Is the dynamic range of CD sufficient?

[Julf]

Ey Julf, So the coding methods differ but the data is the same?

"Same" in the sense of "can be viewed as". Conceptually the delta-sigma process is not producing a pulse width that is proportional to the input signal - it is producing a fixed-rate, fixed-pulse-width stream of 1-bit values. It just happens that for 1-bit data that also results in a waveform that is a discrete-time-value PDM approximation of the input signal.

 

[JackD201]

Got it. Thanks.

 

[jiannone]

Mark,
I suggest you find 25 uninterrupted minutes and watch this video:
http://www.youtube.com/watch?v=cIQ9IXSUzuM

Some people disagree with the presenter's opinions on the use of dither in some cases, but the technical content is correct.

Wow, thank you. I'd like to see some of the criticism. I feel like I learned a lot.

 

[Orb]

Ey Julf, So the coding methods differ but the data is the same?

Sorry this is definitely not my area of expertise. Just looking for clarification. This is over my head not to mention news to me.

For a definitive answer look to something like this Jack, by AP.
Understanding PDM Digital Audio:
http://users.ece.utexas.edu/~bevans...ersion/AP_Understanding_PDM_Digital_Audio.pdf

Cheers
Orb

 

[JackD201]

Thanks Orb. I was searching for context as I followed your exchanges with Julf. Particularly since Julf is a code guy, I had a feeling you two might have been talking past each other to some extent so I asked. Thanks again. I'll read your link.

 

[Orb]

Thanks Orb. I was searching for context as I followed your exchanges with Julf. Particularly since Julf is a code guy, I had a feeling you two might have been talking past each other to some extent so I asked. Thanks again. I'll read your link.

Well here is an old Philips paper.
If you followed many of the previous technical threads you are probably aware of the discussion of modern SDM DAC chips and what is truly native PCM or DSD.
As Julf and JJ mention SDM is integral to 1-bit stream in general (never disagreed with that, and more technical info has already been discussed in many threads here), however DSD by Sony utilises PDM encoding.
Anyway interesting read but needs to be taken as a whole rather than bits for arguments (really needs to be used in conjunction with the AudioPrecision "Understanding PDM" document I linked earlier), anyway the Philips paper focuses upon SDM but mentions aspects of PCM,PDM,etc in relation to DSD:
http://www.gearslutz.com/board/atta...portable-recorder-derkreefman_dsd_wp-2323.pdf

Cheers
Orb

 

[Julf]

Particularly since Julf is a code guy

Started out in hardware (and even studied music for a while), did a fair bit of software/code, but have recently done quite a bit of embedded and HW stuff again. The stuff we are discussing here is neither software or hardware, but mathematics and information and signal processing theory.

 

[JackD201]

I was thinking PWM, looked it up and what I found said PDM which I again looked up and found to be a type of PWM.

I was surprised as well when JJ said DSD was PCM. That and confused. I still am.

 

[JackD201]

Started out in hardware (and even studied music for a while), did a fair bit of software/code, but have recently done quite a bit of embedded and HW stuff again. The stuff we are discussing here is neither software or hardware, but mathematics and information and signal processing theory.

Hence my question to you. Poorly constructed as it may have been.

 

[Julf]

If you followed many of the previous technical threads you are probably aware of the discussion of modern SDM DAC chips and what is truly native PCM or DSD.

Sigh. Again - this has nothing to do with "modern SDM DAC chips".

As Julf and JJ mention SDM is integral to 1-bit stream in general (never disagreed with that, and more technical info has already been discussed in many threads here), however DSD by Sony utilises PDM encoding.

No. PDM modulates the pulse density with the source signal. DSD uses delta-sigma conversion to produce a 1-bit fixed-frequency, fixed-pulse-width encoding of the source signal.

 

[Julf]

I was thinking PWM, looked it up and what I found said PDM which I again looked up and found to be a type of PWM.

Pulse width modulation modulates pulse width. Pulse density modulation modulates pulse density. Thus PWM is a subset of PDM with the constraint that all pulses related to one sample are contiguous.

DSD is a fixed-frequency, fixed-pulse-width data stream. Because the data is only 1-bit, it also incidentally forms a quantized PDM stream - but that is purely incidental. For actual signal analysis, it is better to treat it as 1-bit PCM.

If you look at "DSD-wide", it is a simple variation of DSD where the data values are converted to multi-bit to allow digital signal processing. It is still DSD, but multi-valued instead of 1-bit - and is definitely not corresponding to a PDM stream any more.

 

[Orb]

Sigh. Again - this has nothing to do with "modern SDM DAC chips".



