home *** CD-ROM | disk | FTP | other *** search
- Path: hercules.zrz.TU-Berlin.DE!rawneiha
- From: rawneiha@hercules.zrz.TU-Berlin.DE (Philipp Boerker)
- Newsgroups: comp.sys.amiga.programmer
- Subject: Re: TMapping again!
- Date: 22 Jan 1996 13:38:34 GMT
- Organization: Technical University of Berlin, Germany
- Message-ID: <4e040q$cif@news.cs.tu-berlin.de>
- References: <4d0ou6$835@astfgl.idb.hist.no> <Z31Wx*zA0@mkmk.in-chemnitz.de> <4d42di$9e9@maureen.teleport.com> <4d5lvi$emc@brachio.zrz.TU-Berlin.DE> <4d6v0t$3dt@maureen.teleport.com> <4djj0t$4ni@sunsystem5.informatik.tu-muenchen.de> <4dlo7d$h39@news.cs.tu-
- NNTP-Posting-Host: hercules.zrz.tu-berlin.de
- Mime-Version: 1.0
- Content-Type: text/plain; charset=iso-8859-1
- Content-Transfer-Encoding: 8bit
-
- fischerj@informatik.tu-muenchen.de (Juergen "Rally" Fischer) writes:
-
-
- >|> >|> : Yes, I agree. Sharon, one of the coders of matrix, is actually a
- >|> >|> : PC coder but his polygon engine on a 33 MHz 486 is slower than
- >|> >|> : our (Sharon, Grond, Skyphos all/matrix) engine on a 25 MHz 030!
- >|> >|> : That's the power of a large register set!
- >|>
- >|> >well is you polyengine still faster using 8 planes ? ;)
- >|>
- >|> I was refering to 8bit.
- >|>
- >|> >he got a non-local-bus vga ? ;)
- >|>
- >|> Local bus...
-
- >ooh =:o
-
- >well hm... polygon is just writing 32bit values to vram.
-
- >The 030 will do this at on AGA 7MB/sec. Or do you refer to
- >A3000+gfx-card ? ;)
-
- No, A4000-30-25
- Copying to VRAM/CHIPMEM is not the main speed problem on both platforms...
-
- >how much fps ? is it mem overwritten multiple times (fastmem render ?).
- >then fastmem speeds cares.
-
- Fastmem is faster on PC as well.
-
- >are you sure the PC version is as optmized?
-
- There are two reasons why the engine is faster:
- We have 15 registers during the mapping while PC has
- to load from RAM. And hey, it was Sharon's first engine...
- A lot of algorithmic optimizations have been added as well.
-
- >|> >my friend always argues that you could see the cache as big register
- >|> >set on intel
- >|>
- >|> Then ask him how he does a
- >|> "shift left with mask out and insert" rdest,rsource1,rsource2
- >|> within 1 cycle from datacache...
-
- >can 680x0 do that ?
-
- No, 68k is out of the race anyway. The example above was refering to
- PPC.
-
- > x86 operations with datacache as destination need
- >33 cycles AFAIK. add.l D0,var or in x86 add eax,var
-
- No! Not more than 3 c even on 486!
-
-
- Greets,
- Phil.
- ----------------------------
- grond/matrix
- rawneiha@sp.zrz.tu-berlin.de
- ----------------------------
-
-