home *** CD-ROM | disk | FTP | other *** search
- Path: nntp.teleport.com!sschaem
- From: sschaem@teleport.com (Stephan Schaem)
- Newsgroups: comp.sys.amiga.programmer
- Subject: Re: Amiga programmers in deep shit...
- Date: 10 Feb 1996 22:04:15 GMT
- Organization: Teleport - Portland's Public Access (503) 220-1016
- Message-ID: <4fj4ov$1lo@maureen.teleport.com>
- References: <4ccpsd$d8f@maureen.teleport.com> <Kevin_Phair.040b@bgi.internet-eireann.ie> <4cjimb$4im@maureen.teleport.com> <Kevin_Phair.040n@bgi.internet-eireann.ie> <4cto7l$t4s@maureen.teleport.com> <4ctuqh$1gj@serpens.rhein.de> <4d0dch$cup@maureen.teleport.com> <4d0lji$b3s@serpens.rhein.de> <4d37mb$sbo@maureen.teleport.com> <4ddg32$7h5@serpens.rhein.de> <4de1ok$ren@maureen.teleport.com> <4den0n$apt@serpens.rhein.de> <4dgiqq$9s5@maureen.teleport.com> <peterk.0kbh@combo.ganesha.com> <4f14i8$5ap@maureen.te
-
- NNTP-Posting-Host: kelly.teleport.com
- X-Newsreader: TIN [version 1.2 PL2]
-
- leport.com> <peterk.0kg0@combo.ganesha.com>
- Organization: Teleport - Portland's Public Access (503) 220-1016
- Distribution:
-
- Dr. Peter Kittel (peterk@combo.ganesha.com) wrote:
- : In article <4f14i8$5ap@maureen.teleport.com> sschaem@teleport.com (Stephan Schaem) writes:
- : >Dr. Peter Kittel (peterk@combo.ganesha.com) wrote:
- : >: In article <4dgiqq$9s5@maureen.teleport.com> sschaem@teleport.com (Stephan Schaem) writes:
- : >: >
- : >: > My point was... CBM didn't think about speed a bit when designing
- : >: > the grpahics.library.
- : >
- : >: You are rather fast with sweeping accusations, aren't you? Never
- : >: heard that one great games programmer and graphics wizard named
- : >: Chris Green worked there?
- : >
- : > Chris Green was not the one that design the graphics.library,
- : > I remember electronicly talking to him in 88 and he was not
- : > working for CBM yet, but the .library was already designed then.
-
- : And you never heard that especially graphics became faster over time?
- : You mean that happened by itself, or perhaps because some experts
- : optimized the code?
-
- after near a decade of optimization by expert its pretty far from
- beying fast... at least as efficient as it should be. But maybe
- those expert done the most they could... I wish more experts worked
- on the workbench code on 3.x!
-
- Just analize what happen when you open a wb window and scroll down
- into it (first if you move the mouse (HW sprite pointer) while the WB is
- sorting the sorting will take ~10% more time to complete, and that can be
- a long time even on a 25mhz 32bit cpu) then the display itself is very
- slow (caused by wb code, and or the fact that the blitting might not be
- done in parallel... busy wait loop before exiting gfx function)
- And the worst , the scrolling is soo slooow & buggy, jumping back and
- forth... I guess thats why the intuition expert 'removed' realtime
- WB window scroll....
- Anyway, I dont use the WB so I dont suffer from all this slow and buggy
- code.
-
- About gfx, it was intereesting to see that Chris Green require the user to
- have a tmpras when its not even used! In all my code I alloc a 4k tmpras
- (even so its NEVER USED) so I can 'fake' gfx function to return without
- doing a busy loop CPU waitblit. During the process of finding that out
- (This is not mentioned in ANY doc) I found that 3.x still use function
- C glue and push register on the stack and jump to the C function that
- get them from the stack...
-
- Stephan.
-