PCM to DSD conversion
#1 PCM to DSD conversion
I am still considering PCM to DSD conversion at some point. With HQ player there are many options/filters etc and I tried to get a little bit more information about everything. So I ended up reading this article: http://archimago.blogspot.com/2019/02/m ... r-325.html.
I was a bit shocked when I read the following:
“There are clearly measurable objective differences of course, and as usual, conversion of the 24-bit PCM signal to DSD results in added noise at varying levels which were not intrusive in my listening. Remember, the addition of noise to the signal when we use DSD is inevitable and chasing after a "better" sound is essentially about listening for the "euphonic distortion" of the conversion process. That preference would be idiosyncratic to specific listeners and perhaps DACs. There is no "new information" being added that was not there already in the PCM data.
Could someone with some practical PCM to DSD conversion experience please comment on this.
I was a bit shocked when I read the following:
“There are clearly measurable objective differences of course, and as usual, conversion of the 24-bit PCM signal to DSD results in added noise at varying levels which were not intrusive in my listening. Remember, the addition of noise to the signal when we use DSD is inevitable and chasing after a "better" sound is essentially about listening for the "euphonic distortion" of the conversion process. That preference would be idiosyncratic to specific listeners and perhaps DACs. There is no "new information" being added that was not there already in the PCM data.
Could someone with some practical PCM to DSD conversion experience please comment on this.
#2 Re: PCM to DSD conversion
Why "shocked" Wolfgang? One could argue that "...a "better" sound is essentially about listening for the "euphonic distortion"..." is a theme at the very heart of our hobby and not something peculiar to DSD playback - consider your EF86 buffer project for example.
The graphs show that the noise is very low level and he says "...were not intrusive in my listening.". As a generalisation 'euphonic' means something along the lines of 'pleasant sounding' so perhaps I have an idiosyncratic tendency in preferring the sound of my DSD decoders to my PCM DACs, though they (PCM) still sound very good too - we're not talking chalk and cheese.
I've lots of practical experience of PCM/DSD conversion (previously I did a lot of comparison of the same tracks in PCM and DSD using my ESS9018 DAC, which set me on the DSD road). As you know, I'm not the most technically savvy hobbyist and I don't have a suite of test equipment (or the understanding of how to use it) so that my comparisons are based purely on my personal test equipment.
Its an interesting read though and he aligns with my own experience that the different options in HQP are subtle at best (and sometimes I can't hear a difference at all).
BTW, I'm not aware of anyone claiming that HQP (or any of the other resampling tools) adds 'new information'?
In practical terms with my DSD decoder projects, a core element of the hardware is the filtering of the ultransonics with output filters and that works well. I have had some noise issues with my most recent valve project but that is hardware related around the matching of the valves and metastability of the clocking but with adjustment I have now achieved a result of just about silence (very small amount of 'hiss' type noise if you put your ear next to the speaker cone).
The graphs show that the noise is very low level and he says "...were not intrusive in my listening.". As a generalisation 'euphonic' means something along the lines of 'pleasant sounding' so perhaps I have an idiosyncratic tendency in preferring the sound of my DSD decoders to my PCM DACs, though they (PCM) still sound very good too - we're not talking chalk and cheese.
I've lots of practical experience of PCM/DSD conversion (previously I did a lot of comparison of the same tracks in PCM and DSD using my ESS9018 DAC, which set me on the DSD road). As you know, I'm not the most technically savvy hobbyist and I don't have a suite of test equipment (or the understanding of how to use it) so that my comparisons are based purely on my personal test equipment.
Its an interesting read though and he aligns with my own experience that the different options in HQP are subtle at best (and sometimes I can't hear a difference at all).
BTW, I'm not aware of anyone claiming that HQP (or any of the other resampling tools) adds 'new information'?
