Stockfish 8 SE (Special Edition)

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

Moderator: Ras

peter
Posts: 3358
Joined: Sat Feb 16, 2008 7:38 am
Full name: Peter Martan

Re: Stockfish 8 SE (Special Edition)

Post by peter »

Jimbo I wrote: (Odd, TinyUpload didn't ask me for any credit card information.)
Neither it did in my case.
Anyhow, the engine is great, thanks to Michael and to Joseph Ellis. Did I get that right, that the code for the "Study" mode is the same as for MateFinder?
:!:
Peter.
tpoppins
Posts: 919
Joined: Tue Nov 24, 2015 9:11 pm
Location: upstate

Re: Stockfish 8 SE (Special Edition)

Post by tpoppins »

MikeB wrote:The link provided above is asking me for credit card info which I am reluctant to provide. Not that I'm saying they are not legitimate , but it's just exposing your credit card info to one more port for a hacker to hack.
TinyUpload is one of the most straightforward and simple-to-use free file hosting sites. The link in my original post should take you to this page

Image

where clicking on the sf-8SE_tp.zip link should start the download immediately (or your browser may ask you first whether you want to download or to open the linked file). There are no intermediate pages, CAPTCHAs, ads, or prompts to log in or register. I'd do a thorough scan of your box for malware if I were you, Michael.

peter wrote: Did I get that right, that the code for the "Study" mode is the same as for MateFinder?
Yes, it's in the first post of the thread:
MikeB wrote:New in this update , the "study" option from SF Matefinder by Joseph Ellis was added - based on some of the posts here , many of you should enjoy this option.
[...]
One can check out the "study" option from the command line "setoption name Study value true" and then type bench - the bench nodes will be significantly higher. On any position , this will provide a different perspective and, sometimes, a very interesting "second opinion" - very nice work by Joseph Ellis.
fantasmadel50 wrote:Thank you very much for the links, unfortunately gives me network error when downloading.
Another poster (James I) provided a SendSpace link that works for me. If it doesn't work for you let me know a file hosting site you prefer and I'll try to upload the ZIP there.

To all: you should probably ignore the BMI2 compile, it's very likely to be much slower than the POPCNT one, as Jouni Uski reported in another thread.
User avatar
MikeB
Posts: 4889
Joined: Thu Mar 09, 2006 6:34 am
Location: Pen Argyl, Pennsylvania

Re: Stockfish 8 SE (Special Edition)

Post by MikeB »

Correct - the study is exactly the same, matefinder does have a few other changes in its search function which are not in SF SE.
User avatar
MikeB
Posts: 4889
Joined: Thu Mar 09, 2006 6:34 am
Location: Pen Argyl, Pennsylvania

Re: Stockfish 8 SE (Special Edition)

Post by MikeB »

sendspace link worked fine - when I clicked the zipped file in the original link I now get this warning:

Image
User avatar
MikeB
Posts: 4889
Joined: Thu Mar 09, 2006 6:34 am
Location: Pen Argyl, Pennsylvania

Re: Stockfish 8 SE (Special Edition)

Post by MikeB »

Mac and Widows Binary's are now deposited here on Github:

https://github.com/MichaelB7/Stockfish/ ... s/tag/8.SE
User avatar
MikeB
Posts: 4889
Joined: Thu Mar 09, 2006 6:34 am
Location: Pen Argyl, Pennsylvania

Re: Stockfish 8 SE (Special Edition)

Post by MikeB »

minor update for those that want to current, maybe a couple of ELO:

Code: Select all

Rank Name                      Rating   Δ     +    -     #     Σ    Σ%     W    L    D   W%    =%   OppR 
---------------------------------------------------------------------------------------------------------
   1 Cfish 051116 64 POPCNT     3113   0.0    7    7  2400 1253.0  52.2  478  372 1550  19.9  64.6  3098 
   2 Stockfish SE 110516        3107   5.7    7    7  2400 1231.0  51.3  453  391 1556  18.9  64.8  3099 
   3 Cfish 8 64 POPCNT          3105   2.5    7    7  2400 1218.5  50.8  464  427 1509  19.3  62.9  3099 
   4 Stockfish 8se              3103   1.9    7    7  2400 1211.5  50.5  433  410 1557  18.0  64.9  3100 
   5 Stockfish 8 64 POPCNT      3101   1.6    7    7  2400 1203.0  50.1  426  420 1554  17.8  64.8  3100 
   6 Stockfish MateFinder 8 6   3100   1.6    7    7  2400 1198.5  49.9  439  442 1519  18.3  63.3  3100 
   7 Stockfish 8se-Study        3071  28.4    7    7  2400 1084.5  45.2  347  578 1475  14.5  61.5  3105 
------------------
---------------------------------------------------------------------------------------

source: https://github.com/MichaelB7/Stockfish

MacOS exe: https://github.com/MichaelB7/Stockfish/ ... _MacOS.zip
User avatar
MikeB
Posts: 4889
Joined: Thu Mar 09, 2006 6:34 am
Location: Pen Argyl, Pennsylvania

Re: Stockfish 8 SE (Special Edition)

Post by MikeB »

Updated adobe and now do not get the warning message , apparently there was update posted on November 1st.

https://support.apple.com/en-us/HT202681
User avatar
MikeB
Posts: 4889
Joined: Thu Mar 09, 2006 6:34 am
Location: Pen Argyl, Pennsylvania

Re: Stockfish 8 SE (Special Edition)

Post by MikeB »

Juts for kicks , I played SF8se with the throttle option set at the highest "20", no opening book, other than that SF8se was at full strength.

