TalkChess.com

Our next GM handicap match will be in four days, Dec. 16 and 17, on chess.com tv. GM Simon Williams (FIDE 2439 but FIDE rapid 2535, age 36), who will be talking about the game live as did GM Robert Hess. The handicaps will be one game at pawn (f7) and THREE moves (1.e4,2d4, White to move), one at pawn (f7) and two moves (same as three moves but Black to move), and two at Exchange and Move (remove b1 knight and a8 rook, White rook moved to b1). The only change from the Hess match (in which all four games ended in draws) is the one game where the handicap was increased from pawn & two to pawn & three moves. As usual, 45' + 15", Komodo running on my 24 core machine with latest version and a small handicap opening book I made.
We've learned from these matches that it's fairly easy for a GM to draw with a significant handicap, but very hard to win. No one has won a single game from Komodo at less than two pawn or knight odds in 17 tries (counting Exchange as less than two pawns), scoring 9 draws and 8 losses. But f7 and three moves is a huge handicap, more than two pawns, so Williams should win this one, and he would win the match if he does this and draws the other three.

I'll side with the GM winning this match.

Giving three moves is extremely ambitious.

Graham Banks wrote:I'll side with the GM winning this match.

Giving three moves is extremely ambitious.

Yes it is, but it's only one of the four games. Chess.com thought the audience would like to see the human win a game for a change, and since we can't even successfully give knight odds at this time control to FMs, pawn and three moves seemed like the most reasonable way to make a human win likely without it being a slam-dunk.

lkaufman wrote:

Graham Banks wrote:I'll side with the GM winning this match.

Giving three moves is extremely ambitious.

Yes it is, but it's only one of the four games. Chess.com thought the audience would like to see the human win a game for a change, and since we can't even successfully give knight odds at this time control to FMs, pawn and three moves seemed like the most reasonable way to make a human win likely without it being a slam-dunk.

Win what? The human isn't really winning anything. It's like taking a Lamborghini and making it move at 2km to 10km per hour; even a guy riding a bicycle could beat it easy. The example is a little bit absurd of course but still. This whole handicap thing is more like trying to make humans feel like they are still relevant versus an engine. They don't even belong playing with engines. I really don't get this stuff. Humans really just belong playing versus other humans unless in sparring\training contests. What's the point in victory when you handicap the engine to the point where its position has absolutely nothing good to be had? The great beauty in engine play isn't just depth and tactical prowess; it is staying away from allowing opponents to have good positions against them. They literally force their inferior foes into losing positions without mercy.

APassionForCriminalJustic wrote:

lkaufman wrote:

Graham Banks wrote:I'll side with the GM winning this match.

Giving three moves is extremely ambitious.

Yes it is, but it's only one of the four games. Chess.com thought the audience would like to see the human win a game for a change, and since we can't even successfully give knight odds at this time control to FMs, pawn and three moves seemed like the most reasonable way to make a human win likely without it being a slam-dunk.

Win what? The human isn't really winning anything. It's like taking a Lamborghini and making it move at 2km to 10km per hour; even a guy riding a bicycle could beat it easy. The example is a little bit absurd of course but still. This whole handicap thing is more like trying to make humans feel like they are still relevant versus an engine. They don't even belong playing with engines. I really don't get this stuff. Humans really just belong playing versus other humans unless in sparring\training contests. What's the point in victory when you handicap the engine to the point where its position has absolutely nothing good to be had? The great beauty in engine play isn't just depth and tactical prowess; it is staying away from allowing opponents to have good positions against them. They literally force their inferior foes into losing positions without mercy.

It's just entertainment.

GM Williams won the first game at pawn (f7) and three move handicap today, as many expected. Second game at Exchange and White handicap was drawn, a very exciting game. Tomorrow pawn and two moves and then another Exchange odds game.

Larry & Mark,

I watched the the second game. It was so much fun! I thoroughly enjoyed it. Simon Williams gave a great performance on (and off) the board. It was amazing to hear his real-time thought process. I thought at one point he was going to blow it but after some thought he found the only move (Rb3).

High addictive viewing!

Thanks,

Steve

lkaufman wrote:GM Williams won the first game at pawn (f7) and three move handicap today, as many expected. Second game at Exchange and White handicap was drawn, a very exciting game. Tomorrow pawn and two moves and then another Exchange odds game.

Steve Maughan wrote:Larry & Mark,

I watched the the second game. It was so much fun! I thoroughly enjoyed it. Simon Williams gave a great performance on (and off) the board. It was amazing to hear his real-time thought process. I thought at one point he was going to blow it but after some thought he found the only move (Rb3).

High addictive viewing!

Thanks,

Steve

lkaufman wrote:GM Williams won the first game at pawn (f7) and three move handicap today, as many expected. Second game at Exchange and White handicap was drawn, a very exciting game. Tomorrow pawn and two moves and then another Exchange odds game.

Yes, he was quite entertaining, as was the game itself. The game uncovered a book-related problem which we have already fixed. I think these games are more helpful for development than engine vs engine games, because I can understand what's going on much better.

Graham Banks wrote:
Giving three moves is extremely ambitious.

You can't do anything major with 3 moves. But with 4 moves you can get a rook, or even the queen and pawn for a bishop.
This is a funny post. (my post) Still, it's true that anything up to 3 first moves is no damage.

S.Taylor wrote:

Graham Banks wrote:
Giving three moves is extremely ambitious.

You can't do anything major with 3 moves. But with 4 moves you can get a rook, or even the queen and pawn for a bishop.
This is a funny post. (my post) Still, it's true that anything up to 3 first moves is no damage.

How do you win a queen and pawn for bishop with four moves (without crossling midline, else e3,bd3 Qh5+ leads to mate. )) and no f7 pawn? I only know how to win the Exchange and a pawn. Probably with three moves you can pretty much prove winning a second pawn.