[Yuri Korzunov]

Thank you.
 
DSD DAC divide 1 output to 2 lines or it's 2 DSD DACs?

 

[maverickronin]

Note that in most cases with a single output from the original source (DSD in this case), you will need the PCM loop input and output both through the switch box. Essentially, like a tape loop.

 
Depends how complicated you want to make the diagram but that should be enough for the basic premise.
 
After all I left out the amplifier too...

 

[maverickronin]

  Thank you.
 
DSD DAC divide 1 output to 2 lines or it's 2 DSD DACs?

 
It's deliberately simplified, so it's basically assuming either there are two separate outputs or that you split the one output to two lines.  If it only has one output, a tape loop like BlackbeardBen suggested would be the way to go.

 

[BlackbeardBen]

Thank you.

DSD DAC divide 1 output to 2 lines or it's 2 DSD DACs?

A good hardware ABX comparator is going to have a tape out/in function to allow a single input to the comparator looped out for processing and thus comparison with a pass-through.

Depends how complicated you want to make the diagram but that should be enough for the basic premise.

After all I left out the amplifier too...

Right. But it seems there's still some confusion.

Heck, you can even use an ABX comparator with speaker (or amplified headphone) level capabilities for comparing level matched amplifiers!

 

[maverickronin]

Heck, you can even use an ABX comparator with speaker (or amplified headphone) level capabilities for comparing level matched amplifiers!

 
That sounds like a fun kinda project.  A fully automated hardware ABX switchbox.  Need to get an Ardiuno, a screen for it, some high current relays, and a few more connectors...

 

[Yuri Korzunov]

   
It's deliberately simplified, so it's basically assuming either there are two separate outputs or that you split the one output to two lines.  If it only has one output, a tape loop like BlackbeardBen suggested would be the way to go.

 
Thank you. Me seems that scheme fully correct.
 
1. I watch here we compare [certain system PCM ADC+DAC] vs. [source analog signal].
 
I.e. we estimate how impact the [certain system PCM ADC+DAC] to [source analog signal].
 
2. Same experiment we can done with [certain system DSD ADC+DAC] vs. [source analog signal].
 
Like case #1 we estimate how impact the [certain system DSD ADC+DAC] to [source analog signal].
 
 
3. We can realtime switch between case #1 and case #2.
 
However we will compare 2 systems (i.e. devices) not formats.
 
I correctly understand gist of experiment?

 

[Yuri Korzunov]

A good hardware ABX comparator is going to have a tape out/in function to allow a single input to the comparator looped out for processing and thus comparison with a pass-through.
Right. But it seems there's still some confusion.

Thank you for details.

 

[maverickronin]

  However we will compare 2 systems (i.e. devices) not formats.
 
I correctly understand gist of experiment?

 
Since you need different hardware to play DSD and PCM you can only test differences between the formats indirectly.  If DSD is audibly superior to PCM then the addition of the PCM ADC/DAC conversion should degrade the signal and produce audible differences.
 
You can't strictly conclude that that one format is superior or that the they are equivalent with this method because there is different hardware involved, but it's like that with every other kind of test you could do since there's always hardware involved somewhere.

 

[Yuri Korzunov]

   
Since you need different hardware to play DSD and PCM you can only test differences between the formats indirectly.  If DSD is audibly superior to PCM then the addition of the PCM ADC/DAC conversion should degrade the signal and produce audible differences.
 
You can't strictly conclude that that one format is superior or that the they are equivalent with this method because there is different hardware involved, but it's like that with every other kind of test you could do since there's always hardware involved somewhere.

Yes. Even we can replace DSD to PCM DAC (on upper part of scheme). It don't impact anything.
 
Same example in 44 vs. 192 kHz domain: processional DAC very possibly give better result in 44 kHz than consumer DAC at 192 kHz. Diferent apparatus, not advantages of standard here too.

 

[KeithEmo]

Actually, you've got it a bit backwards there..... the filtering, and so the potential to alter the sound as part of the conversion process, is much MORE extreme with an A-to-D or D-to-A conversion than with a relatively simpler digital-to-digital format conversion.
 
When you play that DSD file through the DSD DAC, the DAC is going to alter the sound (because filtering is applied by the DAC), then the PCM ADC process is going to apply filtering, then the PCM DAC is going to apply filtering again .... You've gone through THREE separate conversions, and each of those conversions is going to alter the sound. You're also starting from an assumption that the output of the DSD DAC is at least good enough to count as a "source" for the sake of our test. (What if the DSD recording sounds different, not because it's somehow "capturing something that PCM is missing", but simply because it's coloring the sound? If that's the case, then we'd also like to know that.)
 
(I could even really play devil's advocate and suggest that it just might be that DSD is adding some sort of coloration, and PCM then removes that coloration, thereby "fixing a defect" caused by DSD, rather than removing something beneficial that DSD succeeds in capturing and PCM does not. Even if there really is some "magical difference in sound" between DSD and PCM, we still don't know if it's an improvement or a reduction in quality.... just as deliberately adding second harmonic distortion can make some recordings sound "more realistic" - even though it is clearly added distortion.)
 
The proper way to test both formats, and to compare them, would be to start with an analog source, then convert it into digital and back into analog, using both DSD and PCM, and see how the results compare - to each other and to the original analog source. (Of course, even then, unless you repeat the test lots of times, with lots of different ADCs and DACs, you won't know if any differences you hear - or don't hear - are because of the format or just because of how the specific brands and models of devices you're using perform. It's quite possible that a certain DSD DAC may be better than another certain PCM DAC, or vice versa, for some reason other than because of the format used.)
 
