> Has anyone but me seen the issue of Computer Music magazine with the
> tutorial and samples to MAKE YOUR OWN PIANO? Very cool -- wonderful
> magazine/CD (UK publication, but I get it here in the states no problem).

Does it go into all the estroic little details that make up the piano? We
have a collection of samples (64meg dual velocity pianos - a few I think)
over on Sonik curtesty of Gunter Matejick (heh! I just probably butchered
his name! He'll probably beat me if he still reads this). I love these
samples, and with a lot of work, they would be professional quality.

Still, there are a LOT of things that are done that most people have no
clue about. Those of us with Mike Martin's Piano stuff know that when we
look at the programming behind it. It isn't just keymapping and hoping
that everything will be all right. Its things like: choosing the right
samples. Not every sample recorded in a session will be right for the
program. If you aren't using every note, then it becomes a tedius process
of weeding out which notes with in this range and which ones don't. You
may find your sweetspot only to find that in order to get all the samples
to sound uniform, its time to start all over again.

Again with Mike's Piano: look at the layers. I believe he has 8 layers
JUST for a piano. Why would one do this? Well, if i remember right, this
was to ensure the note polyphony / dropouts worked in a certain way and so
the filters could be tweeked per individial ranges.

Then think of things like Sweetwater's Pianos: When these were recorded,
Dan Fisher went outta his way to digitally edit out the hammer knocks (I
think they tried to do some of this physically as well) so that when
played back, the natural sound wasn't as overpowering. There is a
difference between what you hear when playing a real piano and what you
hear from adding several notes together and the artifacts that come around
because certain mechanical noises are technically a global item, but being
played back polyphonicly. So what did they do? They removed the knocks
and then added a monolayer which added this back in the most palletable
way.

So, what I'm getting at is that a great piano sample ISN'T just a bunch of
samples keymapped together. It is this plus years of programming
experience from dedicated sound engineers. Will it sound like a Piano?
Probably. Some people prefer the crappy sound of the Roland Piano (overly
compressed and easily spotted switch points...yeah, I'm being critical of
it, but I also wish I wouldn't have sold my GM box as it was PERFECT for
pop music...the Kurz ain't anywhere near good for pop without a lot of
compression and overplaying. Then again, I can't imagine the piano
sampled having hardened hammers and all the other tweeks the roland one
probably had physically either :-)

This all said, you can get good samples for a cheap rate anywhere. When I
work in studios, I generally charge between $50 and $75 an hour. By
computer programming rates are close to this as well. This means to me
that if it took more than 4 hours to program a sample from an assemblage
of samples, I'm doing myself a disservice. I've spent more than 4 hours
tweaking even the commercial samples to my liking to even consider trying
them on my own.

BTW: I'm not trying to get anyone from building their own program, I'm
just letting ya know what kinda work is involved. I encourage everyone to
learn how to build their own patches...commerical stuff just ain't going
to cover everything or even be at the style ya want it to be in.

blah

clif

marsiglio.clif.c _ . _ . _ . _ . _ . _ sonik: technology + music + kurzweil
mgr. development . _ . _ . _ . _ . _ . _ . _ . _http://sonikmatter.com/clif
iupui.testing.center _ . _ . _ . _ . _ . _ . _ . _ ccmarsig@...
tel.317.274.2897 . _ . _ . _ . _ . _ . _ . _ . _ . _ . _ . fax.317.274.3400