The tradeoff between DSP and DSD

[sdolezalek]

30 minutes ago, mansr said:

Room correction works up to 50 kHz or so.  For these purposes, even 96 kHz is thus sufficient, and I don't believe resampling to this rate around the correction filters will be of detriment. If you worry that the resampling filters will have and audible effect, you can pick a higher rate like 384 kHz and still reduce the CPU load compared to running the main DSP at DSD rates. A clear benefit of using a lowish intermediate rate is that it enables the use of a more sophisticated correction filter without exceeding available CPU resources.

 Thanks! Given the other work that you have recently shown us with regard to measuring DAC noise and time/frequency smearing, I'm a bit surprised that you are ok with multiple layers of filtering and what that might do to time accuracy/ringing; but I respect your far deeper knowledge and all the measurement work you have been doing; so I have no basis to doubt you.

[mansr]

44 minutes ago, Samuel T Cogley said:

In the case of Roon, this "modulator" is a software implementation?

Yes, presumably the same as is used for PCM to DSD conversion.

[Miska]

2 hours ago, mansr said:

Room correction works up to 50 kHz or so. Speakers and microphones only extend that far, if you're lucky, so for higher frequencies you have no measurements on which to base a correction. For these purposes, even 96 kHz is thus sufficient, and I don't believe resampling to this rate around the correction filters will be of detriment.

 

I usually measure up to 20 kHz and then have rest of the response just flat-out to what ever the Nyquist in question happens to be. So I run the room correction always at the source material rate. So no need to have rate conversions for room correction purposes.

 

However, as usual, I use upsampling. Since highest rate DSD material I have is DSD256 and many of my DACs support DSD512, the output is anyway DSD512. Provides also some extra headroom compared to the original material.

[firedog]

1 hour ago, Samuel T Cogley said:

 

In the case of Roon, this "modulator" is a software implementation?

Yes. In most DACs you have some kind of similar software built into the chip, so it is HW. Nothing wrong with that, but it is inflexible and not upgradeable.

In a software like Roon or HQP you can offer different flavors of software and improve it over time.

[firedog]

2 hours ago, sdolezalek said:

 

I can see that this Holiday Season has put a lot of us into very cheery moods...   On a more serious note, if this is so easy, are you suggesting that Dirac, REW, Acourate and others providing DSP corrections are doing the same to preserve the integrity of DSD input signals?  If not, are you of the view that the benefits of their corrections clearly outweigh any damage they may be doing to the original sound wave?  

 

I'm not trying to be difficult, but I am genuinely interested in the tradeoffs being made in favor of delivering fully flat frequency response.  

 

 

I don’t know specifically about every implementation, but most room  correction devices and software convert everything to some set PCM rate like 24/48 or 24/96 and work from there. They would tell you that there’s no real sonic quality lost by this. Also, working with PCM is easier and takes less computing power. And since almost all setups either don’t work with DSD or convert it to PCM anyway at some point, they’d say no reason to use DSD format. 

[mansr]

1 hour ago, Miska said:

I usually measure up to 20 kHz and then have rest of the response just flat-out to what ever the Nyquist in question happens to be.

Would it make sense to lowpass somewhat below the tweeter breakup frequency?

 

1 hour ago, Miska said:

So I run the room correction always at the source material rate. So no need to have rate conversions for room correction purposes.

There's no need if your computer is fast enough. Depending on the complexity of the correction, performing it at a lower rate can be cheaper even with the overhead of resampling.

[Miska]

1 hour ago, mansr said:

Would it make sense to lowpass somewhat below the tweeter breakup frequency?

 

Yes, certainly it would. I just personally avoid using those peaking dome types that need it...

 

1 hour ago, mansr said:

There's no need if your computer is fast enough. Depending on the complexity of the correction, performing it at a lower rate can be cheaper even with the overhead of resampling.

 

Sure!

[Ron Scubadiver]

On 12/21/2017 at 2:30 AM, Miska said:

 

...and then your DAC chip will do equivalent thing inside (conversion from PCM to DSD-like data), in a not so great way...

 

How do you know it's not so great?

[Fitzcaraldo215]

Love the sound of pure DSD, especially in DSD256 5.1.  Love the sound of 5.1 DSD, even DSD256, converted to 176k PCM wth speaker distance correction, bass management for my sub and Dirac Live EQ even better!