FCT: 3rd Fight for Place 1 with SF7 ?!

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

Moderators: Harvey Williamson, bob, hgm

Forum rules
This textbox is used to restore diagrams posted with the [d] tag before the upgrade.
Post Reply
Frank Quisinsky
Posts: 5155
Joined: Wed Nov 18, 2009 6:16 pm
Location: Trier, Germany
Contact:

FCT: 3rd Fight for Place 1 with SF7 ?!

Post by Frank Quisinsky » Sun Dec 06, 2015 11:38 pm

Hi there,

will be the next SF 7 available in the next 10 days?
If not I will start the 3rd Fight for Place 1 with the available dev. version.

Here are the first and second results:

I should start it directly if Stockfish 7 is available.
I am in waiting position!

Code: Select all

History (engines are sync with opponents / games).
 For the second "Fight for Place 1" I gave Stockfish the "Contemp = 15" setting.

Second "Fight for Place 1"
 1 Komodo 9.2 x64                   : 3173.5   14.5   2018.5    2400   84.1%
 2 Stockfish 20Aug2015 BMI2 x64 C15 : 3162.4   14.2   1998.0    2400   83.3%

First "Fight for Place 1"
 3 Stockfish 26Apr2015 BMI2 x64     : 3159.9   14.7   1865.0    2250   82.9%
 4 Komodo 9 x64                     : 3158.5   14.1   1862.5    2250   82.8%
1:1 at the moment
:-)

For me is important ... SF7 in the next 10 days available or not!
I think the compile from SF7 will be a bit faster and SF can make maybe Elo points more.

Best
Frank
I like computer chess!

stavros
Posts: 160
Joined: Tue Dec 02, 2014 12:29 am

Re: FCT: 3rd Fight for Place 1 with SF7 ?!

Post by stavros » Mon Dec 07, 2015 1:19 pm

Frank Quisinsky wrote:Hi there,

will be the next SF 7 available in the next 10 days?
If not I will start the 3rd Fight for Place 1 with the available dev. version.

Here are the first and second results:

I should start it directly if Stockfish 7 is available.
I am in waiting position!

Code: Select all

History (engines are sync with opponents / games).
 For the second "Fight for Place 1" I gave Stockfish the "Contemp = 15" setting.

Second "Fight for Place 1"
 1 Komodo 9.2 x64                   : 3173.5   14.5   2018.5    2400   84.1%
 2 Stockfish 20Aug2015 BMI2 x64 C15 : 3162.4   14.2   1998.0    2400   83.3%

First "Fight for Place 1"
 3 Stockfish 26Apr2015 BMI2 x64     : 3159.9   14.7   1865.0    2250   82.9%
 4 Komodo 9 x64                     : 3158.5   14.1   1862.5    2250   82.8%
1:1 at the moment
:-)

For me is important ... SF7 in the next 10 days available or not!
I think the compile from SF7 will be a bit faster and SF can make maybe Elo points more.

Best
Frank
in 10 days? lool dont expect sf7 even in 1 month,they have to solve many issues since lazy sf.

Modern Times
Posts: 2604
Joined: Thu Jun 07, 2012 9:02 pm

Re: FCT: 3rd Fight for Place 1 with SF7 ?!

Post by Modern Times » Mon Dec 07, 2015 2:12 pm

I thought about this, and my guess was that we wouldn't see if for another couple of months. Definitely not in 10 days. I'm testing 251115 for CCRL blitz, which is the day K9.3 was released
.

Opinions expressed here are my own, and not necessarily those of the CCRL Group.

Frank Quisinsky
Posts: 5155
Joined: Wed Nov 18, 2009 6:16 pm
Location: Trier, Germany
Contact:

Re: FCT: 3rd Fight for Place 1 with SF7 ?!

Post by Frank Quisinsky » Mon Dec 07, 2015 9:23 pm

Hi Ray,

yes, I think I should start the test.
Will do that December 11th, 2015 after the still running TogaII test-run.

And of course, using a dev. Version from Stockfish from the same day. Larry wrote that K9.3 is available November 27th but it can be that the release Version 9.3 is older as November 27th. I wrote a short mail to Mark for that Information. But I think I should test the same version as in CCRL blitz Rating list.

Thanks for your hint!

Best and my regards
Frank

PS1 ... in my second "Fight for place 1" I gave SF dev contemp 15. For the next I will not do that, SF should play with standard settings.

PS2 ... I am not up to date in testing engines and this Fight Need 2x 5 days. But not important I will have a lot of fun. On the other hand I have which all of tested engines a lot of fun. But I will fix the problem with SF ... not sure bad maybe Contemp 15 was a bad decision.
I like computer chess!

Modern Times
Posts: 2604
Joined: Thu Jun 07, 2012 9:02 pm

Re: FCT: 3rd Fight for Place 1 with SF7 ?!

Post by Modern Times » Tue Dec 08, 2015 7:18 am

Frank Quisinsky wrote:
And of course, using a dev. Version from Stockfish from the same day. Larry wrote that K9.3 is available November 27th but it can be that the release Version 9.3 is older as November 27th.
It was released 27th, which is the important date I think, but yes the code may have been frozen a couple of days before. The Stockfish version I used was the lastest one there on 27th (Timestamp 1448414685 )
.

Opinions expressed here are my own, and not necessarily those of the CCRL Group.

Frank Quisinsky
Posts: 5155
Joined: Wed Nov 18, 2009 6:16 pm
Location: Trier, Germany
Contact:

Re: FCT: 3rd Fight for Place 1 with SF7 ?!

Post by Frank Quisinsky » Tue Dec 08, 2015 10:07 am

Hi Ray,

I can see on dev. site:

Author: Stéphane Nicolet
Date: Wed Dec 2 11:18:26 2015 +0000
Timestamp: 1449055106

Author: Joona Kiiski
Date: Tue Nov 24 17:24:45 2015 -0800
Timestamp: 1448414685

I will used the version Dec 2.
+-7 days should be OK.

Maybe SF have a time advantage from 1 Elo ... I don't know. It will be a good compansation for the slower compile (if I compare with the original versions of SF). I am not a SF expert with compiles or so but I think all should be OK what I do here.

Best
Frank
I like computer chess!

Post Reply