HIARCS Chess Forums Forum Index HIARCS Chess Forums
World Championship winning computer chess software program & downloads for chess database, analysis and play on PC, Mac and iPhone
 
 QuestionsQuestions   SearchSearch   MemberlistMemberlist   UsergroupsUsergroups   RegisterRegister 
 ProfileProfile   Private MessagesPrivate Messages   Log inLog in 

Spacious-Mind Swiss Chess Computer League - Division 3
Goto page Previous  1, 2, 3
 
Post new topic   Reply to topic    HIARCS Chess Forums Forum Index -> Main Lobby / General Discussions
View previous topic :: View next topic  
Author
spacious_mind
Senior Member


Joined: 01 Aug 2007
Posts: 2813
Location: Alabama

PostPosted: Sun Jul 16, 2017 12:23 am    Post subject: Reply with quote

CHESSBASE 30 FIDE 2000/USCF 2060 PLAYER EVALUATION



I ran 15 player games from Chessbase Big Database where both players were rated 2000 ELO. As you can see I picked any game that was moves or above to simulate an average that you also have with dedicated computer games.

Above are the results after I ran the games through Lichess. At the bottom of the chart you can see the average performance of the randomly selected 2000 ELO Male Player.


DIVISION 3 FINAL RANKINGS PREDICTOR AFTER 2 ROUNDS



Here is the updated Predictor after Round 2, taking the average score from both Round 1 and Round 2. I also inserted the average 2000 ELO player which might help as a guide as to the strength of these chess programs entered in Division 3.

At the end of Round 3 I will also add a line for 2100 FIDE/2162 USCF rated players.

Well it looks like Amiga Sargon 3 has made the least mistakes over 2 games and Amiga Checkmate Aggressive mode has scored the best overall so far.

The average performances will obviously change over 12 rounds, but it will be interesting to compare it all in the end.

Best regards
_________________
Nick
Back to top
View user's profile Send private message Visit poster's website
Chessmaster Ireland
Full Member


Joined: 18 Feb 2009
Posts: 509

PostPosted: Sun Jul 16, 2017 8:19 am    Post subject: Reply with quote

Very interesting tournament Nick!
With some wild and weird games too!!
Atari ST Regards
Bryan
Back to top
View user's profile Send private message Send e-mail
Volodymyr
Member


Joined: 08 Apr 2017
Posts: 37
Location: Ukraine,Radyvyliv

PostPosted: Sun Jul 16, 2017 8:28 am    Post subject: Reply with quote

Hi Nick,
Very interesting information.
Mistakes of players in many ways depend on style and opening. One player creates a puzzle, and the other player plays positionally.
Play gambit, and playing closed positions is a different thing.Example style of Tal and Petrosyan.
I often play online. I've seen the "antichess" players.
"Antichess" is the ideal score in the analysis, according to the computer.
But such players do not know how to play at all openly. Without a pawn structure they are lost.
Also many nuances of the person's game influencing this analysis.

I saw more in-depth results.
http://talkchess.com/forum/viewtopic.php?t=64436
What Karpov won, and what won Kasimdzhanov.Or Botvinnik,Ponomariov,Lasker.
A very big difference, but they are close.
Maybe in a hundred years we will not see in the list of the best,not Carlsen,not Kramnik,not Fischer.
Computers very seriously affect the style of players.Yes they play more perfectly,but little creativity.
I'm against such sheets for all time.
It's just a way to compare.
Back to top
View user's profile Send private message Send e-mail
spacious_mind
Senior Member


Joined: 01 Aug 2007
Posts: 2813
Location: Alabama

PostPosted: Sun Jul 16, 2017 2:20 pm    Post subject: Reply with quote

Volodymyr wrote:
Hi Nick,
Very interesting information.
Mistakes of players in many ways depend on style and opening. One player creates a puzzle, and the other player plays positionally.
Play gambit, and playing closed positions is a different thing.Example style of Tal and Petrosyan.
I often play online. I've seen the "antichess" players.
"Antichess" is the ideal score in the analysis, according to the computer.
But such players do not know how to play at all openly. Without a pawn structure they are lost.
Also many nuances of the person's game influencing this analysis.

