Why there is no interest in Computer with odds Vs Humans match?

Discussion of anything and everything relating to chess playing software and machines.

Moderators: bob, hgm, Harvey Williamson

Forum rules
This textbox is used to restore diagrams posted with the [d] tag before the upgrade.
Chessqueen
Posts: 634
Joined: Wed Sep 05, 2018 12:16 am
Full name: Nancy M Pichardo

Re: Why there is no interest in Computer with odds Vs Humans match?

Post by Chessqueen » Wed Jul 03, 2019 2:44 am

Nordlandia wrote:
Wed Jul 03, 2019 2:10 am
The world champions pride is too high to be hurt by losing to Komodo. I don't know if team Komodo have considered a match with him. The assumption that he decline the invitation is clearly there though. Alternatively Maxime Vachier-Lagrave or Caruana is also viable alternative.
GM Caruana will definitively decline, the only top GMs that are NOT Afraid of playing vs Engines are Nakamura and MVL and very soon engine will be able to give 2 pawns odds. Another possibility would be to offer Chinese Young GM like Wei Yi FIDE rated 2741 who is hungry for Money to play some odds game vs Komodo and more likely would be rated in the top 10 very soon or by the end of next year :roll:

Uri Blass
Posts: 8611
Joined: Wed Mar 08, 2006 11:37 pm
Location: Tel-Aviv Israel

Re: Why there is no interest in Computer with odds Vs Humans match?

Post by Uri Blass » Wed Jul 03, 2019 3:57 am

Chessqueen wrote:
Wed Jul 03, 2019 12:29 am
lkaufman wrote:
Tue Jul 02, 2019 10:13 pm
lkaufman wrote:
Tue Jul 02, 2019 8:19 pm
Chessqueen wrote:
Tue Jul 02, 2019 7:31 pm
lkaufman wrote:
Tue Jul 02, 2019 6:54 pm
Chessqueen wrote:
Tue Jul 02, 2019 2:34 pm
lkaufman wrote:
Tue Jul 02, 2019 6:01 am
Chessqueen wrote:
Tue Jul 02, 2019 3:29 am
lkaufman wrote:
Mon Jul 01, 2019 5:03 pm
Nordlandia wrote:
Mon Jul 01, 2019 3:21 pm
Maybe odds matches in Fischer Random is harder for the human side. Humans are used to the standard opening setup. The unfamiliarity of Chess960 may give the engine the upper hand?

Maybe more odds is needed to equalize the chances in Fischer Random.

What is your thought on this one Larry?
Well we already "proved" that this is so, John Meyer won 3 to 1 at regular knight odds vs Komodo, but lost 0-4 at frc knight odds, both 45' + 15". Komodo has no chance giving knight odds in normal chess to strong GMs at rapid (not blitz) time controls, but may have some chance at frc knight odds, depending on the time control and the rating of the GM. But even with frc, knight odds is a lot to give to a GM with enough time to think.
I was also wondering what would be the outcome of pitting Komodo versus The fritz version that played versus Kramnik about 10 years ago Deep Fritz Bahrain with the f3 pawn or the f7 pawn odds in 50 games game in 30 minutes TC. What would your prediction be ?
First of all, Deep Fritz Bahrain played Kramnik in 2002, 17 years ago, not 10, and drew the match on hardware of that time. Presumably on my i7 laptop that same Fritz on 8 threads would be at least a couple hundred elo stronger, over FIDE 3000 surely. But to answer your question, I can't actually do the test on our tester since I don't have that ancient engine and anyway need something that runs in Linux on current hardware, but perhaps I can run some sort of test on something similarly rated on CCRL or CEGT lists. I am quite certain that Komodo would win such a match at f7 handicap (f2 would just be a blowout); there would probably be a fair number of draws, but Komodo would win a lot more than Fritz. Evan at f7 plus second White move handicap Komodo would be favored. Knight odds would be too much though. If someone knows of a suitable engine with a free and reliable Linux version, please advise.

