Author: leonid
Date: 11:51:39 01/27/00
Go up one level in this thread
On January 27, 2000 at 13:21:47, Robert Hyatt wrote: >On January 27, 2000 at 08:16:47, leonid wrote: > >>On January 26, 2000 at 23:47:40, Robert Hyatt wrote: >> >>>On January 26, 2000 at 17:21:36, leonid wrote: >>> >>>> >>>> >>>>800k (nodes per second) on 1600Mhz sound to me as extremely speedy. Maybe Xeon >>>>is also the 64 bits computer. This could explain this speed excess with what I >>>>could expect. I compare your speed with one PC game written on Assembler. Your >>>>speed just shine. >>>> >>>>Leonid. >>>> >>>> >>> >>>I don't think that is fast. IE it is about 200K nps on a single xeon/400mhz >>>processor. Fritz/Junior are _way_ faster. >> >>If single Xeon 400Mhz is almost the same like AMD 400Mhz, your speed is more >>that good. Few days ago I found that Rebel 10 on AMD 400Mhz have average number >>of positions/second around 165k. I counted the speed for the positions where >>expected speed should be the biggest. This is why I am so surprised with your >>speed rate. >> >>Leonid. >> >> > > >Depends on the position. There are plenty of cases where I get less than >500K nodes per second, particularly if neither side has castled, as I spend >a lot of time making sure that I castle when I should, don't castle when I >shouldn't, and castle to the right side when castling in either direction is >legal. > >My speed rangd goes from 450K to about 1M. Or on a single Xeon, take 1/4 of >the above... The same about all the games. Rebel goes between 120000 and 260000 (AMD 400) but average is 165000 for heavy positions. For me heavy position is the position when almost all the pieces are on the board and there are plenty of moves for both sides. Leonid. >>>And no the xeon is not a 64 bit machine. Mine are PII's but with a better >>>L2 cache. PIII xeons are a bit better...
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.