Glaurung 2.2

Discussion of anything and everything relating to chess playing software and machines.

Moderators: hgm, Harvey Williamson, bob

Forum rules
This textbox is used to restore diagrams posted with the [d] tag before the upgrade.
Marc Lacrosse
Posts: 511
Joined: Wed Mar 08, 2006 9:05 pm
Contact:

Re: Glaurung 2.2

Post by Marc Lacrosse » Mon Dec 22, 2008 8:27 pm

Dr.Wael Deeb wrote:[
Hi Marc,
I happen to share your opinion regarding this issue :D
BTW,what other PG books are available out there besides your excellent Performance opening book that you would recommend :?:
Opening theory regards,
Dr.D
Performance.bin is a brave old thing that was good in 2005 but is now outclassed by quite a few newer books. It's just still a nice general book for the average amateur. I saw that a competitor at next CCC will still play with performance (mediocre). This seems a bit suicidal but I wish good luck to him ...

AFAIK there are no publicly available PG books that clearly outperform this oldie.

But I know for sure that there are quite a few private high-class PG books with superior results on Playchess and elsewhere.

Inbetween tools for preparing PG books more easily are coming, thanks to Pascal George (PG books tuning functions in SCID) and Michel Van den Berg (moves adding and ... a brilliant novelty that will soon be public).

So I am pretty sure that PG books will have a brilliant future!

For what regards myself I am not in book cooking at the moment - this could change :-)

you should give it a try Wael !

Regards

Marc

User avatar
Dr.Wael Deeb
Posts: 9635
Joined: Wed Mar 08, 2006 7:44 pm
Location: Amman,Jordan

Re: Glaurung 2.2

Post by Dr.Wael Deeb » Mon Dec 22, 2008 8:33 pm

Marc Lacrosse wrote:
Dr.Wael Deeb wrote:[
Hi Marc,
I happen to share your opinion regarding this issue :D
BTW,what other PG books are available out there besides your excellent Performance opening book that you would recommend :?:
Opening theory regards,
Dr.D
Performance.bin is a brave old thing that was good in 2005 but is now outclassed by quite a few newer books. It's just still a nice general book for the average amateur. I saw that a competitor at next CCC will still play with performance (mediocre). This seems a bit suicidal but I wish good luck to him ...

AFAIK there are no publicly available PG books that clearly outperform this oldie.

But I know for sure that there are quite a few private high-class PG books with superior results on Playchess and elsewhere.

Inbetween tools for preparing PG books more easily are coming, thanks to Pascal George (PG books tuning functions in SCID) and Michel Van den Berg (moves adding and ... a brilliant novelty that will soon be public).

So I am pretty sure that PG books will have a brilliant future!

For what regards myself I am not in book cooking at the moment - this could change :-)

you should give it a try Wael !

Regards

Marc
Thanks Marc,I'll wait for the new tools to arrive and will give it a shot :D
Already tweaking my databases regards,
Dr.D
_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….

Uri Blass
Posts: 8221
Joined: Wed Mar 08, 2006 11:37 pm
Location: Tel-Aviv Israel

Re: Glaurung 2.2

Post by Uri Blass » Mon Dec 22, 2008 9:28 pm

bob wrote:
Spock wrote:After 900 chess960 games now (9 opponents x 100) Glaurung 2.2 is holding a +55 ELO gain over 2.1

http://www.computerchess.org.uk/ccrl/40 ... _pure.html

Still more opponents to play, but I don't expect the rating to change much now. I'd expect it to hold on to a 50+ gain, fingers crossed
Almost every test I run sees major changes after even 1,000 games, so with that few games anything can happen. Never forget the "error bar".
Ray test only FRC and in big majority of cases performance after 100 games against a single opponent gives an error that is smaller than 50 elo
relative to the rating of the program.

My conclusion based on his results is that you can predict the rating of a program in FRC based on 1000 games with an error that is not more than 50 elo with something close to 90% confidence based on 100 FRC games against a single opponent.


http://www.computerchess.org.uk/ccrl/40 ... t_all.html

You can click on every program in the list and look at the last colomn that is the difference between performance against a single opponent and rating.

It seems that in 90% of the cases or something close to 90% you will get number that is smaller than 50 and bigger than -50.

Here is the statistics for the top programs in Ray's list

Rybka3 64 bit -50<performance-rating<50 in 12 cases out of 12
Naum 4 64-bit -50<performance-rating<50 in 9 cases out of 11
Rybka 2.3.2z3 64-bit same in 16 cases out of 17
Deep Sjeng WC2008 64-bit 1CPU same in 13 cases out of 13
Shredder 11 same in 21 cases out of 22
Naum 3.1 64-bit same in 15 cases out of 16
Deep Sjeng 3.0 64-bit 1CPU same in 11 cases out of 13
Naum 3 64-bit same in 12 cases out of 12
hiarcs12.1 same in 11 cases out of 12


Uri

Golem

Re: Glaurung 2.2

Post by Golem » Tue Dec 23, 2008 4:32 pm

Thank you Tord, nice Christmas gift !

Post Reply