Author: Robert Hyatt
Date: 13:52:31 05/22/98
Go up one level in this thread
On May 22, 1998 at 15:12:19, Ernst A. Heinz wrote: >On May 22, 1998 at 13:19:04, Robert Hyatt wrote: > >Bob, > >Great WAC results -- my congratulations! > >>Note that the ALR machine is running approxmiately as fast as the 500mhz >>alpha machine we used in Paris, for comparison. On that machine we were >>hitting around 250K nodes per sec... on this machine roughly a little >>faster, but about 1/4 to 1/3 of the machine is "wasted" due to search >>overhead... So that the newer alpha machines would actually be a bit >>faster on one processor, and way faster with more than one... > >Sounds a bit strange to me if you look at the SPECint95 ratings of >a PPro 200 MHz = 8.7 (i.e., 4x PPro 200MHz = 4x 8.7 = 34.8) and a >500MHz Alpha-21164a = 15.0 ... > >Are you sure about comparing the speeds on the same positions? > >=Ernst= yes. We ran several tests comparing a single pentium pro/200 to the alpha/500 we used in Paris, and got 3.1X speedup on that machine. I attributed part of it to the faster clock speed, part to the fact that the alpha does most of my code more than twice as fast as the P6 because of the 32 to 64 bit problem. IE 64 bit shifts are a pain using 32 bit operations, but they are a single instruction on the alpha... We'll have some real numbers before long for 1 and multiprocessor alphas I hope, as we are trying to line up something fast to play GM Yermolinsky a game in Washington DC next month... I don't think I am over 3x faster on the quad p6 on WAC... The problem is to run fast you have to have good move ordering. To have good move ordering, you need continuity from one move to the next so that the hash table provides better "suggestions." In wac, nothing gets reused from one position to the next. I ran a test like this with CB a couple of years ago and found that our performance in a real game was much better than our performance over disjoint test positions (this was in the JICCA in fact.) Bob
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.