I saw more in-depth results.
http://talkchess.com/forum/viewtopic.php?t=64436
What Karpov won, and what won Kasimdzhanov.Or Botvinnik,Ponomariov,Lasker.
A very big difference, but they are close.
Maybe in a hundred years we will not see in the list of the best,not Carlsen,not Kramnik,not Fischer.
Computers very seriously affect the style of players.Yes they play more perfectly,but little creativity.
I'm against such sheets for all time.
It's just a way to compare.


Hi Volodymyr,

The whole point of this exercise is to show chess programs and human ELO tournament players next to each other. This is meant to show that if for example you are a 2000 ELO player then the programs above you and some slightly below you will be very good opponents for you.

Since Stockfish evaluates them all 25 year old chess program and human in the same way, they are both evaluated based on what Stockfish sees as mistakes. This does not make a difference to two humans playing or two old chess programs playing as they are both blissfully doing their own thing. All that is happening is Stockfish scores the moves and I put them through my rating Smile There is nothing sinister here that is meant to devalue the human or the computer.

If you see what Bryan posted earlier you will see that some these programs play very human like wild west chess with mistakes and blunders. So they are perfect for the human player to play against with lots and lots of variety since there are so many of these programs available to play against.

ps... When I complete the 12 Rounds in Division 3, then this should become a pretty interesting and quite accurate statistic as well as being fun to compare if someone wants to.

@ Bryan, thanks, I am having a lot of fun with them. There are some really great games amongst them too.

Best regards
_________________
Nick
Back to top
View user's profile Send private message Visit poster's website
Volodymyr
Member


Joined: 08 Apr 2017
Posts: 37
Location: Ukraine,Radyvyliv

PostPosted: Sun Jul 16, 2017 4:24 pm    Post subject: Reply with quote

Fully agree.I will watch this topic. Very interesting. Smile
Back to top
View user's profile Send private message Send e-mail
spacious_mind
Senior Member


Joined: 01 Aug 2007
Posts: 2813
Location: Alabama

PostPosted: Mon Jul 17, 2017 11:42 am    Post subject: Reply with quote

HUMAN PLAYER EVALUATION BETWEEN FIDE2400/USCF2469 AND FIDE1400/USCF1274



I continued evaluating Human players listed in Chessbase between 1400 to 2400 ELO. The above chart shows their performance. Between 30 - 40 players were evaluated (15-20 games).

I am hoping to continue this in order to get a bigger sample where I will evaluate 200 players (100 games) per category. Once I am complete with this evaluation, I will remove 10% from the very high end and 10% from the low end to end up with a good average sample 160 players (80 games) per category. Which I hope will provide a good accurate Performance evaluation to compare with computer chess programs and dedicated computers.

DIVISION 3 FINAL RANKINGS PREDICTOR AFTER 2 ROUNDS



This chart has the current human performance ratings included from 1400 to 2400 ELO

Best regards
_________________
Nick
Back to top
View user's profile Send private message Visit poster's website
spacious_mind
Senior Member


Joined: 01 Aug 2007
Posts: 2813
Location: Alabama

PostPosted: Wed Jul 19, 2017 10:48 am    Post subject: Reply with quote

Round 3 Game 1:

DB MChess 1.06 - DB Kallisto 1.83

RESULT: 0-1


CLICK BELOW TO REPLAY GAME

https://lichess.org/tVpr3tPv

CRITICAL POSITION



DB MChess 1.06 had all the early initiative but lost it all by the time the above position was reached. 30. Ne3! is the natural and correct move. Here DB MChess 1.06 blundered with the move 30. h3? after which the game was lost quickly.

