Arduous Victory against Chess Player 2150

[Fernando]

Again I have experimented how interesting and difficult is to win a program from the 80's or 90's, but running in current hardware.
Now It has been the case with Chess Player 2150, a Wittington program that came before Chess Player 2175, that came before Complete Chess System, which at his turn was a predecessor of Tal, the fully innovative engine that came to the market in a DOS and in a Windows version, years ago, too.

Ches Player 2150 was pretty strong even in those times. It won the 1989 British Open Personal Computer Chess Championship (according to a post in ST Report International Online Magazine, November 13, 1992, No.8.45)

But 2150 Elo?

Not quite, however. Probably something around 1900-2000, which is good enough. The program, as far as I have detected, is prone to move his Queen slightly more than what is good in early middle game and that gives chances to make him lose time and retard his development.
Anyway plays a good game, it is dangerous and you must play it with caution, slowly getting the upper hand.

Next experiment: Complete Chess System....
If someone have CP 2175, please let me know where to get it...

Fern

[spacious_mind]

It's a fun program, but it drops off a lot in strength in endgames. Here you can see the Atari ST Version:



http://www.spacious-mind.com/html/atari ... _2150.html

The game it played against Mephisto Academy came to this position where Chess Player 2150 plays Black:

[fen]6k1/p5pp/8/3P4/4PK2/r3N1P1/8/8 w - - 0 45[/fen]

45. ... g5! would have been powerful and easy win. Even 45. ... a5 would probably have won.

[fen]8/5kpp/3P4/p2KPN2/8/6P1/1r6/8 w - - 0 53[/fen]

In this position 52. ... Rd2+ is the only move. Everything else is bad. CP2150 played 52. ... Rb5?

[fen]1r6/2K2kpp/3P4/p3PN2/8/6P1/8/8 w - - 0 55[/fen]

And here the final straw. 55. ... Ra8 might still have drawn. Instead CP2150 plays 55. ... Rb3? and loses.

CP2150 is not so good in endgames.

Best regards

[Fernando]

Atari is lot slower than any average pc
I will chevk that position in mine..

Fern

[spacious_mind]

Atari is lot slower than any average pc
I will chevk that position in mine..

Fern

OK, it would be interesting to see if any of the above moves are improved by faster hardware. The game was played at 30 seconds per move. ST Hardware is approx. the equivalent of a 6502 at 4 MHz whereas Mephisto Academy plays at 5 MHz, therefore 20% faster than CP2150 on Atari ST.

Best regards

[spacious_mind]

I forgot I also have CP2150 for the Commodore Amiga:

http://www.spacious-mind.com/html/amiga ... _2150.html

A test game was also played against Mephisto Academy which you can play back from the above link. Same result as Atari ST with bad ending skills ruining a game that should have really been drawn because of different colored Bishops and passed pawn on a4 offsetting the pawn advantage that Black enjoys.

[fen]8/5p1p/3kb3/5p2/K3p3/P7/1B4PP/8 w - - 0 44[/fen]

The white King on a4 needs to be moved closer to the center immediately with Kb4, for a fairly easy draw. CP2150 seems to forget that the King exists and leaves it sitting there for a long time allowing Black to maneuver his pieces into dominating center positions.

Again after a good battle up to that point, it loses easily in the end game.

This is where CP2150 saw itself in the grand scheme of things of the chess world:



Best regards

[Martin Hertz]

Hello Nick,

I've tested your positions with CP2150 on a HP-1000CX (80186 @ 7.91 MHz) and a i7 @ 3.3 GHz.

[fen]6k1/p5pp/8/3P4/4PK2/r3N1P1/8/8 w - - 0 45[/fen]
g5 found at depth 3

i7 immediately

80186 00:00:22


[fen]8/5kpp/3P4/p2KPN2/8/6P1/1r6/8 w - - 0 53[/fen]
Rd2+ found at depth 4

i7 immediately

80186 00:11:50


[fen]1r6/2K2kpp/3P4/p3PN2/8/6P1/8/8 w - - 0 55[/fen]
Ra8 found at depth 8

i7 00:05:55

80186 not tested

I don't think that CP2150 on the Atari is much faster than the HP-1000CX, but I'm not sure.

A very nice Shannon B type program.

[Fernando]

This proves that CP 2150 is a good engine and that perhaps Atari was not the best platform for him. I believe Atari was lot slower than any current PC, to begin with.

Now I will test strong Socrates by our gone friend Don Daily

Fern

[spacious_mind]

Hello Nick,

I've tested your positions with CP2150 on a HP-1000CX (80186 @ 7.91 MHz) and a i7 @ 3.3 GHz.

[fen]6k1/p5pp/8/3P4/4PK2/r3N1P1/8/8 w - - 0 45[/fen]
g5 found at depth 3

i7 immediately

80186 00:00:22


[fen]8/5kpp/3P4/p2KPN2/8/6P1/1r6/8 w - - 0 53[/fen]
Rd2+ found at depth 4

i7 immediately

80186 00:11:50


[fen]1r6/2K2kpp/3P4/p3PN2/8/6P1/8/8 w - - 0 55[/fen]
Ra8 found at depth 8

i7 00:05:55

80186 not tested

I don't think that CP2150 on the Atari is much faster than the HP-1000CX, but I'm not sure.

