" End of an Era" is coming?

[henk2]

Not sure who is responsible for LC0, but holy crap ... what an incredible achievement! I would have never thought it would be possible to even be in the top 50 engines a few months back! Also what a monster SF has become! It seems like only recently where SF was slowly climbing the engine ladder and now it has become a real beast!

Blondie25 was a top 12/13 engine back in 2006. Blondie25 was a NN checkers engine (Blondie24) ported over to chess.
So it was just a matter of time.

https://ieeexplore.ieee.org/document/4100133

"The results of 12 games played on each side were 1 win, 4 losses, and 7 draws for black, and 2 wins, 6 losses, and 4 draws for white. Fritz 8.0 is rated currently at 2752 (plusmn20) on SSDF (the acronym for the Swedish Chess Computer Association), placing it as the 12th strongest program in the world. At the time of the contest between Blondie25 and Fritz 8.0, Fritz 8.0 was rated #5 in the world. The results are the first case of an evolved chess program defeating a world-class chess program (three times). The performance rating for Blondie25 against Fritz 8.0 was 2635.33, which compares well with the previous performance rating of 2650 against Pocket Fritz 2.0. Blondie25 was then tested against a nationally ranked human chess master, rated 2301. In four games, Biondie25 won three and lost one"

"The weights of the neural network were obtained by an evolutionary algorithm (an approach now called neuroevolution). In this case, a population of Blondie24-like programs played each other in checkers, and those were eliminated that performed relatively poorly. Performance was measured by a points system: Each program earned one point for a win, none for a draw, and two points were subtracted for a loss. Points were earned for each neural network after a multiple of games; the neural networks did not know which individual games were won, lost, or drawn. After the poor programs were eliminated, the process was repeated with a new population derived from the winners. In this way, the result was an evolutionary process that selected programs that played better checkers games."

The only difference is that GPUs weren't as strong in the past so it used alpha/beta pruning instead of a MCTS-like algorithm as it had to rely on CPU power. But NN-engines have had top-tier potential for a while now.

[Nay Lin Tun]

Next CCC tournament is 1+2. My bet is Leela will beat SF in Bullet.

In their hardware SF is approx +20 elo stronger in 30+5 TC,
but in 1+2 Bullet Leela will be approx +20 elo stronger.

(Expecting end of an era in bullet at least)

[Uri Blass]

Next CCC tournament is 1+2. My bet is Leela will beat SF in Bullet.

In their hardware SF is approx +20 elo stronger in 30+5 TC,
but in 1+2 Bullet Leela will be approx +20 elo stronger.

(Expecting end of an era in bullet at least)

I do not consider it end of an era when the programs do not use equal hardware.

Usually if you take an engine that is 200 elo weaker and give it some significant hardware advantage then you will find that the weaker engine wins at fast time control but lose at slow time control.

if lc0 lose at slow time control but win at bullet then the conclusion is that probably lc0 is a weaker software that use a better hardware.

[Leo]

When Alpha Zero defeats Stockfish in a 50 to 100 game match with Stockfish using 44 real cores and using the latest Brainfish book then I will consider Zero the best engine.

[Alexander Schmidt]

Next CCC tournament is 1+2. My bet is Leela will beat SF in Bullet.

In their hardware SF is approx +20 elo stronger in 30+5 TC,
but in 1+2 Bullet Leela will be approx +20 elo stronger.

(Expecting end of an era in bullet at least)

I do not consider it end of an era when the programs do not use equal hardware.

Usually if you take an engine that is 200 elo weaker and give it some significant hardware advantage then you will find that the weaker engine wins at fast time control but lose at slow time control.

if lc0 lose at slow time control but win at bullet then the conclusion is that probably lc0 is a weaker software that use a better hardware.

As soon as the MC search give a statistically relevant result it will not gain much with even more thinking time. That doesn't make it a better or a weaker software. It is just completely different.

[M ANSARI]

I think everyone here assumes that SF will just sit idly by and let NN engines take over. There is nothing that stops SF from also developing a hybrid type of fork to its engine that can also use GPU for NN or MC search. There is no question that SF is the absolute best in AB engines today and my guess is that it will soon start using some sort of hybrid setup that makes use of GPU. At the moment, NN engines are still weak in many areas and it might be a while before those areas are patched up. I think there is still a lot of easy ELO points for LC0 by simply improving how it uses available hardware resources, and that alone might be enough to cover weak areas.

[Jouni]

If Lc0 sometimes goes to SF level or even higher it seems to be almost useless for any serious position analysis: tactics, endgames and mate finding!

[AdminX]

I think everyone here assumes that SF will just sit idly by and let NN engines take over. There is nothing that stops SF from also developing a hybrid type of fork to its engine that can also use GPU for NN or MC search. There is no question that SF is the absolute best in AB engines today and my guess is that it will soon start using some sort of hybrid setup that makes use of GPU. At the moment, NN engines are still weak in many areas and it might be a while before those areas are patched up. I think there is still a lot of easy ELO points for LC0 by simply improving how it uses available hardware resources, and that alone might be enough to cover weak areas.

I believe I read on Discord that 'AntiFish' is actually a tool developed for Stockfish to help them find and correct weaknesses in Stockfish. Not the only tool used to improve Stockfish, but one of them.

[henk2]

...

[jp]

...
I believe I read on Discord that 'AntiFish' is actually a tool developed for Stockfish to help them find and correct weaknesses in Stockfish. Not the only tool used to improve Stockfish, but one of them.

Yeah, looks like an idea to "help improve Stockfish's original evaluation function", but not coming from within the SF community and possibly not getting any traction from either SF or Lc communities, so it may have trouble becoming a real tool.