are. you comparing blind - level matched etc. etc.
What mechanism do you suppose can create this difference.
Comparisons are not blind but with instantaneous switching.
When comparing digital inputs on the same DAC, level matching should not be an issue.
If level matching is necessary, then that would prove something.
I will do that over the weekend, when I have more time.
I have a T8 and Pi5 with Pi2AEs. Roon is configured to link the T8, Pi2AES and native Pi5 USB zones.
The D900 can then be instantaneously switched between inputs.
The change in sound stage is quite obvious.
To some extent, all DACs are black boxes, but these Topping DACs seem more so.
The D900 and D90 III discrete also have the option of apply digital processing (PEQ) for some inputs.
This suggests a different data path. The I2S bypassing this function.
For USB, this data is internally clocked.
For AES and I2S the source clock is used at some level.
The I2S cannot using the onboard PEQ, so that is another difference.
The AES input is an outlier; it is very clear but seems to lose some bass.
It is so dramatic that I suspect a phase issue between channels.
I suppose ultimate certainty would make sense if these DACs were tested with actual signals comparing their inputs in stereo mode.
I also have a Shelly Rely controlled ARX XLR switch that I have used to compare different level matched DACs.
It is possible that this is unique to these Topping DACs.
- Rich