WOW it has been 17 years already! Another program that you can pit is Deep Junior the same that play versus Gary Kasparov back in 2003 which also ended in a Draw. Since DJ and Kasparov were close in rating back in 2003 and it is hard to get a good Grandmaster that wants to play versus Komodo. and Carlsen will never agree since he is too proud to lose versus a computer. We can get Deep Junior with a similar hardware that can hit the same MNPS Millions nodes per second and give Deep Junior a handicap of the f2 pawn, or the f7 pawn odds. If Komodo wins easy then 2 pawns odds which to me it would be very hard for Komodo to even get a draw versus another engine, with a human they have the tendency to get nervous and sometimes blunder versus computer.
https://www.youtube.com/watch?v=WWYaLg21Dss
https://en.chessbase.com/post/kasparov- ... -sacrifice
Okay, I've started the following test. I don't have any engines as weak as those old Fritz and Junior versions on our Linux tester, the weakest one I have on the tester is Arasan 18.0, which is of course not weak at all (2860 CCRL 40/40 single thread). I am reasonably sure that with 1/4 of the time on one thread it is stronger than the two mentioned old engines on whatever number of threads they ran on back then. They are too old to find on the list, but if you extrapolate backward you can see that they should be somewhere below 2700, so this should be correct. I picked a time limit of 5' + 2", which if quadrupled would give games of about the same total time as 30' sudden death. I am running latest Komodo on four threads against Arasan 18.0 on one thread, with Komodo giving the f7 pawn handicap (which in self-testing at one minute + .4" inc. is nearly a 500 elo handicap based on 700 games). Note that this test is very unfair to Komodo (compared to the one you ask for) because Komodo is also only getting 5' + 2", it should really get quadruple this time. Also Komodo is only given four threads, but in an actual match with a top GM we would use maybe 47 threads on a 48 core machine. Also of course Arasan (and Komodo) are running on modern hardware, not 2002/2003 hardware, which makes a huge difference. On the other hand, a human GM would know that he was playing a much stronger entity and would avoid complications and try to simplify, which we can't expect an older engine to do. So maybe these factors balance out. I'll report results when I have them.
I was thinking if you were going to use Linux about Demolito ==> https://github.com/lucasart/Demolito

I am also going to do some test using this version of Araran ==> https://www.arasanchess.org/

Using GCC on Linux, type: make. Otherwise, just have a look at the makefile, nothing fancy there. You should also be able to compile it on any platform (POSIX or Windows), using a C11 capable compiler (eg. GCC or Clang but not MSVC). But you'll have to figure out the exact commands for yourself, depending on your compiler, your libc, and your system.
Why would you want to use the latest Arasan for this rather than an older one like 18.0? Even 18.0 is way stronger than the engines of 2002/2003. Maybe you meant to use Arasan 14?
Anyway, the test above surprised me; Komodo won by an incredible 87 wins to 2 losses, with 33 draws, for a plus 320 elo result despite the nearly 500 elo handicap! So I raised the handicap to pawn and two moves (f7 off, e4 played, WTM) and this made a big difference, but still Komodo won decisively (54 to 13 with 41 draws, plus 144 elo). All this without even raising Contempt from the default value! So now I'm also giving Arasan the same four threads as Komodo, with the pawn and two move handicap; this should be enough for Arasan to close the gap. But then this is a simulation of playing against an entity roughly a class stronger than the Fritz and Junior in question.
With both Komodo and Arasan 18 running on four threads at pawn and two move handicap, the result was close, Komodo winning by 116 wins to 105 losses with 101 draws, +12 elo. Now I'm running the same match with Contempt = 50 to see if that helps.
I guess you need to find out if Komodo still can win with a Knight for a pawn odds that would be interesting to find out :shock:
It is strange to noticed that the only Chess Champions that benefited from playing versus Computer engines where Kramnik and Kasparov the most since he got paid top money first playing versus Deeper Blue and later playing versus Fritz XD3 and Deep Junior,

PS: I wonder why Carlsen does not benefit to accept the challenge of playing Odds versus Engines, definitively he would make more money than playing in any top humans tournament :roll:
Why do you think carlsen is going to make more money by playing against engines releative to playing against humans?

I guess that it is not correct.

Uri Blass
Posts: 8611
Joined: Wed Mar 08, 2006 11:37 pm
Location: Tel-Aviv Israel

Re: Why there is no interest in Computer with odds Vs Humans match?

