LCZero: Progress and Scaling. Relation to CCRL Elo

Discussion of anything and everything relating to chess playing software and machines.

Moderators: hgm, Harvey Williamson, bob

Forum rules
This textbox is used to restore diagrams posted with the [d] tag before the upgrade.
yanquis1972
Posts: 1762
Joined: Tue Jun 02, 2009 10:14 pm

Re: LCZero: Progress and Scaling. Relation to CCRL Elo

Post by yanquis1972 » Sat Jun 02, 2018 1:08 am

i was giving you the benefit of the doubt, in which case the point you're trying to make is very clear. i doubt ignoring you will be necessary for much longer, but thanks for the advice.

glennsamuel32
Posts: 84
Joined: Sat Dec 04, 2010 4:31 pm
Location: 223

Re: LCZero: Progress and Scaling. Relation to CCRL Elo

Post by glennsamuel32 » Sat Jun 02, 2018 1:20 am

8 games played from the point we agreed on...

"8/2p2p1p/5p2/p7/P1k4p/2p2p1P/K1P2P2/B7 b - - 0 5"

Stockfish vs Houdini (alternating)
TC 5+5
3 cpus each
1gb hash
6 man TB

All were 3-fold repetitions.

The black king has to just move between the B and C files...
How the hell is that a forced white mate ?

https://lichess.org/IaowAVbQ#26

https://lichess.org/Gh01SUTp#16

https://lichess.org/NoBBEvgM#29

Albert Silver
Posts: 2829
Joined: Wed Mar 08, 2006 8:57 pm
Location: Rio de Janeiro, Brazil

Re: LCZero: Progress and Scaling. Relation to CCRL Elo

Post by Albert Silver » Sat Jun 02, 2018 2:12 am

glennsamuel32 wrote:
Sat Jun 02, 2018 1:20 am
8 games played from the point we agreed on...

"8/2p2p1p/5p2/p7/P1k4p/2p2p1P/K1P2P2/B7 b - - 0 5"

Stockfish vs Houdini (alternating)
TC 5+5
3 cpus each
1gb hash
6 man TB

All were 3-fold repetitions.

The black king has to just move between the B and C files...
How the hell is that a forced white mate ?

https://lichess.org/IaowAVbQ#26

https://lichess.org/Gh01SUTp#16

https://lichess.org/NoBBEvgM#29
Because White wins by zugzwang. Four actually. Each time Black will be forced to play one of his pawn moves, then White begins the long trek to g1-h2-h1 to repeat the zugzwang, until Black runs out of pawn moves and is forced to let White's king in and after that it is an easy game over.
"Tactics are the bricks and sticks that make up a game, but positional play is the architectural blueprint."

glennsamuel32
Posts: 84
Joined: Sat Dec 04, 2010 4:31 pm
Location: 223

Re: LCZero: Progress and Scaling. Relation to CCRL Elo

Post by glennsamuel32 » Sat Jun 02, 2018 2:41 am

Thanks Albert !!
That's an amazing explanation.
It's obvious I'm nowhere near any expertise with chess, but your solution didn't even need a chess board !!

Are there any chess engines that can actually see these complex moves ?
Doesn't the fifty-move rule apply here ?

Also, my apologies to George and yanquis1972 for doubting the pgn and coming off rude.

carldaman
Posts: 1687
Joined: Sat Jun 02, 2012 12:13 am

Re: LCZero: Progress and Scaling. Relation to CCRL Elo

Post by carldaman » Sat Jun 02, 2018 3:00 am

glennsamuel32 wrote:
Sat Jun 02, 2018 2:41 am

Doesn't the fifty-move rule apply here ?
The 50-move counter is reset after each pawn push (or any capture).

glennsamuel32
Posts: 84
Joined: Sat Dec 04, 2010 4:31 pm
Location: 223

Re: LCZero: Progress and Scaling. Relation to CCRL Elo

Post by glennsamuel32 » Sat Jun 02, 2018 3:07 am

carldaman wrote:
Sat Jun 02, 2018 3:00 am
glennsamuel32 wrote:
Sat Jun 02, 2018 2:41 am

