Hmm.
What will happen if chess got solved?
Sort:
tygxc wrote:
Only now in the Finals they have reached perfection: 112 draws out of 112 games.
Those games are incredibly far from being perfect.
Do you seriously believe that those guys would have even the slightest chance against the 2100 ICCF world champion assisted by an engine from the year 2100 running on a computer from the year 2100?
@181
"Those games are incredibly far from being perfect." ++ Based on what?
"Do you seriously believe that those guys would have even the slightest chance against the 2100 ICCF world champion assisted by an engine from the year 2100 running on a computer from the year 2100?" ++ Yes, they would draw just the same.
tygxc wrote:
@181
"Those games are incredibly far from being perfect." ++ Based on what?
Based on your answer on the next question. But your answer is completely insane, so I just give up at this point.
@184
"We do not know if chess is a win or a draw" ++ We know it is a draw.
"claiming draw rate as proof of perfect play"
++ 112 draws out of 112 games in the strongest chess of the planet.
"Draw rate would still not prove perfect play"
++ Not draw rate, but 112 draws out of 112 games.
@187
"Where is the proof chess is a forced draw"
- White has the initiative, an advantage of +1 tempo = + 1/3 pawn, not enough to win.
- Each further move dilutes the +1 tempo advantage: 0-0, 1-0, 1-1, 2-1, 2-2, 3-2, 3-3...
- 112 draws out of 112 games in the strongest chess on the planet, 17 different entities human+engine from 11 different countries playing, average 5 days/move, twin servers of 90 million positions/second each.
"I showed a match with 120 draws out of 120 games"
++ These may be perfect games. The white opening choice made it easy for black to draw.
That is another human contribution: choice of openings with the help of data bases.
"We do not know fact if chess is a win or a draw." ++ We know it is a draw.
Deslising wrote:
Chess can't be solved.
This is false. Look up "chess tablebases". That project has succeeded in solving a giant amount of chess positions by decades of intense computing work.
At the same rate, chess will be solved in roughy 10000000000000000 years. Or, as tyxcg would say, "in 5 years".
DesperateKingWalk wrote:
tygxc wrote:
@168
"Draws do not mean perfect play" ++ But perfect play means draw.
"the humans are a non factor" ++ The humans and the engines complement each other.
The engines calculate tactics, what humans are weak at. The humans take care of long term strategy and planning beyond the horizon of the engines, where engines are weak.
"the ICCF players just use the strongest chess engines"
++ Yes, they use engines, but the humans make the difference.
That is why 17 qualified through Preliminaries, Semifinals, Candidates to the World Championship Finals and most did not, though engines are about the same.
Even more: Russians have worse hardware because of sanctions,
but 4 of the 17 ICCF World Championship Finalists are Russian.
"That is why ICCF tournaments and players. Are not taken seriously"
++ You are not taken seriously to say so.
Sign up for an ICCF World Championship Preliminary and get yourself beaten.
"What are these ICCF ****** going to give to the computer that is rated 1000 Elo greater then Magnus Carlsen." ++ Long term strategy and planning beyond the horizon of the engine.
"Stockfish is not capable, or designed to play perfect chess" ++ But human + engine @ average 5 days / move now play perfect chess: 112 draws from 112 games.
Again that is why you are a low IQ B.S. artist. And you just make up things because you think that is how things work in the world of computer chess.
Let us just shut down your B.S. And do a demonstration of just how silly and easy it is achieve 112 draws. By showing 120 draws in computer match play at 1 minute +0.1 seconds.
I have been involved with computer chess since the early 1980's. And know and can demonstrate what a insane joke your claims are that draws means perfect play....
Hey look everyone my computer can play perfect chess at 1 second a move. I do not need 5 days.
Here is a match played between two recent versions of Stockfish at 1m + 0.1s. Match length was 120 games.
The results was 120 straight draws.....
Result:
------------------------------------------------------------------------------------
# name games wins draws losses score elo + -
1. Stockfish dev-20240705 120 0 0.0% 120 100.0% 0 0.0% 60 2700 21 21
2. Stockfish dev-20240706 120 0 0.0% 120 100.0% 0 0.0% 60 2700 21 21
Cross table:
------------------------------------------------------------------------------------
# name score games 1 2
1. Stockfish dev-20240705 60 120 x ========================================================================================================================
2. Stockfish dev-20240706 60 120 ======================================================================================================================== x
Tech:
------------------------------------------------------------------------------------
Tech (average nodes, depths, time/m per move, others per game), counted for computing moves only, ignored moves with zero nodes:
# name nodes/m NPS depth/m time/m moves time
1. Stockfish dev-20240705 9330K 9923049 42.7 0.9 52.0 48.9
2. Stockfish dev-20240706 9830K 10222808 43.6 1.0 52.0 50.0
all --- 9355K 10074538 43.1 1.0 52.0 49.4
Here are is the first and last game of the match....