FINAL POSITION



GAME QUALITY SUMMARY

DB MChess 1.06, 30S. (2137)
2 Inaccuracies
5 Mistakes
1 Blunders
44 Average centipawn loss

DB Kallisto 1.83, 60/30. (2167)
2 Inaccuracies
1 Mistakes
0 Blunders
22 Average centipawn loss

Best regards
_________________
Nick
Back to top
View user's profile Send private message Visit poster's website
Chessmaster Ireland
Full Member


Joined: 18 Feb 2009
Posts: 509

PostPosted: Wed Jul 19, 2017 7:27 pm    Post subject: Reply with quote

Hi Nick

I have to give you a big "Thank You"

Why, because of this thread I started looking through Google for various old chess console programs and there in the Google Images was a photo of the front cover of the "Sinclair User" magazine from June 1984 which I thought I recognised from somewhere. Looking at it more closely I realised that this 33 year old magazine contained my very first published article about computer chess.

Basically it was a report on a twelve game match between the Sinclair Spectrum 48K program, "Cyrus-IS-Chess" versus the Fidelity Sensory 9. The mini match ended with a 8-4 victory for the Fidelity.

When I realised that it was the issue with my article in it (Page 51-52). I searched the internet to see if by any chance it was still available anywhere to buy and amazingly it was. Originally I paid 85p but now it cost me ú6 to get another copy of it.

Here are a few photos of it, thanks again Nick I had completely forgotten about it!!







Back to top
View user's profile Send private message Send e-mail
spacious_mind
Senior Member


Joined: 01 Aug 2007
Posts: 2813
Location: Alabama

PostPosted: Thu Jul 20, 2017 2:53 am    Post subject: Reply with quote

Chessmaster Ireland wrote:
Hi Nick

I have to give you a big "Thank You"

Why, because of this thread I started looking through Google for various old chess console programs and there in the Google Images was a photo of the front cover of the "Sinclair User" magazine from June 1984 which I thought I recognised from somewhere. Looking at it more closely I realised that this 33 year old magazine contained my very first published article about computer chess.

Basically it was a report on a twelve game match between the Sinclair Spectrum 48K program, "Cyrus-IS-Chess" versus the Fidelity Sensory 9. The mini match ended with a 8-4 victory for the Fidelity.

When I realised that it was the issue with my article in it (Page 51-52). I searched the internet to see if by any chance it was still available anywhere to buy and amazingly it was. Originally I paid 85p but now it cost me ú6 to get another copy of it.

Here are a few photos of it, thanks again Nick I had completely forgotten about it!!



Hi Bryan,

Thanks I am happy that I could bring back some memories for you. Yes over the years I have seen many reports that you have written in various publications. You were also mentioned in publications like the German CS&S. Therefore finding you in a Sinclair Spectrum 48K magazine is not surprising to me Smile

Cyrus is chess is actually pretty good all considering. I wouldn't mind actually seeing it play with its normal chess clock speed but running full blast on an emulator. Which is exactly what Sargon 3 is doing on the Amiga. It is running full blast yet still retaining its normal internal clock. Amazing really. On Sinclair I have tried it in the past as well with Cyrus and Colossus 4 and Chris Whittington's Superchess versions. But you can only achieve a comparison by increasing the levels to say 20 Hours per move to simulate 30 seconds per move at full speed for example and I find that less satisfactory. Amiga does it well. Everything else does not really do it as well. With Atari you have a more accurate simulation where it will only play 32 MHz for example at 68060.

Best regards
_________________
Nick
Back to top
View user's profile Send private message Visit poster's website
spacious_mind
Senior Member


Joined: 01 Aug 2007
Posts: 2813
Location: Alabama

PostPosted: Sat Aug 12, 2017 9:27 pm    Post subject: Reply with quote

Round 3 Game 2:

AT68060-32 Psion Chess 2.00 - AT68060-32 Psion Chess 2.01

RESULT: 1-0


CLICK BELOW TO REPLAY GAME

https://lichess.org/YXmsz0yu

CRITICAL POSITION



Well, when a version 2.00 plays a version 2.01 in a match, you mostly expect it to be a draw or for the higher version number to win. And this game should have been a draw. However from the above position onwards Atari v. 2.01 made some bad mistakes and loses a game that should never have been lost.