Post by Uri Blass » Wed Jul 03, 2019 4:00 am

lkaufman wrote:
Wed Jul 03, 2019 1:05 am
Chessqueen wrote:
Wed Jul 03, 2019 12:29 am
lkaufman wrote:
Tue Jul 02, 2019 10:13 pm
lkaufman wrote:
Tue Jul 02, 2019 8:19 pm
Chessqueen wrote:
Tue Jul 02, 2019 7:31 pm
lkaufman wrote:
Tue Jul 02, 2019 6:54 pm
Chessqueen wrote:
Tue Jul 02, 2019 2:34 pm
lkaufman wrote:
Tue Jul 02, 2019 6:01 am
Chessqueen wrote:
Tue Jul 02, 2019 3:29 am
lkaufman wrote:
Mon Jul 01, 2019 5:03 pm
Nordlandia wrote:
Mon Jul 01, 2019 3:21 pm
Maybe odds matches in Fischer Random is harder for the human side. Humans are used to the standard opening setup. The unfamiliarity of Chess960 may give the engine the upper hand?

Maybe more odds is needed to equalize the chances in Fischer Random.

What is your thought on this one Larry?
Well we already "proved" that this is so, John Meyer won 3 to 1 at regular knight odds vs Komodo, but lost 0-4 at frc knight odds, both 45' + 15". Komodo has no chance giving knight odds in normal chess to strong GMs at rapid (not blitz) time controls, but may have some chance at frc knight odds, depending on the time control and the rating of the GM. But even with frc, knight odds is a lot to give to a GM with enough time to think.
I was also wondering what would be the outcome of pitting Komodo versus The fritz version that played versus Kramnik about 10 years ago Deep Fritz Bahrain with the f3 pawn or the f7 pawn odds in 50 games game in 30 minutes TC. What would your prediction be ?
First of all, Deep Fritz Bahrain played Kramnik in 2002, 17 years ago, not 10, and drew the match on hardware of that time. Presumably on my i7 laptop that same Fritz on 8 threads would be at least a couple hundred elo stronger, over FIDE 3000 surely. But to answer your question, I can't actually do the test on our tester since I don't have that ancient engine and anyway need something that runs in Linux on current hardware, but perhaps I can run some sort of test on something similarly rated on CCRL or CEGT lists. I am quite certain that Komodo would win such a match at f7 handicap (f2 would just be a blowout); there would probably be a fair number of draws, but Komodo would win a lot more than Fritz. Evan at f7 plus second White move handicap Komodo would be favored. Knight odds would be too much though. If someone knows of a suitable engine with a free and reliable Linux version, please advise.

WOW it has been 17 years already! Another program that you can pit is Deep Junior the same that play versus Gary Kasparov back in 2003 which also ended in a Draw. Since DJ and Kasparov were close in rating back in 2003 and it is hard to get a good Grandmaster that wants to play versus Komodo. and Carlsen will never agree since he is too proud to lose versus a computer. We can get Deep Junior with a similar hardware that can hit the same MNPS Millions nodes per second and give Deep Junior a handicap of the f2 pawn, or the f7 pawn odds. If Komodo wins easy then 2 pawns odds which to me it would be very hard for Komodo to even get a draw versus another engine, with a human they have the tendency to get nervous and sometimes blunder versus computer.
https://www.youtube.com/watch?v=WWYaLg21Dss
https://en.chessbase.com/post/kasparov- ... -sacrifice
Okay, I've started the following test. I don't have any engines as weak as those old Fritz and Junior versions on our Linux tester, the weakest one I have on the tester is Arasan 18.0, which is of course not weak at all (2860 CCRL 40/40 single thread). I am reasonably sure that with 1/4 of the time on one thread it is stronger than the two mentioned old engines on whatever number of threads they ran on back then. They are too old to find on the list, but if you extrapolate backward you can see that they should be somewhere below 2700, so this should be correct. I picked a time limit of 5' + 2", which if quadrupled would give games of about the same total time as 30' sudden death. I am running latest Komodo on four threads against Arasan 18.0 on one thread, with Komodo giving the f7 pawn handicap (which in self-testing at one minute + .4" inc. is nearly a 500 elo handicap based on 700 games). Note that this test is very unfair to Komodo (compared to the one you ask for) because Komodo is also only getting 5' + 2", it should really get quadruple this time. Also Komodo is only given four threads, but in an actual match with a top GM we would use maybe 47 threads on a 48 core machine. Also of course Arasan (and Komodo) are running on modern hardware, not 2002/2003 hardware, which makes a huge difference. On the other hand, a human GM would know that he was playing a much stronger entity and would avoid complications and try to simplify, which we can't expect an older engine to do. So maybe these factors balance out. I'll report results when I have them.
I was thinking if you were going to use Linux about Demolito ==> https://github.com/lucasart/Demolito