Yup one of stockfishes main issues is it always goes for a draw if they can't see the winning moves stockfish doesn't realy know how to take riskes in order to win even if it's better than people at chess
tygxc wrote:
@181
"Those games are incredibly far from being perfect." ++ Based on what?
"Do you seriously believe that those guys would have even the slightest chance against the 2100 ICCF world champion assisted by an engine from the year 2100 running on a computer from the year 2100?" ++ Yes, they would draw just the same.
So you are saying the best chess playing ability today would draw the best chess playing ability from the year 2100. Which means in the next 75 years there will be no improvements to chess computers. We've reached the end of the road when it comes to chess.
To me that seems unlikely. But who knows, maybe you are right. Maybe it's not possible for technology to advance any farther.
@194
"the best chess playing ability today would draw the best chess playing ability from the year 2100" ++ Yes, at 5 days/move.
"in the next 75 years there will be no improvements to chess computers"
++ No, what they do today in 5 days in 75 years they will do in 5 milliseconds.
"Maybe it's not possible for technology to advance any farther." ++ It is possible to advance further: do the same faster and before 2100 strongly solve chess to a 32-men table base.
tygxc wrote:
@194
"the best chess playing ability today would draw the best chess playing ability from the year 2100" ++ Yes, at 5 days/move.
"in the next 75 years there will be no improvements to chess computers"
++ No, what they do today in 5 days in 75 years they will do in 5 milliseconds.
"Maybe it's not possible for technology to advance any farther." ++ It is possible to advance further: do the same faster and before 2100 strongly solve chess to a 32-men table base.
Apparently they already do that in bullet today and draw consistently I highly doubt computers at bullet don't miss stuff
We already know how they miss stuff in blitz games
It's not about them drawing consistently it's about an unknown amount of errors
@193
"stockfish doesn't realy know how to take riskes in order to win"
++ The human ICCF player knows how to take risks to try to win.
They play sharp stuff: Najdorf, Catalan, Open Ruy Lopez..., but this year all draws result.
tygxc wrote:
@193
"stockfish doesn't realy know how to take riskes in order to win"
++ The human ICCF player knows how to take risks to try to win.
They play sharp stuff: Najdorf, Catalan, Open Ruy Lopez..., but this year all draws result.
But then again if the same things happen in bullet games ...
@196
"We already know how they miss stuff in blitz games"
++ I do not talk about engine bullet or blitz,
I talk about the ICCF World Championship Finals at average 5 days/move.
tygxc wrote:
@196
"We already know how they miss stuff in blitz games"
++ I do not talk about engine bullet or blitz,
I talk about the ICCF World Championship Finals at average 5 days/move.
If bullet and blitz is usually and they draw every game in bullet what does that make classical chess dispite "better play "?
It probably means the engines have made errors we do not know about just like in faster time controls
tygxc wrote:
@194
"the best chess playing ability today would draw the best chess playing ability from the year 2100" ++ Yes, at 5 days/move.
"in the next 75 years there will be no improvements to chess computers"
++ No, what they do today in 5 days in 75 years they will do in 5 milliseconds.
"Maybe it's not possible for technology to advance any farther." ++ It is possible to advance further: do the same faster and before 2100 strongly solve chess to a 32-men table base.
Except that's not what has happened in the past. Do you think the very best chess playing ability from 1980 would draw the very best chess playing ability from 2020? In your world they would, because todays computers aren't better, they are just faster. So if a 1980 computer had enough time it would draw or match, the ability of todays computer, right?
Maybe someone with more computer knowledge than me could answer if that's what would happen. It shouldn't be too hard to dig up a computer from 1980, give it 5 days per move and see how it does against todays computers. Has anyone tried it?
@176
"you are a low IQ B.S. artist" ++ Speak for yourself.
"how things work in the world of computer chess" ++ No, how things work in ICCF.
"computer match play at 1 minute +0.1 seconds"
++ A blitz match is ridiculous. ICCF is average 5 days / move.
In previous years there were decisive games.
In Preliminaries, Semifinals, Candidates there are decisive games.
Only now in the Finals they have reached perfection: 112 draws out of 112 games.