home *** CD-ROM | disk | FTP | other *** search
- Path: comma.rhein.de!serpens!not-for-mail
- From: mlelstv@serpens.rhein.de (Michael van Elst)
- Newsgroups: comp.sys.amiga.programmer
- Subject: Re: Demo/game to OS friendly part II
- Date: 1 Feb 1996 02:30:46 +0100
- Organization: dis-
- Message-ID: <4ep546$hpr@serpens.rhein.de>
- References: <john.hendrikx.47u5@grafix.xs4all.nl> <PETERM.96Jan29164036@tui.maths.irl.cri.nz> <4ek6bo$84n@xmission.xmission.com> <4ekl5d$51@serpens.rhein.de> <PETERM.96Feb1123338@tui.maths.irl.cri.nz>
- NNTP-Posting-Host: serpens.rhein.de
-
- peterm@maths.grace.cri.nz (Peter McGavin) writes:
-
- >Current high-level gfx calls have a queue limit of 1 (or less than 1).
-
- Right. But that's as good as N.
-
- >BTW: What is the correct way of waiting for an async gfx operation to
- >complete?
-
- There is no correct way :-(
-
- The best approximation is WaitBlit() and that is used everywhere. Using
- WaitBlit() for synchronization is a BIG obstacle on the way towards RTG.
-
- >One could use OwnBlitter()/WaitBlit()/DisownBlitter(), but
- >it's not always clear whether the blitter is used.
-
- WaitBlit() works without owning the blitter.
-
- >I agree. However task switches are not a problem for large gfx
- >operations like copying an entire bitmap.
-
- Right. I am thinking of a flexible solution that queues blit nodes
- for large blits.
-
- >This is akin to a large IO
- >request in DOS. Perhaps the system should have a mechanism for
- >batching trivial gfx operations, something like buffered IO in DOS.
-
- All you need is a proper model for synchronization. This should be
- done on either a bitmap or a rastport basis. Graphic operation can then
- be completely asynchronous as the driver decides.
-
- Regards,
- --
- Michael van Elst
-
- Internet: mlelstv@serpens.rhein.de
- "A potential Snark may lurk in every tree."
-