On Mon, 13 May 2013, James Stone wrote:
Just a quick poll of opinion - I'm thinking of buying a cheap rompler
- Roland JV-1080 or Proteus 2000. I kind of think that the things
these offer should be possible with software, but beyond getting some
soundfonts, I'm not really sure if this is true. Problems I see with
the purely computer based approach is all the samples I have found so
far (admittedly only looking at what's available free) is not really
anywhere near what is available in these kind of hardware modules.
Sure the size of samples etc. available now is much larger, but the
musicality and tweakability/ synthesis options are not so much there
(although possible I guess).
So, am I wrong?
Well, my studio is mostly embracing hardware too, but for totally
different reasons. I've found that if you look, it's amazing what's out
there as SoundFonts. I've found all the kinds of things I used to look
for as libraries for rack samplers and been disappointed, either because
it didn't exist or the quality was poor. (For example, I've bought a
SoundFont of a Persian santur string instrument that's just amazing...
I like sample libraries of exotic string instruments because they can't
be faked by synthesizers and you'd likely never learn to play the real
thing ever.) So SoundFonts have actually been exciting for me and not a
disappointment at all.
Here's where I do still embrace hardware though:
* In the sampling arena, the thing that any hardware sampler still does
better than LinuxSampler or FluidSynth is having good resonant filters
that act like resonant filters. Originally, the notion behind inventing
the sampler was that with traditional analog synthesizers, you might
have an architecture that looks (simplified) like this:
Oscillators -> Filters -> ADSR (envelopes) -> AUDIO OUT
The filters provided most of the magic that made the synthesizer sound
cool, but the oscillators at the start of the chain were basically just
tone generators, that made simple waveforms like triangle waves,
sawtooth waves, square waves, etc. Filters can do magic with that, but
eventually you get to the point where it's just boring. What if the
waveform going into the filters could be literally anything you wanted
it to be? What if it could be acoustic sounds from the real world, and
you could use those sounds as if they were a signal from an oscillator?
And then feed *that* into your filters...wow! A super-synth!
And the old hardware samplers gave us that, but it wasn't long before
people forgot that was what we originally wanted them for. It soon
became sufficient for a sampler to simply play back sounds without doing
anything terribly interesting to the waves, filter-wise or otherwise.
Linux sampler programs seem to be in this state, where the
implementation of real resonant filters that behave like what you hear
in hardware Midi equipment and can be made an integral part of a
performance/patch just isn't there (the filters sound more like mere EQ
cuts). They play back sounds, and they may even have ADSR/envelope
control, but they're not quite what I'd call synthesizers that use
samples for oscillator tones. If you have a real sampler around, you'll
still have access to this world. If it's from the 80's, the filters
will likely even be real analog.
* In the plugin arena (effects, EQ's, compressors, etc.), let's face it:
Though we obviously can't complain about software that's being written
for us for free or with voluntary contributions, it still must be said,
there aren't exactly a lot of LADSPA/LV2 plugins out there you could
sell in a commercial retail bundle for over $500 USD like some of the
stuff that's out there as VST's for Mac/Windows. Frankly, the quality
just isn't there. Some of these EQ's actually make the audio sound
worse than it did untreated nomatter what setting you use. The
compressors are flaky, and that's without even getting into comparing
with what you can get from a real one built with real tubes. Let's not
even get into the situation with reverb. The reason they're able to
charge so much for plugins in the Mac/Windows world is because those
plugins are often just as good as your rack effects, but easier to use
because they run inside your DAW. They have the same value as some
hardware gear. Until Linux plugins can say the same, you really have
two options: Run Windows VST's under Linux DAW's (which can be very
tricky, and an additional technical odyssey to go distracting you from
your actual music), or...just do what we used to do in the old
days...get some rack gear!
* Personality... I find that even when plugins and virtual instruments
have been done well, they still seem like vampires without a soul.
There's nothing like working on real instruments that you have to mic,
record, and treat with respect. It sounds better, it feels better, and
in the end, you feel a lot less like you tried to paint the Mona Lisa
with a mouse.
[end of rant...some things I feel strongly about I guess...]
--
+ Brent A. Busby + "We've all heard that a million monkeys
+ Sr. UNIX Systems Admin + banging on a million typewriters will
+ University of Chicago + eventually reproduce the entire works of
+ James Franck Institute + Shakespeare. Now, thanks to the Internet,
+ Materials Research Ctr + we know this is not true." -Robert Wilensky
_______________________________________________
Linux-audio-user mailing list
Linux-audio-user@xxxxxxxxxxxxxxxxxxxx
http://lists.linuxaudio.org/listinfo/linux-audio-user
