[Superunknown2]

Technically a modulator and noise shaper are sort of the same thing, though modulator specifically implies modulating to a lower bit depth.

Basically, when you convert a higher bit depth piece of info to a lower bit depth piece of info, you encounter a 'quantization error'. Simple example of this:

- You have a device with a maximum output of 3V
- You need to output exactly 2V
- This can be accurately represented by a 2-bit system, because two bits can be '11' (100%/3V), '10' (66%/2V), '01' (33%/1V) or '00' (0%/0V). So you just choose '10' and you're good.
- If you now wanted to get this output with a 1-bit system, you can't do it accurately. A 1-bit system can only be '1' (100%/3V) or '0' (0%/0V). And so picking the closest value of '1' (3V), you now have a quantization error of 1V, the difference between the value you intended and the actual value.

This can be shown visually too. Here for example is an image on the far left, represented with 8-bit accuracy, meaning there are 255 different possible shades from black/white that can be used to represent each pixel.



If you now wanted to represent each pixel with just 1-bit, meaning ONLY pure black or pure white, you can do so by just throwing away the bottom 7 bits and keeping only the top bit '1' (black) or '0' (white). But you've thrown away all the finer detail information about the particular shade and this leaves a fairly inaccurate result due to what is called truncation distortion.

We can address this by dithering, which is the process of intentionally adding some random noise to prevent any strong signal-correlated truncation distortion errors. This can get us closer to a more perceptually ideal result, but it's still not perfect.

We can take things a step further by applying dithering in a more advanced way, called noise-shaping. This means that rather than noise being added in a more random fashion and evenly distributed, a feedback loop system is employed to correct remaining noise/quantization error and provide more dynamic range within the area we want or the area of perceptual relevance, and shape the quantization error out to areas where it's not as much of an issue.

In audio, this means that the noise shaper is designed in a way which provides higher dynamic range in the audible band itself, and shapes the quantization errors out to higher frequencies, where they can just be filtered out with an analog filter.

This image shows a 1khz signal, represented by 768khz 8-bit PCM. No dithering was applied, and so we can see unwanted extra content, truncation distortion.



Now if we apply some standard TDPF dither, this adds a small amount of random noise, but in doing so removes the truncation distortion and actually results in better effective dynamic range:



We can then try with a noise shaper, HQPlayer's LNS15 noise shaper in this instance, which prioritises dynamic range in the audible band, and shapes the noise out to higher frequencies and gives us even better dynamic range up to about 100khz, and the higher noise above 100khz can just be filtered out by the DACs analog filter:




Noise shaping can be applied when upsampling just in PCM as shown above, because the new samples generated as part of the upsampling are only as accurate as the bit-depth of the output allows. They can't perfectly represent the ideal real value, so quantization error is still a concern even though you're not actually reducing bit depth.

But when doing upsampling + modulation, such as converting 16 bit PCM info to higher rate, 1-bit DSD info, you have enormous quantization error, and thus there's a lot of work for the modulator to do to try to get a high dynamic range output in the area you want.

SUPER TLDR:
Even if something can only be 'on' at 100% or 'off' at 0%, you can still get an output of say 80% by turning it off and on a hundred thousand times rapidly, with it being 'on' about 80% of the time, and then filtering out the excess high-frequency noise.

Thank you for that explanation. That makes it seem like there is no benefit to DSD over PCM in HQPlayer, because of all that potential error. What is the benefit of DSD over PCM? Is there any benefit of using DSD instead of PCM on a NOS DAC like a Holo Spring/May?

 

[genefruit]

Thank you for that explanation. That makes it seem like there is no benefit to DSD over PCM in HQPlayer, because of all that potential error. What is the benefit of DSD over PCM? Is there any benefit of using DSD instead of PCM on a NOS DAC like a Holo Spring/May?

some prefer the sound of DSD while others prefer the sound of PCM

 

[Superunknown2]

some prefer the sound of DSD while others prefer the sound of PCM

But why would that be, in theory? I notice a difference, though it is hard to say if it is better. If I look hard at it, I can see some reasons why maybe it might be better, though I don't know if that is just placebo effect. It is a lot of trouble to deal with all that processing, so I am wondering why in theory it would sound better.