I am also going to do some test using this version of Araran ==> https://www.arasanchess.org/

Using GCC on Linux, type: make. Otherwise, just have a look at the makefile, nothing fancy there. You should also be able to compile it on any platform (POSIX or Windows), using a C11 capable compiler (eg. GCC or Clang but not MSVC). But you'll have to figure out the exact commands for yourself, depending on your compiler, your libc, and your system.
Why would you want to use the latest Arasan for this rather than an older one like 18.0? Even 18.0 is way stronger than the engines of 2002/2003. Maybe you meant to use Arasan 14?
Anyway, the test above surprised me; Komodo won by an incredible 87 wins to 2 losses, with 33 draws, for a plus 320 elo result despite the nearly 500 elo handicap! So I raised the handicap to pawn and two moves (f7 off, e4 played, WTM) and this made a big difference, but still Komodo won decisively (54 to 13 with 41 draws, plus 144 elo). All this without even raising Contempt from the default value! So now I'm also giving Arasan the same four threads as Komodo, with the pawn and two move handicap; this should be enough for Arasan to close the gap. But then this is a simulation of playing against an entity roughly a class stronger than the Fritz and Junior in question.
With both Komodo and Arasan 18 running on four threads at pawn and two move handicap, the result was close, Komodo winning by 116 wins to 105 losses with 101 draws, +12 elo. Now I'm running the same match with Contempt = 50 to see if that helps.
I guess you need to find out if Komodo still can win with a Knight for a pawn odds that would be interesting to find out :shock:
It is strange to noticed that the only Chess Champions that benefited from playing versus Computer engines where Kramnik and Kasparov the most since he got paid top money first playing versus Deeper Blue and later playing versus Fritz XD3 and Deep Junior,

PS: I wonder why Carlsen does not benefit to accept the challenge of playing Odds versus Engines, definitively he would make more money than playing in any top humans tournament :roll:
Setting Contempt to 50 increased Komodo's winning margin at pawn and two move odds by 22 elo (from 12 to 34 elo, 353 games). I started a new match at two pawn odds (c2 and f2); I predict that Komodo will win by more at this handicap than at pawn and two moves; the difference of two moves with the f pawn gone at the start is even bigger than the value of the "c" pawn, I think. I don't have a way to test knight for pawn in the tester now, I can't add a new type of test to it myself. I could do knight odds, but I would need a weaker opponent than Arasan 18. As for Carlsen, who is going to pay him big money to play a computer? There is only slight financial benefit to a chess software company like Komodo to prove what everyone already knows, that it is way beyond human level. The amount we pay to superstars like Nakamura and MVL for rapid or blitz matches is pretty small compared to what they get for tournaments, and even if we paid Carlsen twice as much that would still be true. The top players understand that they have nothing to lose and conversely that we have little to gain, so the fees are small. Not at all like 15 years ago.
You can also use knight odds with arasan 18.
I understand that komodo is going to lose the match but the question is still if it is going to lose every game or maybe get some draws or even a win.

User avatar
Ovyron
Posts: 2837
Joined: Tue Jul 03, 2007 2:30 am

Re: Why there is no interest in Computer with odds Vs Humans match?

Post by Ovyron » Wed Jul 03, 2019 4:04 am

Do I gather that chess stopped being interesting so people are finally moving to other variants, and the most interesting one is chess with material odds? What happened here? :|
Great spirits have always encountered violent opposition from mediocre minds.

lkaufman
Posts: 3768
Joined: Sun Jan 10, 2010 5:15 am
Location: Maryland USA
Contact:

Re: Why there is no interest in Computer with odds Vs Humans match?

Post by lkaufman » Wed Jul 03, 2019 4:27 am

Ovyron wrote:
Wed Jul 03, 2019 4:04 am
Do I gather that chess stopped being interesting so people are finally moving to other variants, and the most interesting one is chess with material odds? What happened here? :|
Chess did not stop being interesting, just chess between computers and humans with no handicap. That is why it is the topic of this thread.
Komodo rules!

lkaufman
Posts: 3768
Joined: Sun Jan 10, 2010 5:15 am
Location: Maryland USA
Contact:

Re: Why there is no interest in Computer with odds Vs Humans match?

