IS it possible for CSTal to be improved a great deal more than it was, being that its system of choosing unsound moves might not stand up to modern programs, however much it is improved?
Or maybe, certain aspects can still be used in a much more sophisticated manner, and it might then often beat even Rybka (I mean, VERY often)?
questions re cstal (Thorsten?)
Moderator: Ras
-
S.Taylor
- Posts: 8514
- Joined: Thu Mar 09, 2006 3:25 am
- Location: Jerusalem Israel
-
fern
- Posts: 8755
- Joined: Sun Feb 26, 2006 4:07 pm
Re: questions re cstal (Thorsten?)
Stuart:
The core of your question to T. has no sense at all because CSTAL was thought and made NOT to play computer and get the best results against them. It was to play us, the people, real people hat buy them.
If some extra work could be done to CSTAL should be to enhance that features, not to get points against Rybka.
We already have too many engines capable of counting beans and crush all of us regards
Fernando
The core of your question to T. has no sense at all because CSTAL was thought and made NOT to play computer and get the best results against them. It was to play us, the people, real people hat buy them.
If some extra work could be done to CSTAL should be to enhance that features, not to get points against Rybka.
We already have too many engines capable of counting beans and crush all of us regards
Fernando
-
cornetmike
Re: questions re cstal (Thorsten?)
I agree with Fernando!fern wrote:Stuart:
The core of your question to T. has no sense at all because CSTAL was thought and made NOT to play computer and get the best results against them. It was to play us, the people, real people hat buy them.
If some extra work could be done to CSTAL should be to enhance that features, not to get points against Rybka.
We already have too many engines capable of counting beans and crush all of us regards
Fernando
regards
Mikey
p.s. I like CSTal just the way it is. IMO it is much more enjoyable to play than any other engine. "Improving" CSTal might simply remove the "enjoyment."
-
mclane
- Posts: 18899
- Joined: Thu Mar 09, 2006 6:40 pm
- Location: US of Europe, germany
- Full name: Thorsten Czub
Re: questions re cstal (Thorsten?)
yes i think CSTAL has this potential. its search is from 1999.
with a new search it should be possible to go way higher.
with a new search it should be possible to go way higher.
-
Dr.Wael Deeb
- Posts: 9773
- Joined: Wed Mar 08, 2006 8:44 pm
- Location: Amman,Jordan
Re: questions re cstal (Thorsten?)
Whenever you have a working new version of CSTII,pass it to me at once,I'll be more than glad to test itmclane wrote:yes i think CSTAL has this potential. its search is from 1999.
with a new search it should be possible to go way higher.
Until then,I'll continue to test the oldie but goodie one
_No one can hit as hard as life.But it ain’t about how hard you can hit.It’s about how hard you can get hit and keep moving forward.How much you can take and keep moving forward….
-
fern
- Posts: 8755
- Joined: Sun Feb 26, 2006 4:07 pm
Re: questions re cstal (Thorsten?)
I wonder if a guy as Martin Bryant, that knows a lot about searching techniques, could one day become interested in such a task OR in the sheer idea of developing a CSTAL kind of Colossus.
Yes, we are dreaming regards
Fernando
Yes, we are dreaming regards
Fernando
-
Dr.Wael Deeb
- Posts: 9773
- Joined: Wed Mar 08, 2006 8:44 pm
- Location: Amman,Jordan
Re: questions re cstal (Thorsten?)
I do believe that one day it will happen,don't underestimate Thorsten,he's full of surprisesfern wrote:I wonder if a guy as Martin Bryant, that knows a lot about searching techniques, could one day become interested in such a task OR in the sheer idea of developing a CSTAL kind of Colossus.
Yes, we are dreaming regards
Fernando
_No one can hit as hard as life.But it ain’t about how hard you can hit.It’s about how hard you can get hit and keep moving forward.How much you can take and keep moving forward….
-
fern
- Posts: 8755
- Joined: Sun Feb 26, 2006 4:07 pm
Re: questions re cstal (Thorsten?)
Hi doc:
I do not underestimate T. , but as far as I know he is not a chess programmer at all. But, yes, he could give verbal ideas to a programmer and get the thing done, why not.
Still dreaming regards
Fernando
I do not underestimate T. , but as far as I know he is not a chess programmer at all. But, yes, he could give verbal ideas to a programmer and get the thing done, why not.
Still dreaming regards
Fernando
-
Dr.Wael Deeb
- Posts: 9773
- Joined: Wed Mar 08, 2006 8:44 pm
- Location: Amman,Jordan
Re: questions re cstal (Thorsten?)
Hi,fern wrote:Hi doc:
I do not underestimate T. , but as far as I know he is not a chess programmer at all. But, yes, he could give verbal ideas to a programmer and get the thing done, why not.
Still dreaming regards
Fernando
Exactly what i want to say,he'll do it sooner or later
_No one can hit as hard as life.But it ain’t about how hard you can hit.It’s about how hard you can get hit and keep moving forward.How much you can take and keep moving forward….
-
Stan Arts
Re: questions re cstal (Thorsten?)
In some way, most of it's legacy is visible in other programs now.
Most preprocessors and piece-square-table-evaluators have vanished.
Many engines today do full complicated slow leaf-evaluations, with many kingsafety patterns and other heavy evaluation being done. (Even an amateur engine like mine, although my engine is a little biased toward evaluation (crappy/slow search to go with it) it even manages to find a lot of moves in positions mentioned in CSTal articles on low searchdepths, even though it's hardly speculative or in any way tuned for them.)
Programs throughout the ratinglists have become much less materialistic then the time when CSTal was actively developed. It's not just search that stops CSTal from winning most games nowadays, it's the other programs that have filled most evaluation-holes CSTal could take advantage of.
So just let CSTal be, that way it will stay shrowded in legend and "what could of been"'s. ..
You'll find many CSTal successors already, just look around a bit.
Stan
Most preprocessors and piece-square-table-evaluators have vanished.
Many engines today do full complicated slow leaf-evaluations, with many kingsafety patterns and other heavy evaluation being done. (Even an amateur engine like mine, although my engine is a little biased toward evaluation (crappy/slow search to go with it) it even manages to find a lot of moves in positions mentioned in CSTal articles on low searchdepths, even though it's hardly speculative or in any way tuned for them.)
Programs throughout the ratinglists have become much less materialistic then the time when CSTal was actively developed. It's not just search that stops CSTal from winning most games nowadays, it's the other programs that have filled most evaluation-holes CSTal could take advantage of.
So just let CSTal be, that way it will stay shrowded in legend and "what could of been"'s. ..
You'll find many CSTal successors already, just look around a bit.
Stan
fern wrote:I wonder if a guy as Martin Bryant, that knows a lot about searching techniques, could one day become interested in such a task OR in the sheer idea of developing a CSTAL kind of Colossus.
Yes, we are dreaming regards
Fernando