Doesn't the fifty-move rule apply here ?
The 50-move counter is reset after each pawn push (or any capture).
Thanks...I didn't know it applied to pawn movements too :oops:

I'll shut up for good now...

jp
Posts: 702
Joined: Mon Apr 23, 2018 5:54 am

Re: LCZero: Progress and Scaling. Relation to CCRL Elo

Post by jp » Sat Jun 02, 2018 4:14 am

glennsamuel32 wrote:
Sat Jun 02, 2018 3:07 am
carldaman wrote:
Sat Jun 02, 2018 3:00 am
The 50-move counter is reset after each pawn push (or any capture).
Thanks...I didn't know it applied to pawn movements too :oops:
I'll shut up for good now...
Don't worry. Take comfort from the knowledge that this rule has changed many times in history, including in this century (not for P movement, but...).

If the computers are good enough (they are not) it makes sense to drop the rule. (Tablebases show wins in 100s of moves.)

duncan
Posts: 9990
Joined: Mon Jul 07, 2008 8:50 pm

Re: LCZero: Progress and Scaling. Relation to CCRL Elo

Post by duncan » Sun Jun 03, 2018 11:18 am

duncan wrote:
Thu Apr 26, 2018 10:57 pm
so elo 2552 on april 14th

now april 26 it is elo 2803. so a 251 elo gain in 12 days and 21 points a day gain. (including regression). To catch stockfish at 3563, another 760 points needed which should take about 5 weeks unless it slows down. so beginning of June. just maybe.
3rd june 'only' 3121 . so a 318 gain in 5 weeks so 9 points a day gain. To catch stockfish at 3563, another 442 points needed which should take about 7 weeks unless it slows down. so beginning of august. just maybe.

Milos
Posts: 3383
Joined: Wed Nov 25, 2009 12:47 am

Re: LCZero: Progress and Scaling. Relation to CCRL Elo

Post by Milos » Sun Jun 03, 2018 2:22 pm

duncan wrote:
Sun Jun 03, 2018 11:18 am
duncan wrote:
Thu Apr 26, 2018 10:57 pm
so elo 2552 on april 14th

now april 26 it is elo 2803. so a 251 elo gain in 12 days and 21 points a day gain. (including regression). To catch stockfish at 3563, another 760 points needed which should take about 5 weeks unless it slows down. so beginning of June. just maybe.
3rd june 'only' 3121 . so a 318 gain in 5 weeks so 9 points a day gain. To catch stockfish at 3563, another 442 points needed which should take about 7 weeks unless it slows down. so beginning of august. just maybe.
Man, how come you don't get tired of constantly writing all this kind of nonsense?
I'm starting to think you are really at intellectual level of Czub.

duncan
Posts: 9990
Joined: Mon Jul 07, 2008 8:50 pm

Re: LCZero: Progress and Scaling. Relation to CCRL Elo

Post by duncan » Sun Jun 03, 2018 4:30 pm

Milos wrote:
Sun Jun 03, 2018 2:22 pm
duncan wrote:
Sun Jun 03, 2018 11:18 am
duncan wrote:
Thu Apr 26, 2018 10:57 pm

so elo 2552 on april 14th

now april 26 it is elo 2803. so a 251 elo gain in 12 days and 21 points a day gain. (including regression). To catch stockfish at 3563, another 760 points needed which should take about 5 weeks unless it slows down. so beginning of June. just maybe.
3rd june 'only' 3121 . so a 318 gain in 5 weeks so 9 points a day gain. To catch stockfish at 3563, another 442 points needed which should take about 7 weeks unless it slows down. so beginning of august. just maybe.
Man, how come you don't get tired of constantly writing all this kind of nonsense?
I'm starting to think you are really at intellectual level of Czub.
possibly that is a step up. thanks

Post Reply