Post by lkaufman » Wed Jul 03, 2019 4:34 am

Uri Blass wrote:
Wed Jul 03, 2019 4:00 am
lkaufman wrote:
Wed Jul 03, 2019 1:05 am
Chessqueen wrote:
Wed Jul 03, 2019 12:29 am
lkaufman wrote:
Tue Jul 02, 2019 10:13 pm
lkaufman wrote:
Tue Jul 02, 2019 8:19 pm
Chessqueen wrote:
Tue Jul 02, 2019 7:31 pm
lkaufman wrote:
Tue Jul 02, 2019 6:54 pm
Chessqueen wrote:
Tue Jul 02, 2019 2:34 pm
lkaufman wrote:
Tue Jul 02, 2019 6:01 am
Chessqueen wrote:
Tue Jul 02, 2019 3:29 am
lkaufman wrote:
Mon Jul 01, 2019 5:03 pm
Nordlandia wrote:
Mon Jul 01, 2019 3:21 pm
Maybe odds matches in Fischer Random is harder for the human side. Humans are used to the standard opening setup. The unfamiliarity of Chess960 may give the engine the upper hand?

Maybe more odds is needed to equalize the chances in Fischer Random.

What is your thought on this one Larry?
Well we already "proved" that this is so, John Meyer won 3 to 1 at regular knight odds vs Komodo, but lost 0-4 at frc knight odds, both 45' + 15". Komodo has no chance giving knight odds in normal chess to strong GMs at rapid (not blitz) time controls, but may have some chance at frc knight odds, depending on the time control and the rating of the GM. But even with frc, knight odds is a lot to give to a GM with enough time to think.
I was also wondering what would be the outcome of pitting Komodo versus The fritz version that played versus Kramnik about 10 years ago Deep Fritz Bahrain with the f3 pawn or the f7 pawn odds in 50 games game in 30 minutes TC. What would your prediction be ?
First of all, Deep Fritz Bahrain played Kramnik in 2002, 17 years ago, not 10, and drew the match on hardware of that time. Presumably on my i7 laptop that same Fritz on 8 threads would be at least a couple hundred elo stronger, over FIDE 3000 surely. But to answer your question, I can't actually do the test on our tester since I don't have that ancient engine and anyway need something that runs in Linux on current hardware, but perhaps I can run some sort of test on something similarly rated on CCRL or CEGT lists. I am quite certain that Komodo would win such a match at f7 handicap (f2 would just be a blowout); there would probably be a fair number of draws, but Komodo would win a lot more than Fritz. Evan at f7 plus second White move handicap Komodo would be favored. Knight odds would be too much though. If someone knows of a suitable engine with a free and reliable Linux version, please advise.

WOW it has been 17 years already! Another program that you can pit is Deep Junior the same that play versus Gary Kasparov back in 2003 which also ended in a Draw. Since DJ and Kasparov were close in rating back in 2003 and it is hard to get a good Grandmaster that wants to play versus Komodo. and Carlsen will never agree since he is too proud to lose versus a computer. We can get Deep Junior with a similar hardware that can hit the same MNPS Millions nodes per second and give Deep Junior a handicap of the f2 pawn, or the f7 pawn odds. If Komodo wins easy then 2 pawns odds which to me it would be very hard for Komodo to even get a draw versus another engine, with a human they have the tendency to get nervous and sometimes blunder versus computer.
https://www.youtube.com/watch?v=WWYaLg21Dss
https://en.chessbase.com/post/kasparov- ... -sacrifice
Okay, I've started the following test. I don't have any engines as weak as those old Fritz and Junior versions on our Linux tester, the weakest one I have on the tester is Arasan 18.0, which is of course not weak at all (2860 CCRL 40/40 single thread). I am reasonably sure that with 1/4 of the time on one thread it is stronger than the two mentioned old engines on whatever number of threads they ran on back then. They are too old to find on the list, but if you extrapolate backward you can see that they should be somewhere below 2700, so this should be correct. I picked a time limit of 5' + 2", which if quadrupled would give games of about the same total time as 30' sudden death. I am running latest Komodo on four threads against Arasan 18.0 on one thread, with Komodo giving the f7 pawn handicap (which in self-testing at one minute + .4" inc. is nearly a 500 elo handicap based on 700 games). Note that this test is very unfair to Komodo (compared to the one you ask for) because Komodo is also only getting 5' + 2", it should really get quadruple this time. Also Komodo is only given four threads, but in an actual match with a top GM we would use maybe 47 threads on a 48 core machine. Also of course Arasan (and Komodo) are running on modern hardware, not 2002/2003 hardware, which makes a huge difference. On the other hand, a human GM would know that he was playing a much stronger entity and would avoid complications and try to simplify, which we can't expect an older engine to do. So maybe these factors balance out. I'll report results when I have them.
I was thinking if you were going to use Linux about Demolito ==> https://github.com/lucasart/Demolito

