Author: Derek Paquette
Date: 14:12:35 08/31/05
Go up one level in this thread
On August 31, 2005 at 17:04:32, Uri Blass wrote: >On August 31, 2005 at 16:45:06, Derek Paquette wrote: > >>On August 31, 2005 at 15:41:34, Ray Banks wrote: >> >>>On August 31, 2005 at 15:32:58, Ray Banks wrote: >>> >>>>Seems with the exception of the new Zappa which isn't yet available, you've >>>>already tested against the strongest engines ? So perhaps Toga will remain >>>>completely unbeaten >>> >>>Perhaps Fritz 9 will be able to do the job ? >>>Hiarcs 10 perhaps too when released >> >>I honestly don't see it. >> >>Hiarcs would have to add another 100+ elo too its strength, and judging from the >>current trend of deep searching engines (shredder zappa toga junior) which are >>ontop of the world right now, hiarcs who doesn't manage to search past depth 12 >>in a 2 hour game is not going to do it in my opinion, > >If you assume that the programmer will learn nothing from Fruit-toga that is >free source code program, Well the hiarcs team have been programming this hiarcs 10 for two years, hiarcs 9.6 has been out for awhile, fritz has completely re-developed their program from scratch and then BANG, fruit comes out and now everyone can look and see how it works, these programers are going to have to go BACK and do it over again, not from scratch but they worked so hard, maybe they just want to get another version out. > >> >>I honestly don't think superior knowledge is the way of the game anymore, it >>seems that search depth and spectacular pruning is doing it for programmers. > >You cannot get pruning and search depth without knowledge. >Fruit-Toga has a lot of knowledge in the code that smart programmers may learn >from it. thats true. > >Knowledge is of course not only about evaluation. > > As >>for fritz 9, they are also behind the times, their track record is horrible, >>fritz 7 to deep fritz 7 to fritz 8 to deep fritz 8, very small elo gains, > >Again there was no strong free source code that they could learn from it in the >time of Fritz7 and Fritz8. > Well that is true, but that is why we call them professional programs, you expect THEM to be better, except they aren't now, so professional no longer has a place with playing strength. >> >>my bet if I were to make one is that hiarcs and fritz won't even have enough >>mustard to beat Shredder 9. > >I will not talk specifically about Fritz and Hiarcs but my guess is that we are >going to see clear improvement in the level of chess programs thanks to Fabien. > >Uri yes i agree, but these guys have THEIR own programs, i'm sure they'll add, and improve but a 100 elo improvement ontop of their current program architecture, we'll see!
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.