Author: Ed Schröder
Date: 04:23:28 02/01/02
Go up one level in this thread
On February 01, 2002 at 06:05:22, Steve Maughan wrote: >Ed, > >>It will be even slightly stronger because surprise surprise the Windows version >>runs 11-12% faster than the DOS version due to the use of a new compiler. > >I thought Rebel's engine was 100% assembler, in which case the compiler would >make little or no difference to the execution speed. What am I missing? I am as surprised as you. I assume the speed-up is related to the compiler putting the assembler code more well placed in memory (alignment). Another factor could be the IO (the keyboard, the mouse, the display info on the screen), maybe under Windows this go a lot faster than under DOS? So far I haven't put much energy in finding its reason, I just enjoy the unexpected speed gain. Although DOS this has always worked for me: whenever a version is ready for release the last step I do is optimizing the engine for speed. The tricks I used a) swap include files b) put the bigger tables you use in some other parts of the source code and recompile. The effect is that your code and data are put elsewhere in memory in the hope for a better alignment. Mostly I get a 2-4% speedup, but I have seen one case having a 6% speed-up just switching one big table. Not bad for a couple of hours work. I have no idea if this kind of tricks work under Windows too but I will find out soon. Ed >Regards, > >Steve
This page took 0 seconds to execute
Last modified: Thu, 15 Apr 21 08:11:13 -0700
Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.