
GUS Musician's Digest       Mon, 1 Nov 93   007 MST      Volume 2: Issue   1  

Today's Topics:
                           Csound ftp site
                        ground loop eliminator
                              piano2.pat
                             Piano20.zip
                     Sesamet.669 - it's too big!
                             Some queries

Standard Info:
	- Meta-info about the GUS can be found at the end of the Digest.
	- Before you ask a question, please READ THE FAQ.

----------------------------------------------------------------------

Date: 31 Oct 1993   03:26:10 PST
From: chrisw <chrisw@leland.Stanford.EDU>
Subject: Csound ftp site

------------

Since the topic has come up a few times lately, Csound for the PC 
is available by ftp from ftp.bath.ac.uk in directory pub/jpff.

For those who don't know, Csound is a software synthesis program.
It takes in two text files: one containing instructions on the sounds to use
and one containg information on which notes to play and when to play them,
and puts out .wav files (in this PC version). It does not run in `real time'.

Beware: this is not a program for novices. Also, I'm not willing to answer
questions on it. Try it at your peril...

Chris.

------------------------------

Date: Mon, 01 Nov 1993 09:39:27 +1000
From: David Vu <ccdvu@cc.uq.oz.au>
Subject: ground loop eliminator

Hello GUSers,

There was a mention of this Ground Loop Eliminator that you 'd use
between a line out and an amp to eliminate ground noise.
My local Tandy doesn't know what it is and I doubt if other electronics
shops know about it.  So what is the device, how does it work, and
how to construct one - I've got my soldering iron ready :-)

-David-

------------------------------

Date: Sun, 31 Oct 93 9:36:04 EST
From: dmcintyr@muselab.ac.runet.edu
Subject: piano2.pat

You're right...  It sounds much more mellow than the original one you did.

I'm sorry we irritated you by being such critical bastards.  I was only
trying to be helpful.

BTW, I tried the .mid on a friend's JV-80 and it sounded just fine with
that piano patch.  We never could figure out exactly what was wrong with
the patch, and se we lengthened the notes by 25% to try to get rid of the
choppiness.  It didn't work.

I think acpiano sucks though, so I'm definitely going to keep these two
patches.
--Michael-- << dmcintyr@muselab.ac.runet.edu >>

------------------------------

Date: Sun, 31 Oct 93 16:00 EST
From: kildayb@erau.db.erau.edu (B. Kilday)
Subject: Piano20.zip

Has anyone successfully gotten piano2.0 (piano20.zip) to record in a sequencer?

I have tried to do this (unsuccessfully) in calkwalk 2.0 for win, midisoft recording session, and the three shareware midi sequencers i could find on epas.  Although I can use my midi keyboard for this, i find the ability to record quick ideas with my computer keyboard usefull when I am taking a break from some other work.  

I did set the driver up in the control panel btw.

------------------------------

Date: Sun, 31 Oct 1993 12:33:50 -0800 (PST)
From: mikebat@netcom.com (Mike Batchelor)
Subject: Sesamet.669 - it's too big!

I have a 669 file - Sesamet.669 (Sesame Street Rave).  Under any DOS
Module player I care to try - even PMP - it sound garbled, with the wrong
sounds at the wrong times, partial sounds in most places, and the
occasional complete sample, but at the wrong place.

None of the DOS players say anything is wrong.

I finally got around to building a Linux kernel with Ultrasound support
this weekend.  Everything went smoothly, and I was able to get both gmod
and Tracker compiled and operating right away.  I thought the problem with
Sesamet.669 was that there was just not enough memory under DOS to load
the song.  Under Linux, this would not be a consideration.  I tried
sesamet.669 with gmod, and it was garbled, same as with DOS.  I tried it
with Tracker, and for the first time, got an error message that explains
everything:  Sesamet.669's samples are too large to fit in the 1Mb GUS
memory!  The file itself is 1.2Mb, by the way.

I suppose it is a credit to the sound driver and/or Tracker authors that
one or both of them caught this problem, and reported it, rather than
winging it.  Now the question is, how can I reduce the size of the
samples?  Am I going to have to pull some samples out of the song, reduce
their size (lower freq, or...?), and put them back into the song?  Or is
there some way to cut back via software as the song loads?

The song is 1.2Mb in size, so I would guess that only a couple of samples
need to be shrunk to get it to fit (200K worth).

I was frankly surprised that a lowly module file would be beyond the
capability of the GUS.  Presumably, it plays fine on SB's and PAS's, but
takes a lot of CPU overhead to shove the samples at the DAC.  Is there no
way to operate the GUS in "CPU intensive" mode, playing samples out of
main system memory?

