Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!utgpu!water!watnot!watmath!clyde!rutgers!husc6!husc7!hadeishi
From: hadeishi@husc7.UUCP
Newsgroups: comp.sys.mac
Subject: Blitter and graphics performance
Message-ID: <1353@husc6.UUCP>
Date: Mon, 9-Mar-87 12:10:31 EST
Article-I.D.: husc6.1353
Posted: Mon Mar  9 12:10:31 1987
Date-Received: Tue, 10-Mar-87 04:02:10 EST
References: <422@yabbie.rmit.oz> <425@yabbie.rmit.oz> <5083@mit-eddie.MIT.EDU>
Sender: news@husc6.UUCP
Reply-To: hadeishi@husc7.UUCP (Mitsuharu Hadeishi)
Organization: Harvard Univ. Science Center
Lines: 22


	I just realized that the Mac II does not have a blitter.
Can individual video cards have blitters?  It seems to be that in
order for >8-plane video to work you DEFINITELY need a blitter,
perhaps several blitters.  Color Suns have one blitter per bit plane.
A 68020 is a pretty good blitter in itself (with cache memory and all)
but it doesn't quite hack it when compared to an 020/blitter coprocessor
combination (since the blitter can use the unused bus cycles for memory
DMA.)  Or perhaps the blitter can somehow manage its own information on
the video board without causing bus contention with the CPU?  Anyway,
the Mac II seems to draw graphics a tad slower than blitter-equipped
machines such as the Sun or the Amiga.  With more than 8 bit planes
I think QuickDraw would have to be modified to use a blitter or
risk incredibly slow performance on these mythical 48-bit plane devices
(or even 16-bit plane devices).

	Also a blitter can be used to do high-speed area fills,
line drawing, etc., which still seems a little slow on a full-blown
8-bit plane Mac II display.  This of course is nitpicking, but for
CAD/CAM/CAE this might be an issue.

			-Mitsu