Y-ellow All and thanks for the kind responses.
Out of interest. Yes the name does attract the ladies. Right up
until they find out I'm actually a geek. It's all down hill from there.
I was deliberately brief. Or as brief as possible with my first message. So
I'll try to elaborate a little without getting too wordy. Which I am
inclined to do.
Apart from the many crazy things I do, I'm something of an electronics
engineer. I know very little about coding except for machine code on small
micros. In fact, truth be known, I can barely compile a kernel to save
myself. Sad but true. And whilst my lab is currently in a state of
disarray, I am mainly a hardware person. Apart from a little bit of machine
code on small micros, I couldn't cobble together "hello world" in a pink fit.
The problem is that slackware has no native package loader. It sometimes
works with RPMs and DEBs but I will probably require a lot of hand-holding
to get this to work. But I'm more familiar with slackware than anything
else. I'm the kind of animal that dives in as root and damn the
consequences. I never did get along with ubuntu for that reason. :(
I have a Yamaha A4000 sampler but nothing will talk to it's proprietary
file system. There use to be some software for BOSH (AKA Bastard Operating
System from Hell. AKA windows. ) called A3Kdisky. This is apparently the
only thing out there which will understand the FS. However it's totally
made of unobtainium these days. I've been searching for it for months. The
A4K would otherwise be perfect. Nice filters. 3 FX chains. Solid hardware.
Perhaps a little too solid as it's big and weighs a ton. Best of all, I've
got one right here right now. And it's fully tricked out. But if I can't
talk to it, it's really just a boat anchor.
Some ten years ago, before my musical career took a detour into 3D
graphics, I use to use a lowly Creati-Flabs AWE32 sound card as a sampler.
This was surprisingly serviceable. It was limited but I still managed to
record albums, master and play live with it. Scary as that thought might
seem. I liked the work-flow in fact. There were lots of trade-offs but
apart from anything else, it's ISA and can't be pressed into service these
It seems none of the good ol' wave-table sound cards are of any use these
days. Even the old SAM7 based Terratech cards. That was then and this is now.
So this would be the basic spec I'd be looking for...
*32 voices or more polyphonic.
*16 part polytimbral or greater
*Multi-mode filters per voice.
*Stereo sampling (If possible) Although 2 mono layers would be fine.
*Decent envelopes. (IE: something better than an ADSR) Filter and amplitude
*Syncable, multi-wave-form LFOs
*3 effects chains. (Rev/Chor/flange/del etc)
*Should have low latency so it can be played live. And an S/PDIF out.
Being a computing platform it may as well have some wave editing software
on-board as well. But otherwise be a turn-key system. With as much
non-essential stuff stripped out of the OS as possible. I'll still need
networking and SAMBA though. And if it could do MIDI over Ether, it would
be the ant's pants. Otherwise it would have to talk to a generic, (Read
cheap Chinese made) USB MIDI interface.
Hardware wise I'd like to keep it as small as possible. In fact if I could
squeeze the whole thing into a 1U 19" rack boxen I'd be over the moon.
Regardless of which it has to live in a 19" rack for gigging. I was
thinking that perhaps I could find an old lap top with a broken screen for
nix. Tear it apart and, as a friend of mine once put it. "Modify it beyond
manufacturer's specification." This of course means it would be using a
5400RPM disk drive. Although perhaps with a little cable jockying, I could
run a 3.5" drive in the box. It would be nice if I could get it to boot
from FLASH but just as long as it's relatively physically quiet.
On the other hand, I'm not sure it be possible to get the latency down to
an acceptable (Read real-time) rate with on-board sound. Unless there's a
trick I can do under linux? Most Mo-bo sound systems have a buffer which
can't easily be tweaked. Therefore it may be necessary to use a PCI card of
some kind. And I need to figure this out before I even start thinking about
nailing a test-bed to a piece of wood.
Which brings me to another, perhaps stupid question. What about the notion
of getting something like a Soundblaster Live or Audigy card and using
that? When I say hardware based, I mean that the wave-table engine is based
in hardware rather than software. So that the software need only control
the hardware. This is a far more robust solution. But it would still have
to meet the above specs (Or something close) and I would still need to be
able to edit/tweak and play it as a synth. It also means that the CPU
wouldn't have to be as fast. Probably even a PIII would suffice. (Were it
not for the GUI.) Has anyone done this kind of thing to any great effect?
I've got a few VIA based Mo-bos. I don't know how fast the smaller ones are
but they're tiny. I have been thinking for a long time how nice it would be
to build a linux based synth with an actual front panel. This is a long
term goal but it would be rather cool. Design the front panel fairly
generically and shove the board out on the net. Then people could chose the
synth system they want to build, whack the front panel onto a Mo-bo and
build a proper rack synth. Only a synth that can be upgraded and tweaked.
For now though, I'm rather desperate for a sampler I can get together and
gig with. So I'm open to suggestions.
Thanks in advance. Very much appreciated.
Be absolutely icebox.
________ _ _
/ ____|| | // Disgusting -> http://all-electric.com
/ /| |____ | |// The trusted _-_|\
/ __ ____|| < name in / \
/ / | |____ | \ \ non-sequitur \_.-*_/
/ / |______|| |\ \ entertainment v
All Electric Kitchen Music-> http://all-electric.com/music_downloads.html
Linux-audio-user mailing list