On Monday 15 July 2002 4:43 am, Rommel Palma Feria wrote:
> If you cannot afford to get a new computer, get VMWare instead.

VMWare would be cool, but I hear that it's resource-intensive. I fear that the 
VMWare layer over Linux would greatly slow things down.

Does VMWare support direct access to sound card drivers and hardware? Direct 
hardware access is necessary in improving latency and minimizing MIDI timing 
delays. A few milliseconds wasted makes the difference between a clean, tight 
grooves and a sloppy one. Everyone hates listening to dance grooves that 
skip, pauses, and jumps, thanks to MIDI timing delays. 

I do a lot of techno (Nine-Inch Nails, Depeche Mode), electropop (Pet Shop 
Boys, Fat Boy Slim), and sometimes orchestral (Bach, Mozart, Haydn, 
Beethoven, Steven Roach, Herbie Hancock, Yanni) MIDI sequencing with my 
soundcard. I enjoy layering lots of fat patches (sound banks, to be 
specific). On the average, I always max out the 32-polyphonic limit of my 
sound card. Taking advantage of my Pentium 733 processing power, I do 
tortuous audio to MIDI synchronization in Cakewalk, inserting long audio 
clips inside MIDI sequences. Pretty neat, but the timing always leaves me 
wanting for more. 

I would love to do a song entirely in audio sequenced with Cakewalk, a single 
MIDI track serving as a metronome timing device (like the persistent salt 
shaker in the Cindy Lauper track "Time after time"). Four dedicated, stereo 
(that's a scary two simultaneous wave files per track at 44.1 khz CD-sampling 
rate). 1 audio track for voice, 2 audio tracks for lead and backup guitars, 1 
audio tracks for drums. No audio tracks for keyboards (thank God it's MIDI). 

For now, this is what I currently do. Do my MIDI sequencing in Windows, fixing 
tight grooves here and there to eliminate MIDI latency problems. Sometimes I 
rigidly sync notes to the nearest 32th or 64th note to avoid MIDI "hiccups" 
(I don't like rigid syncs, sounds very mechanical). When I'm satisfied with 
the mix, I convert the whole sequence to wave file audio format. Rebooting 
into Linux later, I crunch this wave file to mp3 or ogg vorbis, then burn my 
tracks to CD with X-CD-roast. 

I've tried to encode wave files to mp3 before with BeOS. BeOS was a very fast 
OS. It could simultaneously encode 10 mp3 from wave files, and I still have 
plenty of cpu cycles left to do word-processing. Too bad BeOS has gone belly 
up. Their audio hacks are some of the best OS hacks on the planet.

I am salivating at the possibilities of audio processing with Linux. Say you 
run a MIDI sequencer in the foreground. You can simulatenously plug in your 
trusty Les Paul into the audio input of the card, route it to a DSP (digital 
signal processing) program running in the background and do realtime effects 
such as flanging, wah wah, distortion. Route the processed signal back to the 
MIDI sequencer in the foreground. Pretty weird but cool. :-)

Someday I'll do my MIDI grooves in Linux. Since Linux is a great multitasking 
machine, I am confident it will greatly reduce timing problems. If not, then 
there's always a Korg Trinity Workstation (P 60,000) that I am forever unable 
to purchase (sigh). Better yet, stop doing danze traks and shift to New Age, 
where cool ethereal synth sounds are deliberately slow.

mikol
_
Philippine Linux Users Group. Web site and archives at http://plug.linux.org.ph
To leave: send "unsubscribe" in the body to [EMAIL PROTECTED]

To subscribe to the Linux Newbies' List: send "subscribe" in the body to 
[EMAIL PROTECTED]

Reply via email to