sting-sf-3 release
Moderators: hgm, Harvey Williamson, bob
Forum rules
This textbox is used to restore diagrams posted with the [d] tag before the upgrade.
This textbox is used to restore diagrams posted with the [d] tag before the upgrade.
sting-sf-3 release
As earlier, Sting-sf-3 is based on Stockfish-211 and doesn't use any code of next versions.
Sting works as "One brain in one head" engine (play and analysis in the same way).
Sting is to prove that chess engines are able to solve many difficult positions (zugzwangs, long sacrifice, endgames, tactical) without tricks (!) and play on a very high level.
Sting-sf-3 contains many of changes in very important parts of the Stockfish's code.
In my tests new Sting seems to be stronger than the previuos version, but it is no main goal of my work.
I hope Sting can be a very good tool for chessplayers and study composers.
Many thanks for the Stockfish's team for open source code and Jim Ablett for nice logo.
Sorry, but Sting-sf-3 needs 64-bit-compilation-exe files, if I may ask, I need the help.
I prepared two 32-bits windows compilations only: http://www.mediafire.com/download.php?bnryd9jib3p0zye
If you run Sting-sf-3 bench, it should return 6450727 nodes.
New features let Sting to recognize zugzwangs delayed by pawn moves:
5K2/k7/5p1p/1p3p2/bP6/1p6/1P5p/7B w - - bm Ke7;
or solve quickly tough endgames like this:
1b6/3kB3/8/1KpPp2p/4Pp1P/5P2/8/8 w - - bm Bf8;
But many of such positions needs a very strong computer, though.
As earlier Sting is able to find such a sacrifice:
8/5pKN/5Pp1/5kP1/8/Bb6/8/7q b - - bm Qxh7;
Another feature can be very useful in some tactical positions too.
Source code: http://www.mediafire.com/download.php?d3xjiqm6yg3aud6
Sting works as "One brain in one head" engine (play and analysis in the same way).
Sting is to prove that chess engines are able to solve many difficult positions (zugzwangs, long sacrifice, endgames, tactical) without tricks (!) and play on a very high level.
Sting-sf-3 contains many of changes in very important parts of the Stockfish's code.
In my tests new Sting seems to be stronger than the previuos version, but it is no main goal of my work.
I hope Sting can be a very good tool for chessplayers and study composers.
Many thanks for the Stockfish's team for open source code and Jim Ablett for nice logo.
Sorry, but Sting-sf-3 needs 64-bit-compilation-exe files, if I may ask, I need the help.
I prepared two 32-bits windows compilations only: http://www.mediafire.com/download.php?bnryd9jib3p0zye
If you run Sting-sf-3 bench, it should return 6450727 nodes.
New features let Sting to recognize zugzwangs delayed by pawn moves:
5K2/k7/5p1p/1p3p2/bP6/1p6/1P5p/7B w - - bm Ke7;
or solve quickly tough endgames like this:
1b6/3kB3/8/1KpPp2p/4Pp1P/5P2/8/8 w - - bm Bf8;
But many of such positions needs a very strong computer, though.
As earlier Sting is able to find such a sacrifice:
8/5pKN/5Pp1/5kP1/8/Bb6/8/7q b - - bm Qxh7;
Another feature can be very useful in some tactical positions too.
Source code: http://www.mediafire.com/download.php?d3xjiqm6yg3aud6
Maybe, I can't be friendly, but let me be useful.
Re: sting-sf-3 release
Thank you Marek!
I gonne send you a PM.
JP.
I gonne send you a PM.
JP.
- Graham Banks
- Posts: 30733
- Joined: Sun Feb 26, 2006 9:52 am
- Location: Auckland, NZ
Re: sting-sf-3 release
Thanks Marek.lech wrote:As earlier, Sting-sf-3 is based on Stockfish-211 and doesn't use any code of next versions.......
Hopefully somebody can provide a 64-bit version as well.
Graham.
My email addresses:
gbanksnz at gmail.com
gbanksnz at yahoo.co.nz
gbanksnz at gmail.com
gbanksnz at yahoo.co.nz
Re: sting-sf-3 release
Both the compilation was prepared under Windows-7 (i7). First (better) can return some problems under Windows-XP.lech wrote:I prepared two 32-bits windows compilations only: http://www.mediafire.com/download.php?bnryd9jib3p0zye
The compilation is ~ 15% slower than Jim Ablett's.