FINAL POSITION



GAME QUALITY SUMMARY

AT68060-32 Psion Chess 2.00, 30S. (2137)
2 Inaccuracies
2 Mistakes
3 Blunders
37 Average centipawn loss

AT68060-32 Psion Chess 2.01, 30S. (2137)
7 Inaccuracies
4 Mistakes
3 Blunders
53 Average centipawn loss


Round 3 Game 3:

AM68060 Checkmate Normal - AM68060 CP2150

RESULT: 1/2-1/2


CLICK BELOW TO REPLAY GAME

https://lichess.org/bQ2AlWza

FINAL POSITION



This was a solid game played by both opponents.


GAME QUALITY SUMMARY

AM68060 Checkmate Nor, 30S. (2113)
4 Inaccuracies
0 Mistakes
0 Blunders
10 Average centipawn loss

AM68060 CP2150, 30S. (2290)
4 Inaccuracies
0 Mistakes
0 Blunders
10 Average centipawn loss



Round 3 Game 4:

DB Psion Chess 1.01 - AM68060 Sargon 3

RESULT: 1-0


CLICK BELOW TO REPLAY GAME

https://lichess.org/fbfq9HJR

CRITICAL POSITION



DB Psion 1.01 sacrificed its Knight for 2 pawns in the above position by playing 11. Nxe5? This was very unsound and you are not going to get away with that kind of play against Amiga 68060 Sargon 3. Open games are right up Sargon 3's alley.

FINAL POSITION



No problems in beating DB Psion 1.01 for Amiga 68060 Sargon 3.

GAME QUALITY SUMMARY

DB Psion Chess 1.01, 30S. (2137)
9 Inaccuracies
6 Mistakes
0 Blunders
32 Average centipawn loss

AM68060 Sargon 3, 30S. (2137)
0 Inaccuracies
0 Mistakes
0 Blunders
12 Average centipawn loss

Best regards
_________________
Nick
Back to top
View user's profile Send private message Visit poster's website
spacious_mind
Senior Member


Joined: 01 Aug 2007
Posts: 2813
Location: Alabama

PostPosted: Sun Aug 13, 2017 11:31 am    Post subject: Reply with quote

Round 3 Game 5:

Millennium Chess Genius - AT68060-32 Checkmate Normal, 30S. (2113)

RESULT: 1-0

CLICK BELOW TO REPLAY GAME

https://lichess.org/tnQrQHWi

CRITICAL POSITION



AT68060-32 Checkmate Normal messed up quickly in this game. 15. ... Bxb5? loses a piece for a pawn and makes this a quick and easy win for Millennium Chess Genius.

FINAL POSITION



I could have stopped this game much earlier but in order to obtain a good evaluation through enough played moves at Lichess I let the game play past 40 moves. A minimum of 40 moves is what I want to achieve in the games to allow me later to compare well against humans.

GAME QUALITY SUMMARY

Millennium Chess Genius, 30S. (2224)
6 Inaccuracies
0 Mistakes
0 Blunders
19 Average centipawn loss

AT68060-32 Checkmate Nor, 30S. (2113)
4 Inaccuracies
5 Mistakes
0 Blunders
42 Average centipawn loss

Best regards
_________________
Nick
Back to top
View user's profile Send private message Visit poster's website
Display posts from previous:   
Post new topic   Reply to topic    HIARCS Chess Forums Forum Index -> Main Lobby / General Discussions All times are GMT
Goto page Previous  1, 2, 3
Page 3 of 3

 
Jump to:  
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum


Powered by phpBB © 2001, 2005 phpBB Group
Protected by Anti-Spam ACP