 

[GoldenSound]

Thank you for that explanation. That makes it seem like there is no benefit to DSD over PCM in HQPlayer, because of all that potential error. What is the benefit of DSD over PCM? Is there any benefit of using DSD instead of PCM on a NOS DAC like a Holo Spring/May?

DSD and PCM are not inherently better than one or the other, they're just fundamentally different ways of doing things.

PCM is high bit depth low sample rate, DSD is low bit depth high sample rate.

The idea (and the whole premise on which delta sigma DACs operate even though most aren't strictly going to 1-bit, most are 5-7bit) is that you can trade off native accuracy/dynamic range for speed. I'd refer back to the TLDR example in the previous post:

Even if something can only be 'on' at 100% or 'off' at 0%, you can still get an output of say 80% by turning it off and on a hundred thousand times rapidly, with it being 'on' about 80% of the time, and then filtering out the excess high-frequency noise.

Making a native PCM converter (R2R) that is highly linear is really really hard. You need components to be matched to insanely accurate tolerances. They get more accurate by having more precise resistance values.

Making a native 1-bit converter that runs at insanely high speeds is also quite hard, and there's a lot of work to be done in terms of the modulator/noise shaper. They get more accurate by running at higher speeds and/or differences in modulator design.

Most modern DACs take a middleground approach, modulating to a high-ish speed, moderate bit depth. The newest AKM chips for example run at typically 11Mhz 7-bit (though can run at 22Mhz). This means that some of the more extreme challenges in either side can be avoided, you don't need an insanely accurate converter natively, but you also don't have to run at as high a speed or have as much quantization error to handle.

This makes a lot of sense, though some people find they prefer the resulting sound of a 'pure DSD' DAC or a native PCM R2R DAC.

There are inherent challenges to going full one way or the other, with no clear answer as to what is 'best'. It's all tradeoffs

Personally I prefer PCM on the Holo DACs, but some prefer DSD. Worth trying both and seeing which suits your preferences. You've essentially got two fundamentally different DACs in one box

 

[Superunknown2]

DSD and PCM are not inherently better than one or the other, they're just fundamentally different ways of doing things.

PCM is high bit depth low sample rate, DSD is low bit depth high sample rate.

The idea (and the whole premise on which delta sigma DACs operate even though most aren't strictly going to 1-bit, most are 5-7bit) is that you can trade off native accuracy/dynamic range for speed.

Making a native PCM converter (R2R) that is highly linear is really really hard.
Making a native 1-bit converter that runs at insanely high speeds is also quite hard, and there's a lot of work to be done in terms of the modulator/noise shaper.

Most modern DACs take a middleground approach, modulating to a high-ish speed, moderate bit depth. The newest AKM chips for example run at typically 11Mhz 7-bit (though can run at 22Mhz).

There are inherent challenges to going full one way or the other, with no clear answer as to what is 'best'.

Personally I prefer PCM on the Holo DACs, but some prefer DSD. Worth trying both and seeing which suits your preferences. You've essentially got two fundamentally different DACs in one box

Ah that makes perfect sense, thank you. One is high bit, low sample rate, while the other is the reverse. DSD seems to me to have less slam behind it than PCM but more soundstage depth, making it more relaxed and potentially better with slower music.

 

[GoldenSound]

Ah that makes perfect sense, thank you. One is high bit, low sample rate, while the other is the reverse. DSD seems to me to have less slam behind it than PCM but more soundstage depth, making it more relaxed and potentially better with slower music.

Yep I'd agree with that.

Subjectively DSD tends to sound softer to me, and there's some reasons as to why that might be. PCM for example can effectively react instantly to and change to a sample value, it can go from 0 one sample to 100 next sample, whereas DSD you have to have a series of consecutive samples to increase the amplitude significantly.