Maybe, I can't be friendly, but let me be useful.
Re: sting-sf-3 release
First mini test is not bad.
A compilation of "friend" without fanfare but hopefully work on all machines.
How touchstone not bad.
http://www.mediafire.com/?jmlceijx666v25g
Comprising a 64-bit version popcnt Sting SF 3 VE_pop64
and one without instructions popcnt. Sting SF 3 VE_x64
Another OLD_LOCKS 32-bit enabled, Sting SF 3 VE_x32
All Intel Compiler and static linking, no additional libraries needed.
Marek original versions of 32-bit and the source code.
A compilation of "friend" without fanfare but hopefully work on all machines.
How touchstone not bad.
http://www.mediafire.com/?jmlceijx666v25g
Comprising a 64-bit version popcnt Sting SF 3 VE_pop64
and one without instructions popcnt. Sting SF 3 VE_x64
Another OLD_LOCKS 32-bit enabled, Sting SF 3 VE_x32
All Intel Compiler and static linking, no additional libraries needed.
Marek original versions of 32-bit and the source code.
- Graham Banks
- Posts: 30733
- Joined: Sun Feb 26, 2006 9:52 am
- Location: Auckland, NZ
Re: sting-sf-3 release
Thanks.velmarin wrote:First mini test is not bad.
A compilation of "friend" without fanfare but hopefully work on all machines.
How touchstone not bad.
http://www.mediafire.com/?jmlceijx666v25g
Comprising a 64-bit version popcnt Sting SF 3 VE_pop64
and one without instructions popcnt. Sting SF 3 VE_x64
Another OLD_LOCKS 32-bit enabled, Sting SF 3 VE_x32
All Intel Compiler and static linking, no additional libraries needed.
Marek original versions of 32-bit and the source code.
My email addresses:
gbanksnz at gmail.com
gbanksnz at yahoo.co.nz
gbanksnz at gmail.com
gbanksnz at yahoo.co.nz
Re: sting-sf-3 release
Many thanks friend!!!velmarin wrote:First mini test is not bad.
A compilation of "friend" without fanfare but hopefully work on all machines.
How touchstone not bad.
http://www.mediafire.com/?jmlceijx666v25g
Comprising a 64-bit version popcnt Sting SF 3 VE_pop64
and one without instructions popcnt. Sting SF 3 VE_x64
Another OLD_LOCKS 32-bit enabled, Sting SF 3 VE_x32
All Intel Compiler and static linking, no additional libraries needed.
Marek original versions of 32-bit and the source code.
Maybe, I can't be friendly, but let me be useful.
Re: sting-sf-3 release
Indeed (Win XP 32-bit)lech wrote:First (better) can return some problems under Windows-XP.
First:
I get "Entry point of ReleaseSRWLockExclusive procedure cannot be found in KERNEL32.dll library"
Maybe something has to be added in the same directory...
Second:
seems OK, but bench or Sting-sf-3 bench commands don't work (=> Unknown command)
But those glitches will soon be overcome, thanks Marek!
Re: sting-sf-3 release
Ernest,ernest wrote:Indeed (Win XP 32-bit)lech wrote:First (better) can return some problems under Windows-XP.
First:
I get "Entry point of ReleaseSRWLockExclusive procedure cannot be found in KERNEL32.dll library"
Maybe something has to be added in the same directory...
Second:
seems OK, but bench or Sting-sf-3 bench commands don't work (=> Unknown command)
But those glitches will soon be overcome, thanks Marek!
All is OK. Use my 2nd compilation. You haven't to run bench.
Marek
Maybe, I can't be friendly, but let me be useful.
Re: sting-sf-3 release
Another OLD_LOCKS 32-bit enabled, Sting SF 3 VE_x32 - it doesn't work under XP.velmarin wrote:First mini test is not bad.
A compilation of "friend" without fanfare but hopefully work on all machines.
How touchstone not bad.
http://www.mediafire.com/?jmlceijx666v25g
Comprising a 64-bit version popcnt Sting SF 3 VE_pop64
and one without instructions popcnt. Sting SF 3 VE_x64
Another OLD_LOCKS 32-bit enabled, Sting SF 3 VE_x32
All Intel Compiler and static linking, no additional libraries needed.
Marek original versions of 32-bit and the source code.
I recommend "my" two 32-bit compilations. One of them should work.
Maybe, I can't be friendly, but let me be useful.