I am also going to do some test using this version of Araran ==> https://www.arasanchess.org/

Using GCC on Linux, type: make. Otherwise, just have a look at the makefile, nothing fancy there. You should also be able to compile it on any platform (POSIX or Windows), using a C11 capable compiler (eg. GCC or Clang but not MSVC). But you'll have to figure out the exact commands for yourself, depending on your compiler, your libc, and your system.
Why would you want to use the latest Arasan for this rather than an older one like 18.0? Even 18.0 is way stronger than the engines of 2002/2003. Maybe you meant to use Arasan 14?
Anyway, the test above surprised me; Komodo won by an incredible 87 wins to 2 losses, with 33 draws, for a plus 320 elo result despite the nearly 500 elo handicap! So I raised the handicap to pawn and two moves (f7 off, e4 played, WTM) and this made a big difference, but still Komodo won decisively (54 to 13 with 41 draws, plus 144 elo). All this without even raising Contempt from the default value! So now I'm also giving Arasan the same four threads as Komodo, with the pawn and two move handicap; this should be enough for Arasan to close the gap. But then this is a simulation of playing against an entity roughly a class stronger than the Fritz and Junior in question.
With both Komodo and Arasan 18 running on four threads at pawn and two move handicap, the result was close, Komodo winning by 116 wins to 105 losses with 101 draws, +12 elo. Now I'm running the same match with Contempt = 50 to see if that helps.
I guess you need to find out if Komodo still can win with a Knight for a pawn odds that would be interesting to find out :shock:
It is strange to noticed that the only Chess Champions that benefited from playing versus Computer engines where Kramnik and Kasparov the most since he got paid top money first playing versus Deeper Blue and later playing versus Fritz XD3 and Deep Junior,

PS: I wonder why Carlsen does not benefit to accept the challenge of playing Odds versus Engines, definitively he would make more money than playing in any top humans tournament :roll:
Setting Contempt to 50 increased Komodo's winning margin at pawn and two move odds by 22 elo (from 12 to 34 elo, 353 games). I started a new match at two pawn odds (c2 and f2); I predict that Komodo will win by more at this handicap than at pawn and two moves; the difference of two moves with the f pawn gone at the start is even bigger than the value of the "c" pawn, I think. I don't have a way to test knight for pawn in the tester now, I can't add a new type of test to it myself. I could do knight odds, but I would need a weaker opponent than Arasan 18. As for Carlsen, who is going to pay him big money to play a computer? There is only slight financial benefit to a chess software company like Komodo to prove what everyone already knows, that it is way beyond human level. The amount we pay to superstars like Nakamura and MVL for rapid or blitz matches is pretty small compared to what they get for tournaments, and even if we paid Carlsen twice as much that would still be true. The top players understand that they have nothing to lose and conversely that we have little to gain, so the fees are small. Not at all like 15 years ago.
You can also use knight odds with arasan 18.
I understand that komodo is going to lose the match but the question is still if it is going to lose every game or maybe get some draws or even a win.
Okay, i ran that match. Latest Komodo on four threads, Contempt 50, time limit 5' + 2", vs Arasan 18 on one thread, knight odds (alternating b1 and g1). Komodo variety set to 5, with MP providing most of the variety. After 395 games, Arasan won by 232 wins to 112 losses with 51 draws, for a +113 elo result. Quite a bit closer than I had expected. Obviously a lot closer than you expected!
Komodo rules!

peter
Posts: 1786
Joined: Sat Feb 16, 2008 6:38 am
Full name: Peter Martan

Re: Why there is no interest in Computer with odds Vs Humans match?

