so what is a fair GM handicap vs Rybka?

[DomLeste]

Like harvey said best solution is using old hardware. As i stated when kramnik lost to deep fritz.

My condiitons..

1. Book only up to 10 moves
2. Found a old processor 1GHz
3. Endgame databases limited to 4 pieces

No need for material or time handicaps please! Limit the technology like other sports have done to even up the competition.

[Uri Blass]

Maybe none of the above. Give the engine a time handicap and then we get a real game of Chess.

I agree. The best handicap is to cut the computer’s time in 1/2 or maybe 1/4.

I disagree.

I am even not sure if this is a handicap for the computer because the human can think less on the computer time.

rybka with 1/4 of the time beat easily chess programs that performed at 2700 level or better in tournaments.

I remember that old tiger and old shredder already got performance of more than 2700 some years ago.

Uri

Which tournament did rybka win with 1/4 of the time?
Computer vs computer is not the same as man vs computer.
The old tiger and old shredder performance of more than 2700 were not at handicap time.
Has there been man vs computer with time handicap?
Before we write off time handicap, lets have a man vs computer at ¼ the time and see what happens.
Mark

These tournaments were at relatively slower hardware than the hardware that rybka is using(tiger used only one processor) and I have no reason to think that computer vs computer is different than computer vs human.

If rybka with 1/4 of the time can beat tiger when both use one processor then I have no reason to think that the situation is going to be different against humans and that tiger is going to perform better against them.

Uri

Can you tell me where I can find more information about these time handicap tournaments?
When and where did they take place?
How many games were played?
What was the score?
Mark

You do not understand
The games were not handicapped time game but based on the fact that the hardware today is faster the programs did not play better than they can play today with 1/4 of the time with the best hardware so giving programs 1/4 of the time today is not going to make them weaker and only going to make the human weaker because the humans will have less time to think on the opponent move.

Uri

[ozziejoe]

..... just the same as you could never get strong enough to lift an elephant over your head with one arm. In theory there is no limit to the expansion of your myocytes, but no matter how much weight training occurs a man will never be able to lift a grown elephant over his head.

My dear christopher, have you so little faith in science? Already, we at university are begining to bread an elephant with a beagle. The result will be what we like to call a "beaglephant."" I can assure you that you will be able to lift this creature over your head, even with one arm

[mschribr]

Maybe none of the above. Give the engine a time handicap and then we get a real game of Chess.

I agree. The best handicap is to cut the computer’s time in 1/2 or maybe 1/4.

I disagree.

I am even not sure if this is a handicap for the computer because the human can think less on the computer time.

rybka with 1/4 of the time beat easily chess programs that performed at 2700 level or better in tournaments.

I remember that old tiger and old shredder already got performance of more than 2700 some years ago.

Uri

Which tournament did rybka win with 1/4 of the time?
Computer vs computer is not the same as man vs computer.
The old tiger and old shredder performance of more than 2700 were not at handicap time.
Has there been man vs computer with time handicap?
Before we write off time handicap, lets have a man vs computer at ¼ the time and see what happens.
Mark

These tournaments were at relatively slower hardware than the hardware that rybka is using(tiger used only one processor) and I have no reason to think that computer vs computer is different than computer vs human.

If rybka with 1/4 of the time can beat tiger when both use one processor then I have no reason to think that the situation is going to be different against humans and that tiger is going to perform better against them.

Uri

Can you tell me where I can find more information about these time handicap tournaments?
When and where did they take place?
How many games were played?
What was the score?
Mark

You do not understand
The games were not handicapped time game but based on the fact that the hardware today is faster the programs did not play better than they can play today with 1/4 of the time with the best hardware so giving programs 1/4 of the time today is not going to make them weaker and only going to make the human weaker because the humans will have less time to think on the opponent move.

Uri

Uri, I am not sure I understand. I think shorter sentences would be better.

Are you saying the games were not time handicapped? Today’s computers are faster. Giving the program ¼ the time on faster computers, it still plays as good as strong slower computer with full time. So it’s worse for the human. The human will get less time thinking on opponents time.

On second thought I don’t think making the computer weaker makes for an interesting match. More interesting would be to play full strength and removing a piece handicap.
Mark

[Harvey Williamson]

Maybe none of the above. Give the engine a time handicap and then we get a real game of Chess.

I agree. The best handicap is to cut the computer’s time in 1/2 or maybe 1/4.

I disagree.

I am even not sure if this is a handicap for the computer because the human can think less on the computer time.

rybka with 1/4 of the time beat easily chess programs that performed at 2700 level or better in tournaments.

I remember that old tiger and old shredder already got performance of more than 2700 some years ago.

Uri

http://www.chessbase.com/newsdetail.asp?newsid=3718

Looks like a time handicap match has just been played!?

