Sjeng 2.7 released

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

Moderator: Ras

User avatar
Graham Banks
Posts: 45256
Joined: Sun Feb 26, 2006 10:52 am
Location: Auckland, NZ

Re: Sjeng 2.7 released

Post by Graham Banks »

Werner wrote:
chessfurby wrote:Werner as said above:
Eval learned from a grandmaster database (human gms). It was a first experiment with which GCP was happy.

So there is no talk about engine here ...
Yes correct,
but I have 2 interesting results here:

Code: Select all

1   Deep Sjeng 2.7 2CPU  +22/-11/=17 61.00   30.5/50
2   Fritz 10             +11/-22/=17 39.00   19.5/50

Code: Select all

1   Shredder 10          +16/-14/=20 52.00   26.0/50
2   Deep Sjeng 2.7 2CPU  +14/-16/=20 48.00   24.0/50
In our list Fritz 10 and Shredder 10 are nearly equal strong.

Congrats to that good results!
Hi Werner,

I've been very impressed with the Deep Sjeng 2.7 games I've observed so far also.
In Ray's FRC testing, Deep Sjeng 2.7 is currently performing +50-70 better than Deep Sjeng 2.5 did.

The CEGT rating will be very interesting and I am looking forward to seeing it.

Regards, Graham.
gbanksnz at gmail.com
Spock

Re: Sjeng 2.7 released

Post by Spock »

Yes, after 400 games at 40/4 FRC, I have Deep Sjeng 2.7 exactly 70 ELo higher than Deep Sjeng 2.5 (both 1CPU). Many more games to go, so let's see if that holds