In practical terms with my DSD decoder projects, a core element of the hardware is the filtering of the ultransonics with output filters and that works well. I have had some noise issues with my most recent valve project but that is hardware related around the matching of the valves and metastability of the clocking but with adjustment I have now achieved a result of just about silence (very small amount of 'hiss' type noise if you put your ear next to the speaker cone).
Sorry, I couldn't resist!
#3 Re: PCM to DSD conversion
Ultrasonic noise has always been part of DSD and SACD's for the reason that they use DSD. By definition the conversion from PCM to DSD will be lossy as the data rates are reduced. 16 * 44100 = 705600 bits per second, so to encode CD with a single bit lossless delta sigma format would require a data rate of 706kb/s
Whenever an honest man discovers that he's mistaken, he will either cease to be mistaken or he will cease to be honest.
#4 Re: PCM to DSD conversion
Maybe my understanding about the ideal of ADC-DAC is no longer correct then (or never was?). I thought the idea was to get as close as possible to the original analog curve and upsampling would be one possible way to achieve this. From that understanding the DSD format should be theoretically even better and recordings made in native DSD probably are better. "Listening for the euphonic distortion of the conversion process" is something else IMO. This doesn't mean that it cannot sound good and couldn't be preferred over listening to PCM!
The 2nd order/even distortions of tubes belong to a different category unless the DSD conversion process would also produce mainly even distortions.
HQP changes the sound. I can clearly hear the subtle influence of filters - I actually like this feature very much - and I can also hear the difference (more than expected) of original vs upsampled. I have collected quite some HiREs downloads of all kind of sample/bit rates over the years. Because I hear pops when the DAC switches sample rates (I am still using one of the first Soekris modules which had that problem) I converted all HRes files down to 44.1/24 with iZotope RX6. When I play them back upsampled at 352.8kHz with HQP they all sound dramatically better, even better than upsampled CDs.
So obviously we cannot really say anymore what's what or assume we would know or could predict results as everything can turn out as something unexpected for the better or the worse.
The 2nd order/even distortions of tubes belong to a different category unless the DSD conversion process would also produce mainly even distortions.
HQP changes the sound. I can clearly hear the subtle influence of filters - I actually like this feature very much - and I can also hear the difference (more than expected) of original vs upsampled. I have collected quite some HiREs downloads of all kind of sample/bit rates over the years. Because I hear pops when the DAC switches sample rates (I am still using one of the first Soekris modules which had that problem) I converted all HRes files down to 44.1/24 with iZotope RX6. When I play them back upsampled at 352.8kHz with HQP they all sound dramatically better, even better than upsampled CDs.
So obviously we cannot really say anymore what's what or assume we would know or could predict results as everything can turn out as something unexpected for the better or the worse.
#5 Re: PCM to DSD conversion
Well, upsampling cant add information, but it may make the filtering simpler.upsampling would be one possible way to achieve this
I don't understand why you think that?From that understanding the DSD format should be theoretically even better
DSD is different to PCM, but it involves reduced information to achieve practical data rates. A simple way to think of DSD is a delta sigma DAC but with the front and end parts done in different places at different times. The noise shaping is just another name for digital filtering and its used to get the data rate down.
Personally every time I have heard DSD I hear a signature sound that I assume some prefer, but its there never the less.
Whenever an honest man discovers that he's mistaken, he will either cease to be mistaken or he will cease to be honest.
#6 Re: PCM to DSD conversion
Me neither if I think about it. It's indeed not based on solid ground.I don't understand why you think that?
It's more based on a feeling that DSD should/could theoretically be less prone to “errors” because of the simpler filtering and maybe less filtering artifacts in the audible range.
#7 Re: PCM to DSD conversion
Sorry, I couldn't resist!
-
- Shed dweller
- Posts: 2299
- Joined: Fri Aug 03, 2007 10:57 am
- Location: From the land of the Bodgers
#8 Re: PCM to DSD conversion
Wasnt DSD meant to be a simple analogue to digital archival format for transferring tapes before they ‘died’? You can’t mix or master in DSD it has to be converted to PCM which sort of defeats the point. I don’t see the point of it other than to sell you the same music again in a different format....
