home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.sys.amiga.programmer
- Path: sparky!uunet!pipex!warwick!coventry!csg019
- From: csg019@cck.coventry.ac.uk (-~=Zaphod=~-)
- Subject: Blitter vs 680x0
- Message-ID: <BxM2CM.9E5@cck.coventry.ac.uk>
- Sender: news@cck.coventry.ac.uk (news user)
- Nntp-Posting-Host: cc_sysk
- Organization: Coventry University
- Date: Thu, 12 Nov 1992 16:12:20 GMT
- Lines: 26
-
- All this OS vs Hardware stuff has got me thinking about the actual speed of
- the blitter. I read somewhere that it runs at 28mz. This sounds all well and
- good for speedy graphics, but what with the 68030 and 68040 picking up real
- speed, how much long is it going to be, before using the blitter is slower
- than using the CPU?
-
- Obviously with lots of small blits, the CPU will be faster, due to the
- need to set the blitter up, but at what point does the blitter become
- faster?
-
- Also i read in one of the Abacus book (great book, full of non-relocatable
- code and hardware pokes ;) ) that it may be faster to draw lines with the
- processor than to draw them with the blitter.
-
- I suppose after a certain point, the blitters DMA would make it faster than
- the CPU. How does the new blitter fair (4 x bandwidth?) ?
-
- Does the speed of the new blitter affect old programs not taking them into
- account (ie, downwardly compatable).
-
-
- --
- ******************************************************************************
- * /// O O A member of S.H.I.T. (Super High Intelligence Team) /// *
- * /// u Fight, defeat and kill organized laming. /// *
- ****\\\X//****\-/********csg019@uk.ac.cov.cck ok ?**************** \\\X//*****
-