Author: Otello Gnaramori
Date: 08:52:42 09/09/01
Go up one level in this thread
On September 09, 2001 at 11:00:21, Günther Simon wrote: > >Please take a look again at the former post of K.Burcham and you >would see that it took Shredder 5 (in the position when F6 grabbed >at g4) 41! minutes on a 1Gig Cpu to avoid it here. Probably the algorithms of Shredder aren't yet well tuned against this kind of traps... >Maybe it would take another 41 minutes in the NEXT move to avoid this >and your statement cannot be maintained when all of this is even >far beyond tournament time controls because of the incredible depth >of plys necessary to see the apocalypse comin'... >(Yes the horizon effect even nowadys is not yet faded - in certain >positions it is obvious!) The 'Horizon' effect is what happens when it doesn't calculate variations deep enough to understand the tactics (and usually happens if you don't give the machine enough time). BTW Nowadays several programming techniques has been developed to overcome this effect at a reasonable extent.. >Btw I did some Nemeth-Gambits too and you can truly believe me that >with only having the vague idea of how this systems work, I crushed >F6 in my first game doing so [of course agin it was just a 30min game:))] See my previous post on blitz games. >without any retry.(The other side of the coin is that I believe this >makes my human chess worser though...) >Last to say is that the whole opening theory is at least nothing else than >try and error and if someone can reproduce his training in serious >games there is no difference at all. But using "trial and error" against comps "after" the opening phase is unfair IMHO, since they will probably repeat the same moves. >I would appreciate if you for yourself would give this kind of play a try >because I had the same attitude towards this phenomenon and was rather >sceptic seeing all this posts(and maybe you are right: too much) about easily >defeating modern comps by a player not much better than me. >(around 1900 ICC: 2100) I have no doubt that comps can be beaten easily using this kind of techniques at blitz timings, but things change when you turn the level to regular timings, as already explained. >It is like developing AntiComp opening theory and there are not only >Trojans but some other methods I wont tell by now ;). I know few of them , e.g. Dutch Stonewall, Closed Positions , etc. >Eduard knows a bunch of them more...and I fear he wont show his best here. > Personally I'm looking forward to see these kind of techniques applied at regular timings, since to me it's a bit unfair to do it at blitz. I'm convinced that Eduard, as I said in previous post, will be in that sense of great help for the programmers , discovering the weaknesses of the programs and helping these programmers to improve their creatures. Regards. > >Regards, >Günther
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.