DSD is better than PCM!

[Galliardist]

I agree. DSD makes sense for digitalising analogue masters, either of old recordings or new ones made with analogue equipment (e.g. Cookie Marenco/Bluecoast Records). Or perhaps "direct to master" un-edited recordings (e.g. PlayClassics, although Mario would not be able to "calibrate" the mic feeds).

If 99% of ADCs and DACs use a ΣΔ modulator then DSD is probably better (at least in theory).

No. The only thing to be said for DSD is that it is not audibly inferior in playback. As soon as we accept that it is not "better", any case to continue using it for any modern recording or processing disappears, because PCM is inherently more flexible, if nothing else.

I could cut down and chop up a tree with a flint axe, if I have one. At the end of the day, the tree is still cut down. But does that make it in any way a sensible choice if I also have a modern axe and a chainsaw to hand?

 

[Audiofire]

Anyway, the title of the thread is "DSD is better than PCM!" and I think we kinda nailed down and crucified that idea by now...

A technical argument in favor of DSD, or vacuum tubes for that matter, seems to be that we are biological organisms that could easily prefer the sound of distortion or another parameter that is not directly related to electronic measurements.

 

[Miska]

No. The only thing to be said for DSD is that it is not audibly inferior in playback. As soon as we accept that it is not "better", any case to continue using it for any modern recording or processing disappears, because PCM is inherently more flexible, if nothing else.

Too bad current DAC chips have inferior handling for PCM. But luckily number of chips offer way to bypass the inferior on-chip DSP and go straight to the D/A conversion section by using DSD.

Or even better, many DAC are available that have a discrete D/A conversion stage without trying to cram everything on a tiny $10 chip.

 

[Miska]

How many of these are truly single bit? How many of the state of art designs are truly single bit?

For example TI PCM4202 and PCM4204 chips, used on many professional "PCM" ADCs. Including RME devices. Likely you have listened to a lot of "PCM" recordings made with these chips. Which is actually result of "DSD128" to PCM conversion.

 

[Miska]

ADCs aren't single bit these days either. Haven't been for over a decade.

Yes there are many, even today you can buy brand new ADCs with those chips.

You can also buy true 1-bit DSD recorders, such as TASCAM DA-3000 and KORG MR-2000S.

But I make my DSD256 recordings with Merging Hapi (8 channels) and RME ADI-2 Pro (stereo).

 

[Miska]

It looks like all current and recent RME's use either Cirrus or mostly AKM's which are multi-bit.

At least I have some RME devices that use PCM4202. The most obvious one where it is visible is HDSPe AIO. And what I remember the 4-channel PCM4204 model has also been in some other ADC models.

The AK5385 and AK5388 used on some other models (Fireface) is 2-bit ADC (Dual Bit as AKM calls it), kind of running 1-bit modulator for each differential pair. Same for other AK4621 chip they use.

E-MU has used PCM1804 which is older version of the PCM4204 architecture, capable of only DSD64.

 

[krabapple]

It's worth remembering that the Lipshitz and Vanderkooy paper was published right in the middle of the SACD vs DVD-A war. They weren't materially involved in that war, as far as I know, but the paper got a lot of one-sided publicity at the time including claims that went well beyond what it actually says. It doesn't consider the full effects of noise-shaping or further upscaling of the DSD format. As for space, I don't have that problem (most of what I play is PCM, only converted to DSD in the player anyway).

Correct. The problem L&V identified was basically 'solved' in succeeding years by eg "DSD wide" (aka 'PCM narrow')

So where does that leave us? DSD does what it does. SACD is actually STILL the main way to buy surround music, until multichannel streaming takes over at some point in the future. For those of us using stereo, we don't have to go anywhere near DSD unless we want the particular mastering of a DSD stereo layer. There is neither magic nor horror to be seen here. Just accept that and move on.

+1!

and adding a reiterative 'it's the mastering, stoopid'

 

[krabapple]

Well some Sony engineers said sacd was to create a new format Sony owned as patents were expiring on CD.

the creation of SACD and the creation of DSD had two different motives.

 

[krabapple]

View attachment 186438

the appropriate response to such nonsense:

 

[krabapple]

What I'd like to know is where the heavy, deep, abiding faith in the superior nature of DSD comes from no matter what evidence is presented.

Well the usual audiophile rag suspects, and certain industry folk connected to DSD/SACD , promoted the idea that PCM was problematic, and that DSD/SACD was the solution.

It's of a piece with Bob Stuart's (Meridian) relentless promotion of the idea the consumers needed 'hi rez' delivery formats, starting in the 1990s. The questionable premises re: audibility were never an obstacle for them.

