1K Real Chess for ZX80

Any discussions related to the creation of new hardware or software for the ZX80 or ZX81
Moggy
Posts: 3231
Joined: Wed Jun 18, 2008 2:00 pm

Re: 1K Real Chess for ZX80

Post by Moggy »

Hi Stefano.

Results of a game with the spectrum playing white (IS chess level 1) and ZX80 playing black(4ply game)

White won easily with ZX80 taking between 5 and 40 seconds per move. Yesterday I played a game where the ZX80(black) forced a draw under the threefold repetition rule and was heading to force a draw anyway under the 50 move rule, but sadly I didn't log this one.

I have made a little numbered grid that fits to the monitor so I can easily read the ZX80's moves and transfer the notation when using the ZX81 as an opponent which is what we set out to do and board orientation is no longer a problem so will go back to the ZX81 as the opponent. :D
Attachments
chess.zip
(48.23 KiB) Downloaded 189 times
Moggy
Posts: 3231
Joined: Wed Jun 18, 2008 2:00 pm

Re: 1K Real Chess for ZX80

Post by Moggy »

A quick result .

ZX81 ZXSuperchess 2 (white)level 0 against ZX80 1k 4ply game ( black)

Mate in 9 moves!

Had to force last ZX80 move due mate bug.

ZX80 average time 10 secs per move except in check positions up-to 2 mins ZX81 2 secs per move


Played this game twice same result.
Attachments
chess2.zip
(872 Bytes) Downloaded 173 times
User avatar
marste
Posts: 250
Joined: Sun Aug 10, 2014 9:58 pm
Location: Italy
Contact:

Re: 1K Real Chess for ZX80

Post by marste »

I was trying to follow the first game moves and make a pgn but I found some illegal ones and even "interpreting" what the move could be at the end probably I did something wrong and the result is not complete...

So I decided to do a test myself (before seeing the 9 move shame game! :( ).

Spectrum Chess IS Level 2 (0 is the first ;)) playing white and the current debugging version of 1k Real Chess playing black: the super micro one unfortunately lost, even if it was "resisting" for 37 moves!...

For the records attached the PGNs (the first incomplete).

Before testing more I should try to debug and improve the little baby, even if for casual players might be good as it is! (there is a hope to win! :lol: )

This weekend I'm away. Hope for the next!

Happy Easter!
Attachments
test-4a-debug-n02.pgn.txt
(686 Bytes) Downloaded 174 times
test-4a-debug-n01.pgn.txt
(525 Bytes) Downloaded 155 times
Moggy
Posts: 3231
Joined: Wed Jun 18, 2008 2:00 pm

Re: 1K Real Chess for ZX80

Post by Moggy »

Stefano let me be the first to say all your work on this is fantastic. :ugeek:

One thing to bare in mind is that humans make illogical moves unlike computers so the results of ZX81 v ZX80 are not really fair. When up against a human opponent the results are far different to the cold logic of another computer and as a result are more enjoyable. The 9 move game from the ZX81 I never expected as IS chess took so long and is supposed to be tournament level.

I have had more pleasure from this than you could know and will still keep logs for you if needed.



Best regards.

Moggy.
User avatar
marste
Posts: 250
Joined: Sun Aug 10, 2014 9:58 pm
Location: Italy
Contact:

Re: 1K Real Chess for ZX80

Post by marste »

For the record here also the Superchess game
Attachments
test-4a-debug-n03.pgn.txt
(273 Bytes) Downloaded 166 times
User avatar
marste
Posts: 250
Joined: Sun Aug 10, 2014 9:58 pm
Location: Italy
Contact:

Re: 1K Real Chess for ZX80

Post by marste »

Moggy wrote: Fri Mar 30, 2018 10:18 pm Stefano let me be the first to say all your work on this is fantastic. :ugeek:

One thing to bare in mind is that humans make illogical moves unlike computers so the results of ZX81 v ZX80 are not really fair. When up against a human opponent the results are far different to the cold logic of another computer and as a result are more enjoyable. The 9 move game from the ZX81 I never expected as IS chess took so long and is supposed to be tournament level.

I have had more pleasure from this than you could know and will still keep logs for you if needed.
Very glad to hear that!! :D

For now I would stop engine testing. Clearly some improvement is needed! Let's see how it will be possible..

Thank you!
Moggy
Posts: 3231
Joined: Wed Jun 18, 2008 2:00 pm

Re: 1K Real Chess for ZX80

Post by Moggy »

marste wrote: Fri Mar 30, 2018 10:27 pm
Moggy wrote: Fri Mar 30, 2018 10:18 pm Stefano let me be the first to say all your work on this is fantastic. :ugeek:

One thing to bare in mind is that humans make illogical moves unlike computers so the results of ZX81 v ZX80 are not really fair. When up against a human opponent the results are far different to the cold logic of another computer and as a result are more enjoyable. The 9 move game from the ZX81 I never expected as IS chess took so long and is supposed to be tournament level.

I have had more pleasure from this than you could know and will still keep logs for you if needed.
Very glad to hear that!! :D

For now I would stop engine testing. Clearly some improvement is needed! Let's see how it will be possible..

Thank you!

Understood. :D
Moggy
Posts: 3231
Joined: Wed Jun 18, 2008 2:00 pm

Re: 1K Real Chess for ZX80

Post by Moggy »

A small progress.

ZX81 playing White using the simplest 16k chess ZX Chess level 1.

ZX80 4ply 1k playing black.

Stalemate by threefold repetition rule.

No illegal moves or both computers would flag this up.

The log is a direct printout from ZX81 and is very basic unfortunately but easy to follow.
Attachments
chess.zip
(2.39 KiB) Downloaded 180 times
User avatar
marste
Posts: 250
Joined: Sun Aug 10, 2014 9:58 pm
Location: Italy
Contact:

Re: 1K Real Chess for ZX80

Post by marste »

Good! :)
(for the record I made the pgn file attached)

I saw also in this game moves that doesn't convince me...
Can you try one game also with the "35" version? (with black smooth a bit the initial moves still without logic...)

Thank you!
Attachments
test-4a-debug-n04.pgn.txt
(632 Bytes) Downloaded 171 times
Moggy
Posts: 3231
Joined: Wed Jun 18, 2008 2:00 pm

Re: 1K Real Chess for ZX80

Post by Moggy »

OK Stefano.

Something to consider.

I played the ZX80 4 ply against ZX81 chess level 2. The ZX80 played white but I made the opening move of d2-d4 instead of the ZX80 choosing its own move which is always knight g1-h3 and by doing this the logic performed better and the game ended in stalemate with both sides having lost most pieces except a few pawns and a knight for white and a few pawns and a rook for black remaining.

I do think think some kind of simple opening book move should be considered because when the ZX80 makes its own opening moves it stops attacking and tends to endlessly move a rook side to side or do nothing till black advances well up the board.


I will try what you ask and report back.

Moggy.
Post Reply