Computer ratings vs humans

This forum is for general discussions and questions, including Collectors Corner and anything to do with Computer chess.

Moderators: Harvey Williamson, Steve B, Watchman

Forum rules
This textbox is used to restore diagrams posted with the fen tag before the upgrade.
User avatar
Steve B
Site Admin
Posts: 10140
Joined: Sun Jul 29, 2007 10:02 am
Location: New York City USofA
Contact:

Post by Steve B »

Steve B

Interesting approach
can I ask you...
how did you verify you were really playing against a human?
scandien wrote:
I can't be sure that i plaid only humans of course! but I think that if a playe rwas using computer chess program , the older machine would have been really crushed ... So i am quite confident with this .
Sorry but i think you are missing the point
the fact that the dedicated computers won some games is not the point
its the games the dedicated computers LOST that are the issue
my guess is a high percentage of the losses were to players cheating with chess engines resulting in a lower overall rating for the dedicateds
keep in mind that a player does not have to use an engine for the entire match or even for an entire game
he can start out on his own then switch to an engine during crucial or highly tactical moments of the game ..have the engine make some moves then switch back
even some of the draws might be suspect
i have seen players start out a game playing on their own and then switch to an engine when way behind on material thereby getting a draw
if you can not verify that every one of the games the dedicated computers played were all against humans then you have an invalid result
you have a study with a mix of human/computer opponents
its for this reason that i never thought online play was a good idea for dedicated computers
the incidence of cheating on the servers is just too high

Polluted Rating Pool Regards
Steve
User avatar
scandien
Member
Posts: 206
Joined: Mon Sep 12, 2011 1:15 pm
Contact:

Post by scandien »

i am not agree with you but i think we cannot prove who is right !

I guess that if a guy is cheating and is using computer to play , then he will have a better level at FICS ( average level weas nearly 1800-1900 , is it a computer based level ?).

I checked manualy the games ( i get them and i can of course publish them if yo want), and i saw nothing really wrong with them....

And it is a fact that FIDE Rating and standard FICS rating are rather the same (for level under expert ) . So i don't think that FICS is really invade by cheater ( my feeling).

The differents matches were only performed to do a test and the result seems to be quite realistic and compliant with 80's result ;).


Just to get an idea about dedicated chess level ;)
Br

Nicolas
Reinfeld
Member
Posts: 486
Joined: Thu Feb 17, 2011 3:54 am
Location: Tacoma, WA

Post by Reinfeld »

This is a beautiful thread, with a wealth of thoughtful contributions.

Observations, no particular order:

1. I'm dying to see results of the tourney envisioned by Mike Watters:

http://www.chesscomputeruk.com/html/strong.html

2. I want more detail from Ivengo about this, and I want to see games:
Im at least equal vs Mephisto Atlanta, but Novag Star Diamond made me loose a lot with very few wins/draws, so I think I score about 15% against it...
3. I agree with M. Plastique about this:
My feeling (unfortunately not substantiated in any way) is that humans today are significantly stronger than similarly rated humans in years gone by when the influences of computer strength upon human play was not a significant factor - at least for club level players and higher (obviously not novices)
I don't think he has to substantiate. Computers teach discipline. They remind us to sweat the small stuff. They've taught us to be more vigilant about mid-level tactical forays. They've taught us how to manage the horizon effect. That's why I also agree with this point:
I myself admit to exploiting computer weakness to win (in fact just in the game I played last week where I left a pawn en prise knowing that if the computer took it, the resulting combination was outside of it's selective search horizon).
You know what? Winning is winning. If Tarrasch overvalued the two bishops and certain types of moves, Lasker was right to exploit it. It's not the *human's* fault that the dedicated never learns. Purity arguments on this front are bs. Winning is winning.

4. I agree with Steve on this matter:
Well 600 games seems a bit harsh to me
to this day the SSDF rating lists are one of the most highly respected rating lists in the world(not to mention the longest continuously published list)..many of their rating's are based upon less then 600 games
if you review their lists you can see that Generally speaking ...
600 games will produce a standard rating deviation of about +-30 Elo
200 games will produce a standard rating deviation of about +-50 Elo

for me...knowing that a dedicated chess computer has a rating that is accurate within 50 Elo pts is more then sufficient
5. Steve is also dead right on this bit - you can't track cheaters:
my guess is a high percentage of the losses were to players cheating with chess engines resulting in a lower overall rating for the dedicateds
keep in mind that a player does not have to use an engine for the entire match or even for an entire game
he can start out on his own then switch to an engine during crucial or highly tactical moments of the game ..have the engine make some moves then switch back
even some of the draws might be suspect
i have seen players start out a game playing on their own and then switch to an engine when way behind on material thereby getting a draw
if you can not verify that every one of the games the dedicated computers played were all against humans then you have an invalid result
you have a study with a mix of human/computer opponents
its for this reason that i never thought online play was a good idea for dedicated computers
the incidence of cheating on the servers is just too high
- R.
"You have, let us say, a promising politician, a rising artist that you wish to destroy. Dagger or bomb are archaic and unreliable - but teach him, inoculate him with chess."
– H.G. Wells
User avatar
ricard60
Senior Member
Posts: 1285
Joined: Thu Aug 09, 2007 2:46 pm
Location: Puerto Ordaz

Post by ricard60 »

I allways have been atracted by the relation between human elo (FIDE ELO) and dedicated chess machines elo´s. On this link and in the year 2007 a tournament between GM IM and FM against strong dedicated chess machines was held. Probably one of the strongest tournament ever between humans and dedicated chess machines. http://www.meca-web.es/torneos/reyes2007.htm

In this link there are other tournaments http://www.meca-web.es/torneos.htm

In the section (torneos celebrados en vivo-live tournaments) there are more tournaments between humans rated players and dedicated chess machines. Some time ago i took all the results from this site between human rated players and dedicated chess machines and got the following relation:

If a machine has 2000 elo its FIDE elo will be more or less 2070 this was from about 100 games on this site.
Post Reply