Theory suggests that especially at higher DSD rates this should be far too small to be perceived, but many people seem to find similar results regardless of the DAC so idk

Rob Watts has commented on this a few times and that's one of his main gripes with DSD, he feels that it cannot react quickly enough to strong transients

 

[Superunknown2]

Yep I'd agree with that.

Subjectively DSD tends to sound softer to me, and there's some reasons as to why that might be. PCM for example can effectively react instantly to and change to a sample value, it can go from 0 one sample to 100 next sample, whereas DSD you have to have a series of consecutive samples to increase the amplitude significantly.

Theory suggests that especially at higher DSD rates this should be far too small to be perceived, but many people seem to find similar results regardless of the DAC so idk

Rob Watts has commented on this a few times and that's one of his main gripes with DSD, he feels that it cannot react quickly enough to strong transients

That makes sense. Do you still prefer sinc-L and sinc-Mx for both DSD and PCM? Or do you prefer a different filter? Also, what modulator do you prefer for DSD?

 

[GoldenSound]

That makes sense. Do you still prefer sinc-L and sinc-Mx for both DSD and PCM? Or do you prefer a different filter? Also, what modulator do you prefer for DSD?

Yep I use sinc-L most of the time and then Sinc-Mx on tracks that aren't produced too well. (the stuff I listen to a lot I run through PGGB)

for dsd I use ASDM7EC-Super

 

[Superunknown2]

Yep I use sinc-L most of the time and then Sinc-Mx on tracks that aren't produced too well. (the stuff I listen to a lot I run through PGGB)

for dsd I use ASDM7EC-Super

I listen to a lot of rock, and it seems to me that Sinc-Mx is better on quicker, more aggressive music. Sinc-L seems to be better with music that isn't as aggressive. I will try that modulator. I tried it before but my older computer couldn't handle it.

 

[Superunknown2]

Somewhat randomly when I try to run HQPlayer DSD on my Macbook Pro, nothing happens. I check the system, and it does not look like it is choking on what HQPlayer is trying to do, HQPlayer just isn't doing anything. HQPlayer and Roon are using little of the processor and little RAM when this happens. If I change HQPlayer's DSD settings to an easier modulator and a lower bit rate, the same thing happens. I never have this issue with PCM. Does anyone know what might be causing this?

 

[genefruit]

Somewhat randomly when I try to run HQPlayer DSD on my Macbook Pro, nothing happens. I check the system, and it does not look like it is choking on what HQPlayer is trying to do, HQPlayer just isn't doing anything. HQPlayer and Roon are using little of the processor and little RAM when this happens. If I change HQPlayer's DSD settings to an easier modulator and a lower bit rate, the same thing happens. I never have this issue with PCM. Does anyone know what might be causing this?

Change output to ensure DSD is using DoP

 

[Superunknown2]

It is. See my screenshot. Anything else I should set? Maybe change the DSD sources on the SDM tab?

 

[Erick Muller]

Somewhat randomly when I try to run HQPlayer DSD on my Macbook Pro, nothing happens. I check the system, and it does not look like it is choking on what HQPlayer is trying to do, HQPlayer just isn't doing anything. HQPlayer and Roon are using little of the processor and little RAM when this happens. If I change HQPlayer's DSD settings to an easier modulator and a lower bit rate, the same thing happens. I never have this issue with PCM. Does anyone know what might be causing this?

Try this settings on HQP, I had the same problem, you're probably trying to upsample a 48khz file and HQP is unable to do so because you don't have the correct settings.

 

[Superunknown2]

Try this settings on HQP, I had the same problem, you're probably trying to upsample a 48khz file and HQP is unable to do so because you don't have the correct settings.

That seems to have fixed it, thank you. Should the "48k DSD" and "adaptive rate" always be checked, or should they only be checked when the song requires it to play?

 

[Erick Muller]

That seems to have fixed it, thank you. Should the "48k DSD" and "adaptive rate" always be checked, or should they only be checked when the song requires it to play?

Depends on your DAC, but in my case that fixed it completely, there where entire records and songs that I couldn't play until I checked mark those boxes, I never made the connection that it was because this where native 48khz files until a few weeks ago lol!!!! and also people here where a huge help in this forum.