Only the Sith deal in absolutes.
#9 Re: PCM to DSD conversion
DSD (Direct Stream Digital) is Sony's (and I guess Philips) brand name for their SDM (Sigma Delta Modulation) technology that emerged with SACD (Super Audio Compact Disc) product lines. Sigma-Delta/Delta-Sigma (they seem to be interchangable) has been around for a number of decades longer.
To me, the point of it is simply that, from my personal perspective, it sounds better than PCM playback.
To me, the point of it is simply that, from my personal perspective, it sounds better than PCM playback.
Sorry, I couldn't resist!
#10 Re: PCM to DSD conversion
I remember being surprised when I looked at this some time ago I couldn't find any definition of the encoding or decoding process used by Sony. I don;t know what the software that people use now have as their filter definitions, but I would be interested in finding out where they got it from or if its not based on the encoding Sony used.
Whenever an honest man discovers that he's mistaken, he will either cease to be mistaken or he will cease to be honest.
#11 Re: PCM to DSD conversion
According to this SOS article you can.You can’t mix or master in DSD it has to be converted to PCM which sort of defeats the point.
https://www.soundonsound.com/sound-advi ... master-dsd
"Processing native DSD streams in a DAW environment is possible in a fashion, but isn't easy and has to contend with some significant technical issues. As I understand it, there are currently only two DAWs that can process (ie. mix, equalise, control dynamics of, and so on) DSD data directly,(...)".
But that article was written in 2009. Who knows what's possible today?
#12 Re: PCM to DSD conversion
Nick, I've had a quick look but can't find it at the moment but, IIRC, there was some sort of working group for the topic - I'm sure Jussi Lasko (= Signalyst = HQ Player) was part of it. Wasn't Sony/Philips proprietary take on 'DSD/SACD' more to do with copy protection than the fundamental process?Nick wrote: ↑Fri Sep 04, 2020 3:43 pm I remember being surprised when I looked at this some time ago I couldn't find any definition of the encoding or decoding process used by Sony. I don;t know what the software that people use now have as their filter definitions, but I would be interested in finding out where they got it from or if its not based on the encoding Sony used.
Sorry, I couldn't resist!
#13 Re: PCM to DSD conversion
It would seem that I've had the capability all the time, I just didn't realise it......I haven't used the studio setup for a while but I'm sure going to have a fiddle now.....This will allow me a diversion from current problems as soon as the Topping E30 arrives...
I haven't looked in detail, but I am aware that it's not DSD direct.
I haven't looked in detail, but I am aware that it's not DSD direct.
There's nowhere you can be that isn't where you're meant to be
#14 Re: PCM to DSD conversion
It was, which is probably the reason for the lack of publications. I just wonder if the actual info is behind a NDA so that begs the qyestion about the public converters equivalence to the SACD specification.Wasn't Sony/Philips proprietary take on 'DSD/SACD' more to do with copy protection than the fundamental process?
Whenever an honest man discovers that he's mistaken, he will either cease to be mistaken or he will cease to be honest.
#15 Re: PCM to DSD conversion
I would be very surprised if there was not some DSD to PCM and back involved in that, I have no idea how you could add two DSD streams and get a meaningful result.Wolfgang wrote: ↑Fri Sep 04, 2020 4:07 pmAccording to this SOS article you can.You can’t mix or master in DSD it has to be converted to PCM which sort of defeats the point.
https://www.soundonsound.com/sound-advi ... master-dsd
"Processing native DSD streams in a DAW environment is possible in a fashion, but isn't easy and has to contend with some significant technical issues. As I understand it, there are currently only two DAWs that can process (ie. mix, equalise, control dynamics of, and so on) DSD data directly,(...)".
But that article was written in 2009. Who knows what's possible today?
Whenever an honest man discovers that he's mistaken, he will either cease to be mistaken or he will cease to be honest.