DSD comparison to PCM.

[Ragnar]

There's much more to being PCM format than merely not being DSD. Most people listen to PCM nowadays on non-PCM DACs and which has been recorded with non-PCM ADCs. For example, your Sound Devices 722 (having checked it out) doesn't use a pure PCM ADC so when you say that its not sounding as good as DSD, you're not actually listening to a PCM system.

Looks like we're back to: the only pure PCM system around is the Pacific Microsonics Model 1; the Model 2 if you allow for high level dither and decimation.
Unless the MSB ADC is a flash or successive approximation type. Their dac is a ladder type last I knew.

Bruce, can you comment?

 

[opus111]

I looked over the MSB's user manual and they're coy, as usual. They don't say explicitly they're using an SAR type ADC but do suggest they're applying the same principles to ADC design as they've done to DACs. And they do have a steep AAF (extended filter) which tends to indicate its not an S-D type ADC. Presumably Chesky switched from their previous S-D type ADC for good reason so the indications (not conclusive) are that this MSB ADC is the real deal.

 

[LL21]

...In any event, you can get great results from either format (and conversions from one to the other are pretty easy). I just prefer DSD.

What I do know from experience is that 16/44 is entirely inadequate for proper capture of an audio event.

How much does this apply to the recording/mastering side vs final format production side? 50-50? IOW, how much does one capture if using hires PCM from recording thru mastering and then put onto 16/44 CD?

 

[Lee]

How much does this apply to the recording/mastering side vs final format production side? 50-50? IOW, how much does one capture if using hires PCM from recording thru mastering and then put onto 16/44 CD?

You are still downrezzing and that brings its own problems!

You are throwing away valuable musical data. You really need hirez to get the full event.

 

[Lee]

There's much more to being PCM format than merely not being DSD. Most people listen to PCM nowadays on non-PCM DACs and which has been recorded with non-PCM ADCs. For example, your Sound Devices 722 (having checked it out) doesn't use a pure PCM ADC so when you say that its not sounding as good as DSD, you're not actually listening to a PCM system.

You are splitting hairs here. By definition you are throwing away all common implementations of PCM. I am less concerned about technical minutiae-I want to know what sounds best in practice and that is DSD.

 

[LL21]

You are still downrezzing and that brings its own problems!

You are throwing away valuable musical data. You really need hirez to get the full event.

Thanks for that. I gathered as much...the question is: if we compare: A) hires from recording to mastering to putting it onto disc in hires format...vs B) the same hires track all the way thru the same hires mastering until the very final transfer which is 'down-res'd' to 16/44.1 disc...what % of the total quality do you think we've lost between A and B? Thanks for any guidance.

 

[Lee]

Thanks for that. I gathered as much...the question is: if we compare: A) hires from recording to mastering to putting it onto disc in hires format...vs B) the same hires track all the way thru the same hires mastering until the very final transfer which is 'down-res'd' to 16/44.1 disc...what % of the total quality do you think we've lost between A and B? Thanks for any guidance.

30% to 50% in my view/experience. I hear more width and depth to the soundstage and instrument realism is dead on at 24/176 or higher.

 

[opus111]

You are splitting hairs here. By definition you are throwing away all common implementations of PCM.

I'm by no means throwing them away, just not calling them pure PCM because they have elements of non-PCM.

I am less concerned about technical minutiae-I want to know what sounds best in practice and that is DSD.

I have no quibbles with your preferences for SQ, just pointing out that you're not comparing DSD to the best that PCM has to offer - an uncompromised PCM signal chain. If you're at all curious to know why a technically compromised format (DSD64) sounds better to you than what you've heard of PCM, that's why - you have a technically compromised version of PCM.

 

[LL21]

30% to 50% in my view/experience. I hear more width and depth to the soundstage and instrument realism is dead on at 24/176 or higher.

Wow...thanks! I am surprised to hear that...and that is great to know. I thougt that would be the case but mainly with great hires material. my experience of hi res is extremely limited...maybe even a false experience. We had a few hires versions on server, SACD to compare to CD...and found very little if any improvement in the hires versions. System was Wadia s7i, Esoteric X01SE and ARC CD8 with seprver and transporta. That said, I could believe the potential for hires is great. I also just see so little software out there compared to CD...and CDs are so cheap, you can build a huge collection a few bucks each CD.

