cvs.gedasymbols.org/archives/browse.cgi   search  
Mail Archives: djgpp/1999/11/01/06:16:47

From: jdominy AT mars DOT cs DOT unp DOT ac DOT za (James Dominy)
Subject: sound blaster code
Newsgroups: comp.os.msdos.djgpp
Organization: CS and IS, Univ of Natal, PMB
Keywords: Sound Blaster, programming
X-Newsreader: TIN [version 1.2 PL2]
Message-ID: <941451872.692273@mars.cs.unp.ac.za>
Cache-Post-Path: mars.cs.unp.ac.za!jdominy AT mars DOT cs DOT unp DOT ac DOT za
X-Cache: nntpcache 2.3.2.1 (see http://www.nntpcache.org/)
Cache-Post-Path: Raven!news AT mars DOT cs DOT unp DOT ac DOT za
X-Cache: nntpcache 2.3.3 (see http://www.nntpcache.org/)
Cache-Post-Path: banana.uni.net.za!unknown AT raven DOT und DOT ac DOT za
X-Cache: nntpcache 2.3.3 (see http://www.nntpcache.org/)
Date: 01 Nov 1999 10:24:34 GMT
NNTP-Posting-Host: 196.10.251.2
X-Trace: wits.uni.net.za 941451874 253 196.10.251.2
Lines: 34
To: djgpp AT delorie DOT com
DJ-Gateway: from newsgroup comp.os.msdos.djgpp
Reply-To: djgpp AT delorie DOT com

Hi there,

I relise this might be a bit off topic, but the sound blaster newsgroup is
completely unhelpful, probably because my news feed is screwed.

I'm trying to program my sound blaster, and have had some success. I have
set up an ISR to handle DMA interrupts generated by the card/dma controller,
in auto-initialsed mode. The ISR performs the following steps

1 - copies the pre-mixed buffer into Real Memory Space, for the soundblaster
to access

2 - mixes samples from an array of pointers to samples (i.e. array of
unsigned chars) in blocks of 4K, puting the mixed result into the buffer.

3 - acknowledges the interrupt, and exits

All of the data used is locked, as well as the functions. The DMA will
continue playing a buffer filled with 0's, when there are no samples in the
array.

When the last sample is completed in the middle of a block, the buffer is
completed with 0's. Samples can only start at the beginning of a block.

My problem is that I get static bursts whenever I start playing a sample,
when none were playing previously, and whenever all samples finish playing.

My theory, based on the symptoms, was that padding the buffer with 0's causes
static, but I tried with other values, to no avail. I am now at my wits end
and cannot find a cause, much less a cure for the static!

Any help appreciated
Thanx
James

- Raw text -


  webmaster     delorie software   privacy  
  Copyright © 2019   by DJ Delorie     Updated Jul 2019