I received a copy of Torch 2 just some hours ago from chesscom (thanks to A.Grant!) for testing.
Testrun started. If all works as expected, the result of Torch 2 should go online on my website on monday.
Stay tuned!
Torch 2 is here
Moderator: Ras
-
- Posts: 2648
- Joined: Sat Sep 03, 2011 7:25 am
- Location: Berlin, Germany
- Full name: Stefan Pohl
-
- Posts: 31
- Joined: Sun Mar 10, 2024 1:56 pm
- Full name: Draude Stoalis
Re: Torch 2 is here
Do you know if Torch 2 has a dimension 3072 network? It is very good in other strong engines like Big-SF.
-
- Posts: 1938
- Joined: Tue Apr 19, 2016 6:08 am
- Location: U.S.A
- Full name: Andrew Grant
Re: Torch 2 is here
I like to look at things from a different perspective, in a very literal sense, so I don't see Torch as mirrored-halfka x 2560, but rather as N x 22528. So in some sense you could say that Torch has a dimension of 22528. Which makes us 733% more powerful at extremely balanced positions strictly in low sample size tests, compared to 3072 users.
-
- Posts: 126
- Joined: Tue Oct 29, 2019 4:14 pm
- Location: Canada
- Full name: Ron Doughie
Re: Torch 2 is here
+1
Brilliant explanation, Andrew!
Brilliant explanation, Andrew!
-
- Posts: 3538
- Joined: Wed Mar 08, 2006 8:15 pm
- Full name: Jouni Uski
-
- Posts: 5685
- Joined: Wed Sep 05, 2018 2:16 am
- Location: Moving
- Full name: Jorge Picado
Re: Torch 2 is here
By the time that Torch get 5 elo, Stockfish will get another 5 or more elo, the back wheel on the car will always follow the front, unless the front get rotated to the back

Russian believe that they are special, that they can Kill thousands of Ukranian civilians, but cry like babies when a few Russian...
-
- Posts: 10704
- Joined: Thu Mar 09, 2006 12:37 am
- Location: Tel-Aviv Israel
Re: Torch 2 is here
Torch needs more than it.
https://computerchess.org.uk/ccrl/404FRC/
1 Stockfish 16.1 4065 +13 −13 72.9% −165.1 51.8% 2530
2 Torch v2 4038 +14 −14 68.9% −126.4 56.6% 2073
https://computerchess.org.uk/ccrl/Chess324/
1 Stockfish 16.1 3804 +5 −5 73.5% −137.1 49.9% 2800
100.0%
2 Torch v2 3780 +5 −5 69.8% −111.4 51.7% 2800
-
- Posts: 2080
- Joined: Mon Mar 13, 2006 2:31 am
- Location: North Carolina, USA
Re: Torch 2 is here
Neural networks don't work that way. More neurons and dendrites aren't linearly mapped to greater ability.AndrewGrant wrote: ↑Wed Mar 20, 2024 11:50 amI like to look at things from a different perspective, in a very literal sense, so I don't see Torch as mirrored-halfka x 2560, but rather as N x 22528. So in some sense you could say that Torch has a dimension of 22528. Which makes us 733% more powerful at extremely balanced positions strictly in low sample size tests, compared to 3072 users.
-
- Posts: 1938
- Joined: Tue Apr 19, 2016 6:08 am
- Location: U.S.A
- Full name: Andrew Grant
Re: Torch 2 is here
CCRL's other lists are extremely compressed. 5 elo on those lists is right.Uri Blass wrote: ↑Sun Mar 24, 2024 4:26 pmTorch needs more than it.
https://computerchess.org.uk/ccrl/404FRC/
1 Stockfish 16.1 4065 +13 −13 72.9% −165.1 51.8% 2530
2 Torch v2 4038 +14 −14 68.9% −126.4 56.6% 2073
https://computerchess.org.uk/ccrl/Chess324/
1 Stockfish 16.1 3804 +5 −5 73.5% −137.1 49.9% 2800
100.0%
2 Torch v2 3780 +5 −5 69.8% −111.4 51.7% 2800
-
- Posts: 1938
- Joined: Tue Apr 19, 2016 6:08 am
- Location: U.S.A
- Full name: Andrew Grant
Re: Torch 2 is here
I'm exceptionally aware that Neural Networks don't work like this.CRoberson wrote: ↑Mon Mar 25, 2024 1:28 amNeural networks don't work that way. More neurons and dendrites aren't linearly mapped to greater ability.AndrewGrant wrote: ↑Wed Mar 20, 2024 11:50 amI like to look at things from a different perspective, in a very literal sense, so I don't see Torch as mirrored-halfka x 2560, but rather as N x 22528. So in some sense you could say that Torch has a dimension of 22528. Which makes us 733% more powerful at extremely balanced positions strictly in low sample size tests, compared to 3072 users.