Skip to main content ↑ ↓

Notice

Please note that most of the software linked on this forum is likely to be safe to use. If you are unsure, feel free to ask in the relevant topics, or send a private message to an administrator or moderator. To help curb the problems of false positives, or in the event that you do find actual malware, you can contribute through the article linked here.
Topic: About DSD-native audio editing tools: ALL of them PCM conversions? (Read 29687 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

About DSD-native audio editing tools: ALL of them PCM conversions?

Hi, I put the same thread on head-fi, but it seems I am not going to get an answer.

My question is about this whole DSD madness. Recently from various articles from internet (obviously...) I learned that there is none/very few (if any) editing tools which can handle DSD stream natively. What I heard is pretty much all of editing softwares and hardwares do not alter the stream right away. Instead, they convert the DSD into PCM first, put alternation, then convert the PCM signal to DSD back.

I understand this whole DSD stuffs are... well... useless. But if these DSD streams are fake in the first place, I think we have a bigger problem. I will be glad to hear if anyone who knows about this chime in.

About DSD-native audio editing tools: ALL of them PCM conversions?

Reply #1 –
Quote from: wnmnkh on
I understand this whole DSD stuffs are... well... useless. But if these DSD streams are fake in the first place, I think we have a bigger problem. I will be glad to hear if anyone who knows about this chime in.


Why do you think that?  The specific choice of algorithm used is irrelevant so long as it performs acceptably.

About DSD-native audio editing tools: ALL of them PCM conversions?

Reply #2 –
As far as I know, digital signal processing is done in PCM.    You might be able to cut & splice DSD, but I think mixing, EQ, reverb, etc., requires PCM. 


About DSD-native audio editing tools: ALL of them PCM conversions?

Reply #3 –
Quote from: wnmnkh on
I understand this whole DSD stuffs are... well... useless. But if these DSD streams are fake in the first place, I think we have a bigger problem.
What does “fake” mean here? Real-world sound waves are of arbitrarily large bit-depths, not 1-bit, so if anything can be called fake by an intuitive definition, it’s DSD. I confess to not knowing whether we can ever record directly to DSD, but I suspect not, and in that case, it all has to go through a multi-bit PCM stage at some point in its lifetime.

I found a Q&A at SoundOnSound that namedrops a couple of programs that apparently can edit natively in DSD, although it too emphasises that this ability is very much a rarity.

About DSD-native audio editing tools: ALL of them PCM conversions?

Reply #4 –
Quote from: saratoga on
Quote from: wnmnkh on
I understand this whole DSD stuffs are... well... useless. But if these DSD streams are fake in the first place, I think we have a bigger problem. I will be glad to hear if anyone who knows about this chime in.


Why do you think that?  The specific choice of algorithm used is irrelevant so long as it performs acceptably.


Yes, I am sure most of professional tools perform in acceptable level, but that is not my point.



Quote from: DVDdoug on
As far as I know, digital signal processing is done in PCM.    You might be able to cut & splice DSD, but I think mixing, EQ, reverb, etc., requires PCM.



Quote from: db1989 on
Quote from: wnmnkh on
I understand this whole DSD stuffs are... well... useless. But if these DSD streams are fake in the first place, I think we have a bigger problem.
What does “fake” mean here? Real-world sound waves are of arbitrarily large bit-depths, not 1-bit, so if anything can be called fake by an intuitive definition, it’s DSD. I confess to not knowing whether we can ever record directly to DSD, but I suspect not, and in that case, it all has to go through a multi-bit PCM stage at some point in its lifetime.

I found a Q&A at SoundOnSound that namedrops a couple of programs that apparently can edit natively in DSD, although it too emphasises that this ability is very much a rarity.


Thank you DVDdoug for explanation. With my limited knowledge, I also suspected there is no way to edit stuffs while still in DSD form.

What I feel disgusted about this issue is that those 'DSD' albums should have not have label 'DSD', since they are nothing more than PCM stream converted into DSD. The whole process is lossy, and they should not be even callled 'High Res' at that point. I guess they never solved the issue with DSD introducing noise on upper frequency as well.

