Author: George Tsavdaris
Date: 03:18:33 04/23/04
Go up one level in this thread
On April 23, 2004 at 05:51:56, Francesco Di Tolla wrote: >> Grow up! Why slander the SSDF because they don't use your toy of choice? >>Besides who wants to see another crafty tested ? Everyone already knows it's a >>third rate program not even close to the top, so what's the Point?? > >Actually I think so many copies of Shredder and Fritz do a lot o noise. > >Besides: the results show that there was no growth between version n and n+1 of >many engines in the last times: No. Let suppose the following is true: EngineA N+1 2818 34 -32 EngineA N 2809 24 -23 EngineB N+1 2782 27 -26 EngineB N 2702 24 -23 Then you say an Elo improvement of 2818-2809 = 9 ELO points is nothing. But that's not true because we don't know the exact improvement(if there is any). EngineA N+1 can have 2818+34 ELO points and EngineA N can have 2809-23 ELO points. That means a difference of 9+34+23 = 66 points. A rather satisfying improvement. Then you say an Elo improvement of 2782-2702 = 80 ELO points is good. But that's not true because we don't know the exact improvement(if there is any). EngineB N+1 can have 2782-26 ELO points and EngineB N can have 2702+24 ELO points. That means a difference of 80-26-24 = 30 points. Not so good after all. And all these exist with a probability of 95%. That means EngineA N+1 can be even 150 ELO points stronger than EngineA N. > >1 Shredder 8.0 CB 256MB Athlon 1200 MHz 2818 34 -32 481 70% 2673 >2 Shredder 7.04 UCI 256MB Athlon 1200 MHz 2809 24 -23 967 71% 2648 > >3 Deep Fritz 8.0 256MB Athlon 1200 MHz 2790 26 -25 855 72% 2625 >6 Deep Fritz 7.0 256MB Athlon 1200 MHz 2769 24 -24 898 66% 2655 > >7 Fritz 8.0 256MB Athlon 1200 MHz 2759 22 -22 1043 65% 2649 >10 Fritz 7.0 256MB Athlon 1200 MHz 2744 29 -28 634 63% 2653 > >14 Chess Tiger 14.0 CB 256MB Athlon 1200 2719 30 -30 557 61% 2640 >14 Chess Tiger 15.0 256MB Athlon 1200 MHz 2719 23 -22 983 58% 2659 > >20 Rebel 12.0 256MB Athlon 1200 MHz 2691 34 -33 442 56% 2647 >23 Rebel Century 4.0 256MB Athlon 1200 MHz 2677 29 -29 590 60% 2606 > >22 Ruffian 1.0.1 256MB Athlon 1200 MHz 2679 29 -29 565 53% 2658 >31 Ruffian 2.0.0 256MB Athlon 1200 MHz 2620 54 -55 165 48% 2635 > >27 Gandalf 4.32h 256MB Athlon 1200 MHz 2660 31 -31 514 53% 2637 >29 Gandalf 5.0 256MB Athlon 1200 MHz 2648 44 -45 250 43% 2697 > > >exceptions are: > 4 Junior 8.0 256MB Athlon 1200 MHz 2782 27 -26 750 67% 2654 > 18 Junior 7.0 256MB Athlon 1200 MHz 2702 24 -23 881 54% 2674 > > 9 Hiarcs 9.0 256MB Athlon 1200 MHz 2747 26 -26 744 61% 2667 > 21 Hiarcs 8.0 256MB Athlon 1200 MHz 2684 22 -22 996 53% 2663 > > > > >regards >Franz
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.