lkaufman wrote: Probably the most practical match we could hold from the normal starting position against a top GM would be Komodo on a cellphone (similar to throttling), playing Black, and with no or just three-move book. There is at least some commercial and spectator interest in a cellphone match, much more than in a "throttling" match.
lkaufman wrote:Probably the most practical match we could hold from the normal starting position against a top GM would be Komodo on a cellphone (similar to throttling), playing Black, and with no or just three-move book. There is at least some commercial and spectator interest in a cellphone match, much more than in a "throttling" match.
A phone, coupled with a time handicap, to reduce match length, sounds good. But you'll probably want to use one of the cheapest (slowest) ones.
Engin wrote:such odd handicap games are total irrelevant for chess, i will even say its nonsense chess.
its making more sense to me if one grandmaster is want to play against a similar strength engine like about 2700 elo instead of 3300 elo.
It sounds to me like you would like to see Komodo on a cellphone vs. a 2700 GM in normal chess, perhaps with GM getting White, double time, and perhaps no or minimal book for Komodo. Perhaps we might try that someday.
Engin wrote:such odd handicap games are total irrelevant for chess, i will even say its nonsense chess.
its making more sense to me if one grandmaster is want to play against a similar strength engine like about 2700 elo instead of 3300 elo.
It sounds to me like you would like to see Komodo on a cellphone vs. a 2700 GM in normal chess, perhaps with GM getting White, double time, and perhaps no or minimal book for Komodo. Perhaps we might try that someday.
My understanding is that such a match would not be competitive. Komodo 8 on a cellphone beat Shredder 10 on an i7 5-1 (http://en.chessbase.com/post/komodo-8-t ... -challenge). This suggests Komodo on a cellphone still plays at >3000 strength. I do not think the stated bonuses for the human would be enough to close this gap.
Engin wrote:such odd handicap games are total irrelevant for chess, i will even say its nonsense chess.
its making more sense to me if one grandmaster is want to play against a similar strength engine like about 2700 elo instead of 3300 elo.
It sounds to me like you would like to see Komodo on a cellphone vs. a 2700 GM in normal chess, perhaps with GM getting White, double time, and perhaps no or minimal book for Komodo. Perhaps we might try that someday.
My understanding is that such a match would not be competitive. Komodo 8 on a cellphone beat Shredder 10 on an i7 5-1 (http://en.chessbase.com/post/komodo-8-t ... -challenge). This suggests Komodo on a cellphone still plays at >3000 strength. I do not think the stated bonuses for the human would be enough to close this gap.
Maybe it has to be on a cheap cellphone, and perhaps throw in no pondering on opponent's time and no tablebases. And maybe the opponent needs to be stronger than 2700.
Being able to showcase the 'power' of one's device by comparing it to deep blue 'a specialized supercomputer...' is something a broader range of companies than just cell phone mfg's would be interested. quick brainstorming:
- arduino https://www.arduino.cc/
- raspberry pi: https://en.wikipedia.org/wiki/Raspberry_Pi
- microprocessor companies
- VIA (company making eco-friendly processors)
- companies making embedded systems
- companies making little computers http://www.tinygreenpc.com/
Jesse Gersenson wrote:Being able to showcase the 'power' of one's device by comparing it to deep blue 'a specialized supercomputer...' is something a broader range of companies than just cell phone mfg's would be interested. quick brainstorming:
- arduino https://www.arduino.cc/
- raspberry pi: https://en.wikipedia.org/wiki/Raspberry_Pi
- microprocessor companies
- VIA (company making eco-friendly processors)
- companies making embedded systems
- companies making little computers http://www.tinygreenpc.com/
Basically anything playing Komodo at roughly 0.5 million nodes per move (hardware speed x time control) is competitive with top GMs at 45' + 15'' on regular chess. One can improvise on time control and hardware.
Although the poll shows sentiment about equally divided between Komodo winning the match and Movsesian drawing or winning it, I'm becoming a bit concerned because I'm running a simulation of the f7 games, on four cores, rotating among what I consider the best three first moves for Black (vs. 1e4). White gets 1' + 1", Black gets 30' + 30", which I estimate is about a 450 elo time handicap (I'll pin this down later). So far the results are seven wins for White, six draws, and no losses, which is +209 elo. In the Movsesian match Komodo will be running on 23 cores (keeping one for gui and everything else), so it will be stronger than in my current 4 core test, and Movsesian with 45' + 15" is probably not as strong as 4 core Komodo with 1' + 1", but still the results so far imply that the handicap is about 660 elo. If we count Movsesian as 2700 (a compromise between his current 2658 regular FIDE rating and his 2729 Rapid FIDE rating, since the match time limit is in between standard and rapid), Komodo will have to play above 3350 level to split the four f7 games, based on this simulation. I know that some of the rating lists do have Komodo above that level, but a lot of people doubt whether such ratings are really realistic by FIDE standards.
Human are smarter than computer, so 660 elo can be more vs human. I already said that a 2700 human can easy draw and probably win vs the f7 pawn handicap.
This time the human is high rated, and has time to get prepare. So I except him to win at least one game at this handicap, and draw every others games.