Such depressing that they charge these fraud 25 bucks per album....

About DSD-native audio editing tools: ALL of them PCM conversions?

Reply #5 –
Quote from: wnmnkh on
Quote from: saratoga on
Quote from: wnmnkh on
I understand this whole DSD stuffs are... well... useless. But if these DSD streams are fake in the first place, I think we have a bigger problem. I will be glad to hear if anyone who knows about this chime in.


Why do you think that?  The specific choice of algorithm used is irrelevant so long as it performs acceptably.


Yes, I am sure most of professional tools perform in acceptable level, but that is not my point.


My point is that if processing is "acceptable" then there is no "problem". 

About DSD-native audio editing tools: ALL of them PCM conversions?

Reply #6 –
Quote from: saratoga on
My point is that if processing is "acceptable" then there is no "problem".


Yes, but it is only technically 'no problem'. But it does have problem with these albums being advertised as DSD.

It is like trying to selling used pair of speakers at the price of new ones. Sure, technically those used ones and new ones are pretty much same in technical aspects, but are you going to buy them with same price as new ones?

I hope you understand what I mean.

About DSD-native audio editing tools: ALL of them PCM conversions?

Reply #7 –
Many $500 per meter interconnects probably have "acceptable" performance. The "problem" is that their performance cannot be measured, or other wise distinguished on the audio signal level (as opposed to the eye candy level) as any different than many $5 sets, but they are sold under the deliberately supplied implication that there is a major difference.

I am under the strong impression that too many people have a very hazy, somewhat magical, view of lossy vs lossless. To direct some attention to the use of "lossy" in this discussion, the initial sampling of all major ADCs is DSD, but only PCM is delivered to the recording program. Therfore, by the foggy definition of lossy used here (I don't know just what that definition is), no digital audio that is not pure DSD is lossless, thus pretty much winning the argument for the analogue only advocates.

About DSD-native audio editing tools: ALL of them PCM conversions?

Reply #9 –
Quote from: wnmnkh on
What I feel disgusted about this issue is that those 'DSD' albums should have not have label 'DSD', since they are nothing more than PCM stream converted into DSD. The whole process is lossy, and they should not be even callled 'High Res' at that point. I guess they never solved the issue with DSD introducing noise on upper frequency as well.



This is hysterical audiophile rubbish.

The PCM steps typically operate at extremely high sample rates and at 8 bits or 24 bits instead of 1 bit.  It's not 'lossy' .

If you're going to be this fanatical, you'd better not call any analog recordings released on SACD 'DSD' either, since they very definitely had 'lossy' steps involved -- starting with the capture on magnetic tape.





About DSD-native audio editing tools: ALL of them PCM conversions?

Reply #10 –
Quote from: saratoga on
Quote from: wnmnkh on
Quote from: saratoga on
My point is that if processing is "acceptable" then there is no "problem".


Yes, but it is only technically 'no problem'.


You're discussing math.  Other than a technical sense, how else can you look at it?



You may want to read the rest of my comment you conveniently decided to not quote. Because that portion answers this question.


Quote from: krabapple on
Quote from: wnmnkh on
What I feel disgusted about this issue is that those 'DSD' albums should have not have label 'DSD', since they are nothing more than PCM stream converted into DSD. The whole process is lossy, and they should not be even callled 'High Res' at that point. I guess they never solved the issue with DSD introducing noise on upper frequency as well.



This is hysterical audiophile rubbish.

The PCM steps typically operate at extremely high sample rates and at 8 bits or 24 bits instead of 1 bit.  It's not 'lossy' .

If you're going to be this fanatical, you'd better not call any analog recordings released on SACD 'DSD' either, since they very definitely had 'lossy' steps involved -- starting with the capture on magnetic tape.


Absolutely, no one should call those recordings as "High Res" in the first place if analog tape is used. This is why DDD is supposed to be relevant at this regard.

