Discussion of anything and everything relating to chess playing software and machines.
Moderators: bob, hgm, Harvey Williamson
Forum rules
This textbox is used to restore diagrams posted with the [d] tag before the upgrade.
-
AdminX
- Posts: 5183
- Joined: Mon Mar 13, 2006 1:34 pm
- Location: Acworth, GA
-
Contact:
Post
by AdminX » Tue Feb 19, 2019 1:52 pm
LeelaFish is a modified version of the Lc0 engine that gets help from an outside (auxiliary) engine. Leela uses the AuxEngine's PV to boost the Policy of those moves.
https://github.com/killerducky/lc0/wiki/LeelaFish
Gui: Chessbase/ Fritz
OS: Windows Pro 64
System: i7-4790 CPU @ 3060GHz : 32 GB Ram : 2x GTX 1060
AuxEngine: Stockfish 10 64bit BMI
AuxEngineOptions: Hash=1024;Threads=2
-
Attachments
-

- Engine Settings
- hybrid.png (41.66 KiB) Viewed 7105 times
"Good decisions come from experience, and experience comes from bad decisions."
__________________________________________________________________
Ted Summers
-
Jouni
- Posts: 2024
- Joined: Wed Mar 08, 2006 7:15 pm
Post
by Jouni » Tue Feb 19, 2019 2:01 pm
Which engine is this Eman? It's topping same rating lists in PlayChess! SF clone?
Jouni
-
AdminX
- Posts: 5183
- Joined: Mon Mar 13, 2006 1:34 pm
- Location: Acworth, GA
-
Contact:
Post
by AdminX » Tue Feb 19, 2019 2:06 pm
Jouni wrote: ↑Tue Feb 19, 2019 2:01 pm
Which engine is this Eman? It's topping same rating lists in PlayChess! SF clone?
That is the theory that I subscribed to. I believe it is a Stockfish clone. Maybe someone else will correct me If I am wrong.
"Good decisions come from experience, and experience comes from bad decisions."
__________________________________________________________________
Ted Summers
-
henk2
- Posts: 30
- Joined: Mon Jan 14, 2019 6:55 am
- Full name: Henk Verbaasdonk
Post
by henk2 » Tue Feb 19, 2019 3:03 pm
Why not let Leela play until there are X pieces left and then let Stockfish take over?
No need for an A/B veto early-middle game.
-
Damir
- Posts: 2190
- Joined: Mon Feb 11, 2008 2:53 pm
- Location: Denmark
- Full name: Damir Desevac
Post
by Damir » Tue Feb 19, 2019 3:41 pm
Jouni wrote: ↑Tue Feb 19, 2019 2:01 pm
Which engine is this Eman? It's topping same rating lists in PlayChess! SF clone?
http://eman.zohosites.com/
-
Eduard
- Posts: 224
- Joined: Fri Oct 26, 2018 10:58 pm
- Location: Germany
- Full name: Eduard Nemeth
-
Contact:
Post
by Eduard » Tue Feb 19, 2019 4:34 pm
It seems that this Hybrid works only with GPU (CUDA). Right?
-
AdminX
- Posts: 5183
- Joined: Mon Mar 13, 2006 1:34 pm
- Location: Acworth, GA
-
Contact:
Post
by AdminX » Tue Feb 19, 2019 4:39 pm
Eduard wrote: ↑Tue Feb 19, 2019 4:34 pm
It seems that this Hybrid works only with GPU (CUDA). Right?
I have not tried it with CPU, but I would think that it should work with the CPU version. On second thought, I think you are right.

Only Cuda binary.
"Good decisions come from experience, and experience comes from bad decisions."
__________________________________________________________________
Ted Summers
-
Laskos
- Posts: 9486
- Joined: Wed Jul 26, 2006 8:21 pm
- Full name: Kai Laskos
Post
by Laskos » Tue Feb 19, 2019 5:40 pm
With default settings and SF10 auxiliary on 1 thread, +40 Elo points to pure Lc0 with t30 and t40 nets at bullet TC. Much better tactically. With late t30 nets, strongest engine on my PC.
-
M ANSARI
- Posts: 3411
- Joined: Thu Mar 16, 2006 6:10 pm
Post
by M ANSARI » Tue Feb 19, 2019 6:18 pm
Well ... it was only a matter of time that someone did the obvious. Kudos to whoever did that!
-
Milos
- Posts: 3387
- Joined: Wed Nov 25, 2009 12:47 am
Post
by Milos » Tue Feb 19, 2019 6:44 pm
Laskos wrote: ↑Tue Feb 19, 2019 5:40 pm
With default settings and SF10 auxiliary on 1 thread, +40 Elo points to pure Lc0 with t30 and t40 nets at bullet TC. Much better tactically. With late t30 nets, strongest engine on my PC.
Pity aux engine threads are done in a clumsy way, instead of launching a separate instance of SF for each node on a different thread coz each instance of SF will use very little memory due to low hash requirements of SF search, it launches low-depth SMP SF search that is highly inefficient.