Decided it was an appropriate time for release.
Changes since 2.0:
Various bugfixes mostly due to members of the community pointing out usage issues to me.
Some minor eval smoothing (simple linear interpolation between previous value).
Introduction of NNUE. For now, NNUE is either on or off, there is no hybrid method involved or available.
https://github.com/jhellis3/Stockfish/releases/tag/v3.0
Crystal 3.0
Moderators: hgm, Rebel, chrisw
-
- Posts: 251
- Joined: Sat Dec 02, 2006 10:47 pm
- Location: Toronto
- Full name: Peter Kasinski
Re: Crystal 3.0
Thank you Joseph. I really like using Crystal and look forward to testing this latest release!
Hoping someone can post the various binaries as well.
Peter
Hoping someone can post the various binaries as well.
Peter
-
- Posts: 240
- Joined: Sat Mar 18, 2006 4:01 am
- Location: Cold
-
- Posts: 352
- Joined: Thu Mar 09, 2006 6:00 pm
Re: Crystal 3.0
Thanks Joseph, the source compiled without a hitch.
This post may either be cause or result of misunderstandings.
-
- Posts: 251
- Joined: Sat Dec 02, 2006 10:47 pm
- Location: Toronto
- Full name: Peter Kasinski
Re: Crystal 3.0
Thank you, but getting an error trying to access:perejaslav wrote: ↑Fri Sep 11, 2020 3:14 pm https://pixeldrain.com/u/hu2tnLj4
ChessMan compile net is not embedded
An error occurred during a connection to pixeldrain.com. SSL received a record that exceeded the maximum permissible length.
Error code: SSL_ERROR_RX_RECORD_TOO_LONG
-
- Posts: 3187
- Joined: Sat Feb 16, 2008 7:38 am
- Full name: Peter Martan
Re: Crystal 3.0
Maybe you need to be registered at Outskirts, if it's a link from there?kasinp wrote: ↑Fri Sep 11, 2020 4:07 pmThank you, but getting an error trying to access:perejaslav wrote: ↑Fri Sep 11, 2020 3:14 pm https://pixeldrain.com/u/hu2tnLj4
ChessMan compile net is not embedded
An error occurred during a connection to pixeldrain.com. SSL received a record that exceeded the maximum permissible length.
Error code: SSL_ERROR_RX_RECORD_TOO_LONG
Peter.
-
- Posts: 1536
- Joined: Sun Oct 25, 2009 2:30 am
Re: Crystal 3.0
The problem must be somewhere else, I downloaded and I'm not.
-
- Posts: 251
- Joined: Sat Dec 02, 2006 10:47 pm
- Location: Toronto
- Full name: Peter Kasinski
Re: Crystal 3.0
You are right. In my case whitelisting the site with Malwarebytes did the trick. I wish the error message would have been more helpful, but oh well.Ozymandias wrote: ↑Fri Sep 11, 2020 6:28 pm The problem must be somewhere else, I downloaded and I'm not.
Thank you, and thanks again to Joseph for this great engine.
Peter
-
- Posts: 12566
- Joined: Wed Mar 08, 2006 8:57 pm
- Location: Redmond, WA USA
Re: Crystal 3.0
Wonderful news. Crystal is a terrific problem solver and mate finder.
Taking ideas is not a vice, it is a virtue. We have another word for this. It is called learning.
But sharing ideas is an even greater virtue. We have another word for this. It is called teaching.
But sharing ideas is an even greater virtue. We have another word for this. It is called teaching.
-
- Posts: 240
- Joined: Sat Mar 18, 2006 4:01 am
- Location: Cold
Re: Crystal 3.0
And now it is also a complete gamemode and analysis engine! Just take a look at this:Dann Corbit wrote: ↑Fri Sep 11, 2020 11:33 pm Wonderful news. Crystal is a terrific problem solver and mate finder.
Before
Code: Select all
Score of Crystal 110920 vs Crystal 010820: +175 -1 =24 [0.935]
... Crystal 110920 playing White: 97 - 1 - 2 [0.980] 100
... Crystal 110920 playing Black: 78 - 0 - 22 [0.890] 100
... White vs Black: 97 - 79 - 24 [0.545] 200
Elo difference: 463.2 +/- 72.7, LOS: 100.0 %, DrawRatio: 12.0 %
200 of 200 games finished. TC=60"+0.6"
Code: Select all
Score of Crystal 110920 vs Stockfish 080920: +34 -61 =305 [0.466]
... Crystal 110920 playing White: 20 - 17 - 163 [0.507] 200
... Crystal 110920 playing Black: 14 - 44 - 142 [0.425] 200
... White vs Black: 64 - 31 - 305 [0.541] 400
Elo difference: -23.5 +/- 16.5, LOS: 0.3 %, DrawRatio: 76.3 %
400 of 400 games finished. TC=15"+0.15"