Where are the funs of Leela?

[corres]

I should like to read some interpretation about the results of Leela in TCEC 15. Champion Bonus.
Where is the "New Era"?

[Nay Lin Tun]

sample-size-is-too-small.jpg

[marsell]

Yes you're right! Even hundred games in the final Tcec S15 are too few.

[corres]

sample-size-is-too-small.jpg

If you think that you never refer to results of TCEC!
It is humorous when Leela win you say: "End of Era (of AB engines)!"
If results of Leela are weak "Sample size is too small".
So?

[supersharp77]

sample-size-is-too-small.jpg

If you think that you never refer to results of TCEC!
It is humorous when Leela win you say: "End of Era (of AB engines)!"
If results of Leela are weak "Sample size is too small".
So?

Hahahahaha..........Bravo My Friend...BRAVO!!

[Nay Lin Tun]

Yes you're right! Even hundred games in the final Tcec S15 are too few.

Someone says LOS is 96. Yes, there is 4% chance that SF was really better than Leela and " End of an Era was a scam".

images (1).png

[MikeGL]

I should like to read some interpretation about the results of Leela in TCEC 15. Champion Bonus.
Where is the "New Era"?

As already mentioned by some, this leela is too young and there's huge room for improvement. A bit of patches here and there might make this deadlier than it is now. Compared to AB engines which started since the 70's

[ChiefPushesWood]

This thread is JUST AS SILLY as the 'end of an era' threads. Anyone who has paid attention over the last few months can make the following, fair assessments...
1. Stockfish is STILL a great engine... It is clear that the developers have tweaked it's settings in such a way to punish lower tier'd engines in these competitions. IE. Contempt=24 to stop it from drawing them so much. No surprise at all that they're doing so well.

2. Leela is a great engine... It is clear that Leela's developers, in it's young stage, have built it to beat SF in H2H matchups. And it has done so. Unfortunately, Leela still struggles with draws (and even some losses) against lesser engines. I believe Leela can be really scary over the next year or so.

3. These 'competitions' should be taken with a grain of salt. Issues like hardware, opening line selection (most of which NOBODY would ever play in serious chess) and time controls all serve to color the picture one way or the other.

The weiner-pining by fan-boys from both sides is slightly amusing and highly annoying. Here's the truth... its a GREAT time to be alive and involved with Computer Chess.

Chief

[corres]

This thread is JUST AS SILLY as the 'end of an era' threads. Anyone who has paid attention over the last few months can make the following, fair assessments...
1. Stockfish is STILL a great engine... It is clear that the developers have tweaked it's settings in such a way to punish lower tier'd engines in these competitions. IE. Contempt=24 to stop it from drawing them so much. No surprise at all that they're doing so well.

2. Leela is a great engine... It is clear that Leela's developers, in it's young stage, have built it to beat SF in H2H matchups. And it has done so. Unfortunately, Leela still struggles with draws (and even some losses) against lesser engines. I believe Leela can be really scary over the next year or so.

3. These 'competitions' should be taken with a grain of salt. Issues like hardware, opening line selection (most of which NOBODY would ever play in serious chess) and time controls all serve to color the picture one way or the other.

The weiner-pining by fan-boys from both sides is slightly amusing and highly annoying. Here's the truth... its a GREAT time to be alive and involved with Computer Chess.

Chief

Where is your logic?
If "Stockfish is STILL a great engine..." only the posts were silly what said 'end of an era'.
I think there is no any "end of era" only the competition is between developers of Stockfish and the developers of Leela.
We, users are the beneficiary this competition.
Thanks for the developers and their supporters.

[Alexander Schmidt]

I should like to read some interpretation about the results of Leela in TCEC 15. Champion Bonus.
Where is the "New Era"?

An 3000 ELO Engine will performe better against a older version of itself with 2990 ELO than against a different engine with 2990 ELO. Thus it is not a surprise that in this setup some Stockfish engines took the lead.

Unfortunately this forum transformed from an experts-forum to a fan-forum...