opus101

Austrian economics doesn't start with that assumption, probably one of the reasons it isn't taught. If it was, perhaps economics would begin to lose its dismal reputation.

Multibit SDM has been pretty much the standard for decades now in commercial SDM DAC chips. The multibit DAC is a 'low bit' DAC, typically 4-6 bits and runs heavily oversampled (256X is typical).

'One of the best 3' is suitably vague - I'd interpret that to mean 'its number 3, after those two'. But doesn't say how close it gets to them. If the reviewer mentioned which other DACs he'd heard above Yggy's price point that would set some valuable context.

I severely doubt you'll ever read such a thing in an advertiser-funded magazine. Which is of course the 800lb gorilla in the room.

No, it further clarified that your 'no' applied to my second statement and not the first.

There's isn't an 'in general'. 'No' stands as agreement when the premise is expressed negatively, as mine was. But that wasn't the only statement in my post which is why I asked for clarification. You're quite mistaken that listening is a form of measurement, rather measurement is a subset of observation. So you have it backwards.

That's in agreement with my 'So not....' then?

So not really a form of measurement at all then. I'd say measurement is a form of observation (listening counts as an observation) and observations are by nature qualitative. Measurements are, by their very nature, quantitative.

And there'll still be an argument about what counts as 'proper'. Isn't that rather a subjective word? ( In the background I'm searching another forum for an example of the point I'm making, but so far not coming up with the post I remember. I'll keep searching. Found it : https://www.audio “science” review/forum/index.php?threads/review-battle-of-schiit-audio-dacs.5487/post-121858. Here Amir misuses ADI's measurements to make his point to Schiit's detriment. )

It takes two to argue - just because there's still an argument doesn't mean the matter hasn't been resolved. So the presence of argument tells nothing.

This is wrong and we've been here before. He could demonstrate that there's something Amir missed (for example) or an assumption Amir's made that isn't correct. No need to provide his own measurements.

I've not met any commercial classD HP amps, I guess for the reasons you have already mentioned - nobody needs that many Watts. I've DIYed HP amps using a classD chip - TDA8932 - with an output transformer and gotten very pleasing results. The OPT is required as the chip works best (gives lowest noise and distortion) in BTL mode, as do many other similar chips.

This might come in useful - https://shepazu.github.io/Audio-EQ-Cookbook/audio-eq-cookbook.html

I'd take steps to reduce the number of trafos by specifying a single larger trafo with multiple individual windings.

Nah its definitely only a finite number. It was only five years ago that that number he quoted was 30.