Subject: Re: [linux-audio-dev] [ann] unmatched - a LADSPA amp tone
From: Tim Goetze (tim_AT_quitte.de)
Date: Tue Oct 22 2002 - 16:00:03 EEST
Steve Harris wrote:
>> currently, i'm looking at what a sine wave looks like after
>> it's been handled by a good distortion stomp box. the way it
>> shapes the wave form seems easy to grasp, but as usual, i
>> am hesitant to implement what i don't understand fully ...
>
>I've found that the precise shape of the transfer function is important to
>the sound, do you have a model for the transfer function or are you just
>going to smaple it?
i'll rather try and come up with a formula. a simple mapping
will not work, because the transfer function moves the peaks
of the sine to happen earlier (i guess looking at the derivative
of the signal may play a role here). i have some ideas, but i'll
need to read up on what's actually happening inside the
circuitry. and if it turns out to include fabs() and branching,
i'll try and combine it with a compression effect (if it isn't
some sort of already, which i suspect).
>Sadly heres where my zero knowledge of amps kicks in ;) I wouldn't know a
>fender form a hole in the ground.
they all sound like a stratocaster, whatever instrument is
plugged in. ;) no seriously, their sound is very bright; at
best you'll call it very 'electric'.
>> >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...
>>
>> very interesting thought indeed. do you have a good response
>> for trying this? (sorry, "steve's flat" doesn't qualify,
>> "albert hall" is more like it ;)
>
>I have some, but there not included because there 1) very long 2) legally
>dubious. I dont think that deriving an EQ curve form them can be dodgy
>though - the hard part is more likly to be getting a purly white reverb -
>I suspect that all those allpass's and combs dont even out very well.
i guess (don't know much about reverb design [yet]) that in
order to get a truly 'white' reverb the number of delay lines
approaches infinity. which turns a decent implementation into
a real programming challenge. :) nonetheless, one could start
with what we have -- usually people complain that there's not
enough 'color' in our digital reverbs ('gray'?).
>PS "Steve's flat" was captured with a half knackered monitor and a three
>quarters knackered mic. I should probably do it again, as I have moved :)
and i should really do the fender here and compare the original
to the convolved/iir-filtered emulation. but it's more important
to get the nonlinear parts right first i think.
tim
This archive was generated by hypermail 2b28 : Tue Oct 22 2002 - 16:06:48 EEST