New SMIRF Beta Version 1.49

[Reinhard Scharnagl]

Hello friends of Chess960, 10x8 CRC and other FullChess games,

the new SMIRF Beta has been cleaned from old ballast, and its editing surface
hopefully has been redesigned a little bit more clearly. Moreover now a STL DLL
is no longer used in this package.

Download could be done via http://www.chessbox.de/Compu/schachsmirf_e.html .

The engine itself has not been changed much. But for those, who have missed the
last updates, progresses in 10x8 chess nevertheless should be noticable.

Best Regards, Reinhard.

[L?szl? G?sp]

Hi Reinhardt,

Thank you for the program! I already tried your program several times and I like it. (I played against an earlier version my first ever FRC game and I won:)). This version seems to be limited in strength I think because it moves instantly in standard chess. Am I right? Anyway, I plan to adopt your TMCI interface in my next engine, to be able to play this 10*8 field and even 10*10 field chess variants, not to mention to play engine-engine matches against your one:).
Good luck to your program!

Best regards,
L?szl?

[Reinhard Scharnagl]

Hi L?szl?,

I don't know, why your copy of SMIRF plays instantly. But I prosume,
that you have installed it into a folder containing an old beta. SMIRF
will preserve keys during the installation process and thus would NOT
replace outdated testing keys. Thus it is important to DELETE outdated
keys before installing a new beta.

If this is not the case, please give me more details, e.g. the displayed
key in the key dialog).

SMIRF is playing stronger 10x8 than 8x8 because of different reasons.
One of them is the avoiding of the use of opening libraries.

It would be fine to have more chess programs use the TMCI protocol,
you would actually be the first follower. But because the protocol is only
a draft, it might be productive to write me, before starting. Then I would
send you an actual specification, which could be slightly extended. I am
not sure, whether there should be some modifications, so please write
me your ideas if need be.

Having more TMCI engines would help to motivate me to rewrite the
SMIRF GUI to support engine to engine matches. Actually any used
engine must have the same name like SMIRF's engine, which is of course
not the desired way.

Best, Reinhard.

[Reinhard Scharnagl]

Hi again L?szl?,

it happens, that actually I am also thinking it over to support 10x10 chess
variants in SMIRF, but I am not yet planning to support more piece types.

Any TMCI using chess programmer of course will receive an unrestricted
key to use the most current SMIRF GUI and engine. But I am not willing
to participate in the general Freeware scene, dumping the efforts of chess
programmers without any need.

The SMIRF engine actually is working also as a type of referee. It supports
the GUI e.g. with valid moves and positions. May be it will helpfull to group
the abilities of engines, because not all engines need to be able to do this.
A trusted and aware engine could be linked as a referee. So engines
could be designed simpler in their interface than SMIRF actually is. The
GUI itself is not knowing anything about chess. This keeps the knowledge
about Chess strictly seperated into the engines.

Regards, Reinhard.

[L?szl? G?sp]

Hi Reinhard,

Your advice was good I cleared the old directory and reinstalled the program. Now it is working well, I am testing it.

Thank you for the cooperation in advance! I will notify you as soon as I have a working move generator with simple alpha-beta to try it out. Then I will have a closer look into your TMCI interface.

All the best!

Regards,
L?szl?

[Reinhard Scharnagl]

Actually there is a new Smirf beta engine 1.52 with an enriched GUI (some more variants) available.

Regards, Reinhard.