Will computers ever solve chess? - Chess Forums - Page 68

ptd570 · 2014-01-22T11:57:20-08:00

Will there ever be a computer strong enough to solve chess to the point where white uses its half tempo advantage to always beat black no matter what moves black plays (in otherwords the same computer can never win with black even after a thousand random games against itself) I beleive one day the

ponz111

Jul 10, 2017

0

#1341

JeffGreen333 wrote:

Now I'm tempted to open an alt account and test this theory. I could learn the main lines of the Grob's Attack and start every single game with 1. g4. I'm curious as to what my rating would be, compared to what it is now. I bet I would still beat most players under 1600, opening with 1. g4. I'd probably struggle with players in the 1600-1900 range though. Just a guess. However, I'm really working hard to improve my 1. d4 2. c4 openings right now and don't really want to waste time doing that. It would be a fun experiment though.

All you could possibly prove is that fairly weak players do not know the best lines after 1. g4?

JeffGreen333

Jul 10, 2017

0

#1342

ponz111 wrote:

JeffGreen333 wrote:

Now I'm tempted to open an alt account and test this theory. I could learn the main lines of the Grob's Attack and start every single game with 1. g4. I'm curious as to what my rating would be, compared to what it is now. I bet I would still beat most players under 1600, opening with 1. g4. I'd probably struggle with players in the 1600-1900 range though. Just a guess. However, I'm really working hard to improve my 1. d4 2. c4 openings right now and don't really want to waste time doing that. It would be a fun experiment though.

All you could possibly prove is that fairly weak players do not know the best lines after 1. g4?

I bet I could beat you with it.

JeffGreen333

Jul 10, 2017

0

#1343

Never mind. You'd probably cheat by using Stockfish. Just to try to one-up me.

penandpaper0089

Jul 10, 2017

0

#1344

The danger with 1.g4 is that playing normally tends to lead to Black needing to sac a pawn for an initiative which isn't all that natural for some folks. 1.g4 d5 2.Bg2 c6 is possible I guess but it's easy to just go for the pawn and then miscalculate somewhere. I remember watching Timur Gareev completely rout an IM in this opening in blindfold blitz! Jan Gustafson was just relaying the moves to Gareev and Gareev just beat the IM easily. Black ended up in some +1 position in no time. Gareev's blindfold blitz games can be found on chess24's youtube channel.

ponz111

Jul 10, 2017

0

#1345

JeffGreen333 wrote:

In a previous comment, I also mentioned 2. f3. They are both Fool's Mate, so I didn't think I had to mention it again. lol Now I'm convinced that you just want to argue with me and have nothing of value to add to this topic. So, you are the fool and I just checkmated you. The Bible says, in Proverbs, Go from the presence of a foolish man,
When you do not perceive in him the lips of knowledge.

Oh, so now you call me a "fool"? Guess that is better than how you were calling me a liar before?

And i do not really care what the bible says--chess.com says we are not to talk about religion in these forums. If you wish to discuss religion please go to "Open Discussion".

ponz111

Jul 10, 2017

0

#1346

penandpaper0089 wrote:

The danger with 1.g4 is that playing normally tends to lead to Black needing to sac a pawn for an initiative which isn't all that natural for some folks. 1.g4 d5 2.Bg2 c6 is possible I guess but it's easy to just go for the pawn and then miscalculate somewhere. I remember watching Timur Gareev completely rout an IM in this opening in blindfold blitz! Jan Gustafson was just relaying the moves to Gareev and Gareev just beat the IM easily. Black ended up in some +1 position in no time. Gareev's blindfold blitz games can be found on chess24's youtube channel.

I agree that 1. g4 can be very tricky. It would catch a lot of people who are not prepared.

SmyslovFan

Jul 10, 2017

0

#1347

Has anyone actually bothered to check the variations that the engine is spitting out to give its -.80 eval? There are probably improvements that an engine could find in those lines that will help White.

Yeah, I believe (I don't claim any special knowledge) that 1.g4 isn't losing by force. I don't have a high opinion of it, but my bet is that 1.g4 scores worse than 1.f3 because it's slightly more forcing and tactical. That doesn't make it worse than 1.f3 tho, unless there's a demonstrable win.

I'll wait to see the lines.

SmyslovFan

Jul 10, 2017

0

#1348

Addendum: I just plugged 1.g4 d5 (the computer's first choice) 2.Bg2 into the computer, and the engine dropped the evaluation. I continued just a couple more moves down the main line, 2...Bxg4 3.c4 c6 and the comps still didn't see more than a nice edge for Black.

