NetNews Usenet Archive 1992 #31/NN_1992_31.iso/spool/comp/sys/ibm/pc/soundcar/5851

Text File | 1992-12-29 | 3.5 KB | 64 lines

Newsgroups: comp.sys.ibm.pc.soundcard
Path: sparky!uunet!newsflash.concordia.ca!mizar.cc.umanitoba.ca!janicek
From: [e-mail] (Jana Janicek)
Subject: Re: GUS Programming: The Repost III
Message-ID: <[e-mail]>
Sender: [e-mail]
Nntp-Posting-Host: vogt.cc.umanitoba.ca
Organization: University of Manitoba, Winnipeg, Canada
References: <[e-mail]> <[e-mail]> <[e-mail]> <[e-mail]>
Date: Wed, 30 Dec 1992 02:05:52 GMT
Lines: 51

    Okay, I am pretty sure that patches are not limmited to 64K as I
guessed earlier. I am in fact having the reverse as a problem in my patch
routines. I mean, I cannot seem to make my patches cut off where they
should. They all seem to cycle through the full meg of on board ram. But I
will keep on hacking untill I solve these problems.

Some more observations (GOOD and BAD):

**Remember that my routines are not yet perfected and many of the BAD
stuff is probably based on this fact.

First, the bad. We have heard before some complaints of a slight popping
noice when playing certain MIDI files with the GUS. Well, I thought I
would further investigate this. I listened to all sorts of GUS stuff,
(patch based). MIDIS do indeed (some more than others) seem to have a
popping problem. It is not very noticible at first, but the more you
listen, the more annoying it becomes. I found this same popping in Star
Control II, and again in the Modulatr (which is a BETA I should remind
people). The final place I found popping was in my own patch test
programs. So, I did some experiments.  The patch I am using is the
spook.snd file in the ULTRASND/SOUNDS directory. I played this patch using
USS8, than I saved it from the GUS DRAM with my own routine to disk, and
than the fun began. It seems that whenever start_voice_playing is called,
the first thing it does is generate a pop. I was not 100% certain of this,
so I created a loop, and played the same patch through 15 voices with a
30ms delay between each one. The result was an awsome sound with a
distinct loud buzzzzzzz right at the begining. In other words, alot of
pops being played one after the other. This made my heart sink, could
GRAVIS have overlooked such a flaw in their hardware? I checked the patch
itself to see if there was any noise at the begining, I found a fairly
constant stream of values close to zero. This should not creat such a
distinct buzzing pop noise. I really really really hope this is a bad
patch or there is a fault in my routines that is doing this, but as I
listen to more and more GUS native stuff, I still hear the little pops. If
this is hardware, I hope the next revision addresses this problem.
Remember, I am merely making an observation, I don't want to start a flame
war, I hope I am wrong, but I cannot ignore my ears or those pops any longer.

Now, some good observations. I found that you can do some really cool
stuff with the GUS, one of my favorite things is creating an
echo/reverb/metallic effect by playing the same patch on several voices I
use about 6 to 10 depending on how pronounced I want the effect, with a
very small delay between each one (about 10..20 ms).  This effect would
wave some really neat uses in games where they could use the same patch on
board the card to generate several different sounds. A really neat
effect, must be heard to be believed.
My next project is the Patchinator (cant think of a better name yet)
which will do precisly that mentioned above, using the special abilities
of the GUS to do cool things to samples.

-Later