On Jan 17 2011, Sampo Syreeni wrote:
On 2011-01-14, Dave Malham wrote:
D*mn, I'm just finishing off a (VST/AU) encoder myself using an IIR
filter set based on the analogue all pass filters in the original
Calrec unit (as designed by Geoffrey)- if I'd realised Fons had done
so already, I could have nicked his code and saved myself some
considerable time :-)
The joke aside... Why don't we know this stuff beforehand? I mean, why
aren't all the newest coefficients out in the open in an easy-to-use
form?
Here's some, anyway. I've built a VST with the phase shifters in. I've also
built a goniometer in a VST so I can measure the relative phase shifts,
rather than just look at an XY display. I'm running the whole thing in
Bidule and playing with the sample rate. I've been using the CSound set of
pole frequencies and a chain of non-frequency warped first order all
passes. These pole values are taken, via CSound from Bernie Hutchins
"Musical Engineer's Handbook"
double poles[12] = {0.3609, 2.7412, 11.1573, 44.7581, 179.6242, 798.4578,
1.2524, 5.5671, 22.3423, 89.6271, 364.7914,
2770.1114};
the first 6 are the 0 degree poles, the second the 90 degree ones. These
poles are not used "raw" but multiplied by a "fudge factor" which for
CSound is 15. I could not find a reason for this choice so I've set up a
control in my VST to change this fudge factor to see what happens. For
instance, at 192 kHz sample rate, I get around + 0.3 -0.15 degrees from
20Hz to 22kHz with a "fudge factor" of 5. At 44.1kHz, on the other hand,
it's difficult to keep within the +-0.5 degree above 14kHz without the low
frequencies going out of bounds (that's a fudge factor of around 21) so I
suspect that the symmetrical around fs/4 approach would be much better -
but as I'm lazy I'm just going to oversample 44.1 and 48 k signals :-)
Dave
_______________________________________________
Sursound mailing list
Sursound@music.vt.edu
https://mail.music.vt.edu/mailman/listinfo/sursound
