Author: George Tsavdaris
Date: 05:22:33 07/25/05
Go up one level in this thread
On July 25, 2005 at 07:48:29, Uri Blass wrote: >On July 25, 2005 at 06:26:18, George Tsavdaris wrote: > >>On July 25, 2005 at 03:57:50, Uri Blass wrote: >> >>>On July 25, 2005 at 03:47:13, Andre van Ark wrote: >>> >>>>Hi, >>>> >>>>Fabien has the advantage that he could learn from postings and experiences from >>>>other programmers and make use of existing programs like Crafty. >>>> >>>>The "programmers of the 90's" had to find out "the weel" by themselves, and that >>>>consumes a lot of time. >>>> >>>>Nevertheless, I take my hat off for Letouzey, and off course also all other >>>>chessprogrammers. >>>> >>>>André >>> >>>Fabien could learn from other programmers but the same for the commercial >>>programmers and I do not see big improvement in programs like Tiger2004 or Deep >>>Sjeng or Ruffian that are near the level of Fruit2 and not near the level of >>>Fruit2.1 >>> >>>I also believe that next fruit will be stronger than Shredder9(I do not say that >>>it will be stronger than next Shredder but of course SMK can learn from Fruit >>>when Fabien cannot learn from Shredder). >> >> Nah, it wont be stronger by any means from S9. I think it's not stronger even >>from S8 and i believe it's on the league of the S7.04......... >>I'm not speaking about head to head battles(where it is also not stronger from >>S9 & S8) but for competition with all engines....... > >Note that I talk about next fruit. >I agree that Fruit2.1 is weaker than Shredder9. > >Fabien get improvement near 100 elo every time he release a new fruit. >the difference between Fruit2.1 and Shredder9 is only near 50 elo. I can agree on this in head to head match, as it is around 57% performance for Shredder 9, but i think we are going to around +65 ELO in an overall comparison against other engines too.........OF cource to see Fruit 2.1 being stronger than Fritz 8 Bilbao and Junior 9/8 and Hiarcs 9,etc...was a shock for me and i could never believe that....I still can't. That's why i'm looking forward to Kurt's tests about Fruit 2.1 and for the future SSDF with Fruit 2.1 and for Leo's next edition............Then i will be convinced about this "magic" Fruit 2.1........ > >Simple math tell me that next fruit is going to be stronger than Shredder9(even >if Fabien get this time only 70 elo. Yes math tells this, but only IF this same progress continue and nothing can assure this and of course it's not easy to do. And i would be surprised if he can achieve more than 35-40 ELO impprovement in his next release, assuming a reasonable time for this, as if the next release is after 2 years i can expect such an increase...... Achieving +100 ELO in each release (or even +70) is nearly impossible. >Fabien finished the evaluation of Fruit but there is a lot to improve in the >search(for example extending every check and every single reply is a clear >mistake and I will be surprised if Fabien get less than 70 elo improvement by >this and by other tricks) > >Uri
This page took 0.02 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.