IM pfren

Jul 10, 2017

0

#1349

SmyslovFan έγραψε:

Addendum: I just plugged 1.g4 d5 (the computer's first choice) 2.Bg2 into the computer, and the engine dropped the evaluation. I continued just a couple more moves down the main line, 2...Bxg4 3.c4 c6 and the comps still didn't see more than a nice edge for Black.

-1.40 isn't exactly "a nice edge for Black". And it isn't even "Black has some compensation for the pawn he won", since in the bust line Black returns the pawn.

SmyslovFan

Jul 10, 2017

0

#1350

Pfren, at what depth to you get to -1.40?

Also, there's the passive but possibly playable 1.g4 d5 2.h3 h5 for White. Not great, but not sacking a pawn.

ponz111

Jul 10, 2017

0

#1351

i am 99.99% sure that chess is a draw with optimum play.

i am 74% sure that White should lose after 1. g4

JeffGreen333

Jul 10, 2017

0

#1352

SmyslovFan wrote:

Addendum: I just plugged 1.g4 d5 (the computer's first choice) 2.Bg2 into the computer, and the engine dropped the evaluation. I continued just a couple more moves down the main line, 2...Bxg4 3.c4 c6 and the comps still didn't see more than a nice edge for Black.

Yeah, but that's against a computer. Grob's Attack was created to be played against human beings, not computers (computers didn't exist back in the 1800's). It is specifically meant to take advantage of human weaknesses. It wouldn't work against a computer though, due to it's purely tactical nature (computers excel at tactics). Computers will always see the traps coming. Humans don't. The same could be said of the Cambridge Springs Defense trap. If a human being has never seen it before, they will most likely fall for it the first time. A computer would see it though. I believe that computers have also refuted the King's Gambit, which was played regularly by GM, many years ago.

JeffGreen333

Jul 10, 2017

0

#1353

ponz111 wrote

A - 0.80 disadvantage does not equal a loss by force (against a human being). it does against me and i am a human being. Computers don't count, since they cheat (moving the pieces on multiple, digital boards). computers are machines and do not cheat--they do not know how to cheat. They do not have a mind of their own as of yet.

Even a +3.0 advantage doesn't force a win. K+B vs K is a draw, for example. So, you're full of it, like usual. And yes, computers were programmed to cheat. The way they play chess is considered cheating (by FIDE rules) because they move the pieces on several digital boards, compare the different positions resulting from that and virtually "take back" lots of moves, as a result. They also have complete opening and endgame databases stored inside of them, on chips. So, they cheat by looking at "crib notes" and by violating the touch move rule (since they play out entire sequences in advance, on digital boards, only choose the best one and discard the rest). That's why computers can't be entered into human tournaments. It's an unfair advantage.

SmyslovFan

Jul 10, 2017

0

#1354

JeffGreen, this thread is about perfect play.

ponz111

Jul 10, 2017

0

#1355

s23bog wrote:

There are a lot of opinions expressed, here. Seems strange for something that should be a clear and decisive solution/proof. How can this be actually solved, rather then opined about?

It cannot be "solved" at least not in the lifetimes of humans on earth.

However there is a ton of evidence that chess is a draw with optimum play.

Thus, with the evidence, we are entitled to give an opinion. i am 99.99% sure that chess is a draw with optimum play.

ponz111

Jul 10, 2017

0

#1356

Computers have not refuted the Kings Gambit--there is one line of the Kings Gambit that draws for White.

ponz111

Jul 10, 2017

0

#1357

s23bog wrote:

A 99.99999999% opinion is about as worthless as a 50% opinion.

i am 99.99999999% sure that the sun will not explode in the next 24 hours.

This opinion is based on a whole lot of evidence and is worth more than somebody's 50% opinion.

ponz111

Jul 10, 2017

0

#1358

vickalan

Jul 10, 2017

0

#1359

ponz111 wrote:

s23bog wrote:

A 99.99999999% opinion is about as worthless as a 50% opinion.

i am 99.99999999% sure that the sun will not explode in the next 24 hours.

This opinion is based on a whole lot of evidence and is worth more than somebody's 50% opinion.

That's a good statistical observation. As for solving chess, solved is solved, and everything else is not solved.

JeffGreen333

Jul 10, 2017

0

#1360

ponz111 wrote:

s23bog wrote:

There are a lot of opinions expressed, here. Seems strange for something that should be a clear and decisive solution/proof. How can this be actually solved, rather then opined about?

It cannot be "solved" at least not in the lifetimes of humans on earth.

However there is a ton of evidence that chess is a draw with optimum play.

Thus, with the evidence, we are entitled to give an opinion. i am 99.99% sure that chess is a draw with optimum play.

Me too. At least we agree on this, which is the main point of the thread anyway.