At the very minimum, if we're going to do a PCM recording and playback of a DSD "original", then we should also do a DSD recording and playback of a PCM original. My expectation is that either process is going to alter the sound slightly... and, if so, then this whole "debate" boils down to which coloration you personally find preferable. (If you're going to complain that the PCM recording "doesn't sound exactly like the original", then clearly we also need to know whether the DSD recording and playback does or not as well.)
 
My personal predictions are these:
 
1) Not all DSD DACs, or all PCM DACs, sound the same, and so I wouldn't necessarily expect any two of either type to sound identical.
2) Neither DSD nor PCM will produce a result that is identical to the original source, both will be slightly different from each other, and from the original, and in fact each individual DAC will sound different, but neither format will overall be obviously closer to that original.
3) The differences between DSD and PCM DACs will be about the same magnitude as the differences between different DSD DACs, or different PCM DACs.
 
These are based on the simple fact that all of the scientific analysis of the subject I've seen shows that neither format is superior or different in any meaningful way. There are all sorts of reasons why different DACs, or different ADCs, sound different... and I think that DSD vs PCM is just one more item that might rate being added to that list. I also don't find it in the least surprising that DSD recordings (which are, after all, only produced nowadays "for sale to audiophiles", and which are also much more difficult to edit than PCM) would be mastered and processed differently than PCM recordings. I personally suspect that it is this different mastering that is responsible for all of what people continue to insist is some difference in the sound quality of the format itself. And, finally, I don't find it surprising that a particular DAC might be optimized to sound best with one or the other format - either because the designer considered that one to be "more important", and optimized the filters and other circuitry for it, or simply due to blind luck. (In fact, if I was designing a DSD DAC, and this fact was to be a major selling point for that product, I'd make damned sure that DSD and PCM didn't sound the same on my product - because, after all, that difference in sound is a major selling point for my product.)
 
  Quote:

   
Actually BlackbeardBen is right on this one.  The key phrase is...
 
 
We're not talking about digitally converting a DSD file to PCM, we're taking the output of a DSD DAC, feeding it into a PCM ADC and then out through a PCM DAC.  This sidesteps all the issues about conversion quality and filter optimization since there is not digital format conversion and there are two separate DACs.  If DSD has "something" audible which PCM doesn't, then the ADC/DAC loop will degrade the signal and get rid of that 'something', whatever it may be.  If you can switch the ADC/DAC loop in and out without audible differences then the conclusion will be that PCM is at least as good as DSD.
 
I'd forgotten about it before, but this is is a pretty standard way to test formats so different that they can't be played back with the same hardware.  For example, there have been tests where a PCM ADC/DAC has been switched in and out while listening to vinyl and people couldn't tell the difference either.

 

[RRod]

When you say that not all DACs "sound the same", what do you mean? It doesn't seem like people are able to ABX differences between *competent* consumer-level DACs, so I'd hope that pro-grade stuff is even more transparent.

 

[jodgey4]

  When you say that not all DACs "sound the same", what do you mean? It doesn't seem like people are able to ABX differences between *competent* consumer-level DACs, so I'd hope that pro-grade stuff is even more transparent.

If they all have distortion figures below audible levels, and a full frequency response, and a low enough noise floor - any differences should be inaudible to the human ear. So literally, some say all DACs sound the same (ES9018 no different from a WM8740). Some mean it to say they all sound the same in ABX testing as well... depends on who you ask.

 

[analogsurviver]

 

At long last - a really good post.
 
I'm dead beat from the whole day of recording, but in a nutshell - everything you said does hold water.

 

[RRod]

  If they all have distortion figures below audible levels, and a full frequency response, and a low enough noise floor - any differences should be inaudible to the human ear. So literally, some say all DACs sound the same (ES9018 no different from a WM8740). Some mean it to say they all sound the same in ABX testing as well... depends on who you ask.

 
Yes but he said they *don't* sound the same, so I'm wondering what differences people are hearing that could actually be verified in a DBT.

 

[maverickronin]

 

Actually, you've got it a bit backwards there..... the filtering, and so the potential to alter the sound as part of the conversion process, is much MORE extreme with an A-to-D or D-to-A conversion than with a relatively simpler digital-to-digital format conversion.
 
When you play that DSD file through the DSD DAC, the DAC is going to alter the sound (because filtering is applied by the DAC), then the PCM ADC process is going to apply filtering, then the PCM DAC is going to apply filtering again .... You've gone through THREE separate conversions, and each of those conversions is going to alter the sound. You're also starting from an assumption that the output of the DSD DAC is at least good enough to count as a "source" for the sake of our test. (What if the DSD recording sounds different, not because it's somehow "capturing something that PCM is missing", but simply because it's coloring the sound? If that's the case, then we'd also like to know that.)

 
And people still probably won't be able to tell the difference just like they can't with vinyl.  (I totally forgot that link was in the OP... 

  Oh well, it's been almost 5 years.)
 
Part of the point is basically to cede the high ground to the format under test (DSD) and see if the PCM conversion screws it up.  If it doesn't PCM passes.  If it does, then further tests are called for.
 
No one experiment is going to tell you everything you want to know but this one will tell you something pretty useful.  Having an analog master and making a separate DSD and PCM recording of it it a fine idea too, but it's much less practical. I'd say that's it not even necessary unless someone can first ABX the PCM conversion in the tape loop and then ABX it in revers with a PCM  source and DSD conversion in the middle.