Author: Peter Kappler
Date: 10:27:21 10/02/99
Go up one level in this thread
On October 01, 1999 at 12:34:06, James Robertson wrote: >On October 01, 1999 at 02:35:48, Peter Kappler wrote: > >>On October 01, 1999 at 02:07:07, James Robertson wrote: >> >>>On September 30, 1999 at 16:50:33, Peter Kappler wrote: >>> >>>>On September 30, 1999 at 14:35:25, Charles Unruh wrote: >>>> >>>>> >>>>> In the past i thought faster hardware would benefit slow searchers like >>>>>Chessmaster more than fast searchers like Fritz. Now i'm more inclined to >>>>>believe that it makes more sense that faster hardware benefits fast searchers >>>>>more. For the reason that positional ideas are for the most parts moves made >>>>>from practical experience/knowledge, that we can't always quite calculate. >>>>>However, faster hardware gives programs the ability in many instances to >>>>>actually be able to calculate the result. So although i think Chessmaster is a >>>>>truly awesome engine especially against programs running up to 233Mhz I expct >>>>>that on a P450Mhz it will come in 3rd or 4th. >>>> >>>> >>>>As a general rule, faster hardware should favor the "slow" searchers. This is >>>>especially true at blitz, where search depths are relatively shallow, and every >>>>extra ply makes a huge difference. >>>> >>>>Chess System Tal is a great example of this - it's relatively weak at >>>>blitz(Fritz would kill it) but pretty strong at 40/2 (probably about equal to >>>>Fritz). >>> >>>This kind of falls apart when you see the amazing lightning results Hiarcs 7.32 >>>is having against all other program, Fritz included.... >>> >>>James >>> >> >> >>Hi James, >> >>I'm a big fan of Hiarcs, too, but my experience so far is that Fritz is slightly >>stronger at blitz. This is based on a few engine vs. engine matches I've played >>between them. (~80 games so far at 5+1 time control, and Fritz is ahead by >>about 10 games.) I know this isn't a large enough sample size to draw firm >>conclusions - please share any data you have. > >The only data I have is the lightning (1 0, 1 1) results several people have >posted here.... Hiarcs seems to have a large lead. > >I think your 80 games is a fairly good sample. Most people try to draw >conclusions after 10 games. :) > >James > >> >>By "lightning" are you referring to super-fast time controls, like 2+0? I have >>no idea how Hiarcs performs there, but my money would be on Fritz... >> >>--Peter I decided to test my theory with a 64-game match between Hiarcs and Fritz. Time control was 3 minutes per game, with a 1 second increment per move. Both engines used the Fritz book and had 32MB hashtables, which I think is more than enough for this time control. Hardware was a K6-3/450. Hiarcs 7.32 1½11½1001½½½0½1½00½½0½1½10011110101½½½111½011½½001½1010101½1111½ Fritz 5.32 0½00½0110½½½1½0½11½½1½0½01100001010½½½000½100½½110½0101010½0000½ Looks like I was wrong, as Hiarcs won convincingly, 38-26. Watching these games, I noticed that Fritz regularly gets 9 plies at this time control, while Hiarcs is typically hitting 7. Of course, what really matters is how deeply they search the important sections of the tree, and Hiarcs must be doing one hell of a job there. As for my previous report that Fritz had a 10-game lead in an 80 game match at a 5+1 time control - I should have known better than to try to recall those numbers off the top of my head. It turns out that the total number of games was 56, not 80. And while Fritz did have a commanding lead at one point, it has dwindled to 4 games, 30-26. Also, the time-control was 5+2. My apologies. --Peter
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.