sajunky
Headphoneus Supremus
Yes, we had discussion on Denafrips not being NOS (your allegations) due to:I've had discussions here on this topic several times and usually it just ends up with messages getting removed once the debate goes on too long.
No evidence has ever been provided to substantiate Sajunky's claims regarding 'scrambling' and plenty to the contrary so I'll just leave it at that
1. Denafrips doing linear interpolation, at a x-times base frequency. Not strictly speaking oversampling as it is usually involved with DSP processing, but it was interesting to see your proof on this issue.
2. Denafrips doing DSP sound processing or filtering (which violated a second condition for NOS). You didn't bring any proof for this allegation.
When Denafrips released 12th models with a real NOS you have repeated tests and confirmed that it was in fact NOS-type sample-and-hold decoding method at a base frequency, but there was still data processing, so it was not a real NOS. I was opposing the later, but mods started deleting posts, in result we don't know to this day whether you have provided a proof for your allegation or not.
On the other side you maintain opinion that Holo Audio is a real NOS, while your own tests provided indication of scrambling a sound. A small amount of dither is transparent to the FFT analysers, but the attempt to linearise ladder errors require huge amount dither at approx 15-bit, so it did show as I pointed out on your own screenshot. However Holo Audio matter is off-topic in this thread, so will leave it for now.
Last edited: