iDAC 2 - Quad DSD256 / PCM 384Khz / USB DAC & headphone amp

[mansr]

24 minutes ago, AMR/iFi audio said:

In iDAC 2 micro's 'bitperfect' mode all digital filters are bypassed and the DAC operates at the original sample rate.

What this means, somehow, is that the microcontroller (XMOS) puts the DAC in bypass mode and repeats each sample up to 8 times to reach 384 kHz. I don't see why anyone would want to do this, but that's what it does.

 

24 minutes ago, AMR/iFi audio said:

If any filter is selected, the input signal is upsampled to either 352.8/384kHz or 705.6/768kHz.

The other settings choose between the built-in filters of the DSD1793 chip. This is easily verified with a scope and spectrum analyser.

[mansr]

25 minutes ago, AMR/iFi audio said:

If the bitperfect filter option is selected the DAC is fed directly to the actual DAC part of the chip and converted without any upsampling and digital filtering and without repeating samples or such trickery. 

Something has to repeat samples up to the rate of the sigma-delta modulator. Whether that happens entirely within the DAC chip or externally really doesn't matter.

 

25 minutes ago, AMR/iFi audio said:

It operates equivalent to what has been termed 'Non-Os, Non-Oversampling, Zero-Oversampling' etc. We prefer the term bit-perfect because that describes the operational principle better.

I think it's a rather misleading name. There is nothing "perfect" about zero-order hold upsampling.

[mansr]

5 hours ago, buonassi said:

Specifically for iDAC2 - When you say "minimum/listen" are you referring to minimum ringing or minimum phase?  I think the former is what is meant by these descriptions and also after reading the article on the nano idsd:

https://ifi-audio.com/audio_blog/does-the-nano-idsd-measure-good-bad-or-ugly/

 

Is the "minimum/listen" setting still technically a linear phase filter but with reduced delay of the input signal (less pre and post ringing)?  The impulse response appears to be symmetrical, like a traditional linear phase sharp rolloff filter, but with visibly less ringing before and after the transient (see pic).  At least this is how it is shown in the idsd article.

 

When I think of minimum phase, I think of the filter in the dragonfly red (see other pic), which shows all the ringing post transient.   I'm still learning and may not understand the difference between minimum ringing and minimum phase so please forgive my ignorance. 

There is a more specific mathematical definition, but a minimum phase filter is indeed characterised by having no oscillations before the primary peak. The tail can then be either short or long, steeper filters having longer tails.

 

5 hours ago, buonassi said:

So is the iDAC2 impulse response symmetrical like the iDSD nano or does it dump all ringing after the transient spike like the minimum phase filter in dragonfly red?

The iDAC2 and iDSD Nano use exactly the same filters, specifically the built-in ones of the DSD1793 DAC chip.

 

5 hours ago, buonassi said:

Is it right to call this, technically speaking, a minimum phase filter?

No, that's is patently incorrect. It is a slow roll-off linear phase filter. I find their use of such misleading labels rather disappointing. It makes you doubt everything else they say too.