Xref: news2.mv.net comp.os.msdos.djgpp:8147
From: jb3207 AT eclipse DOT co DOT uk (Jason Barstow)
Newsgroups: comp.os.msdos.djgpp
Subject: Re: Allegro and Win95 time-slicing problem ?
Date: Sun, 01 Sep 1996 09:56:50 GMT
Organization: Jabber Interactive
Lines: 20
Message-ID: <32295d89.0@deimos.eclipse.co.uk>
References: <1996Aug29 DOT 185755 AT uctvms DOT uct DOT ac DOT za> <01bb95d8$5e3ab180$ba27379d AT michmarc2>
Reply-To: jb3207 AT eclipse DOT co DOT uk
NNTP-Posting-Host: deimos.eclipse.co.uk
To: djgpp AT delorie DOT com
DJ-Gateway: from newsgroup comp.os.msdos.djgpp

"Mike Marcelais" <michmarc AT microsoft DOT com> wrote:

>My guess?  Windows cannot maintain an offscreen SVGA video area, so when a
>request comes in to Windows to draw onto an SVGA screen that someone program
>owns that isn't in the foreground, Windows pauses until the screen is in the
>foreground.  Hence, your simulation runs until it tries to draw on the
>graphics screen which causes the program to halt until it returns to the
>foreground.

>solution: Draw to an offscreen buffer and not the screen.

But the same thing will occur when (if) you try to blit the offscreen buffer
to the video buffer.  It would be usefull if you could dynamically test for
the application being a Win95 background task and in this case *not* blit the
offscreen buffer.

Does anyone know if there exists an easy and reliable way of doing this?

Jason.