Post by peter » Wed Jul 03, 2019 4:56 am

Ovyron wrote:
Wed Jul 03, 2019 4:04 am
What happened here?
Two things happened:
Draw death of engine- chesss and black box- death of intelligibilty of engine- chess.
Like Chrilly Donninger used to say, it's like watching the washer washing the laundry.

SF et al already were rather unintelligible in their ways of playing, NNs are fully. And I'm not talking about me and my poor chess- knowledge only, I'm talking about human masters and Grandmasters as well.

I know some of them, more in corr.- chess than in over the board playing, but in both rather different groups the tenor is the same, don't even try to understand the moves and games of SF nor of NN- engines, the latter ones even less then the first ones.

Of course that doesn't give good publicity for a to be well known chess player to say so, not to the audience and not to the opponents, it's more smart to write commentaries, even whole books and have youtube- videos about glorious NN- games and -moves, in one thing chessplayers were always even better then in playing, in commenting post mortem games with full of positional and understanding proving wording.
Funniest thing about that all to me is always reading once and again about "human style" of NN- engines.
:)
With SF you could at least try to understand, what was going on with some Forward- Backward, with NNs you can do as much FW- BW as you like, the "engine" is as "unteachable" and "unconvincible" as for alternative moves as you are yourself with the NN- engines' moves.

With SF et al you could at least try to find out connections between moves and code and parameters to be changed by the programmer and the user.
With NNs you can't, not even as a well trained specialized chess-programmer.

So what to do with these new gods in chess? Not even letting them play against each other and count the points in the end only, forgetting about the games and moves themselves, brings up any excitement anymore, cause number of points depend mainly on the pool of opponents, on the openings and on the given hardware- time.

Progress may go on and on, what do we see and measure of it, if there's much too much draw for the number of games to be played for a significant single decision and full lack of transitivity to other pools of NNs and engines given?

So what?
Peter.

User avatar
Ovyron
Posts: 2837
Joined: Tue Jul 03, 2007 2:30 am

Re: Why there is no interest in Computer with odds Vs Humans match?

Post by Ovyron » Wed Jul 03, 2019 5:04 am

lkaufman wrote:
Wed Jul 03, 2019 4:27 am
Chess did not stop being interesting, just chess between computers and humans with no handicap. That is why it is the topic of this thread.
I agree on that, but there are many handicaps not being considered, like limiting depth (what is the lowest depth that Komodo needs to beat humans?) or nodes (what is the lowest nodes Komodo needs to visit to beat humans?) Couldn't Komodo boast to have some great chess knowledge if it can defeat humans without much search used?

For this I suggest something like Rebel Decade's move delay. You have two Komodos, one to pick the moves, and one to manage the time. So the one with low depth/nodes picks a move instantly, but the other doesn't know, and thinks it's in a normal game, and spends its time finding the best move to play, but when is its time to move, it plays the first Komodo's move.

Material handicap is the laziest one to implement, and it's sad to see people acting as if it's the only handicap left to try...
Great spirits have always encountered violent opposition from mediocre minds.

lkaufman
Posts: 3768
Joined: Sun Jan 10, 2010 5:15 am
Location: Maryland USA
Contact:

Re: Why there is no interest in Computer with odds Vs Humans match?

Post by lkaufman » Wed Jul 03, 2019 5:54 am

Ovyron wrote:
Wed Jul 03, 2019 5:04 am
lkaufman wrote:
Wed Jul 03, 2019 4:27 am
Chess did not stop being interesting, just chess between computers and humans with no handicap. That is why it is the topic of this thread.
I agree on that, but there are many handicaps not being considered, like limiting depth (what is the lowest depth that Komodo needs to beat humans?) or nodes (what is the lowest nodes Komodo needs to visit to beat humans?) Couldn't Komodo boast to have some great chess knowledge if it can defeat humans without much search used?

For this I suggest something like Rebel Decade's move delay. You have two Komodos, one to pick the moves, and one to manage the time. So the one with low depth/nodes picks a move instantly, but the other doesn't know, and thinks it's in a normal game, and spends its time finding the best move to play, but when is its time to move, it plays the first Komodo's move.

