[sdiy] Henry, 256 Bw limited!

[Antti Huovilainen]

On Sun, 27 Dec 2009, karl dalen wrote:

> But are the same ZHO effect the same for variable sample rate systems?

Yes, as long as the samplerate really is variable or close enough that you 
can ignore aliasing (as in PPG which uses very high internal samplerate).

> Assume 8 bit length, no less no more, then this sharp transition
> of bit to bit level adds harmonics (over/under shot swing) further
> increases the spectrum. The faster the settling time between level
> changes (no slew) the broader the bandwidth of the resulting wave!?

Yes. In practise the slewrate is unlikely to be slow enough to have 
audible effect unless it's deliberately limited.

> I get the impression that the ZHO, if so, gets pushed upwards as
> sample rate goes up as well? (aliasing of wave are harmonics).

Yes.

> But where does Andrej muffling apply then?
> Can you really hear a small 3.9db loss at Nyquist?

Not to the minor change in spectrum, yes to the spectral shape extension 
due to mirroring. Without the mirroring very low notes do sound more 
muffled. A110 contains real harmonics only up to 7040 Hz with 128 sample 
wawetable. Any harmonics beyond that are due to mirroring.

> What do you mean by gritty PPG?
> Wave to wave mismatch in table as in MSB integer toggling of bits in wave?
> or gritty as in static cyclic single wave?

The waveforms themselves are far from "ideal". The sine as an example 
contains many higher harmonics than what you would get if you calculated a 
128 point sine and quantized the samples to 8 bits.

Antti

"No boom today. Boom tomorrow. There's always a boom tomorrow"
   -- Lt. Cmdr. Ivanova