You are misreading my post and are writing things I did not say. I just gave common examples - anyone reading my signature sees my current preference is the SOTA ring DAC.
Why should I check your dac preference which has nothing to do with the argument here.
It's not possible to misread your response cause there is no response actually, just plain words IOT respond. No technical explanation or any explanation at all.
The Nyquist frequency is actually 22.05kHz for redbook. An 'ideal' anti-imaging filter would be flat up to 20kHz, and then attenuate sharply to well below -100dB by the Nyquist frequency.
For fun...
Take any 24/44.1 (or 16/44.1) audio file you have. Upsample it to, say, 8x using something like Audacity, which uses a 'sharp' (I would say "correct") filter. So, you now have a 24/352.8 file. Now downsample it back to 24/44.1. Use something like DeltaWave to compare the original to the up-/downsampled file.
Here's one I've just done with Kate Bush's 'Watching You Without Me' (24/44.1):
The original and the up-/downsampled files null to well below -200dB within the passband (<20kHz). There is no DAC on the planet that can resolve anywhere near this level. The files are essentially identical below 20kHz.
Pre- and post-ringing are a total non-issue and just a marketing ploy IMO.
The Nyquist frequency is actually 22.05kHz for redbook. An 'ideal' anti-imaging filter would be flat up to 20kHz, and then attenuate sharply to well below -100dB by the Nyquist frequency.
I know Nyquist frequency is 22.05kHz for 44.1kHz sampling. I wrote 20.5kHz IOT be very optimistic but considering anti-aliasing filtering and as you mentioned it can only reach 20kHz in reality.
The original and the up-/downsampled files null to well below -200dB within the passband (<20kHz). There is no DAC on the planet that can resolve anywhere near this level. The files are essentially identical below 20kHz.
If it's gonna null out itself why upsample and downsample again. It should be like this: upsampling 8X, sharp roll off filtering at 20kHz and downsampling 8X to be proper example of oversampling. For proper evaluation you should check impulse response.
If you're saying there is no such a thing like ringing I suggest you reconsider and check impulse response. But if you are saying, yes, there are pre and post ringing added to the original signal but they are not important to make a difference and nobody can hear them then I ask you a question. How can you convince yourself that digital is capable of replicating all the nuances of analog signal while nuances added by pre and post ringing can not be heard?
You don't need to answer that cause nuances added by ringing can be heard. Digital is not that bad.
The Yggy does not employ NOS mode. The New Bifrost allows you to toggle between the Mega Burito combo filter and NOS mode. On the Yggy you use their proprietary filter which for 44.1 upsamples to 352.8 and 48 to 384. But what do I know as a user and someone who has discussed this with Mike Moffat. Poor me.
The Yggy does not employ NOS mode. The New Bifrost allows you to toggle between the Mega Burito combo filter and NOS mode. On the Yggy you use their proprietary filter which for 44.1 upsamples to 352.8 and 48 to 384. But what do I know as a user and someone who has discussed this with Mike Moffat. Poor me.
C2 errors are almost non existing in good CD players. More than 20 years ago I implemented counters of C1 and C2 errors and checked it. Many people did it - some manufacturers even sold players that included these counters. Anyone using appropriate CD ripping software will confirm it. Bit exact or bit perfect are exactly the same thing.
BTW, occasional interpolation is less invasive than clicks and pops.
It's not possible to misread your response cause there is no response actually, just plain words IOT respond. No technical explanation or any explanation at all.