Page 1 of 4

NXC: autonomous chess program + chess robot for NXT

Posted: 04 Oct 2011, 20:44
by HaWe
my current project:
NXC: chess for NXT
:)
http://www.mindstormsforum.de/viewtopic.php?f=70&t=6790

Re: NXC: chess for NXT

Posted: 12 Oct 2011, 11:58
by spillerrec
How is it going? Can't read anything since it is years ago I was taught German in school and the forum has apparently permanently banned Google Translate...

Re: NXC: autonomous chess program + chess robot for NXT

Posted: 12 Oct 2011, 12:01
by HaWe
please ask your questions in the linked German forum, even in English!

Re: NXC: chess for NXT

Posted: 01 Nov 2011, 07:29
by bungeshea
Google Translate seems to be working fine. http://translate.google.com/translate?s ... 90&act=url

EDIT: Google Chrome (and Chromium)'s built-in translate is also working fine.

Re: NXC: chess for NXT

Posted: 01 Nov 2011, 09:33
by spillerrec
With your link, I still get this page:
mindstormsforum translate.jpg
I can access the page without Google Translate just fine, both from home and school, so it is not my IP which have been blocked. There might be different Google servers though.

Re: NXC: autonomous chess program + chess robot for NXT

Posted: 01 Nov 2011, 09:41
by HaWe
are there any question I can answer?

Re: NXC: chess for NXT

Posted: 01 Nov 2011, 09:44
by mightor
I think they banned Google on those forums. Not very handy, really.

- Xander

Re: NXC: chess for NXT

Posted: 01 Nov 2011, 18:44
by haydenstudios
mightor wrote:I think they banned Google on those forums. Not very handy, really.

- Xander
But how could they have banned google if it works fine for both me, and bungeshea?

Re: NXC: autonomous chess program + chess robot for NXT

Posted: 01 Nov 2011, 19:27
by HaWe
for me it's also working fine (well - whatever Google thinks what's fine):
Google Translate wrote:Beitrag NXC Chess: Chess program for the NXT
Hello,
I'm working on a chess program written in NXC for the NXT.
A real smart program, it is now ntl yet, but he can at least make the correct moves and already knows most of the rules of chess.
You can manually drag (using the arrow keys to select character, then BtnCenter, then select the target field, again BtnCenter: it checks whether train is allowed, if so, again with BtnExit BtnCenter or demolition) -
or
you can automatically pull him (when "auto" flashes: BtnCenter), then he expected from any trains (at the end of the train appears, then press again BtnCenter)


(Note: The latest version further down the thread!)


_________________
Regards,
Ford - "A Kingdom of Heaven if NXC had recursion" - Prefect

my personal translation would be:
NXC Chess: Chess program for the NXT
article by Ford Prefect » 29. Sep 2011, 10:12

Hello,
I'm just working about a chess program, written in NXC for the NXT.
Of course it's not a real smart program yet, but at least you can perform correct moves and the program knows most of the chess rules.
You can perform moves manually (choose the piece using BtnLeft and BtnRight, then press BtnCenter, then move to target square, again press BtnCenter: the program tests if the move is legal, if it is: once more BtnCenter or BtnExit to cancel) -
or
let the program perform moves automatically (if "auto" is blinking: press BtnCenter), then it's calculating different moves (at the end of calculation the move is shown, then BtnCenter to confirm)

...
_________________
Regards,
Ford - "A Kingdom of Heaven if NXC had recursion" - Prefect


(ps:
notice that the chess program is really, really slow: the best move needs several hours to be calculated by a 3-ply-layer deep search. I'm just about to speed this up by parallelizing a couple of NXTs in a BT network. But it's not meant to become a "seriously taken chess program ever", it was just a challenge for me to see if I will be able to understand the underlying chess programm techniques and if I will be able to handle both the chess rules and pseudo-recursive iterative search deepening dispite of the lack of NXC recursions.)

Re: NXC: chess for NXT

Posted: 02 Nov 2011, 13:21
by hassenplug
That's pretty cool. I'd be interested to know a couple things:
1) How does it evaluate the board (to determine if a move is "good")
2) How do you plan to distribute the processing across multiple NXTs.

Steve