Author: Vincent Diepeveen
Date: 15:08:39 08/25/01
Go up one level in this thread
On August 25, 2001 at 16:57:34, K. Burcham wrote: >1. does anyone know what the operator error was in the diep vs crafty game? He used winboard and crafty can't take back a move in winboard. Idem for Zchess. Note i was at +2.0 against crafty already. I was busy mating it, in fact. I'm sure game would have been ended soon anyway as crafty always has problems with its king and tactical it's just not at the same level. So for the end result it wouldn't have mattered. I was at 0.00 against zchess though. >2. did black(crafty) have to forfeit because of the error? > it seems the score was even in this position. >3. what was the rule governing this decision? rule? if you can't take back a move then it's over of course, before you have reentered the game you have forfeited already. >4. were all these computers connected by way of a local area network. what's this for kind of nonsense question? If there would be a protocol to use over the network, then some big company like chessbase, or schroeder or whatever would take advantage of it and cheat of course. That happened with auto232, so i would decline all offers for automatic testing in a WM, considering what happened so far in the past. It would be smart however to add a takeback feature in crafty. idem in zchess. Note also adjusting time is not simple in winboard. Impossible kind of. I also had problems with this, but it's my own interface, that i was losing so much time each game was because of my own interface never played before in history blitz. this World championship was the first test ever where i tried to blitz with diep. >5. if not, how were the programs moved from pc to pc? everyone using his own PC of course. of course i'm not going to move my program from my dual 1.2Ghz palomino machine to a single cpu 1 Ghz machine where Xinix ran at (or whatever machine it ran on). What you do is move your machine of course. That' snot so easy if you are on your own. Lucky i had prepared myself well for this tournament when talking about moving, my case has 4 wheels under it, i can advice everyone using a case with wheels. It's just 20$ and it's great. I bet that everyone who saw me there is going to have wheels themselves soon too! >6. can you tell me how an operator sets his time to achieve 60 moves > in two hours, plus 30 minutes for the balance of the game? In diep you can enter any level of course (3 time controls supported). How others do it i do not know, but i guess they also can set this level. In case of winboard, god only knows (i think you can only enter the level before the game not modify it during the game). Ask the poor fellows how they did it! What i know is that crafty used a textmode interface in the normal games. I definitely think that Michel Langeveld was the first ever operator other as Bob to operate crafty properly in the tournament. Of course in blitz using a textmode interface not many have ever achieved. I think i was one of the few who in the WMCC 2000 was operating in a text mode interface in blitz. >7. if you see your program is taking to long for a move, or you feel your > program is using too much time compared to your opponent, can you adjust > your programs time setting? (some gamers think that there are some wins > from manually managing your programs time. most of these types of > games are at the servers where moves are all made manually). you may not interfere during a move. A few times a game you were allowed to SYNCHRONIZE the internal clock. >8. how can you tell if your program is "thinking during your move", that depends upon your program. In my case i see every second updated 2 lines which shows in the first line the current status of processor 1 and in the second line the current status of processor 2. To see how it looks go to www.aceshardware.com for example. > and not a software freeze? and/or "pondering during your opponents > move" and not a software freeze? i can tell with tiger but some of > the others do not have an indication that the engine is ok. Right tiger sometimes freezes here, but you are allowed 15 minutes for technical problems in a tournament. Lucky this year not too many needed it. Probably junior crashed a few times (did this happen?) and of course tiger had problems. Seems that most people did have problems with their hardware getting too hot. Well it was a very hot room anyway. In my case my harddisk got too hot during round 8. I already i knew that U-ATA100 harddisks from Maxtor have major problems with heat, so i had bought a 40G IBM harddisk (uata100 deskstar). At round 8 i faced a tough opponent and my extensions quickly triggered EGTB after egtb hit short after the opening (as everything could get captured easily in stupid lines). This crashed my machine after round 8 as harddisk became too hot. Note that also the palomino processors are anything but cool either. Even when idling they're 67 degrees. I'm right now looking for a tool to measure how hot they are when running. So before round 9 i had to take out my harddisk and disabled EGTBs after i managed to boot NT 4.0 again (win2000 didn't install). I have brought back this harddisk now and hope to get a new one real soon from IBM. a harddisk in a server case with a 460 watts power supply which is taking away LOADS of air, and considering there is more than enough room for the harddisk inside the case with plenty of air around it, it's a bad shame that you can cook an egg at it! I feel this all is no good sign for the future. All those AMD K7 processors eat loads of power and they get boiling hot, nowadays also the faster IDE harddisks (u-ata called nowadays) get real hot. I heart rumours that P4 at high speeds like 1.7Ghz is having slower clocked caches than 1.7Ghz because it gets too hot. On the other hand i remember the 486dx33 which didn't have any cooling. Nowadays a processor without huge own cooler is unthinkable. Soon it seems we need to extra cool down harddisks too! Where does this all go to? >kburcham
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.