Buggy Whips [was]Re: [sdiy] Should I repair my Fostex, or should I go HD recording?

[Kenneth Elhardt]

Richard Wentk writes:
>>*Anyone* with a half decent ear should be able to tell the difference
between a sampled piano and the real thing without even thinking about
it....Clue: Try sitting close to a double bass section in an orchestra and
tell me there's no acoustic coupling between instruments.<<

By your last sentence I'm starting to think you are comparing a sampled
instrument to sitting infront of a real acoustic instrument.  If that's the
case, no sample or recording sounds as good as sitting infront of a real
acoustic instrument.  I'm comparing a sampled instrument to a recording of
an acoustic instrument since that's where the sample will end up.  Sampling
a note of an instrument (say a trumpet or pipe organ) is going to give you
the same thing as recording a performance of that note.  There can't be any
hidden things in one vs the other since they are both recordings of the same
thing.  If you sample a double bass section and there is something around to
couple with it, then it's there in the sample.

>>(regarding sustain pedal down samples) But they don't work like a piano,
because you'd have to create a different mix for every chord to include the
overtones of each chord in addition to the notes themselves.<<

This is nit-picking.  Each sampled note already has coupled string vibration
within it do to the sustain pedal held down.  There's no need to do a
different mix for every chord since the different notes already have
different coupled string vibrations.  They just add together.  It's real
because it's a sample of a real piano.  The only problem I see in samples is
if you push down the sustain pedal after hitting the keys.

>>I've tried all the recent gigasampled pianos and they can all be described
in one word: crap. They're fine buried in a mix with a ton of other stuff
happening, but they sound like plastic plinky shite when played naked.<<

I'm beginning to think you're trying all of them dry.  Anything sounds lousy
without reverb of some sort.  As soon as reverb is added, the sound is
pushed away and obscured and most of those tiny almost inaudible things
become totally inaudible.  I've heard plenty of sample demos of most
instrumetns that sound like the real thing in a recording.  Demos that
should be able to pass as the real thing to anybody.  Can't sound much more
real than real.

BTW, since most people don't have a Bosendorfer sitting in their house, a
sampled Bosendorfer probably sounds a lot better than what they do have even
if the strings don't behave 100% like the real thing.

Tim Daugard writes:
>>Hard to believe, but I don't play violin. On my bass guitar, (especially
on my acoustic/electric) the three other strings can contribute quite a bit
of harmonics.<<

I would imagine an instrument with electric amplification is going to
amplify those a lot more.  Add to that the much longer strings on a bass
compared to a violin.  And also, what a violinist hears with his ear a few
inches away compared to what a listener hears from a mic placed a few feet
away can be different.

Robotboy writes:
>>(about 8 layers) True... but that many different recordings would
certainly add to a performance feeling more "live", if just that no two hits
in real life are ever truly exactly 100% the same.  I know that most would
say that it doesn't matter so much, but that's one of the things that I
personally feel does matter - even if nobody else can tell a difference, and
even if _I_ can't tell a difference, I still know that it wasn't quite as
interesting as it might have been.<<

Note that 8 velocity cross-switching levels doesn't mean there are only 8
levels of loudness.  Every note will still have 127 different loudness
values, and possibly brightness if a filter is also used.  Sampling works
particularly well for instruments like harpsichord or pipe organ which don't
have velocity responsive keyboards where velocity levels isn't even an
issue.

-Elhardt