The filter code is identical; the benefit of the Hugo M scaler is that the isolation on the dual BNC's has been improved, so you don't need to worry about ferrites. The benefit that the Blu 2 has is being able to play CD's; and all the reports I have seen is that CD replay using the Blu 2 is better than a dedicated server; and the reason for this is the much lower power dissipation and overall lower RF noise as the internal clock speeds of a CD mech is very low, and it's low power too. Many use the CD as a way of calibrating their source.
Oh dear. This won't be easy to answer!
Just to clear things up - from a mathematical or theory perspective, converting the 44.1 kHz to 705.6 kHz is not adding any more information; indeed, it's actually the opposite, in that we are trying to preserve the information content without distortion or inaccuracy, and from a SQ perspective trying to accurately recover the original timing of transients. It's not making things up, but reproducing the original analogue signal that was in the ADC before it was sampled.
Now we know from theory that a sinc function interpolation filter will recover a bandwidth limited signal perfectly, without any change. Conventional filters do not do this; they can have huge instantaneous errors with transients, and this means that the timing of a transient will have an error; it will either be too early, or too late, and this error is hugely important perceptually, as transients are used by the brain to perceive pitch, timbre, soundstage and of course the starting and stopping of notes. So you can imagine how awful and un-musical music would sound if you could not hear pitch, timbre, soundstage and notes starting and stopping. Hence why the M scaler is so important from a musicality POV - because the M scaler is the same as an ideal sinc function filter, to a better than 16 bit accuracy, then all the transients are being reconstructed to a better than 16 bit accuracy all of the time, irrespective of the signal. Conventional DAC's do not do this; as they are the same as an ideal sinc function to 1 bit accuracy (min phase type filters or NOS) or 2 to 4 bits (symmetric filters). So why can't we do this in a PC? Well many claim it can be done, but don't actually deliver the real tap length that is needed, nor the algorithm that is sinc like in it's coefficients.
Note that the M scaler does not sample rate convert - so 44.1 becomes 705.6, and 48 becomes 768, as this would create distortion and noise and would have timing errors.
So in summary - the M scaler is trying to reproduce the original bandwidth limited analogue signal perfectly without any changes - and because it is identical to a sinc function to 16 bit accuracy, it will do this perfectly to a better than 16 bit accuracy.
And yes my DAC's are different - they do not use DSD but pulse array for the DAC conversion - and this is done so that the analogue has no distortion, no small signal non-linearity, and crucially no noise floor modulation, which is highly audible. DSD can't do this at all, nor will it ever be capable of doing this because of fundamental limitations to a 1 bit methodology.
As to DSD - a DSD bitstream looks nothing like the original analogue signal, as it adds huge amounts of out of band noise - with DSD64 you are -20 dB down at 100kHz. Moreover, if you simply low pass filter it, you get large amounts of distortion and noise, as the signal activity (switching) is signal dependent. DAC conversion is very very tough to do with low distortion and noise. Pulse array has constant switching activity, and has no added distortion, and more importantly no noise floor modulation. But pulse array requires PCM for it too work, so DSD must be converted to PCM; and we need to remove the dreadful RF and out of band noise with DSD too. So this noise gets removed, with a 220 dB stop-band filter - and we get complete removal of the HF noise and distortion that DSD creates.
Actually this is the same process in principle to PCM, in that the only way of recovering the original analogue signal before the DSD modulator is to filter it; I am trying to do exactly the same, recover the original analogue signal that was in the ADC....
Rob