-- 
Mike Batchelor      |
mikebat@netcom.com  |                  This space for rent
mikebat@qdeck.com   |

------------------------------

Date: Sun, 31 Oct 1993 12:16:42 +0000
From: Clarke Brunt <CLARKE@lsl.co.uk>
Subject: Some queries

1) Linear volume response.

What exactly does the GUS do, and what
would we like it to do? Certainly what is observed is that
(on a scale of 0-127) the GUS is near inaudible below 64 and most
of the useful range is between 100 and 127, so some MIDs suffer
from near inaudible tracks.

Is it that the GUS plays waveforms with an AMPLITUDE proportional
to the number we tell it (0-127 in MIDI)?  POWER in the sound wave
is proportional to the square of the amplitude, so if the numbers
in the MIDI file represent power, then we need to take its square
root (and re-scale to 0-127) to yield amplitude for the GUS.

A further complication is that perceived sound level is said to
be proportional to the logarithm of the power - that is to say
that to produce apparently equal steps, you need to multiply the
power level by a constant. So are volume levels in MIDI files
supposed to represent power, or perceived volume level?

We often hear mentioned "When we get new Windows drivers..." to
correct for this. Is there any evidence that anyone is producing
any? The present GUS patches have levels set in their enveloeps
that are correct at present - so changes should only affect
MIDI volume (a combination of note velocity, the volume
controller, and the overall level set in Mixer) and not the
actual patch envelopes.

2) What do Windows users out there have set up as system noises?

I only connect my GUS to my amplifier when I am explicitly using
is, so if system beeps are played through it, I am more likely to
get nothing. If you have no wave drivers (WAV=... in SYSTEM.INI)
then you get a beep on the PC speaker. You can set up the
(in)famous PC speaker driver (available by ftp etc.) as default
wave device, changing the GUS to WAV1=..., so system sounds are
played on the PC speaker, but you can still (if you are lucky)
select the GUS for wave output in other apps, but as anyone who
has tried it will know, the PC speaker driver is not very
satisfactory (the machine locks up until the noise finishes
playing). Is there a way of getting system noises to just beep
the speaker (as if no wave drivers were present) even though
the GUS wave driver is installed.

Yesterday, the GUS Mixer app wouldn't start up for me - I found
that I had lost the AUX=.../ultrasnd.drv line from SYSTEM.INI,
so if anyone else gets this problem, that's probably what it is.

3) Patch envelopes.

Following the recent discussion of piano patches, I had another
play with PATCH.EXE. Trying to load Chris's piano.pat into it
crashes my machine with an EMM386 message (maybe changing EMM386
parameters would fix that - that is not my query). Can anyone
out there set out clearly how envelopes work? At the hardware
level, I believe that the GUS is just given a start volume,
a target volume, and a step size - it changes the volume by
the step until the target is reached, and the interrupts to ask
what to do now. The step is not a simple number - some of the
bits are scale factors, to enable attacks and decays on scales from
milliseconds to several seconds. A patch has 6 stages, with a
volume and a step for each. Exactly how to play these is controlled
by (at least?) two bits, one of which means sustain (hold the note
at the 3rd stage of the envelope as long as the note is still on).
The other bit controls what happens when the note is released
and this is where I am less clear what is going on. As someone
said about piano.pat, if you hold a note for a long time, so that
it has nearly died away, and then release it, it gets louder
again before ending. This is because it is playing the final
stages of the envelope, which have a higher level than that to
which the note had already decayed.

So... Has anyone got a proper explanation of what all the control
bits in a patch do? Was there ever a further installment of the
tutorial on this in patchkit.arj?

------------------------------

End of GUS Musician's Digest V2 #1
**********************************

To post to tomorrow's digest:                        <gus-music@dsd.es.com>
To (un)subscribe or get help:                <gus-music-request@dsd.es.com>
To contact a human (last resort):              <gus-music-owner@dsd.es.com>

FTP sites:                archive.epas.utoronto.ca       pub/pc/ultrasound
                          wuarchive.wustl.edu     systems/msdos/ultrasound
                          archive.orst.edu             pub/packages/gravis
FTP mail server:          mail-server@nike.rz.uni-konstanz.de

Hints:
      - Get the FAQ from the FTP sites or the request server.
      - Mail to <gus-music-request@dsd.es.com> for info about other
	GUS related mailing lists (general use, programmers, etc.).


