Crimson 1.1 available

[kranium]

major changes:

eval/material/search parameters tuned
code cleanup/optimization
middlegame -> endgame evaluation/recognition improved
piece square-tables tweaked
verification search removed

http://www.cyclonechess.com/crimson.html

Regards-
Norm

[Graham Banks]

major changes:

eval/material/search parameters tuned
code cleanup/optimization
middlegame -> endgame evaluation/recognition improved
piece square-tables tweaked
verification search removed

http://www.cyclonechess.com/crimson.html

Regards-
Norm

Thanks Norm.

[Olivier Deville]

Hi Norm

Where is the source code ?

When you entered Crimson in my E group, you stated :

I will be releasing it (including source code) very soon...

It is very strange than you changed your mind.

Olivier

[kranium]

Hi Olivier,

i did in fact release the source code, as you well know.

Crimson is not open-source. it's unfortunate you're not pleased.
kindly refrain from pressuring me any further.

Norm

[Olivier Deville]

There is at least a 400 elo gap between 0.2 (open source) and 1.1, which makes the old source meaningless.

I simply won't use it, and I don't care if some gun happy testers do.

Olivier

[kranium]

There is at least a 400 elo gap between 0.2 (open source) and 1.1, which makes the old source meaningless.

I simply won't use it, and I don't care if some gun happy testers do.

Olivier

with all due respect, i think the CEGT, CCRL, and UEL are the best in the field. they are top-notch and highly respected testing organizations, it troubles me somewhat to hear you say this.

creating a functional chess program is not meaningless...it's quite an accomplishment, especially when it plays at a high level...i'd like to see you do it!

400 ELO is not accurate...from my testing (thousands of games during development), i am confident that the ELO difference between 0.2 and 1.0 (i assume you mean 1.0...1.1 was just released last night) is closer to 200, maybe 250 ELO. (of course, i have dozens of even older versions that are weaker still...in the 1600-2000 range, etc. if you're interested in old versions).

how many games did you test? i hope it's more than the handfull that occurred in Group E. please attach them here, i'd like to take a look. did you use the Intel PGO compile I posted? Matthias indicated he encountered some bugs when running/testing 0.2.

i do realize you may still be bitter, i thought we were past that, but obviously, testing is your perogative.

Norm

[Le Fou numerique]

Hi,

Crimson 1.1 or Cyclone 1.1 ?????

The compil in the archive is Cyclone 1.1 !!!!

Strange. In french we say: "Lapsus révélateur".

Patrick

[kranium]

Hi,

Crimson 1.1 or Cyclone 1.1 ?????

The compil in the archive is Cyclone 1.1 !!!!

Strange. In french we say: "Lapsus révélateur".

Patrick

ah non...c'est une erreur par moi...
il a été corrigé maintenent!

all my Intel compiles are created with a .bat file using the name 'cyclone'...didn't change it before archiving.

(mistakes reveal one thing -> we're all human!)

merci Patrick!

[David Dahlem]

Is the source for Crimson 1.0 still available?

I don't see any source download here -

http://www.cyclonechess.com/crimson.html

[Christopher Conkie]

Hi,

Crimson 1.1 or Cyclone 1.1 ?????

The compil in the archive is Cyclone 1.1 !!!!

Strange. In french we say: "Lapsus révélateur".

Patrick

Bonjour Patrick,

I have examined both Crimson 1.0 and Crimson 1.1.

They are both clones of Viper 0.1 by Tord Romstad.

Regards

Christopher