Yeah, coming up with the names is hard. And originally “super” was the “best”. But now you could also think “super heavy”, “light” and “ultra light”.
But over time I discover better and better how different subjective aspects correlate with different aspects in the mathematical formulas. Several decades on these formulas, and still discovering.
You know, I keep going between the modulators over time and finding “this is interesting, I know the mathematical difference, and I know what I hear”. And I know the differences from objective and mathematical perspective. And over time my view on these things develop further. But someone (unnamed) else would tell you that the differences are inaudible. You can decide, it is pretty easy to switch between and try!
Advanced modulators are very complex animals. That is also one reason DAC chip manufacturers and companies making their own modulators say practically nothing about modulators, but babble about digital filter all day long.
As an example, if you look at the table on ESS website:
This is three different modulators, almost nothing said about these. And I can tell the difference is certainly more than 0 dB difference in SINAD and 1 dB difference in DNR.
More interesting is the output resistance which defines the ultimate noise floor level (higher the resistance, higher the thermal noise).
AKM has four modulators in AK4191 front-end chip, but those are just numbered 1, 2, 3 and 4. That’s it.
But it is absolutely interesting domain to work in. And these are precisely things where high clock rates help, because you cannot do many of these things in parallel processors due to mathematical relations. So nothing beats clock cycles (cycles/instructions per output sample). And we have just got past 6 GHz barrier.
I think it’s been 12+ months since I did that. it’s a waste of time
Looks like they just don’t like DSD. period.