No. PDM modulates the pulse density with the source signal. DSD uses delta-sigma conversion to produce a 1-bit fixed-frequency, fixed-pulse-width encoding of the source signal.

If you apply the theory correctly and limit this specifically to your context, BUT this thread IS NOT JUST ABOUT THE ONE POST WHERE JJ DECIDED TO USE 1-BIT DSD AS A COMPARISON TO THE PCM DEBATE AND THE SUBTLY MISUNDERSTOOD APPLICATION OF THE STAIR-STEP BY A POSTER.
As I said from earlier on, you guys are confusing the F out of everyone by using a wrong example and then technically arguing it from a very specific viewpoint, you still are doing that without realising WHAT THE WHOLE THREAD IS ABOUT....

OK Stopping now because I think we are both pissing each other off.
Just wish some were more open minded when dealing with effing threads and the posters initially involved.
I see you still do not accept DSD uses PDM as well sigh, nor what PDM is.
Out.
Orb

 

[Julf]

OK Stopping now because I think we are both pissing each other off.

Fine.

Just wish some were more open minded when dealing with effing threads and the posters initially involved.

Just wish some were more open minded when being wrong, and not dragging on endlessly without actually understanding the theory.

I see you still do not accept DSD uses PDM as well sigh, nor what PDM is.

I see you still don't get it.

 

[Orb]

Fine.



Just wish some were more open minded when being wrong, and not dragging on endlessly without actually understanding the theory.



I see you still don't get it.

No thats right I really do not get it...
How you can argue PDM and PCM encoding are different in other threads and then argue totally different context in this thread
Remember this was originally about DSD compared to PCM.... in a different thread you seem to have agreed with me, how ironic.

 

[mep]

Do we have too many digital experts on this thread or not enough? I'm confused.

 

[Julf]

No thats right I really do not get it...

OK, let's try one more time. What part don't you agree with in "Pulse width modulation modulates pulse width. Pulse density modulation modulates pulse density. Thus PWM is a subset of PDM with the constraint that all pulses related to one sample are contiguous. PCM (and DSD) is a fixed-frequency, fixed-pulse-width data stream (in the case of standard DSD, the data is 1 bit wide)."?

 

[Don Hills]

Wow, thank you. I'd like to see some of the criticism. I feel like I learned a lot.

The main issue for most people was when he said:
"No one ever ruined a great recording by not dithering the final master."

Try a Google search for that phrase (with the quote marks).

 

[Orb]

Do we have too many digital experts on this thread or not enough? I'm confused.

It is because we are mostly arguing technical semantics and scope/context.
None of this has anything to do with the original post that kicked it off by one member mentioning graining and "a stair-step".

OK to keep it at its simplest Julf I am going to quote just one of the application notes (so many variations I can quote) so to remove myself from the debate.

In a PDM signal, specific amplitude values are not encoded into pulses as they would be in PCM.
Instead it is the relative density of the pulses that corresponds to the analog signal's amplitude.

So PCM is designed for discrete samples that have an explicitly defined amplitude value (n-bit binary code), going to 1-bit forces this to change to using PDM or PWM.
This is one fundamental way AP/Merging/various other manufacturers including Sony and Philips (considering they both created DSD) differentiate between PCM and 1-bit noise shaped PDM (or PWM as I have mentioned in the past), or converting from PDM to PCM.

In fact to quote Philips from a paper discussing DSD:

PCM: Pulse Code Modulation. A digital format, used for example in CD, whereby a digital signal is represented by an accurate representation (e.g., 16 bits, meaning that the range -1,+1 is subdivided in 65536 sub-intervals) of the wave form at equidistant points in time (for example, in CD 44100 times per second a 16-bit approximation of the wave form is stored).
Pulse Density Modulation: A form of pulse modulation where a large positive signal is represented by a long series of positive pulses; a zero signal is represented by alernating positive and negative pulses.

Even the bitstream of PCM and DSD are different.
However one might argue from a set point and limited scope that is not related at all to this thread that 1-bit PCM is comparable to SDM, but this is a loose comparison when considering context-semantics and what the definition is for PCM and PDM, but as I said it has absolutely no reference to this thread or what has been discussed before it all kicked off.
Definitely dropping out now as the further technical aspects are not really part of this thread.
Thanks
Orb

 

[JackD201]

Somebody please get Andreas on here.

 

[FrantzM]

methink Julf is correct on all fronts.

I find myself entirely satisfied with Redbook but wouldn't mind having more... You know ... FOMO (Fear Of Missing Out)