home *** CD-ROM | disk | FTP | other *** search
-
- I have just read the your answer to my previous posting. Thanks to everyone
- for the answers.
- Actually my c2p are processor-only converters because I dont understand
- enough about blitter-coding. Besides I am mainly interested in
- 8bpl-converters for my 3d-routines. The original c2p-routines are from
- aminet (I think fastc2p.lha )
- My c2p get a pointer to the chunkydata, one to the (continious) bitplane(s)
- and the plsize. And of may be an comparebuffer. The 3d-routines know
- nothing about the display-enviroment. They simple get the x,y-size, the
- number of colours and 2 functionpointers. One to the db-switchroutine and
- one the the c2p. So its no problem to support gfx-cards.
- I will not send followups to all articles but instead I put all my new
- questions into this posting:
-
- - Stephan Schaem (sschaem@teleport.com) <4cjl2o$4im@maureen.teleport.com>
- >: - are there any very fast c2p which use a comparebuffer ?
- >
- > depand on the aplication I guess... The amiga mac emulator ShapwShifter
- > has a range of driver that use Delta/checksum/mmu buffer in conjunction
- > with the plannar/chunky buffer.
-
- Could you give me a hint what a Delta/checksum/mmu buffer is ?
- ( I use a comparebuffer which check if a block of 8 pixels have changed. If
- yes then the comparebuffer is updated and the converter starts. If not I
- simply step to the next block)
-
-
- - Juergen "Rally" Fischer (fischerj@Informatik.TU-Muenchen.DE)
- <4cjmsk$9aa@sunsystem5.informatik.tu-muenchen.de>
- >: (I have heard this sometimes in this group)
- >: and what are his adventages/disadvantages ?
- >: - are there any very fast c2p which use a comparebuffer ?
- >yes, AFAIK Peter McGavin got most experience data on this.
- >AFAIK the routines get very fast when the soure to convert doesn't
- >change much.
-
- Where could I get this routines ? (could you mail me the source?)
-
-
- - Jyrki Saarinen (jsaarinen@kone.fipnet.fi) <38231876@kone.fipnet.fi>
- >> 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
- >o 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.
-
- Sounds good could you post or mail some asm-source ? I would prefer a
- cpu-only-solution.
-
-
- - Andrew Bennett (ben@gauss.demon.co.uk) <30ef4f99@gauss.demon.co.uk>
- > Assuming that you still have a screenful of planar data from the last
- >conversion,it's very likely that a lot of the new planar data you write
- >will be
- >identical to the old copy and thus doesn't need to be overwritten with the
- >same
- >data!! (Depending on the type of graphics used and the amount of change
- >per
- >frame...)
- >
- > The end result is a c2p routine which takes a variable amount of time to
- >do
- >the conversion,but may well be faster.
- >
- >eg: Assuming that the first four pixels are colours 12,13,220,114. On the
- >next
- >pass let's say they are now 15,12,223,116.
- >
- > If you eor the old set with the new set,and re-order the data into
- >planar,you'll find that only planes 1 & 2 have actually changed... So you
- >can
- >save 6 chip writes and a fair bit of manipulation,but at the the expense
- >of
- >reading twice as much data from (hopefully) fast memory.
-
- Have you any expierences (I mean working code) with this? I suppose that
- this only give advantages if not more than 4 planes have changed. In this
- case you could use a changed 4bpl converter which is a lot faster. But if
- you do so, the code became really big and dont fit into the cache. The
- speed penality for this will probably destroy all advantages.(I think so)
-
-
-
- Also someone wrote that you could do 12 or 13 instructions during chip
- write. Is there somewhere (in aminet) a time-table on this ?
- I found a c2p-converter which use selfmodifying code. This one copy himself
- to a quadword-align memory-position. Does this give any advantages ? (I
- think a 8bpl c2p is small enough to fit into the cache)
-
-
- Cia,
- Sven.
-