Author: Chuck
Date: 19:58:15 07/23/99
Go up one level in this thread
Below are benchmark runs with two compilers on my AMD K6-2 350 Mhz, plus Dann's posted executable. The compilers are all older versions, I may run the test using VC++ 5.0 & 6.0 tomorrow. Dann's posted executable: tscp> on ply nodes score pv 1 21 48 d2d4 2 93 0 d2d4 d7d5 3 996 35 d2d4 d7d5 b1c3 4 5952 5 e2e4 d7d5 f1b5 c8d7 b5d3 5 32924 35 e2e4 b8c6 b1c3 e7e5 g1f3 6 203273 13 e2e4 b8c6 d2d4 d7d5 e4d5 d8d5 time: 4.450000 sec., nodes=203273, NPS = 45679.325843 Dann's posted source compiled with Borland C++ 4.5 tscp> on ply nodes score pv 1 21 48 d2d4 2 93 0 d2d4 d7d5 3 996 35 d2d4 d7d5 b1c3 4 5952 5 e2e4 d7d5 f1b5 c8d7 b5d3 5 32924 35 e2e4 b8c6 b1c3 e7e5 g1f3 6 203273 13 e2e4 b8c6 d2d4 d7d5 e4d5 d8d5 time: 6.228000 sec., nodes=203273, NPS = 32638.567759 The difference noted is great. I compiled with only the -O2 switch. Note the great difference here is probably mostly due to the Borland compiler being old enough that it does not have Pentium Pro optimizations. If so, then Borland C++ 5.0 or later should do much better. Dann's posted source compiled with Symantec C++ 7.2 tscp> on ply nodes score pv 1 21 48 d2d4 2 93 0 d2d4 d7d5 3 996 35 d2d4 d7d5 b1c3 4 5952 5 e2e4 d7d5 f1b5 c8d7 b5d3 5 32924 35 e2e4 b8c6 b1c3 e7e5 g1f3 6 203273 13 e2e4 b8c6 d2d4 d7d5 e4d5 d8d5 time: 6.319000 sec., nodes=203273, NPS = 32168.539326 Again, this compiler is older and does not have Pentium Pro optimizations, just Pentium.
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.