World’s First Valid Comparison of PCM versus DSD

[jap]

http://www.ayre.com/insights_dsdvspcm.htm

 

[audioarcher]

Not sure how this is valid at all. How were these files originally mastered? I would bet that they were not each mastered as optimal for each format. PCM will have an advantage if the original file was not native DSD, and vise versa.

 

[Bruce B]

These are "needle drops" that Charlie did.

Michael did the "World's First Valid Comparison of PCM versus DSD" files over at Audiostream November of last year!

 

[Robin Hood]

Not the world's first and not valid but definitely vendor hype. All this will prove is that the vendor can either make better PCM recordings or better DSD recordings, largely a function of the equipment, processes and skills used. If anyone dismisses either the PCM or DSD formats, then it's their loss.

Perhaps this is just another version of X is better than Y or Y is the best. Substitute analog, digital, tape, vinyl, 24/192 PCM, DSD128 or any other of a number of choices for these variables.

 

[amirm]

Not sure how this is valid at all. How were these files originally mastered?

Using Ayre's Analog to Digital converter from an LP source.

 

[ack]

The Ayre P-5xe is not resolving enough to be able to easily expose differences; so the experiment is not scientific enough. Having said that, I personally don't care about all the noise up in the higher frequencies with DSD; and I have yet to see any valid engineering arguments that DSD is solving problems better than PCM could ever do - and pushing noise out of the audible band does not cut it. For now, to me, DSD remains a solution in need of a problem, but I am willing to learn.

 

[microstrip]

It is surely an interesting experiment, but as valid or invalid as many others. It just compares two particular implementations of both technologies. Pretending it "will settle this issue once and for all," with a few needle drops is ridiculous. Anyway, we now know Charles Hansen opinion - no need to listen to the files for that!

One point - considering the flamed preamble, ideal to propitiate an unbiased the debate, I guess that the issue will soon become as passionate and vitriolic as the analog versus digital debates.

 

[audioarcher]

Whoops. I missed the bit about it being a needle drop. In that case as Robin Hood posted above, it only proves that Charles likes the PCM files better than the DSD files, through his A to D. There are so many factors, including different playback hardware and personal taste, that I doubt there will be any kind of consensus.

I tried to download both types of files but, I was only able to get the DSD files. For some reason my PC would just play the PCM files instead of download them. I will have to try it again on another PC.

 

[microstrip]

(...) For some reason my PC would just play the PCM files instead of download them. I will have to try it again on another PC.

If you use Firefox just open the options of your browser, select applications and mark the .wav type as save. It will save the file instead of playing it.

 

[audioarcher]

If you use Firefox just open the options of your browser, select applications and mark the .wav type as save. It will save the file instead of playing it.

I'm using Google Chrome. I do not see that option in the settings. I have not tried it with IE yet though.

 

[audioarcher]

I have them now. No problem through IE.

There is quite a difference between the sound of the Ayre DSD and PCM files. First the DSD files are at least 6 db louder than the PCM files. After doing my best to level match by ear (I know it's not ideal) the PCM files sound more transparent and accurate. The DSD files sound fatter and more euphoric. The difference is large enough I feel it would be easy to pick them out in a DBT.

From my limited experience of hearing other DSD A/D's I think the Ayre is not at the level of the top DSD A/D converters. I could be wrong (but I don't think so). Only way to know for sure would be to do a direct comparison between them.

 

[Bruce B]

it only proves that Charles likes the PCM files better than the DSD files, through his A to D. .

Guess Charlie needs go build a better DSD A/D converter then!

 

[audioarcher]

Guess Charlie needs go build a better DSD A/D converter then!

Yes, that is my conclusion.

 

[Orb]

TBH the issue is using the same digital architecture for both PCM and DSD IMO.
The only one (at consumer end for playback) that I think "may" be of near equal transparent conversion for both would be the dCS Vivaldi combined with the Upsampler, and even then those who are pure PCM or DSD could argue it is not the simplest architecture for either PCM or DSD. - Maybe another consideration the Chord Electronics with their Pulse Array DAC but this is not a simple solution as well, bear in mind there was an issue with some performance limitation bug at some point with the USB OEM solution that they used like a few other manufacturers did.

Anyway this is even further compounded by the ADC-recording chain where it can be seen the latest next gen top studio products from either Merging Technologies or DAD are going different ways with PCM/DSD; DAD have even come out stating they stopped using ESS Sabre in design of the new AX32 due to technical performance/behaviour minor anomalies.

And finally; what is the perfect upsampling rate (both native recording and chipset)/filter-algorithm-coefficients/etc, deemed the ideal for any such comparison?....Much of that is a major consideration for PCM but some does also apply to DSD.

IMO from a scientific perspective (controlled test,environment and chain) it would be nigh on impossible to do a worthwhile objective comparison-conclusion to what we all discuss on this subject due to the many different variables for each setup (PCM and DSD).
So this will be one of those topics that will be the new pheonix on the internet, forever rising on various forums and threads
On the plus side and the health of the music industry, I guess this means more music sales buying both PCM and DSD
Cheers
Orb

 

[tailspn]

Anyway this is even further compounded by the ADC-recording chain where it can be seen the latest next gen top studio products from either Merging Technologies or DAD are going different ways with PCM/DSD; DAD have even come out stating they stopped using ESS Sabre in design of the new AX32 due to technical performance/behaviour minor anomalies.

Was DAD using the ESS A/D converter chip previously, or are they referring to the 9018 DAC chip used in the D/A converter side? Did they change the A/D side of the unit from the AX-24 at all, other than add 8 more channels? And you're right, DAD and Merging with Horus have gone very different ways.

 

[asiufy]

Guess Charlie needs go build a better DSD A/D converter then!

Or use a proper converter in the comparison! Seems to me even the Korg can do better than his unit...




alexandre

 

[Bruce B]

Or use a proper converter in the comparison! Seems to me even the Korg can do better than his unit...
alexandre

Which is what I did with this comparison.

 

[microstrip]

Which is what I did with this comparison.

That suffers exactly from the same limitations and problems, except that the source material (master-tape) seems better suited - IMHO you were mostly comparing the Korg used as encoder with the Mytek used as decoder, not comparing the formats in an absolute way.

 

[asiufy]

I think a more proper comparison would be using both sides' best units. So, what's the best DSD-capable ADC out there? And which one's the best for PCM?
Use them both, and then play the files back on 3 different DACs, like a Vivaldi, an MSB and a Playback Designs.

microstrip,

Assuming no ADC or DAC is 100% transparent, there is no way to compare the formats without taking into account the "signature" of the device itself, no? Do you see a way that the format itself can be compared, isolating the device?


alexandre

 

[Bruce B]

That suffers exactly from the same limitations and problems, except that the source material (master-tape) seems better suited - IMHO you were mostly comparing the Korg used as encoder with the Mytek used as decoder, not comparing the formats in an absolute way.

The Korg was the only digital component used. You can't get more simpler than that!