When it becomes more mainstream, i will look much more into hires digital.

 

[Lee]

I have no quibbles with your preferences for SQ, just pointing out that you're not comparing DSD to the best that PCM has to offer - an uncompromised PCM signal chain. If you're at all curious to know why a technically compromised format (DSD64) sounds better to you than what you've heard of PCM, that's why - you have a technically compromised version of PCM.

I've heard the DSD comparison across a wide variety of DACs, not just the Sound Devices.

 

[Lee]

Wow...thanks! I am surprised to hear that...and that is great to know. I thougt that would be the case but mainly with great hires material. my experience of hi res is extremely limited...maybe even a false experience. We had a few hires versions on server, SACD to compare to CD...and found very little if any improvement in the hires versions. System was Wadia s7i, Esoteric X01SE and ARC CD8 with seprver and transporta. That said, I could believe the potential for hires is great. I also just see so little software out there compared to CD...and CDs are so cheap, you can build a huge collection a few bucks each CD.

When it becomes more mainstream, i will look much more into hires digital.

Because the musicians we record mostly don't have DVD-Audio players, we carefully downrez to CDs for them. So for each 24/176 recording we do of these classical ensembles, we also hear a CD version based on the 24/176 digital file. It's remarkable how much worse the CD sounds on every recording even when we make a copy and play it back on a reference level CD player.

 

[opus111]

I've heard the DSD comparison across a wide variety of DACs, not just the Sound Devices.

Any pure PCM DACs? - they'd be ones with multibit converters (PCM1704 is one example, TDA1541A another). Any pure PCM ADCs (Pacific Microsonics and, most probably, MSB)? Your 'reference level' CD player is?

 

[tailspn]

I would suggest that the majority of classical music (IMO the only acoustical events that a recording's fidelity can be judged) recording engineers who use DSD, that their experience mirrors Lee's. I personally attended numerous Boston Symphony concerts recorded simultaneously in both 64fs DSD and 88.2KHz PCM, and the differences always favored the DSD session masters.

These engineers may not have been using the "best" PCM A/D converters, but I don't think that would alter the results. There are just too many experienced recording engineers that have heard the differences between DSD and PCM, with the converters that can be bought today, to create a market for ADC manufactures to build professional PCM converters like the PM Model one and two.

Here's an easy test. Go get a copy of any Reference Recording production (recorded with the PM ADC), and compare it to any recent (Grimm ADC) Channel Classics recording. You be the judge.

 

[opus111]

These engineers may not have been using the "best" PCM A/D converters, but I don't think that would alter the results. There are just too many experienced recording engineers that have heard the differences between DSD and PCM, with the converters that can be bought today, to create a market for ADC manufactures to build professional PCM converters like the PM Model one and two.

So your argument is - because so many recording engineers have heard current implementations (not pure ones) of PCM and that those suck, that there's no market for PCM done properly? If so I can't follow the logic - you seem to be saying that professional recording engineers are all closed minded, that they've decided to jettison PCM to history and no listening impressions with decent PCM will ever change their minds. If that is indeed your argument that doesn't reflect very well on the mindsets of the engineers you're citing.

 

[manisandher]

Here's an easy test. Go get a copy of any Reference Recording production (recorded with the PM ADC), and compare it to any recent (Grimm ADC) Channel Classics recording. You be the judge.

Haha, not so easy in practice. You'll need a true PCM 24/192-capable DAC (not many around!) to play back the 24/176.4 RR files. And of course, you'll need a [Edit: didn't see reference to CC recordings] high quality DSD DAC to play back the DSD files. I'm capable of doing the former but not the latter. Quite honestly, if I were not happy with the former I would order a Grimm AD1 and EMM/Meitner/Playback-Designs DAC (essentially replicating Bruce's setup). BUT... having compared DSD vs. 'pure' PCM, I'm happy enough with the results I'm getting with PCM to not feel the need for any further improvement.

Mani.

