home *** CD-ROM | disk | FTP | other *** search
- Welcome back Laurent,
-
- > I'm back from school, and I'll stay here 3 months :-)
-
- :-)
-
- I guess you've seen from my earlier email that the list is now open for
- everyone. Not very many have joined yet, but I think we are about 10 now.
-
- > First, I'm still working on Bad mood, I've got a first version working
- > in assembler, I'm actually optimising it as much as I can.
-
- Could you, please, tell us what kind of speed you've gotten from it?
-
- > I'd be interested in the map displaying version, if you could send it to me ?
- > (And the c code of course !!)
-
- I've been planning to add the wall check (I've got a simple sector height
- check working), but I never seem to get it done. The map drawing could be
- useful as well as nice looking, though, so I'll send it RSN(TM).
-
- > I'm actually optimizing loadseg, and addwall.
- > Which functions are taking the most CPU time ?
- > I think you did tests about it.
-
- Here's the relevant information, copied from an old letter:
- -----
- I've incorporated some simple statistics collection in DVIEW which gave
- some numbers that are interesting for the optimization. The figures are
- approximate averages over more than 200 drawings from five different levels.
-
- Function Calls Internal loops
- (1) (2)
- Addfloor 150 1100
- Addwall 50 1700
- Coldraw 600
- Rowdraw 1000
- Drawnode 200
- Drawssector 100 300
- Loadseg 150 1600 400
-
- Time for
- structure setup: 0.28
- actual drawing: 0.08
- chunky to planar: 0.11
- -----
-
- As you can see, I did not try to get any timing info for the separate
- functions. That would have been very difficult, if at all possible. The
- number of calls and loop iterations should be almost as useful, though.
- I don't have the code available at the moment so I can't give any more
- detailed information on what's going on, but I'll get back on that.
-
- > I've moved away the singlestep option, we won't need it in the definitive
-
- OK, the C version will have to do for high level debugging for the time being.
-
- > version, and now, the maze is correctly displaying.
-
- Uh, do you mean that you get none of the old errors?
- Have you implemented the 'direct' drawing?
-
- --
- 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)
-
-