[Father]

In NFL football, games are made theoretically even by give one team points. Example, a bad team may be given a 21 point handicap to make the contest even.
.
with similar time controls to the current match, what handicap would the GM need so the match is considered even?

...after taking a look at the past against the scope of the same question, to understand the present a little and catalyze the uncertain hypothetical future, I was completely surprised...

[lkaufman]

In NFL football, games are made theoretically even by give one team points. Example, a bad team may be given a 21 point handicap to make the contest even.
.
with similar time controls to the current match, what handicap would the GM need so the match is considered even?

...after taking a look at the past against the scope of the same question, to understand the present a little and catalyze the uncertain hypothetical future, I was completely surprised...

In the year 2007, Rybka on a good computer was already too strong for any human player with just a modest time handicap like 4 to 1. Fair odds vs. strong GMs (2600 FIDE level) was one pawn (Joel Benjamin narrowly lost a match of 8 games getting one pawn removed, all eight pawns in turn, alternating colors). That was "slow Rapid" (45' + 15"). So basically we've gone from one pawn to one knight against the same opponent under fairly similar conditions.

[DomL77]

I asked Grok 3, what hardware it would need to bring stockfish 17 down to super GM levels.

Rating: Stockfish 17 on a 486 DX-33 (33 MHz, 1 CPU, no tablebases) in classical chess would likely be ~2650–2750 Elo.

Reasoning: It’s crippled to ~12,000–15,000 nps, reaching 17–19 ply, losing ~800–900 Elo from its modern peak, but NNUE keeps it above 1990s engines and competitive with 1960s–70s GMs (e.g., Spassky’s 2690 peak). At 2700ish, it rivals Tal or Petrosian, not Kasparov (2851).

This assumes Stockfish 17 compiles and runs on a 486 (possible with tweaks, though RAM—e.g., 4 MB—might choke NNUE).

[Uri Blass]

I asked Grok 3, what hardware it would need to bring stockfish 17 down to super GM levels.

Rating: Stockfish 17 on a 486 DX-33 (33 MHz, 1 CPU, no tablebases) in classical chess would likely be ~2650–2750 Elo.

Reasoning: It’s crippled to ~12,000–15,000 nps, reaching 17–19 ply, losing ~800–900 Elo from its modern peak, but NNUE keeps it above 1990s engines and competitive with 1960s–70s GMs (e.g., Spassky’s 2690 peak). At 2700ish, it rivals Tal or Petrosian, not Kasparov (2851).

This assumes Stockfish 17 compiles and runs on a 486 (possible with tweaks, though RAM—e.g., 4 MB—might choke NNUE).

Grok3 is clearly wrong and stockfish17 at 12000-15000 nodes per second clearly lose less than 800-900 elo.

with these number of nodes it can calculate 1M nodes per move at tournament time control and we have data about stockfish15 with 1M nodes per move.

https://tcec-chess.com/bayeselo.txt

56 Stockfish_15_1M 3277 9 9 2250 52% 3261 55%
If you want super GM level you need to go down in the list and you have

78 Stockfish_15_30k 2781 10 10 2112 51% 2770 29%
81 Stockfish_15_10k 2532 12 12 2104 55% 2474 19%

Considering the fact that stockfish17 is probably better than stockfish15 you may need the hardware to be 100 times slower than 486 to get super GM strength.

[DomL77]

I asked Grok 3, what hardware it would need to bring stockfish 17 down to super GM levels.

Rating: Stockfish 17 on a 486 DX-33 (33 MHz, 1 CPU, no tablebases) in classical chess would likely be ~2650–2750 Elo.

Reasoning: It’s crippled to ~12,000–15,000 nps, reaching 17–19 ply, losing ~800–900 Elo from its modern peak, but NNUE keeps it above 1990s engines and competitive with 1960s–70s GMs (e.g., Spassky’s 2690 peak). At 2700ish, it rivals Tal or Petrosian, not Kasparov (2851).

This assumes Stockfish 17 compiles and runs on a 486 (possible with tweaks, though RAM—e.g., 4 MB—might choke NNUE).

Grok3 is clearly wrong and stockfish17 at 12000-15000 nodes per second clearly lose less than 800-900 elo.

with these number of nodes it can calculate 1M nodes per move at tournament time control and we have data about stockfish15 with 1M nodes per move.

https://tcec-chess.com/bayeselo.txt

56 Stockfish_15_1M 3277 9 9 2250 52% 3261 55%
If you want super GM level you need to go down in the list and you have

78 Stockfish_15_30k 2781 10 10 2112 51% 2770 29%
81 Stockfish_15_10k 2532 12 12 2104 55% 2474 19%

Considering the fact that stockfish17 is probably better than stockfish15 you may need the hardware to be 100 times slower than 486 to get super GM strength.

Thanks for the clarification


Dom