 

[PeterSt]

FWIW :

A couple of years ago I made a home recording of the drumkit we have hanging around here. Not the smallest with all sorts of cymbals (10 or so) and rototoms and such. I did it with two Behringer measurements mics through a stupid RME Fireface 800 at 16/44.1 (also at 24/96 and 24/192).
This is by infinitely far the best I *ever* heard from any drums on recordings (over 30,000 albums here). But I could also say : this 100% resembled the drumkit in all its aspects. Totally live sounding cymbals with the right color and singing and length and everything. One thing maybe : the sound pressure of the kickdrum could be better, but then this needs something like 4 15" drivers per channel, while I have one only.

I often show this recording after first showing the live drumkit in its room (obviously), played back in a completely different room.
100%, including the aspects of the room where it was recorded in.

There is no single way any recording I own comes even close to this recording, because cymbals sound not long enough, are too metal (not enough colour) and all what we know of recordings and how they not resemble reality.

Oh, Mani heard it too. So, he may respond with some other things heard in the "not 100% realm" and I am fine with that. But the infinitely better than all we know should come out of it or otherwise I'll eat something.
... of course with the disclaimer that it is impossible that I could do this better than anyone else, and to be honest two weeks ago I found an album in the same "leage", but point is :

We all (including me) can't judge merits as long as all recordings are so far from reality in the first place.


Keep in mind, this was 16/44.1 recorded through an SDM and playback is through sheer PCM (the 16/44.1 filtered/upsampled to 24/705.6 and no ringing).
So, reading through the last posts about what (combination) is best, I'd say that at least recordings through an SDM (not DSD) but played back through (1704) PCM is able to do it. Whether the "no decimation" is in order here I can't tell (dunno what the RME is doing inside) but at least the hires recordings did not sound better.

I mentioned the 15" drivers for a reason, because while the kickdrum was the only element of which can be said "hey, *there* you can hear it's not real live", the deep-down-earth slam which is needed for good kickdrum representation really needs some speakers. And to that regard Mani knows that too by now (referring to Bert's Orelo's).

As I said, FWIW. But for someone who is day in day out working on playback improvement as a goal for life, it really tells me something. Sadly it also tells me that I can try what I want, but as long as recordings are so far off from what can be (proven !), what to do next ...

Peter


PS: We may know that drums start to sound better through playback when the toms can be differentiated nicely. The next level of that is how you can hear how much tension is put to the skins. Most of the recordings miss that at all.

 

[tailspn]

So your argument is -

That if the recording engineering community at large believed "PCM done right" converters were a significant contributor to better sound, there would be numerous manufactures building them. The word gets around, and except for a few evangelists, there is no push in that direction that I'm aware of. Just the opposite. The move is towards native DSD recording for acoustic music recorded in a natural environment. It just sounds more like the analog mic feeds.

 

[Ragnar]

That if the recording engineering community at large believed "PCM done right" converters were a significant contributor to better sound, there would be numerous manufactures building them. The word gets around, and except for a few evangelists, there is no push in that direction that I'm aware of. Just the opposite. The move is towards native DSD recording for acoustic music recorded in a natural environment. It just sounds more like the analog mic feeds.

Perhaps........

"Humankind cannot bear very much reality." T.S. Eliot

 

[Orb]

That if the recording engineering community at large believed "PCM done right" converters were a significant contributor to better sound, there would be numerous manufactures building them. The word gets around, and except for a few evangelists, there is no push in that direction that I'm aware of. Just the opposite. The move is towards native DSD recording for acoustic music recorded in a natural environment. It just sounds more like the analog mic feeds.

But even that is only half the story because I am sure you will find it is also being pushed that DSD is processed in the Studio/DAW as DXD-PCM 352.8/32bit and from this to whatever is required whether 192/24 or DSD.
So I am not sure about the native DSD being anything more than very niche when one considers the studio/DAW, especially as there also seems weight behind DXD (which is a PCM format).

Edit:
to clarify DXD is 24bit but mentioning 32bit with regards to possibility with DAW.
Cheers
Orb

 

[PeterSt]

Dxd

Maybe it is time that someone tells that DXD is the most ordinary 24/352.8 (PCM).
So I just did.