Computers think in zeros and ones while we humans waste our time thinking in tens and hundreds. Who knows how much more we could achieve if we just made things simpler and used binary?
Yes, it's apparent that the world would have evolved much faster, had the New Testament being written in binary form, and stored in a memory stick.
It would be interesting a match between Carlsen and Houdini. And Houdini has to play for the win only. If a draw occured the point would go for Carlsen.
What would happen there?!
Computers think in zeros and ones while we humans waste our time thinking in tens and hundreds. Who knows how much more we could achieve if we just made things simpler and used binary?
Yes, it's apparent that the world would have evolved much faster, had the New Testament being written in binary form, and stored in a memory stick.
...and then brought next to a large magnet. Maybe save the library of Alexandria?
I wonder why it isn't mentioned that when Ivonov cheated with Houdini that his GM opponent rated "only" 2600 got a drawn position and drew it easily... until Ivanov didn't have enough time to get help from the computer played a dumb move and the GM won.
It seems modern computers aren't as invincible as people tend to think. And this GM was "only" 2600.
rybka 4.1 beat the houdini ....
i see both engine match rybka 4.1 beat houdini 3
He isn't even guaranteed to beat strong grandmasters every time, in fact his recent tournament results aren't all that great so he has no chance against Houdini running on multiple cores. Its actually irrelevant because Houdini 'cheats' by having perfect access to the most up to date opening theory and endgame tablebases while playing, most of us could play better if we were looking things up as we went along.
if you think the opening book isn't a big deal try playing against an engine with the opening book turned off, they play 500 ELO worse.
You may have a point in that. But even if an engine with 3200 ELO is down 500 ELO without opening book it is still a formidable opponent
Taken from wikipedia: "These ratings, although calculated by using the Elo system (or similar rating methods), have no direct relation to FIDE Elo ratings or to other chess federation ratings of human players."
Engine ratings are relative to other engines, not humans. The time of the computer is near, but has not come yet.
The only way to beat Houdini is to close it down!
Stockfish 3 vs Houdini lost 23-25 in a long time control match. The match kept swinging back and forth though, and either side could have won in the end. Stockfish beat Houdini from both sides of the French!
Someone stop bumping this thread so that the painful title capitalization ceases to annoy.
Do you think Carlsen have any chances vs H3 at this setup below as black.
Well to offset the exchange H3 get the white pieces and bishop pair. So in in reality black is just around -0.75 pawn ahead from the initial setup.
I find it surprisingly that H3 Pro draws Critter 1.6a in such setup as above.
It would be interesting a match between Carlsen and Houdini. And Houdini has to play for the win only. If a draw occured the point would go for Carlsen.
What would happen there?!
That would be really interesting. Faced with such a challenge, the Houdini programmers would probably have to go through all available games played by Carlsen and try to figure out any patterns as to what kinds of mistakes Carlsen tends to make that might make him lose (especially with the white pieces), and then to program Houdini to try to steer Carlsen into positions where he will be especially likely to make such mistakes - even at the cost of taking some risk of Carlsen winning by capitalizing on the deviation from perfect play that this strategy forces Houdini into. After Carlsen learns how Houdini is programmed to try to lure him out onto slippery ice, Houdini will have to take even bigger risks, and Carlsen will probably win an increasing proportion of the games played, since humans are - as for now at least - better at spotting patterns than computers are. To defeat Carlsen more often than draw/lose, Houdini will then have to develop true human-level artificial intelligence of a broader kind than merely the good old chess calculations, so that it can hypnotize Carlsen into making crucial mistakes, as that will finally be the only way to make Carlsen lose more often than draw/win. To hypnotize a human into that, the computer probably has to develop general human-level AI and beyond, rather than "merely" an ability to play "perfect" chess.
He isn't even guaranteed to beat strong grandmasters every time, in fact his recent tournament results aren't all that great so he has no chance against Houdini running on multiple cores. Its actually irrelevant because Houdini 'cheats' by having perfect access to the most up to date opening theory and endgame tablebases while playing, most of us could play better if we were looking things up as we went along.
if you think the opening book isn't a big deal try playing against an engine with the opening book turned off, they play 500 ELO worse.
You may have a point in that. But even if an engine with 3200 ELO is down 500 ELO without opening book it is still a formidable opponent
not really cheating. Consider this, sure, houdini uses an opening book, guess it's an option? Anyways, It's data programmed into the engine to know how to play against an opponent.
Now, is it cheating is YOU study and, so-called, program openings into your data base(your brain)? Plus 500elo point difference? Like you really believe that's gonna make a huge difference?
I play against Houdini a lot. But not on it's strongest level. Why kid myself?? But it does have strength adjustments. I work my way up the elo rating for both white and black. I make sure I win on both sides before advancing to the next level. As of now, I'm trying to get over the 1950 elo.
I have an upcoming tournament and I need to program my data base before the meeting next month. I'm a student at MATC and we're playing against another collage. Wish me luck!
carlsen will lose against houdini
but GM Vishy Anand can easily beat houdini
Yes like carlsen crush Vishy :p
Maybe he can draw one game out of very many, but not win...
Absolutely impossible nowadays even for a 2800+
i think carlsen can draw houdini and no only carlsen few player on top can do it too
This reminds me of the movie "2001: A Space Odyssey". In the movie, computers had gotten so strong that it was virtually impossible for a human to beat them. The computers had to be programmed to "throw" away the occasional game to keep humans interested in playing.
I'm not sure what movie you're describing, but it's not 2001: A Space Odyssey.
One of the astronauts on the ship is shown playing a game of chess against
the HAL 9000 computer. You can find the entire game by googling it :)
The only way to beat the machine is to play unorthadox stratedgy. Kasparov did this against Big Blue. But, that was 20 years ago. Now, forget about it. Magnus would get crushed. Computers are better at chess than humans now and by a significant margin. He would tie some games but would lose a lot of them; even with White. Maybe, just maybe, he might win A game out of 15. And if he did win, A game, the programmers would 'fix' the gap in the program making it impossible to do it again.
Why is it sure that the best of computers have something to gain from humans? Let's say it's 500 years in the future and chess has been solved. Houdini 9000 plays perfectly. By definition, Houdini 9000 has nothing to gain from any human.
This can in fact be seen right now. Checkers has been solved, and Chinook plays perfectly. Chinook has nothing to gain from any human players. In this case, Chinook + Human = Chinook unless the human is stupid enough to play a non-Chinook move, in which case Human + Chinook < Chinook.
http://www.chessbase.com/newsdetail.asp?newsid=3997
EDIT: And Chinook also only thinks in binary.