On Mon, Oct 21, 2002 at 07:11:41 +0200, Tim Goetze wrote: > Steve Harris wrote: > > >On Mon, Oct 21, 2002 at 02:39:05 +0200, Tim Goetze wrote: > >> pleased to announce the release of 'unmatched'. > >> > >> 'unmatched' is a simple effort to recreate some aspects of > >> the tone shaping of a real instrument amplifier. unlike > > > >Excellent. Can I ask how you did the FIR -> IIR mapping? Is it neccesary > >or significantly better to use doubles for the coefficents? > > a quick test (%s/double/float/g) shows the cpu usage doubling, > but i'm unsure what may cause this huge performance drop.
Lets just put it down to chache issues and ignore it ;) > doubles should, on average, mean we seldom hit the denormal number > bounds, or at least less frequently than with floats. i also expect > doubles to be beneficial to the filter stability by minimizing > round-off error, though i may be wrong here. lastly, x87 uses 80 bits Yes, though can manually truncate the floats, and floats will give half the cache footprint, so play better with other software. I dont really have a feel for when doubles are neccesary for filters yet. > >Have you experimented with adding a delay line and LP filter to simulate > >reflections off the back wall of the cabinet? > > actually the next thing i'd have done do to come closer to the > original impulse would be an added IIR operating behind a delay > line, yes. the current plugin response does not capture the later > parts of the original response so well (now that you mention it, > i think it must be those reflections). however personally i'll > probably do more research on nonlinear effects before refining > this method. Yes, you can always pep things up with a nonlinear effect of somekind before the cabinet+speaker sim. In FFT convolvers you take several impulses at different amplitudes and shift between them. I wouldn't worry too much aboout being very close the the impulse, the're only for a particular input amplitude and I can't remember when they came from so may not be fantastic. I'm wondering if this technique can be used for reverbs too - generate a purely "white" synthetic reverb tail, and apply an IIR the aproximate shape of the rooms impulse to it to make it sound more real... - Steve
