We were forced to make a quick release of the next version of Glass because of a couple important bugfixes
- opening book is random again
- losses on time will occur less frequently
- minimal search depth introduced
- some book lines corrected
as a bonus, Enhanced Transposition Cutoff is added.
strength gain at longer controls should be minimal, but at short ones there will be some improvement.
please visit www.marittima.pl/glass and download a new version, and enjoy
Glass 1.1
Moderators: hgm, Harvey Williamson, bob
Forum rules
This textbox is used to restore diagrams posted with the [d] tag before the upgrade.
This textbox is used to restore diagrams posted with the [d] tag before the upgrade.
Glass 1.1
Pawel Koziol
http://www.pkoziol.cal24.pl/rodent/rodent.htm
http://www.pkoziol.cal24.pl/rodent/rodent.htm
Re: Glass 1.1
ok here's a bug and pretty big one too:
I used Glass 1.1 UCI in Fritz. Hours after I closed the entire program down I noticed my machine was sluggish. When I checked task manager it said Glass was still running even though the Fritz GUI had long been closed!
I used Glass 1.1 UCI in Fritz. Hours after I closed the entire program down I noticed my machine was sluggish. When I checked task manager it said Glass was still running even though the Fritz GUI had long been closed!
Re: Glass 1.1
I have never had this problem with recent Glass versions. As soon as it receives the command "quit" it terminates. No matter what it was just doing.
The GUI (Fritz in your case) is supposed to send "quit" when it wants the engine to exit.
Are you sure that you didn't close Fritz in an unusual way, so that it wasn't able to deal with this issue?
The GUI (Fritz in your case) is supposed to send "quit" when it wants the engine to exit.
Are you sure that you didn't close Fritz in an unusual way, so that it wasn't able to deal with this issue?
-
Matthias Gemuh
- Posts: 3234
- Joined: Thu Mar 09, 2006 8:10 am
- Contact:
Re: Glass 1.1
Codeman wrote:I have never had this problem with recent Glass versions. As soon as it receives the command "quit" it terminates. No matter what it was just doing.
The GUI (Fritz in your case) is supposed to send "quit" when it wants the engine to exit.
Are you sure that you didn't close Fritz in an unusual way, so that it wasn't able to deal with this issue?
Glass 1.1 does not remain in memory under ChessGUI.
It is easier for testers if an UCI engine takes pawn hash and eval hash from the Hash amount defined in GUI, instead of eating up extra memory as Glass and some other engines do.
Under ChessGUI, users have the option to let the GUI subtract such extra memory from the Hash size sent to UCI engines.
This option is for the sake of fairness towards engines that take pawn hash, eval hash, etc from the Hash amount the GUI tells them to use.
Nalimove cache and bitbases (as extra memory consumers) are a more complex issue.
Matthias.
My engine was quite strong till I added knowledge to it.
http://www.chess.hylogic.de
http://www.chess.hylogic.de