DSP Robotics Support

MyCo - what did changed (in terms of differences produced by audio output) in optimizations you did on this filter pack? Or is this 1:1?

source filters (Complete Dozius Filter Pack):
http://synthmaker.co.uk/forum/viewtopic ... 564#p59720

your mods:
http://synthmaker.co.uk/forum/viewtopic ... =15#p70109

I'm gathering a package of filters (for a project) that would produce good output when used with modulators and noises.

I wrote it in the post:

I removed the samplerate parts and some useless multiplications from the Linkwitz-Riley filter.

I'm getting old, can't remember much from that time... 3 Years ago...

I guess we are in the same or similar age.

I asked, because looking at the codes - I can't tell whether it was added something "obvious not worth to mention" (denormals algos?) or changed something that will produce different rounding (thus - audible results if these variations play the role).

Doing side by side audio comparisons, I had the impression that at least one of these filters sounded a little bit different (gain + something). But maybe I was tired - will check on this one later.

...and now the best part. i just realized, that I was watching at totally different filters, made and optimized by someone else

http://synthmaker.co.uk/forum/viewtopic ... a&start=30

I'm trying to figure out what is the relationship between provided "slope" and real resonance "Q" in shelf filters...

I've used a SVF lp filter optimized by myco on left and right channels, and it creates stereo artifacts (mono source). Seems a bit of precision has lost with optimization, am i right?

could it be a 'timing' difference between the 2 codes running ?

RJHollins wrote:could it be a 'timing' difference between the 2 codes running ?

Solved:
MyCo uses a hop(128) on a part of the code. Its a bit high so I set it to 8 and you can't hear any annoying effect by ear.