home *** CD-ROM | disk | FTP | other *** search
- Path: news.eunet.fi!fipnet!kone!jsaarinen
- Newsgroups: comp.sys.amiga.programmer
- X-NewsReader: IntuiNews 1.2b (31.7.94)
- References: <4cj57f$7qe@irz210.inf.tu-dresden.de>
- From: "Jyrki Saarinen" <jsaarinen@kone.fipnet.fi>
- Date: Sat, 6 Jan 96 12:45:09 UT
- Comments: Illegal date header - new date added by quicknews
- X-Original-Date: Sat, 06 Jan 96 14:24:31
- MIME-Version: 1.0
- Content-Type: text/plain; charset=iso-8859-1
- Content-Transfer-Encoding: binary
- Distribution: world
- Subject: Re: C2P
- Message-ID: <38231876@kone.fipnet.fi>
-
-
- > - what are the fastest ones (ideas, theory, sources) ?
-
- The idea situation is to do as many c2p passes with the CPU,
- depending on the CPU speed, but in such a way that the
- chunky2planar CPU process is as fast as normal copying
- from fast ram to chip ram. So the c2p passes are "free"
- between the chip writes.
- The rest passes are done with the blitter using QBlit().
-
- Starting from the 040, blitter is not needed since
- 040 is able to done all c2p passes completely free!
-
- > - what the hell is a scrambled (maybe spelled wrong) buffer ?
- > (I have heard this sometimes in this group)
- > and what are his adventages/disadvantages ?
-
- Pass1 is skipped by writing vertical colums in non-linear
- order. Only good for vertical drawing, like Wolf3D.
-
- Pass1 only rearranges bytes.
-
- > Are there really fast 6 bitplane c2p-converter ?
- > My actual approachs are (50Mhz 030, OS-conform (except screens of
- > gfxcards), 320x256)
- > 8bpls : 45ms
- > 6bpls : 43ms
- > 4bpls : 30ms
- > These c2p-converter(s) use LONG-writes to CHIPmem.
-
- What routine this is? I guess 030/50 should be able
- to get c2p free -->รก320x256x8 screen should be about 20ms.
- This could be perhaps achieved by doing three passes
- with CPU and the last one with the blitter.
-
- -- _
- a Stellar programmer _ //
- "Amiga - back for the future" \X/
-