TCEC 10

op12no2 · Post by **op12no2** » Fri Nov 24, 2017 4:44 pm

I saw that too and was wondering the same. I just asked and somebody told me Mark L said no problem, but depends if we were talking about the same message time-frame because he has (also) said that pre numa message.

Jouni · Post by **Jouni** » Fri Nov 24, 2017 6:13 pm

Or is it only display bug? Sometimes 1970 Komodo version displays suddenly up to 100% faster nps!

Leo · Post by **Leo** » Fri Nov 24, 2017 6:22 pm

I am surprised by this news. The Komodo team has always been good with many cores. I hope they let Komodo make the adjustment. I am assuming ML knows how to fix it.

Damir · Post by **Damir** » Fri Nov 24, 2017 6:35 pm

No adjustments. Komodo must suffer. The version Komodo is playing with should have been tested before the super final, and found out if it was buggy or not... Komodo should not have any kind of special treatment

whereagles · Post by **whereagles** » Fri Nov 24, 2017 11:53 pm

Well.. rules allow one update during stage to fix play limiting bugs.

Geonerd · Post by **Geonerd** » Sat Nov 25, 2017 12:10 am

It looks like Komodo is now running ~20% slower than in prior rounds.

How many ELO is that supposed to be worth?
I can't imagine K's NUMA bug is entirely responsible for the thrashing it's receiving.

Leo · Post by **Leo** » Sat Nov 25, 2017 12:32 am

About Houdini overall, there is something I like about its piece sacrifices and willingness to take calculated chances.

syzygy · Post by **syzygy** » Sat Nov 25, 2017 12:54 am

Geonerd wrote:It looks like Komodo is now running ~20% slower than in prior rounds.

How many ELO is that supposed to be worth?
I can't imagine K's NUMA bug is entirely responsible for the thrashing it's receiving.

Indeed. Bad luck probably plays a bigger role.

Milos · Post by **Milos** » Sat Nov 25, 2017 3:38 am

syzygy wrote:
Geonerd wrote:It looks like Komodo is now running ~20% slower than in prior rounds.

How many ELO is that supposed to be worth?
I can't imagine K's NUMA bug is entirely responsible for the thrashing it's receiving.
Indeed. Bad luck probably plays a bigger role.

Not such thing as bad luck. It is currently +5=20-0 which translates to
70+/-56Elo (2 sigma error bars).
Houdini is simply a stronger engine.

shrapnel · Post by **shrapnel** » Sat Nov 25, 2017 4:50 am

I read from TCEC chat, that ML confirmed Komodo 1970 has NUMA bug causing slow speed

That is just loser whining, nothing else.
Komodo is a good Engine and in some ways even better than Houdini, but Lefler is simply paying the price of being over-cautious with his ridiculous Contempt 0.
In Chess, as in Life, sometimes chances have to be taken to achieve something worthwhile.
He who Dares, Wins !

About Houdini overall, there is something I like about its piece sacrifices and willingness to take calculated chances.

That's EXACTLY what positive Contempt is about !

TCEC 10

Re: TCEC 10

Re: TCEC 10

Re: TCEC 10

Re: TCEC 10

Re: TCEC 10

Re: TCEC 10

Re: TCEC 10

Re: TCEC 10

Re: TCEC 10

Re: TCEC 10