Author: Eugene Nalimov
Date: 12:15:33 03/30/04
Go up one level in this thread
On March 30, 2004 at 13:43:38, Gerd Isenberg wrote: >On March 30, 2004 at 02:34:21, Slater Wold wrote: > >>Arasan 7.4 32-bit: >> >>D12 00:18 @ 729.047 N/sec (starting position) >> >>Arasan 7.4 64-bit: >> >>D12 00:17 @ 771,932 N/sec (starting position) >> >> >>64-bit Arasan compiled with: -O2 -Ot -Og -Oi -Gr -GX -Oy -Ob2 -GL >>32-bit Arasan would NOT compile on my machine under W2003, so the binary used is >>from the Arasan home page. > >Hi Slater, > >may you supply a small table with all engines you tested so far? >Speedup factor would nice and whether an engine is bitboard based or not ;-) > >As i saw so far, the 64-bit speedup is not as good as expected. >I can't imagine that the signed index issue is that expensive. >With similar optimization abilities the 64-bit code should be much faster, >simply due to the eight additional registers and (per default? fastcall instead >of cdecl or pascal parameter passing). Alignment and Load/Store forewarding >stalls are important issues too, but even true for 32-bit code. > >My guess is, that the 64-bit ms-compiler you use is yet too safe or paranoid >with aliasing issues and tends to write locals inside registers to memory >permanently. However, it would be interesing to have a look at the generated >assembly for some middle sized (20-30 lines c-code) functions with some locals, >for 32-bit and 64-bit each. > >Thanks, >Gerd Post here the (short) code, I'll post compiler output... Thanks, Eugene
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.