Hard Drive Recording Kit, Pre-release 1/27/92

			     Written by Dan Hollis
		    Copyright (C) 1992 MicroImages Software


Ok, I finally decided to send out a pre-release version of my
Hard Drive recording program. I haven't done anything to it for
the past few months, but with some prodding or new ideas I might.

Note: Yes, the programs are only 1k in size!!
Believe me, the files they create really make up for it!! They
also only require about 64k of RAM to run.

Ok, some instructions:

What these two programs do is let you use your sampling cartridge
and hard drive for recording sound. Using the programs is pretty
simple. To record, double-click HDRECORD.PRG, and follow the
prompts. No flashy oscilloscopes or spectrum analyzers here :^)
Recording will stop when either ESC is pressed, or the drive
runs out of space.
To play, double-click HDPLAY.PRG, and sit back and enjoy. Output
is through the ST-REPLAY output, not the monitor, BTW.

Requirements: ST-REPLAY 4, a hard drive with plenty of free space.

Limitations: You're welcome to try it on floppy, lemme know what
happens ;^) Sampling speed is hard-wired at 20khz (for now). The
filename for the sample is stuck at "sample.spl" (for now).
For best results, stick these programs in their own folder
and run them from there.
Make sure you have plenty of free disk space! A 3 minute sample at
20khz will take 3,600,000 bytes!
If you have a slow hard drive, it might be wise to optimize it, i.e.
make all the files contiguous. I use Hard Drive Sentry to do this, it
works well enough - but there are many such programs that do just as
well. If your hard drive can't keep up, you will not notice any
problems on recording, just on playback, where the music will sound
like it's skipping around.

Un-limitations: No longer is your maximum sample length limited by
how much RAM you have, but rather by how much diskspace you have.
This was the general idea, since I was getting pissed with sampling
programs limiting the sample length to available RAM. My favorite
songs just wouldn't fit into RAM. Steve Vai's "ballerina" is one
example. They sure fit now!! :^)

Technical notes: The technique used is "double buffering", i.e.
while sampling takes place in one half of the memory buffer,
the other half is being saved. Then, it switches over to the
other half, and vice-versa. Luckily the Atari ST has DMA for
hard drive I/O, which makes my work a LOT simpler =:^)
STe's should be much better with this technique, I should
even be able to write a multitasking version (i.e. Desk Accessory)
which runs in the background. That is, assuming I had an STe.
Also, stock STs are limited to a 20khz sample rate, this is due to
the speed of the CPU. Higher speeds may be possible on processor
accelerators (i.e. Adspeed and the SST) and the Atari TT, but I
don't have any of these, so until I do, it won't go any higher.

Before I continue any work on this program, I'd like to find
out what kind of interest there is. I'd also like some suggestions.

So, I can be reached at:

Fnet	 : Daniel Hollis @ Node 181 (Bitbucket) and Node 3 (Atari-OH!)
InterNET : dhollis@bitsink.uucp or bitsink!dhollis@moe.rice.edu
	   dhollis@nuchat.sccsi.com (As a last resort)
Snail	 : P.O. Box 580448/Houston, TX 77258/U.S.A.