[pgn][Event "Computer Chess Game"]
[Site "Mac-Pro.local"]
[Date "2016.11.05"]
[Round "-"]
[White "michaelbyrne"]
[Black "Stockfish SE 110516"]
[Result "1-0"]
[TimeControl "600+10"]
[Annotator "6... -1.62"]

1. d4 e6 {-0.25/1} 2. e4 Qh4 {+0.81/1 0.1} 3. Nc3 Bb4 {-0.30/1 0.1} 4. Bd3
Nc6 {-0.76/1 0.1} 5. Nf3 Qg4 {-1.23/1} 6. O-O d5 {-1.62/3 0.1} 7. e5 Nxd4
{-1.02/1} 8. Nxd4 h5 {-3.60/2 0.1} 9. Be2 Qg6 {-4.28/2 0.1} 10. Bd3 Qg4
{+0.00/1 0.1} 11. Be2 Qg6 {+0.00/1} 12. Bb5+ c6 {-1.84/2 0.1} 13. Nxc6 Bc5
{-4.72/3 0.1} 14. Nxa7+ Kf8 {-4.85/1 0.1} 15. Bd3 Qg4 {-4.91/6 0.1} 16.
Nab5 Bd7 {-4.64/4 0.1} 17. Nd6 Bd4 {-4.61/4 0.1} 18. Be2 Qh4 {-4.17/3 0.1}
19. g3 Qh3 {-9.33/6 0.1} 20. Qxd4 Ne7 {-9.13/1 0.1} 21. Qf4 Be8
{-8.90/4 0.1} 22. Bf3 h4 {-8.40/2 0.1} 23. Bg2 Ng6 {-12.56/1 0.1} 24. Qxf7+
Bxf7 {-7.83/4 0.1} 25. Bxh3 Nxe5 {-7.72/2 0.1} 26. Bf4 hxg3 {-8.72/3 0.1}
27. fxg3 Rxh3 {-8.74/1} 28. Bxe5 Rh7 {-13.44/2 0.1} 29. Rxf7+ Kg8
{-13.55/3} 30. Raf1 b6 {-13.14/3} 31. a3 g6 {-13.18/3 0.1} 32. Ne2 Rxf7
{-14.32/5 0.1} 33. Rxf7 Ra4 {-14.53/4 0.1} 34. Nf4 Rxf4 {-57.03/8 0.1} 35.
Rxf4 g5 {-15.34/2 0.1} 36. Rf7 g4 {-16.14/2 0.1} 37. Rg7+ Kf8 {-16.86/1}
38. Nf7 Ke8 {-17.56/6 0.1} 39. Ng5 Kd8 {-17.81/1 0.1} 40. Nxe6+ Kc8
{-18.96/6 0.1} 41. Rc7+ Kb8 {-17.02/2 0.1} 42. Nd8 Ka8 {-99.96/11 0.1} 43.
Nc6 d4 {-99.98/7 0.1} 44. Ra7#
{White mates} 1-0[/pgn]
yanquis1972
Posts: 1766
Joined: Wed Jun 03, 2009 12:14 am

Re: Stockfish 8 SE (Special Edition)

Post by yanquis1972 »

thanks for this, but can someone explain what the "study" feature does?
User avatar
MikeB
Posts: 4889
Joined: Thu Mar 09, 2006 6:34 am
Location: Pen Argyl, Pennsylvania

Re: Stockfish 8 SE (Special Edition)

Post by MikeB »

it solves certain position much quicker , e.g, this mate in 5:

[d]qb6/3N2p1/r2p4/pP1k2p1/1pp3R1/6BB/2P1P3/2N1K3 w - -

Takes standard SF-8se 35 seconds and a depth of 31 to see the mate in 5.

Switch to Study mode and it is solves at depth 24 and 8 seconds.

in this one:

[d]8/3p4/3N4/2P1k1Kb/4p3/p3P3/B7/8 w - -

takes SF-8se 4.5 seconds and in study mode almost instantly in 0.03 seconds to find the mate-in-6.

In the position below, it is slower:

[d]1B6/4p3/N3Kp2/2NP4/P2k1P2/1P2pr2/1Pn1Bppb/4nrq1 w - -

1.4 seconds for SF-8se and 16 seconds with study mode.

and