About DSD-native audio editing tools: ALL of them PCM conversions?

Reply #11 –
It's certainly possible to record direct to DSD. It's also possible to edit a DSD recording converting only the length of the edit (typically around 20 - 40ms for straight cuts) to PCM and back - and the PCM used is 24 bits and around 350kHz, at least in Pyramix.

If you want to do any other processing - EQ, compression or indeed mixing(!), you'd have to either convert the entire stream to PCM, or do it in the analogue domain before hitting the 1-bit ADC.

I'm not sure what would be considered fake about any of these processes.

About DSD-native audio editing tools: ALL of them PCM conversions?

Reply #12 –
Quote from: Cubist Castle on
I'm not sure what would be considered fake about any of these processes.

It's the belief that DSD is audibly superior, that is fake false.

About DSD-native audio editing tools: ALL of them PCM conversions?

Reply #13 –
As far as I get it, so called "Hi Res" is the label that says a resolution of the delivered material is higher than usual 44/16. Even those "regular" 44/16 releases were usually captured using higher resolution in order to have a margin for signal processing. One could capture old master tape using 192kHz 24 bits, process it and deliver to the consumer in the same format (or convert it to DSD and put on a SACD). Why such material is not Hi Res?
If age or weaknes doe prohibyte bloudletting you must use boxing

About DSD-native audio editing tools: ALL of them PCM conversions?

Reply #14 –
Quote from: eevan on
As far as I get it, so called "Hi Res" is the label that says a resolution of the delivered material is higher than usual 44/16. Even those "regular" 44/16 releases were usually captured using higher resolution in order to have a margin for signal processing. One could capture old master tape using 192kHz 24 bits, process it and deliver to the consumer in the same format (or convert it to DSD and put on a SACD). Why such material is not Hi Res?



It is a matter of philosophy. Do you believe that painting wings on a pig enhances its ability to fly, Yes or No?

An analog tape digitized with even just 44/16 is an example of painting wings on a pig. As far as DSD, 24/48, 24/96, or 24/192 goes, we're just talking fancier painting.

Of course I say this all tongue in cheek, except that the second sentence in the first line of this post is absolutely true, and fortunes and significant portions of lives have been spent pursuing that odd process which in the final analysis (as well as a number of earlier ones) amounts to being painting wings on a pig and expecting it to fly.

About DSD-native audio editing tools: ALL of them PCM conversions?

Reply #15 –
Quote from: krabapple on
If you're going to be this fanatical, you'd better not call any analog recordings released on SACD 'DSD' either, since they very definitely had 'lossy' steps involved -- starting with the capture on magnetic tape.

Audio AD and DA converters use fixed-point integer representations of digital audio. Many audio dsp operations are carried out using floating-point representations. There is a conversion back and forth. Oh the horror... :-)

-k

About DSD-native audio editing tools: ALL of them PCM conversions?

Reply #16 –
Quote from: wnmnkh on
Absolutely, no one should call those recordings as "High Res" in the first place if analog tape is used. This is why DDD is supposed to be relevant at this regard.



Leaving aside why anyone should care about 'High Res' in the first place, given the limitations of human hearing and real listening conditions....

Why shouldn't a production foray from (2.8224 MHz SR, 1-bit) to DSD-Wide aka 'PCM Narrow' (2.8224 MHz SR, 8-bit) be considered 'High Res'?

About DSD-native audio editing tools: ALL of them PCM conversions?

Reply #17 –
Please forgive me if I'm wrong but I understood that 1 bit DSD is the lingua franca of current generation, best in class, sigma-delta DAC and ADC chips?

So everything you hear via an audio reproduction system has been in DSD format at least twice. Sorry. Only once if it's never been processed in any way.

About DSD-native audio editing tools: ALL of them PCM conversions?

Reply #18 –
Quote from: krabapple on
Quote from: wnmnkh on
Absolutely, no one should call those recordings as "High Res" in the first place if analog tape is used. This is why DDD is supposed to be relevant at this regard.



