[d]4k3/4p3/3pPp2/2pP1Pp1/1pP3Pp/pP4RP/P5BR/7K b - - 0 1
Even without special knowledge, (most) ab-engines avoid hxg3 after a short think.
Lc0 0.22.0 with weights 42854 struggles here, searching 75M nodes.
- Is there a weights file, which doesn't fall into this trap?
- Are there other engines with the same issue?
Lc0 weak spot position from the past
Moderators: hgm, Rebel, chrisw
-
- Posts: 247
- Joined: Tue Apr 13, 2010 10:41 am
Lc0 weak spot position from the past
Hope we're not just the biological boot loader for digital super intelligence. Unfortunately, that is increasingly probable - Elon Musk
-
- Posts: 6442
- Joined: Tue Jan 09, 2007 12:31 am
- Location: PA USA
- Full name: Louis Zulli
Re: Lc0 weak spot position from the past
320x24.J13-swa-410000Max wrote: ↑Mon Aug 19, 2019 12:46 pm [d]4k3/4p3/3pPp2/2pP1Pp1/1pP3Pp/pP4RP/P5BR/7K b - - 0 1
Even without special knowledge, (most) ab-engines avoid hxg3 after a short think.
Lc0 0.22.0 with weights 42854 struggles here, searching 75M nodes.
- Is there a weights file, which doesn't fall into this trap?
HALTED Nodes: 7.0M, N/s: 75.9k, Time: 1m 32s
6.3% Kf8 Bf3 Kg7 Rc2 Kg8 Rc3 Kh7 Rc1 Kg7 Rb1 Kh6 Kg1 Kh7 Kf2 (N: 5.5M, P: 33.41%)
4.9% Kd8 Bf3 Ke8 Kg2 Kf8 Rh1 Kg7 Rg1 Kh6 Rf1 Kh7 Be4 Kh6 (N: 974k, P: 33.94%)
4.5% hxg3 Be4 gxh2 Bg2 Kf8 Be4 Kg7 h4 Kf8 Kxh2 Kg7 hxg5 fxg5 (N: 492k, P: 32.67%)
-
- Posts: 247
- Joined: Tue Apr 13, 2010 10:41 am
Re: Lc0 weak spot position from the past
Yep, 320x24.J13-swa-410000 is great. From the beginning hxg3 is the third choice.
Hope we're not just the biological boot loader for digital super intelligence. Unfortunately, that is increasingly probable - Elon Musk
-
- Posts: 247
- Joined: Tue Apr 13, 2010 10:41 am
Re: Lc0 weak spot position from the past
Looks like the later T40 nets struggle here (i.e. 42850). Weights file 56215 does fine too.
Hope we're not just the biological boot loader for digital super intelligence. Unfortunately, that is increasingly probable - Elon Musk