Next Komodo vs. GM handicap match - "Opening Blunders&q

[JJJ]

Komodo score 3 points. I thought it could have been 3,5. But I was close.

Human won't win any single game without pieces advantage. Position advantage are way too easy for computer to deal with because humain won't practice 6 month with a specific opening advantage to convert the win.

[lkaufman]

Hi Larry,

shortly an idea in additional for you ...

John Shaw wrote a book to the Topic: The King's Gambit. I am absolutely sure you know about it.

We know that the King's Gambit is complicated and most grandmasters thinking generally the opening is bad. Often in interesting lines engines started with more as +- 2 eval.

Fact is, that many of good known lines are playable and John Shaw wrote a lot about these lines. Grandmaster waiting many years of the book John created because they know that he is working many years on his book. John collected all these interesting material to "King's Gambit" and so many interesting comments you can find here.

One of the most populare newer books about chess opening in the last years.

ISBN Paperback: 978-1-906552-71-8
ISBN Handkower: 978-1-906552-74-9

I think it can be a good idea to contact John for his opinion to the most of critical "The Kings Gambit" lines.

Sure about it that this topic in combination with your idea will reach that a lot of very strong players will follow the games between Komodo vs. GM. In my opinion it's much more interesting to create such matches with opening lines, generally in focus between the grandmasters. The "blunders" idea are nice too but I am sure that such an idea will go around the Grandmaster World.

Furthermore, I love the book John Shaw wrote. It's my favorite chess book and I try out in selfplay vs. Zarkov and AnMon on my Notebook a lot of such lines. Generally, I like Kings Gambit and like the good old school chess. With your energy, maybe in combination with helps by John Shaw you can be sure that "King's Gambit" will get a Booster and maybe a resurrection can be celebrate.

Best for your and thanks for the work you do for us!

Best
Frank

The problem is that the lines you are talking about are undoubtedly too good for a Komodo vs GM match. What we saw in this match and others is that even playing ridiculous moves and outright blunders is only enough to give GMs drawing chances, so any gambit that can plausibly be called playable would just be too good for a competitive game. I think that interesting gambit lines are most suitable for engine vs. engine games with colors switched. Even if they are slightly dubious, as long as the inferior side has decent drawing chances they are fine for such "reversal" testing.
It seems that only with a two nonedge pawn handicap or the equivalent can an ordinary GM hope for a win against Komodo.

[lkaufman]

Komodo score 3 points. I thought it could have been 3,5. But I was close.

Human won't win any single game without pieces advantage. Position advantage are way too easy for computer to deal with because humain won't practice 6 month with a specific opening advantage to convert the win.

Well, don't forget that GM Petr Neuman scored three wins and two draws at the two pawn handicap c2 + f2, so clearly a piece handicap isn't needed. But it does appear that at least two non-edge pawns or the equivalent is the minimum handicap for an ordinary GM to win games. The only other GM win was by Simon Williams at odds of pawn f7 plus three moves, which is more than a two pawn handicap. I still find it strange that Komodo lost so badly to Neuman, a rather low-rated GM, but never lost another GM match.

[JJJ]

Komodo score 3 points. I thought it could have been 3,5. But I was close.

Human won't win any single game without pieces advantage. Position advantage are way too easy for computer to deal with because humain won't practice 6 month with a specific opening advantage to convert the win.

Well, don't forget that GM Petr Neuman scored three wins and two draws at the two pawn handicap c2 + f2, so clearly a piece handicap isn't needed. But it does appear that at least two non-edge pawns or the equivalent is the minimum handicap for an ordinary GM to win games. The only other GM win was by Simon Williams at odds of pawn f7 plus three moves, which is more than a two pawn handicap. I still find it strange that Komodo lost so badly to Neuman, a rather low-rated GM, but never lost another GM match.

Sorry, by piece, I meant at least a pawn. I agree with you. Human can maybe win with pawn advantage but not position advantage.

[Nordlandia]

To Larry Kaufman:

Many of the audience want to witness Komodo playing with time odds.

Someone proposed Carslen with 10 hours and Komodo 20 minutes.

Just to put it into perspective.

This is based on the 1/30 time odds formula.

[JJJ]

To Larry Kaufman:

Many of the audience want to witness Komodo playing with time odds.

Someone proposed Carslen with 10 hours and Komodo 20 minutes.

Just to put it into perspective.

This is based on the 1/30 time odds formula.

I m not sure Carslen would play with 10 hours thinking

[Frank Quisinsky]

Hi Larry,

OK, absolutely evident.
Indeed a bad idea from myself.

It's an other niveau with different things I do at home (means with my Kings Gambit experiments vs. "weaker" engines in selfplay). But such things are very nice for the work on my opening book. I love openings I don't understand for self playing. Must do that or I don't understand my own opening book I am working since a longer time.

I follow your events beside my work on my FCP Rating List.
Thanks for it ...

Best
Frank

[whereagles]

Many of the audience want to witness Komodo playing with time odds.

Time odds would probably see Komodo playing a couple suboptimal moves. Not sure that's what you want. Material/moves odds make it play as strongly as it is possible while maintaining it a challenge for both competitors.

But yeah it might be interesting to see that. Current "komodo odds match" time format might favor something like 60 min vs 1 or 2 min (plus increments), for instance.

My gut feeling is 60-2 might be too good for K and 60-1 just about par.

[Nordlandia]

My gut feeling is 60-2 might be too good for K and 60-1 just about par.

Against Carlsen / Nakamura?

With or without ponder in addition to increment.

These factors also have to be taken into consideration.

[lkaufman]

To Larry Kaufman:

Many of the audience want to witness Komodo playing with time odds.

Someone proposed Carslen with 10 hours and Komodo 20 minutes.

Just to put it into perspective.

This is based on the 1/30 time odds formula.

We might consider big time odds (maybe 2 hours + 1 min vs 2 min + 1 sec) combined with White pieces for human, no opening book (except maybe one or two moves for variety), and no ponder. But this only makes sense if we get another top-ten player like Nakamura, and even then Komodo would be heavily favoured unless we also specify Komodo must run on laptop computer.