home *** CD-ROM | disk | FTP | other *** search
- Path: taurus.zrz.TU-Berlin.DE!rawneiha
- From: rawneiha@taurus.zrz.TU-Berlin.DE (Philipp Boerker)
- Newsgroups: comp.sys.amiga.programmer
- Subject: Re: doubling pixels horizontally
- Date: 15 Feb 1996 18:32:32 GMT
- Organization: Technical University Berlin, Germany
- Message-ID: <4fvu80$4tg@brachio.zrz.TU-Berlin.DE>
- References: <4f4ibc$gl9@news.cs.tu-berlin.de> <591.6610T1165T2102@login.eunet.no><1045.6611T753T2256@vip.cybercity.dk><4faoe1$47@sunsystem5.informatik.tu-muenchen.de><2991.6612T1034T625@vip.cybercity.dk> <576.6613T1070T1730@login.eunet.no> <1257.6614T5
- NNTP-Posting-Host: taurus.zrz.tu-berlin.de
-
- patrick.hanevold@login.eunet.no (Patrick Hanevold) writes:
-
-
- >>>And the new one is even 40% faster. :) (Soon to be released.)
- >>This one will be interesting! I have done a routine like you did. It consumes
- >>8.75 papercycles/pixel on the CPU side of the algo (2 passes for free on 50
- >>MHz 030} and does the 2nd part with the blitter. I don't think that yours is
- >>much different so I would like to know where you get that extra 40 % from ...
- >>What does "released" mean? are you uploading your routines to aminet?
- >Hmm. Our new one does the CPU passes %40 faster than the earlier released
- >version. Maby we use the same algo? :) Yes, it will show up on aminet.
-
- >Your one is linear, right?
-
- Of course! But no comparebuffer involved...
-
- Greets,
- Phil.
- grond/matrix
-
-