Material handicap is the laziest one to implement, and it's sad to see people acting as if it's the only handicap left to try...
There have actually been a huge number of games played already on chess.com in three different events in which grandmasters played against Komodo handicapped primarily by search depth. The levels below level 20 on Komodo (level 25 in future versions) are handicapped by depth and by randomness, but the randomness is only significant below the top few levels. For all practical purposes, levels 17 thru 19 are just depths 8 thru 10 with a little variety. First Nakamura, then MVL, and then many teams of players (including GMs) played against the levels, working their way up. Most of the games were played at 5' + 2" I believe. At this time limit, I would summarize the results by saying that depth 8 is weak GM, depth 9 is strong GM, and depth 10 is super-GM (Naka/MVL). Regarding your delay suggestion, is this just to create the illusion of a normal game? Chess.com wanted Komodo to play as quickly as its search would allow, not wasting anyone's time. It's up to the human player to adapt to this.
I personally prefer material and move handicaps, because limiting depth or nodes has no meaning to ordinary chessplayers, only to chess computer geeks like us. You don't need to know anything about computers to know that if it beats a title contender giving him f7 and two moves, or two pawns, it has vastly outclassed him. But if it beats him giving him 1000 to 1 time odds (either explicitly or via nodes/depth) people just say that shows how fast computers are. But we'll continue to have events with limited depth handicaps. You can play Komodo online yourself with this depth-limit handicap on chess.com. If you want to play depth 10, just challenge Komodo 19.
Komodo rules!

User avatar
Laskos
Posts: 9545
Joined: Wed Jul 26, 2006 8:21 pm
Full name: Kai Laskos

Re: Why there is no interest in Computer with odds Vs Humans match?

Post by Laskos » Wed Jul 03, 2019 7:35 am

lkaufman wrote:
Wed Jul 03, 2019 5:54 am
Ovyron wrote:
Wed Jul 03, 2019 5:04 am
lkaufman wrote:
Wed Jul 03, 2019 4:27 am
Chess did not stop being interesting, just chess between computers and humans with no handicap. That is why it is the topic of this thread.
I agree on that, but there are many handicaps not being considered, like limiting depth (what is the lowest depth that Komodo needs to beat humans?) or nodes (what is the lowest nodes Komodo needs to visit to beat humans?) Couldn't Komodo boast to have some great chess knowledge if it can defeat humans without much search used?

For this I suggest something like Rebel Decade's move delay. You have two Komodos, one to pick the moves, and one to manage the time. So the one with low depth/nodes picks a move instantly, but the other doesn't know, and thinks it's in a normal game, and spends its time finding the best move to play, but when is its time to move, it plays the first Komodo's move.

Material handicap is the laziest one to implement, and it's sad to see people acting as if it's the only handicap left to try...
There have actually been a huge number of games played already on chess.com in three different events in which grandmasters played against Komodo handicapped primarily by search depth. The levels below level 20 on Komodo (level 25 in future versions) are handicapped by depth and by randomness, but the randomness is only significant below the top few levels. For all practical purposes, levels 17 thru 19 are just depths 8 thru 10 with a little variety. First Nakamura, then MVL, and then many teams of players (including GMs) played against the levels, working their way up. Most of the games were played at 5' + 2" I believe. At this time limit, I would summarize the results by saying that depth 8 is weak GM, depth 9 is strong GM, and depth 10 is super-GM (Naka/MVL). Regarding your delay suggestion, is this just to create the illusion of a normal game? Chess.com wanted Komodo to play as quickly as its search would allow, not wasting anyone's time. It's up to the human player to adapt to this.
I personally prefer material and move handicaps, because limiting depth or nodes has no meaning to ordinary chessplayers, only to chess computer geeks like us. You don't need to know anything about computers to know that if it beats a title contender giving him f7 and two moves, or two pawns, it has vastly outclassed him. But if it beats him giving him 1000 to 1 time odds (either explicitly or via nodes/depth) people just say that shows how fast computers are. But we'll continue to have events with limited depth handicaps. You can play Komodo online yourself with this depth-limit handicap on chess.com. If you want to play depth 10, just challenge Komodo 19.
Taking moves back by the GM was not explored. We don't even know what conditions these take backs can take. As many times as he wishes? How far he can take back (say up to 2-3 moves)? 20 takebacks total in one game (several for the same move being allowed)? I guess that taking back would help more against SF or Komodo than against Leela, as Leela can outplay a GM positionally, without that GM even realizing quickly what happens to him and realizing much too late.

Post Reply