample rate conversion, it is just computationally more expensive because of the unusually high input sample rate. PCM to DSD is more difficult because of the noise shaping.

 

[Yuri Korzunov]

 
Converting DSD to PCM is not really much trickier than any other sample rate conversion, it is just computationally more expensive because of the unusually high input sample rate. PCM to DSD is more difficult because of the noise shaping.

 
Sample rate conversion is enought complicate processing As example my alphaC sample rate conversion algorithm I began in 2009 and permanently improves today. Also I have many unresrearced area there. Including by suggestions of users of my software.
 
Of course with well known scheme of sigma delta modulator, as said one manufacturer in discussion about DSD, there "devil in details".
 
However DSD decoder also important part of processing. It can give enought sound advantages.
 
In the articles we can see what for different combinations of encoders/decoders we can get different results (more 20 dB - see results both articles together)
 
http://archimago.blogspot.ru/2014/04/analysis-comparison-of-dsd-encoders.html
 
http://archimago.blogspot.ru/2015/04/analysis-dsd-to-pcm-2015-foobar-sacd.html?showComment=1428926408403#c6196240508548468877

 

[limpidglitch]

   
Sample rate conversion is enought complicate processing As example my alphaC sample rate conversion algorithm I began in 2009 and permanently improves today. Also I have many unresrearced area there. Including by suggestions of users of my software.
 
Of course with well known scheme of sigma delta modulator, as said one manufacturer in discussion about DSD, there "devil in details".
 
However DSD decoder also important part of processing. It can give enought sound advantages.
 
In the articles we can see what for different combinations of encoders/decoders we can get different results (more 20 dB - see results both articles together)
 
http://archimago.blogspot.ru/2014/04/analysis-comparison-of-dsd-encoders.html
 
http://archimago.blogspot.ru/2015/04/analysis-dsd-to-pcm-2015-foobar-sacd.html?showComment=1428926408403#c6196240508548468877

 
Sure, 20dB difference, but when the worst performer already has a noise level 150-160dB below full-scale, I think it's about time to readjust your priorities.