Author: Eugene Nalimov
Date: 22:07:29 11/12/99
Go up one level in this thread
Once again: my generator can be converted to produce DT(conversion or pawn move) instead of DTM in several hours of work (my estimate is 2-3 hours). After that some time will be necessary to re-generate affected TBs (we can easily leave majority of TBs DTM, as they are today), but as long as we are talking about 5 man tables, time will be be measured in days, several weeks at most. Some modifications on the engine side will be necessary, but that's not my problems :-). And I expect that those modifications also can be done in several hours. Of course much more interesting task will be to produce DTM table that takes into account 50 moves rule, but unfortunately I don't see good algorithm for that. The simplest approach would be to add additional counter to the each position, but that would mean 2 times more memory when compared to the first version of my generator, and 16 times when compared to the current (not published yet) version. If that is acceptable, modification can be done in a day or so. So I don't see any problems here at all. Now speaking about 50 moves rule in general. I see good reasons both pro and contra it when programs are playing. If majority will decide that it's not possible to lift it when program can demonstrate forced TB win - fine, I'll do what is necessary. As I already noted, that will greatly speed up generation of 6 man TBs :-). Eugene On November 12, 1999 at 22:44:53, Bruce Moreland wrote: >On November 12, 1999 at 14:21:51, Eugene Nalimov wrote: > >>50 moves rule is "rule of play", not "rule of game". That means that it can be >>changed for the particular event by the TD (and of course that should be writteh >>in the event rules before it starts), and played game will still be named >>"chess". At least it was that way ~10 years ago, when FIDE rules said "50 moves, >>with exception of 75 moves there, and there, and there, and TD can add more >>exceptions". >> >>Of course I can easily modify my generator to produce DTC, not DTM, for those >>endings where that is relevant. It will be much faster, also - 50 iterations >>will be enough (right now it's on 160th iteration on KRNKNN, and I believe ~100 >>more will be necessary). > >What I am concerned about is that people who are taking your work and >integrating it into their things, with expense of very little creativity and >energy upon their part, will not have the world bend to conform to the quirks >evident in the implementation they mooch. > >There is no rational reason why the 50-move rule should be suspended for >computer play, other than that a few of us have made endgame databases that >conveniently ignore this rule. We do not deserve to benefit from our laziness. > >I have the same bug. I see it as my responsibility to fix it. If I don't, and >I lose a half-point because of it, tough luck for me. > >If someone else is going to be in a position where they take your code, and then >argue that they should have to take the bugs inherent in your code, and should >just have half-points handed to them because they are too lazy to write their >own thing, or too lazy to sense this problem and get you to fix it, they are >going to get a big argument from me. > >If I fix this bug, I deserve to benefit, and everyone who uses your thing can go >die unless either you or they are willing to fix it. > >You probably never expected to be a competitor at these events, but you are. > >bruce
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.