 

[CtheArgie]

As this started as a PS Audio "think piece", I find it very ironic that Paul claims there is inherent superiority to DSD and that he can keep the chain completely clean by recording in DSD and his DAC converting to analog. But in another post he states that he uses a miniDSP as a crossover in his IRS speakers (though he now uses the ones he is selling). I asked how he can claim DSD purity if the miniDSP will convert the signal to PCM and back for the crossover. He made no comment.

 

[Blumlein 88]

Yes there are many, even today you can buy brand new ADCs with those chips.

You can also buy true 1-bit DSD recorders, such as TASCAM DA-3000 and KORG MR-2000S.

But I make my DSD256 recordings with Merging Hapi (8 channels) and RME ADI-2 Pro (stereo).

So when you make the DSD 256 recordings they are the type where you don't do any processing? And do you think the DSD sounds different than say 192/24?

 

[Blumlein 88]

At least I have some RME devices that use PCM4202. The most obvious one where it is visible is HDSPe AIO. And what I remember the 4-channel PCM4204 model has also been in some other ADC models.

The AK5385 and AK5388 used on some other models (Fireface) is 2-bit ADC (Dual Bit as AKM calls it), kind of running 1-bit modulator for each differential pair. Same for other AK4621 chip they use.

E-MU has used PCM1804 which is older version of the PCM4204 architecture, capable of only DSD64.

Yeah, I missed that the AKM's have Dual bit and that it is sort of balanced one bit DSD the way it works.

PS-that is why deleted that comment, unfortunately you posted in between me doing the deletion. Wasn't trying to mess up the thread.

 

[Galliardist]

Too bad current DAC chips have inferior handling for PCM. But luckily number of chips offer way to bypass the inferior on-chip DSP and go straight to the D/A conversion section by using DSD.

Or even better, many DAC are available that have a discrete D/A conversion stage without trying to cram everything on a tiny $10 chip.

Simple question. Is playback actually affected by the "inferior handling" on these chips? If it is, then why not convert to DSD externally and then feed the DAC? i've no problem with that. If the problem is inaudible, then the question is whether it's worth worrying about in practice.

I allowed for this in an earlier answer to a question about the "HQPlayer way". I simply assume there's no magic here, and that HQPlayer is a useful tool in different circumstances, not an ideology. If using it improves the sound, use it.

As far as DSD goes, I'll reiterate that everything that plays on my main system has been processed to DSD. I have no problem with DSD whatsoever. My rather pointed analogy was aimed at ADC output and studio processing of audio.

As an aside, I don't diss the flint axe, either. We wouldn't be where we are today without the invention and use of the flint axe.

 

[pinpoint_oxford]

As this started as a PS Audio "think piece", I find it very ironic that Paul claims there is inherent superiority to DSD and that he can keep the chain completely clean by recording in DSD and his DAC converting to analog. But in another post he states that he uses a miniDSP as a crossover in his IRS speakers (though he now uses the ones he is selling). I asked how he can claim DSD purity if the miniDSP will convert the signal to PCM and back for the crossover. He made no comment.

I admit I haven't read the piece, but on the surface, wouldn't the miniDSP just be when listening to the product and not used in "mastering" of the track?

 

[Miska]

So when you make the DSD 256 recordings they are the type where you don't do any processing? And do you think the DSD sounds different than say 192/24?

I do processing with my own DSP tools as necessary. 192/24 is pretty far from that and I don't see a reason to use such.

But for example if I look noise profile of 705.6/32 PCM and DSD256 from ADI-2 Pro ADC both look pretty much the same. I would say practically difference between the two is fairly small. But D/A conversion of the 705.6k requires further rate conversions in most cases.

 

[Miska]

Simple question. Is playback actually affected by the "inferior handling" on these chips?

Measurable yes, which is point of this forum.

 

[CtheArgie]

I admit I haven't read the piece, but on the surface, wouldn't the miniDSP just be when listening to the product and not used in "mastering" of the track?

The point is that he keeps discussing the benefits of going from DSD to analog without any kind of "digital" but then uses the miniDSP before the speakers but doesn't mention that this requires a ADC-DAC processing. So it is not "virginal and pure".

 

[pinpoint_oxford]

The point is that he keeps discussing the benefits of going from DSD to analog without any kind of "digital" but then uses the miniDSP before the speakers but doesn't mention that this requires a ADC-DAC processing. So it is not "virginal and pure".

I see your point now. Technically though it did go DSD > Analog before the dsp .

 

[Galliardist]

Measurable yes, which is point of this forum.

Audible? - which is the point at which intervention is needed