home *** CD-ROM | disk | FTP | other *** search
- Path: informatik.tu-muenchen.de!fischerj
- From: fischerj@informatik.tu-muenchen.de (Juergen "Rally" Fischer)
- Newsgroups: comp.sys.amiga.programmer
- Subject: Re: >>>>>>> The Future: Amiga goes PC
- Date: 26 Jan 1996 17:03:37 GMT
- Organization: Technische Universitaet Muenchen, Germany
- Distribution: world
- Message-ID: <4eb1h9$drc@sunsystem5.informatik.tu-muenchen.de>
- References: <38231985@kone.fipnet.fi> <judas.0hhx@tomtec.abg.sub.org> <38232130@kone.fipnet.fi>
- NNTP-Posting-Host: hphalle5.informatik.tu-muenchen.de
- Originator: fischerj@hphalle5.informatik.tu-muenchen.de
-
-
- In article <38232130@kone.fipnet.fi>, "Jyrki Saarinen" <jsaarinen@kone.fipnet.fi> writes:
- |>
- |> > I had the chance to take a look at Xtreme-reacing, which is incredible
- |> > fast. Their blitter C2P is super fast, even on a stock A1200. The CPU C2P
- |> > needs a 30/25Mhz to appear that smooth.
- |>
- |> Blitter c2p is never "super fast" because it involves very
- but on 020 it's faster ;)
-
- |> much blitting. I guess you are talking about XTRs blitterscreen
- |> mode where blitter c2p on 160x128 screen takes very little
- |> time, 50fps possible if scrambled buffer is used (very likely
- |> in XTRs case) or 25fps in linear buffer is used.
-
- linear buffer needs 1.5 frames vs 1 frame on scrambled, so 33 fps
- should be possible.
-
- |>
- |> > Gloomdeluxe offers a nice way to test C2P-codes. My experiences show a
- |> > 040 C2P to be as fast as a 020+Akiko. The C2P codes for gloomdeluxe from
- |> >áPeter MGavin show a similar result. (aminet game/demo)
- |>
- |> Yep, because CD32 with a 14MHz 020 has better chipram bandwidth
- |> than a A4000/040. If the chipram speeds would be equal,
- |> Akiko would be slower because it involves more memory moving.
-
- I don't understand. yes, the 020 got 7mb/sec acess, but when just
- copying fast->chip it drops to about 4mb/sec.
-
- even if cd32 got fastmem, akio c2p should be fast->fast + fast->chip,
- which should be only about 2.6mb/sec!
-
- no way of akiko beeing better, the only possibility is it is maybe
- slightly better on chiponly machines vs. blitterassistance.
-
- but then AKIKO on chiponly, chip->fast, fast->chip, that's 2mb/sec
- only, 020+blitter should achieve about same on chiponly Amiga.
-
- |>
- |> What is your point? This whole thread started when you
- |> stated Akiko would be very useful in all Amigas or
- |> something like that..
-
- Yes, some believe AKIKO + 040 would do it better than chipmem bandwidth ;)
-
- the only advantage of akiko is imho it's easy to program, so
- CD32 conversions are more likely to be not slow even if developpers
- don't know Amiga hardware so well.
-
- |>
- |> -- _
- |> a Stellar programmer _ //
- |> "Amiga - back for the future" \X/
- ------------------------------------------------------------------------
- fischerj@Informatik.TU-Muenchen.DE (Juergen "Rally" Fischer) =:)
-
-