high rate DSD and DSP systems

[radix]

Hi all,

What's the point of using high-rate DSD if one uses DSP systems that are going to sample at a much lower bit rate? For example, Genelec 8351B seems to use up to 192kHz sample rate (but at variable word size?). A miniDsp 4x10 HD active crossovers are 24-bit @ 96 kHz.

So, if one is using a DSP active crossover, for example, does anything beyond DSD64 or maybe DSD128 make a difference? Also, these systems support AES3 or SPIDF, which I think are 20 bits @ 48 kHz, maybe 24-bit @ 48 kHz?

To make this a bit more concrete, how does a DSD512 over USB to a Topping 90SE make an improvement over a lower-rate source, if one is going to feed it through a DSP system? Why not stick with AES3/SPIDF rates?

Personally, I'd assume (wildly guess) that having the active room correction from a DSP would give a much bigger benefit than high-rate DSD.

If there are already posts that go over this, please help me find them. Or if you know of a good article or technical reference, that would be great.

Thanks,
Marc

 

[digitalfrost]

I agree with you. See also.

https://archive.is/sYerA

https://productionadvice.co.uk/high-sample-rates-make-your-music-sound-worse/

 

[retro]

Personally, I'd assume (wildly guess) that having the active room correction from a DSP would give a much bigger benefit than high-rate DSD.

I think you summed it up pretty nicely there. DSP and DRC gives you sooo much more than DSD ever could. Even in a "perfect" room.

 

[Kal Rubinson]

What's the point of using high-rate DSD if one uses DSP systems that are going to sample at a much lower bit rate?

Quite likely, no point at all.

So, if one is using a DSP active crossover, for example, does anything beyond DSD64 or maybe DSD128 make a difference?

None at all since all DSD at any sample rate must be converted to PCM in order to apply any DSP. FWIW, my default setup is to convert everything to 24/194 for DiracLive although, even before that, I preferred using Audyssey (which in the vast majority of devices converts everything to 24/48) with almost all sources.

Personally, I'd assume (wildly guess) that having the active room correction from a DSP would give a much bigger benefit than high-rate DSD.

It is true and I (and others) have been stating this for years.

If there are already posts that go over this, please help me find them. Or if you know of a good article or technical reference, that would be great.

I stated this in print many times and have and will post such when asked but, frankly, this is not a new situation.

 

[radix]

Thanks everyone for the confirmations. @digitalfrost, I liked that first article. @Kal Rubinson I'll try looking up some of your other posts.

So it seems like a system that's, say, a miniDSP SHD w/ usb or spidf inputs to 2x sub and stereo amp would probably be a good way to go, then correct with Dirac. I don't see why I would want to have a DAC before a DSP.

I'm dealing with an untreated living room, so my goal is that most positions sould OK and not many sound bad, rather than try to optimize one spot to be perfect.

Marc