Re: [iiwusynth-devel] Filter: Implemented, fixed

[Peter Hanappe]

On Tue, 2002-04-02 at 12:17, Christian Nentwig wrote:
> Hello,
> 
> found it. I hadn't tested sounds, which change the filter frequency
over
> time: recalculating the filter coefficients produced some
> 'zipper'-noise. But please let me know, if it still misbehaves.
> 
> The updated version is again at
>
http://prdownloads.sourceforge.net/millisampler/Filterimplementation_tar.bz2

Hi Markus,

I tested and integrated your filter. It's a major improvement over what
we had. My keyboard seems to be out of order so I could only do some
simple tests. The Warbling Bird test (preset 117 in
VintageDreamWaves.sf2) is still not as it should be but doesn't produce
any noise any more. In one song I had some clicks but overall, all the
MIDI files I tested sound much better and the mix is much clearer. 

I integrated it in the code and only made some esthetic changes and the
changes to the center frequency mentioned below. Now in CVS.

BTW: you commented out iiwu_profile() and voice->status = ... in the
code below. Did you ran into any problems with it?


  if ((voice->volenv_section == 5) && (amp < MIN_AMP)) {
    iiwu_profile(IIWU_PROF_VOICE_RELEASE, voice->ref);
    voice->status = IIWU_VOICE_OFF;
    goto post_process;    
  }


> There is also a sound font test file, which sweeps the filter to the
> limit.
> Patch 00: Slow filter sweep
> Patch 01: Fast filter sweep
> Patch 02: Very fast filter sweep ('typewriter-sound')

Patch 0 and 1 work fine. Patch 2 also, but has a low volume. Maybe
because of a high Q value?


> Speaking of it, there is something odd about the fcenter-information
> that is produced by iiwusynth: A linear sweep (cents per second)
should
> reduce the frequency on a logarithmic scale. But the frequency change
> seems to be linear (it reaches 0, that's impossible when sweeping on a
> logarithmic scale).

Yup. You found another bug :) The fres had to be converted from cents to
Herz! I fixed it. Sounds better, too!


> Currently the filter coefficient calculation happens every
IIWU_BUFSIZE
> (64) samples once for each voice, and it includes a sin(x) and cos(x).
> Has anybody an opinion on how CPU-intensive those are? Does it make
> sense in the days of FPUs everywhere to use lookup tables?

They are calculated every 64 samples *if* the center frequency changes.
About the sin() and cos(), large lookup table can also spoil the memory
cache. So I leave it as is right now. 

Wonderful job! Thanks!

Peter



> Regards
> 
> Markus
> 
> PS: Nice sound font. My current favourite: El-Cheapo Organ :-)
> 
> 
> 
> 
> Josh Green schrieb:
> > 
> > On Mon, 2002-04-01 at 10:44, Markus Nentwig wrote:
> > > Hello again,
> > >
> > > now the filter is implemented. I have put a tar file with the
> > > implementation on sourceforge for discussion (+backup...):
> > >
http://prdownloads.sourceforge.net/millisampler/Filterimplementation_tar.bz2
> > > Have a look at it, I'm happy about any any comments.
> > > I have tested it using a high-Q 440 Hz setting and the key A3, it
works
> > > just fine (and it's rather obvious that pianos sound darker with
the
> > > filter).
> > >
> > > I can merge the code into an up-to-date file version some day.
> > >
> > > Regards
> > >
> > > Markus
> > >
> > > PS:I'll be offline for a day or two.
> > >
> > 
> > I gave it a try, but didn't really like the result :( Seems like the
> > filter effect that is currently in iiwusynth sounds closer to what
would
> > be heard on SB hardware, but I haven't compared them directly. It
did
> > cause a bunch of cool radio tuning feedback sounds when upping the Q
> > though :) There is just not right with the parameters perhaps? I
tested
> > some instruments in the Vintage Dreams sound font (what I do a lot
of
> > testing with). Some of them don't sound right at all. Seems like you
> > have a pretty good grasp on the math behind it though, so I'm sure
its
> > just something wrong with your parameters or ?? Its nice to see that
> > there is some work being done with the filter though. I like the
sound
> > and all, but it seems a bit unstable (have you figured out anything
with
> > this Peter?) Cheers!
> >         Josh Green
> > 
> > BTW I just checked in some new changes to libsoundfont and Swami.
The
> > sample viewer is now working again and I changed libsoundfont's
sample
> > management to keep track of multiple storage methods for the same
sample
> > data (a sound font file and RAM for instance).
> > 
> > _______________________________________________
> > iiwusynth-devel mailing list
> > address@hidden
> > http://mail.freesoftware.fsf.org/mailman/listinfo/iiwusynth-devel
> 
> _______________________________________________
> iiwusynth-devel mailing list
> address@hidden
> http://mail.freesoftware.fsf.org/mailman/listinfo/iiwusynth-devel
>