A very nice Shannon B type program.

Hi Martin,

Thanks and interesting. Depth 3 it should have found maybe at 30 seconds per move. But I also see you are at 7.1 MHz. 68,000 at 8 Mhz is the equivalent of 6502 with 4 MHz. You took if I read it correctly 22 seconds. I suspect 30 seconds on the Atari is not enough time to find it. I suspect that the 80186 has similar speed as a 6502 therefore I suspect if I tried it again Atari ST would have probably needed around 40 seconds to find this move.

Depth 4 maybe not enough time it took you 11+ minutes to find it. Whittington's programs were always slow searchers.


Depth 8 would take hours and hours to reach therefore no chance.

@ Fernando, yes you are partially right, in that its endgame improves with higher speed. But comparatively speaking it still drops off in endgames it is a slow searcher and your reasoning for improvement would apply to pretty much every decent dedicated computer if allowed to run at PC speed.

Taking 5:55 minutes at 3.3 GHz to reach only 8 ply means endgame problems continue. It means at tournament level it still would have missed the opportunity to draw the game on modern day hardware.

But please don't get me wrong the program is good and plays competitively as seen in these Mephisto Academy examples who after all is also pretty good dedicated computer.

Can't expect wonders...regards

[Fernando]

of course in any case Cp2150 drops lot of its strength in endings, no doubt, as any old program. My point was just to note that CP could not play so bad as appeared in Atari, but lot better in a PC, as in fact does.
BTW, as I love to play long game and give the engines lot of time, I enjoy playing even old machines that are capable of reaching a decent depth and so a decent ending too...

Fern

[spacious_mind]

of course in any case Cp2150 drops lot of its strength in endings, no doubt, as any old program. My point was just to note that CP could not play so bad as appeared in Atari, but lot better in a PC, as in fact does.
BTW, as I love to play long game and give the engines lot of time, I enjoy playing even old machines that are capable of reaching a decent depth and so a decent ending too...

Fern

I totally agree, I have found from old PC programs and Atari and Amiga that computers that reach 4 ply in search depth play most of the time pretty good chess. The same really applies to dedicated computers as well.

It is when they hang around 2 or 3 ply they make more of the bad mistakes.

best regards

[Fernando]

of course in any case Cp2150 drops lot of its strength in endings, no doubt, as any old program. My point was just to note that CP could not play so bad as appeared in Atari, but lot better in a PC, as in fact does.
BTW, as I love to play long game and give the engines lot of time, I enjoy playing even old machines that are capable of reaching a decent depth and so a decent ending too...

Fern

I totally agree, I have found from old PC programs and Atari and Amiga that computers that reach 4 ply in search depth play most of the time pretty good chess. The same really applies to dedicated computers as well.

It is when they hang around 2 or 3 ply they make more of the bad mistakes.

best regards

4 ply is what WE play when we are lazy.In a good day of decent concentration, my search in middle game gets more or less 8-10 ply in two lines and perhaps 5-6 in a third line if there is one.
So In can be competitive with dedicated units but NOT with engines that goes beyond 20 ply just like that.
I am, we are, totally out searched

[spacious_mind]

of course in any case Cp2150 drops lot of its strength in endings, no doubt, as any old program. My point was just to note that CP could not play so bad as appeared in Atari, but lot better in a PC, as in fact does.
BTW, as I love to play long game and give the engines lot of time, I enjoy playing even old machines that are capable of reaching a decent depth and so a decent ending too...

Fern

I totally agree, I have found from old PC programs and Atari and Amiga that computers that reach 4 ply in search depth play most of the time pretty good chess. The same really applies to dedicated computers as well.

It is when they hang around 2 or 3 ply they make more of the bad mistakes.

best regards

4 ply is what WE play when we are lazy.In a good day of decent concentration, my search in middle game gets more or less 8-10 ply in two lines and perhaps 5-6 in a third line if there is one.
So In can be competitive with dedicated units but NOT with engines that goes beyond 20 ply just like that.
I am, we are, totally out searched

You know there was a study done a long time ago and it is a myth that Grandmasters search deeper than say good club players. Their advantage is more understanding and intuition of grasping a situation at a glance that gives them their big advantage over us mere mortals.

I think what you said is absolutely right humans don't search too deep as we approach situations and grasp things differently to computers.

If we would have to search 8-10 ply deep on every move I guarantee you we would make many more mistakes. Humans tend to pick more positions where they concentrate harder where they see opportunities to either plan an advantage or a cunning defense. But it cannot be sustained for every move, not even Grandmasters are capable of this.

You just have to look at the general high quality of their games in Simultaneous matches or Blitz matches to know that they don't search deeply. It is much more about developed intuition either maybe naturally gifted (but very rare and unlikely) or improved through lots of experience from practice and study.

Best regards

[Fernando]

of course in any case Cp2150 drops lot of its strength in endings, no doubt, as any old program. My point was just to note that CP could not play so bad as appeared in Atari, but lot better in a PC, as in fact does.
BTW, as I love to play long game and give the engines lot of time, I enjoy playing even old machines that are capable of reaching a decent depth and so a decent ending too...

