Author: Charles Unruh
Date: 20:35:42 02/07/00
Go up one level in this thread
On February 07, 2000 at 21:26:35, John Warfield wrote: >On February 07, 2000 at 20:56:38, Charles Unruh wrote: > >>On February 07, 2000 at 20:42:26, John Warfield wrote: >> >>> >>> >>> Unless someone can produce some evidence to the contrary, I still believe >>>rebel is clearly better against humans than the other programs. I think the >>>combined aegon results as well as the GM Challenge prove this. Computer Vs >>>Computer chess proves very little, Evidence is that Annand was able to crush >>>fritz but could do very little against Rebel, Yet Fritz consistently defeats >>>Rebel, what does all this tell you? >> >>Well firstly i haven't seen Anand play Fritz in a 40/2. Further i seriously >>doubt(no offense intended) that you can get 5 people around here that will >>support the idea that Rebel is stronger than Fritz period in 5 minute games. >>Also from looking at actual games, Rebel falls not even for fantastically >>brilliant combos but really quite simple ones that are well in the capability of >>GM's. To me if you play good chess you wont get absolutely blown out as Rebel >>has in the cadaques tournament. > > > I don't think you are referring to Rebel Century! It is every bit as tactically >adept as Fritz, I am indeed referring to Rebel Century, as for rebel being tactically adept as fritz in your own words "Come on now!", The reason that rebel gets spanked so bad in comp vs comp is just that, it doesn't see tactics as good as most of the other programs, let alone seing the fantastic tactics of fritz. You are entitled to your opinion though of course. now I will admit that rebel 10 may have failed tactically but >certainly not Century. All enriques tournament shows is that Ed is correct about >the auto-player putting rebel at a disadvantage. I don't even really believe in this auto-232 problem since my rebel century plays the same moves as were played in enriques tournament, and i'm not using auto232. I don't want to get into that can of worms though. So we'll just go with evidence i seriously doubt that Century will do as well on the combinational portion of any test suite as fritz, and as mentioned above pretty much everyone will agree that Rebel loses precisely because of its tactical inabilities. > > > > I think Rebel is a good solid program, and >>definitely above 2500 on fast hardware (P500s and above). Recently Enrique said >>that J6 was the only program not to fall into line with his test suite, and this >>was because it's the only progam who's positional strength surpassed its >>tactical strength. I then told Enrique that "if Rebels positional strength >>didn't surpass its tactical strength then rebel was in trouble". Enrique >>replied that rebel was indeed in trouble :)(Though he made mention of Ed's >>patch). As an analysis engine of my games i think Fritz has the clearest >>vision.
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.