Author: Robert Hyatt
Date: 06:53:53 08/30/99
Go up one level in this thread
On August 29, 1999 at 14:32:13, Andrew Dados wrote: >On August 29, 1999 at 13:52:58, Robert Hyatt wrote: > >>On August 29, 1999 at 12:22:58, Frank Quisinsky wrote: >> >>> >>>I think this is all correct, but Robert ... >>>When the engine correct play on one PC without ponder (engine-engine) why is >>>ponder 50-100 ELO. >>> >> >>I don't know what you mean there. I didn't say anything about 50-100 elo, >>although I believe that is accurate. >> >> >> >>>example ... >>> >>>Crafty thinking for move 28 in the game >>>02:58 13/02 move Ka1 without ponder >>>02:20 13/04 move Ka1 with ponder >> >>that makes no sense. pondering saved 38 seconds? It should save more like >>2 minutes there. >> >> >> >> >>> >>>In move 29 in this game >>>04:45 11/04 move Ka2 without ponder >>>05:38 11/05 move Ka2 with ponder >> >> >>ditto... it depends on how long the opponent thinks _after_ crafty >>starts pondering... If it thinks for the normal amount of time, crafty >>gets that much think-time _free_. And I've _never_ seen the prediction >>rate below 50% against a computer, more commonly it is well above 50%. >>The log file will show how many moves it correctly predicted, which will >>tell how many times it could potentially save time. >> >>But you are totally missing the point Ed raised and I seconded: if one >>program has been tested and tuned for ponder=off play, and the other has >>not, then that program has a significant advantage. Tough luck, you say? >>Of course... but then your results don't have anything to do with how the >>two programs would perform on separate machines. >> >>That is why we keep saying "don't run games on one computer... the results >>are not always as meaningful as you might assume..." >> >> >> >>> >>>This is 50-100 ELO ? >>> >>>Robert 30% ponder hints in the game ! >>> >>>OK, 100 moves, in this 100 moves 30 ponder hints >>>And how many other moves play Crafty from this 30 ponder hints ? >>> >>>I think 2-5 moves, OK 5 moves but 2 from 5 moves are = or not better ! >>> >>>But in 0:30 second without ponder Crafty play not blunders ! >>>In 0:45 second Crafty can found an better move ! >>> >>>Is this for the statistic relevat when I play 500 Crafty games without ponder ? >>> >>>Robert, for 4 years we play with an 486/66. Is this system bad for you ? >>>What is better Robert, games on this 486/66 with ponder or an game on an AMD >>>K6-3 450 without ponder ? I think Crafty play stronger on an AMD K6-3 450 >>>without ponder, or ? >>> >>>Today you play on an Penitum III 500 Dual (I think). In four years you play on >>>an Pentium V 3000 MHz ! And then ? Is then ponder importent for you and the >>>games from the DUAL Pentium III are bad ? >> >>you are missing the point. my time allocation _depends_ on saving time by >>pondering. You are not allowing it to do that. Which is the problem with >>this... nobody would argue that _all_ engines are 50-100 elo stronger with >>ponder=on than they are with ponder=off. That is easily testable on a chess >>server. But the issue here is whether a program is tested with ponder=off or >>not. Mine isn't. Ed's isn't. >> >> >>> >>>I mean you can play without ponder with an AMD K6-3 450 and have the power from >>>an AMD K6-3 350-400 MHz, 20-40 ELO that is it ! >> >> >>generally 2x faster is 70 Elo better. Pondering has the potential to make >>a program act like it is twice as fast... >> >> >> >>> >>>>The problem is all about whether the program has been tested/tweaked to run >>>>well in that environment. It would probably be better to play on one machine >>>>using pondering instead of turning it off. And even that has problems... >>> >>>Yes I understand this and I understand you entry about ponder and time control >>>but I can not see 50-100 ELO or I am chess blind ! >>> >>>It`s an good example from Ed with forumla 1 and 2. I think forumla 2 (that is >>>right) with Schumacher in position 1 with full power and not Schumacher with an >>>defect. And Schumacher win this finish in forumla 1 and in formula 2 ! >>> >>>The car´s run in forumla 1 with 300 km/h and in forumla 2 with 260 km/h ! >>> >>>Kind regards >>>Frank >> >>But suppose you take his car, and suddenly make him run with rain tires when he >>hasn't in the past. How do you think he'd do then? No testing? He'd be pretty >>unlikely to even finish the race. This is a common NASCAR problem in the USA. >>There are many good rain tires, and some NASCAR races are on wet tracks. But >>the drivers don't use the rain tires because to quote one this week "It would >>be on-the-job-training, because we can't have rain when we need it to test..." >> >>That is the point with chess. You are testing the programs in a mode where _we_ >>don't test them. Poor performance is not unexpected... > > Byt you 'can have rain when you need it to test' here. And I would bet some 90% >matches between engines were/are/will be done on one-cpu computer. Just because >that's what most people have at home (and if they have two - then usually they >are of different CPUs). And I bet few chessplayers test their opening lines by >playing engine-engine from a given position, too. Maybe it's worth considering >adjusting time management to work in that mode then. Or keep repeating same >arguments every once in a while... but that will not make people go out to the >store and buy second same machine (or dual), I bet. >-Regards- > Andrew I don't argue that point. However, I test my program on dry pavement, because that is the way I am going to race in _every_ important event (ponder=on). If you want to race my program in the rain, there is little I can do to stop you, other than to tell you it is not going to perform as well as it does on dry pavement. And it might do worse if the opponent has done a bit of testing in the rain. I simply don't have time to test/tune 'weak' versions (ponder=off). I try to work on the mode that is 'strongest'...
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.