Fern

I totally agree, I have found from old PC programs and Atari and Amiga that computers that reach 4 ply in search depth play most of the time pretty good chess. The same really applies to dedicated computers as well.

It is when they hang around 2 or 3 ply they make more of the bad mistakes.

best regards

4 ply is what WE play when we are lazy.In a good day of decent concentration, my search in middle game gets more or less 8-10 ply in two lines and perhaps 5-6 in a third line if there is one.
So In can be competitive with dedicated units but NOT with engines that goes beyond 20 ply just like that.
I am, we are, totally out searched

You know there was a study done a long time ago and it is a myth that Grandmasters search deeper than say good club players. Their advantage is more understanding and intuition of grasping a situation at a glance that gives them their big advantage over us mere mortals.

I think what you said is absolutely right humans don't search too deep as we approach situations and grasp things differently to computers.

If we would have to search 8-10 ply deep on every move I guarantee you we would make many more mistakes. Humans tend to pick more positions where they concentrate harder where they see opportunities to either plan an advantage or a cunning defense. But it cannot be sustained for every move, not even Grandmasters are capable of this. You just have to look at the general high quality of their games in Simultaneous matches to know that they don't search deeply. It is much more about developed intuition either maybe naturally gifted (but very rare and unlikely) or improved through lots of experience from practice and study.

Best regards

Yes,I have read about that, BUT all the rest being equal, deeper search tends to be best.
Of course there are many nuances. it could be that a deeper search gives rooms for more mistakes.
In my case I have found that my game are always much better if I look a bit more to the sides, more width instead of just deep... And any extra minute of time you think, you will get a better move. I have played games where in a position I have been long time without finding the key to continue and so in a current day i would play anything and perhaps lose, BUt when I am really on it I still continue and finally , perhaps after 15 minutes, i find the answer. This is too a game of patience

[spacious_mind]

of course in any case Cp2150 drops lot of its strength in endings, no doubt, as any old program. My point was just to note that CP could not play so bad as appeared in Atari, but lot better in a PC, as in fact does.
BTW, as I love to play long game and give the engines lot of time, I enjoy playing even old machines that are capable of reaching a decent depth and so a decent ending too...

Fern

I totally agree, I have found from old PC programs and Atari and Amiga that computers that reach 4 ply in search depth play most of the time pretty good chess. The same really applies to dedicated computers as well.

It is when they hang around 2 or 3 ply they make more of the bad mistakes.

best regards

4 ply is what WE play when we are lazy.In a good day of decent concentration, my search in middle game gets more or less 8-10 ply in two lines and perhaps 5-6 in a third line if there is one.
So In can be competitive with dedicated units but NOT with engines that goes beyond 20 ply just like that.
I am, we are, totally out searched

You know there was a study done a long time ago and it is a myth that Grandmasters search deeper than say good club players. Their advantage is more understanding and intuition of grasping a situation at a glance that gives them their big advantage over us mere mortals.

I think what you said is absolutely right humans don't search too deep as we approach situations and grasp things differently to computers.

If we would have to search 8-10 ply deep on every move I guarantee you we would make many more mistakes. Humans tend to pick more positions where they concentrate harder where they see opportunities to either plan an advantage or a cunning defense. But it cannot be sustained for every move, not even Grandmasters are capable of this. You just have to look at the general high quality of their games in Simultaneous matches to know that they don't search deeply. It is much more about developed intuition either maybe naturally gifted (but very rare and unlikely) or improved through lots of experience from practice and study.

Best regards

Yes,I have read about that, BUT all the rest being equal, deeper search tends to be best.
Of course there are many nuances. it could be that a deeper search gives rooms for more mistakes.
In my case I have found that my game are always much better if I look a bit more to the sides, more width instead of just deep... And any extra minute of time you think, you will get a better move. I have played games where in a position I have been long time without finding the key to continue and so in a current day i would play anything and perhaps lose, BUt when I am really on it I still continue and finally , perhaps after 15 minutes, i find the answer. This is too a game of patience

Yes, agreed but not sure if that can be called search depth as in computers. I think you take more time to evaluate the situation based on your experience that you have developed through practice and time and you take more time to acknowledge the threats and opportunities and to discard most of the ones you would consider irrelevant allowing you to focus on the ones you perceive as opportunity or defense.

The danger for all of us is that sometimes in our excitement to follow our plan we misunderstand or totally ignore the opponents plan. A wise player of course takes much more time to consider his opponents options before deciding on his move and his plan. I think when you want to play more seriously you conform more to the principles of being a wise player and therefore you are more successful of course.

I think we all improve as we slow down and remember what we have learned as principle to follow

Interesting topic.

Best regards

[Fernando]

Yes, there is such previous musing about general features of the position, what we should do, where are the weaknesses of both sides, the imbalances, etc. At last you pick a thread and then begin the tactical search. Clearly, we do not do that as the comp. That previous philosophical part if just ours. And it is in this part where the professional player has advantage due to his experience. We have some, bit not that much...

Fern