Leaving aside why anyone should care about 'High Res' in the first place, given the limitations of human hearing and real listening conditions....

Why shouldn't a production foray from (2.8224 MHz SR, 1-bit) to DSD-Wide aka 'PCM Narrow' (2.8224 MHz SR, 8-bit) be considered 'High Res'?



As we both know all sound quality justifications for DSD are fantasy anyway, so what does reason or logic have to do with this discussion, anyway? ;-)

Vanderkooy and Lipshitz uncovered the fact that a pure, PCM-less DSD recorder is inherently broken, and Sony then admitted that they circumvent this by putting a little PCM of a kind into the recording chain...

http://sjeng.org/ftp/SACD.pdf

About DSD-native audio editing tools: ALL of them PCM conversions?

Reply #19 –
Quote from: RonaldDumsfeld on
Please forgive me if I'm wrong but I understood that 1 bit DSD is the lingua franca of current generation, best in class, sigma-delta DAC and ADC chips?

So everything you hear via an audio reproduction system has been in DSD format at least twice. Sorry. Only once if it's never been processed in any way.



The best in class sigma-delta parts run > 1 bit at their lowest level.

http://www.analog.com/static/imported-file...ials/MT-023.pdf

About DSD-native audio editing tools: ALL of them PCM conversions?

Reply #20 –
While it is true that more than 1 bit is commonly used these days, is it not also true that the sampling is more DSD than PCM? It is nothing like the old fashion resistor or capacitor ladders, or similar devices, that actually tried to measure the signal voltage at the input.

About DSD-native audio editing tools: ALL of them PCM conversions?

Reply #21 –
Quote from: AndyH-ha on
It is nothing like the old fashion resistor or capacitor ladders, or similar devices, that actually tried to measure the signal voltage at the input.


Do you really think a modern analog to digital voltage converter doesn't measure voltage?  Nonsense.

Modern devices use multiple bits, oversampling and noise shaping to precisely measure voltage.  This can then be resampled or converted to whatever output format is desired. 

About DSD-native audio editing tools: ALL of them PCM conversions?

Reply #22 –
Quote from: krabapple on
DSD-Wide aka 'PCM Narrow' (2.8224 MHz SR, 8-bit)
Is… Is this a real thing that actually exists

because this qualifies as irony, right? not to mention, perhaps, embarrassing backpedalling

Does anyone think they might, eventually, tacitly, admit that PCM at 24 bits for editability and 96 kHz for noise shaping and anti-imaging (not frequency response) was fine all along?

About DSD-native audio editing tools: ALL of them PCM conversions?

Reply #23 –
Quote from: db1989 on
Quote from: krabapple on
DSD-Wide aka 'PCM Narrow' (2.8224 MHz SR, 8-bit)
Is… Is this a real thing that actually exists

because this qualifies as irony, right? not to mention, perhaps, embarrassing backpedalling

Does anyone think they might, eventually, tacitly, admit that PCM at 24 bits for editability and 96 kHz for noise shaping and anti-imaging (not frequency response) was fine all along?


It's been really long time since so called 2bit, 4bit, 6bit and 8bit DSD exist. Otherwise there is no other good way to eliminate/reduce upper distortion noise if DSD is only 1bit.

About DSD-native audio editing tools: ALL of them PCM conversions?

Reply #24 –
Quote from: AndyH-ha on
While it is true that more than 1 bit is commonly used these days, is it not also true that the sampling is more DSD than PCM? It is nothing like the old fashion resistor or capacitor ladders, or similar devices, that actually tried to measure the signal voltage at the input.


I would characterize the currently used coding technique as being Sigma-Delta assisted PCM.

The low resolution PCM converter at the core of this process is usually called a "Flash (PCM) Converter" and the literature describes the trimming of a resistor ladder within it.

The measuring of the input voltage or the generation of the output voltage is as real as it comes!

If you are standing outside the black box it just works so what is inside the box is only of academic/pedantic interest.