TOPPING D90 III Sabre DAC Review

[maruko]

SNR is the ratio of full-scale output amplitude to output amplitude during silence.

Nothing is playing while the Noise part is being measured.

Ok, got it. Thanks again.

 

[mike7877]

D90 III power consumption at different sampling rates (into XLR into 10k ohm)

176.4 || 0.0855A
192 || 0.0860A
44.1 || 0.0880A
48 || 0.0890A
88.2 || 0.0890A
96 || 0.0895A
352.8 || 0.0910A
384 || 0.0920A

Interesting how 192 is least because 8x = 384

 

[Billy Budapest]

D90 III power consumption at different sampling rates (into XLR into 10k ohm)

176.4 || 0.0855A
192 || 0.0860A
44.1 || 0.0880A
48 || 0.0890A
88.2 || 0.0890A
96 || 0.0895A
352.8 || 0.0910A
384 || 0.0920A


View attachment 453937


Interesting how 192 is least because 8x = 384

What is the native rate for the chip’s delta-sigma modulation? 1-bit, 2.8MHz? My guess is that sample rates with the greater processing overhead to decimate & oversample, or Fourier transform, or whatever needs to happen computationally to get them to the native rate would result in higher power consumption.

 

[mike7877]

What is the native rate for the chip’s delta-sigma modulation? 1-bit, 2.8MHz? My guess is that sample rates with the greater processing overhead to decimate & oversample, or Fourier transform, or whatever needs to happen computationally to get them to the native rate would result in higher power consumption.

I think it's quite a bit higher than that, actually.
It's 48/96/192/384 x 64/128/256/512/1024 and
44.1/88.2/176.4/352.4 x 64/128/256/512/1024

I think default target is to hit 45.1584/49.1520MHz (ie. 48 * 1024 = 49,152)
There might be an alternate mode that makes the target half (22.579MHz and 24.576MHz)
That's the idea I get from the datasheet, though I'm not 100%