home *** CD-ROM | disk | FTP | other *** search
- > First, I don't want to code some parts in c and others in assembler, coz I
- > think pure assembler can let us optimise the code to its limits.
-
- I guess the lack of a C compiler is a good reason as well.
-
- > Now, I didn't convert the .WAD loaders to speed it, but to fill the different
- > structure in an assembler way. I mean, it prepares the hard code. You can be
- > confident in my work, I've been coding assembler for a long time now, I've
- > coded many routs ( 3d, mapping, plasmas, gouraud shading, starscrolls and so
- > on, I've coded two demos...), I know what I do :-) , and I've got experience
- > in optimisations for demos, I'll use this experience for bad_mood. I also think
- > that I code assembler quite fast now.
-
- It was just a question. Since you're the one doing the conversion work you're
- quite welcome to rewrite TOS. ;-)
-
- > A last thing, I know that doom on pc is running with assembler code :-)
-
- Yes, but only a few (like two) routines down in the TM parts according to
- some posted email from Id I read once.
-
- > For the ARGV stuff, I've done it in less than 10 minutes yesterday, so it
- > doesn't take much time to convert.
-
- That's good to hear.
-
- > This week, I'll start to convert the bsp, and if I've got the time, the
- > display routs.
-
- I hope you manage to get some significant speedup even at this stage.
-
- > PS: Thanks for the cleaned code.
-
- No problem. Regarding the missing header files, I did say that I only
- included the relevant files. I saw no changes that would be useful to do
- to the headers. There are a couple of C files that were not included in
- the archive I sent you either (only timer.c and hash.c, I think).
-
- --
- Chalmers University | Why are these | e-mail: d8klojo@dtek.chalmers.se
- of Technology | .signatures | rand@cd.chalmers.se
- | so hard to do | www/ftp: rand.thn.htu.se
- Gothenburg, Sweden | well? | (MGIFv5 and QLem)
-
-