Posted   by 

Deep Shredder For Mac

Update from Deep Shredder 11 Mac. If you already own the predecessor version Deep Shredder 11 Mac you can update for 79.99 EUR to Deep Shredder 12 Mac. Update from Shredder 11 Mac. If you already own Shredder 11 Mac you can update for 89.99 EUR to Deep Shredder 12 Mac. Upgrade from Shredder Classic 4 Mac. CleanMyMac is mainly used to deep clean your Mac to help you free up more disk space, the app also comes with a small utility — Shredder, which you can use to remove data quickly from your Mac hard drive without leaving any trace. The app is perfect for those of you who want to keep your Mac clean and fast, while also protect your privacy by permanently removing some files that are sensitive. Current Mac versions are Shredder Classic 3 Mac, Shredder 11 Mac and Deep Shredder 11 Mac. Deep Shredder 11 Mac combine extremely powerful chess play with easy handling. The level of play of the program can be adjusted to any chess playing strength, so that everyone can find an adequate chess partner.

  1. Deep Shredder For Mac Pro
  2. Deep Shredder 13
  3. File Shredder Mac
  4. Deep Shredder For Mac Computers

Home * Engines * Shredder

Deep Shredder 13 default design and window layout [1]

Shredder,
a chess engine by Stefan Meyer-Kahlen (SMK). Shredder won an astonished number of tournaments and titles, beside various World Computer Chess Championships, World Microcomputer Chess Championships and World Chess Software Championships, the IPCCC, the International CSVN Tournament, the CSVN Programmers Tournament, and more. Shredder had its debut at the WMCCC 1995 in Paderborn, and soon after winning the first title at the WMCCC 1996 in Jakarta, Shredder went commercial, distributed by Millennium 2000 GmbH - since Shredder 3 along with the Millennium Chess System (MCS), since Shredder 5 also developed by SMK supporting the UCI protocol, which was co-designed and implemented by the same author. Therefore Shredder may use any UCI-compatible graphical user interface, in particular its own Shredder Classic GUI. The long-awaited Shredder 13 and its multiprocessor counterpart Deep Shredder 13 were released on October 30, 2016.

  • 1Photos & Games
  • 14Forum Posts
  • 15External Links

WCCC 1999

WCCC 1999, round 6: Ferret - Shredder, Stefan Meyer-Kahlen and Bruce Moreland[2][3]

WCSC 2015

WCSC 2015, Shredder is winning the first blitz play-off game with Black vs. Ginkgo operated by Ingo Bauer

Quote from Shredder Chess[4] :The Shredder chess programs by Stefan Meyer-Kahlen provide a game playing and analysis tool for everybody and are accepted as one of the best chess programs of the world. Since 1996 Shredder has won twenty titles.

  1. 1996 World Micro Computer Chess Champion in Jakarta
  2. 1999 World Micro Computer Chess Champion in Paderborn
  3. 1999 World Computer Chess Champion in Paderborn
  4. 2000 World Micro Computer Chess Champion in London
  5. 2001 World Micro Computer Chess Champion Single CPU in Maastricht
  6. 2002 World Computer Chess Blitz Champion in Maastricht
  7. 2003 World Computer Chess Champion in Graz
  8. 2003 World Computer Chess Blitz Champion in Graz
  9. 2004 World Computer Chess Blitz Champion in Ramat Gan
  10. 2005 World Computer Chess Blitz Champion in Reykjavik
  11. 2006 World Computer Chess960 Champion in Mainz
  12. 2007 World Computer Chess Blitz Champion in Amsterdam
  13. 2009 World Computer Chess Champion in Pamplona
  14. 2009 Computer Olympiad (no hardware limit) in Pamplona
  15. 2009 World Computer Chess Blitz Champion in Pamplona
  16. 2010 World Computer Chess Blitz Champion in Kanazawa
  17. 2010 World Computer Chess Software Champion in Kanazawa
  18. 2013 World Computer Chess Blitz Champion in Yokohama
  19. 2015 World Computer Chess Software Champion in Leiden
  20. 2017 World Computer Chess Software Champion in Leiden

Description given in 1999 from the ICCA tournament site [5] :

Since Shredder is written in C, it is available for various hardware platforms, such as PC, Macintosh, and PDAs, and various operating systems like Windows, Linux[6] and Mac OS[7] .

Clever & Smart was an experimental triple-brain version of Shredder which played the IPCCC 1998.

Deep Shredder, first released in 2001 based on Shredder 5[8], is the multiprocessor version of Shredder. The most recent version, Deep Shredder 13[9] released on October 30, 2016 [10], is the further developed 2015 World Computer Chess Software Champion, which was a complete rewrite based on bitboards. Deep Shredder 13 has improved more than 300 Elo over Shredder 12 [11].

Shredder and Deep Shredder come with a clear and reliable graphical user interface which evolved from the Millennium Chess System - supporting the universal chess interface[12], also compatible with other UCI engines running under Windows, Linux or Mac OS. Sims for the mac download. The Shredder GUI is favoured by many programmers playing over the board tournaments with their engines, for instance Johannes Zwanzger and his WCCC 1995 winner Jonny.

Jonny's final WCCC 2015 screen in the game versus HIARCS using the Shredder GUI

Pocket Shredder is a version for WindowsPDAs, such as iPAQ[13] .

Shredder Mobile runs on most mobile phones with Java[14] .

Shredder for Android

Shredder for Android runs on all phones and tablets with at least Android 4 as well as on all Kindle Fire phones and tables since the 2nd generation[15].

Shredderbases are compact, proprietary endgame bitbases for up to 5 men with WDL information, suited to use inside the search[16].

  • Shredder 1.0: 1996
  • Shredder 2.0: 1998
  • Shredder 3.0: 1999
  • Shredder 4.0 : 2000
  • Shredder 5.0 : 2001
  • Shredder 5.32 : June 2001
  • Shredder 6.0 : December 2001
  • Shredder 7.0 : December 2002
  • Shredder 7.04 : 2003
  • Shredder 8 : February 2004
  • Shredder Classic 1.1 : October 2004
  • Shredder 9 : February 2005
  • Shredder 9.1 :
  • Shredder 9.12 : December 2005
  • Shredder Classic 1.3 : ?
  • Shredder 10 : June 2006
  • Shredder 10.1 : April 2007
  • Shredder 11 : October 2007
  • Shredder 11 SE
  • Shredder WM Edition Bonn : October 2008
  • Shredder 12: October 2009
  • Shredder 12 SE
  • Shredder Classic 4 : November 2011
  • Deep Shredder 13: October 30, 2016
  • Tom King (2000). Shredder wins the 17th World Microcomputer Chess Championship. ICGA Journal, Vol. 23, No. 3 » WMCCC 2000
  • Theo van der Storm (2004). Shredder Shows Strength in International CSVN Tournament. ICGA Journal, Vol. 27, No. 3 » ICT 2004
  • Eric van Reem (2006). Shredder Wins Second Chess960 Computer-Chess World Championship. ICGA Journal, Vol. 29, No. 4 » Chess960CWC 2006
  • Matej Guid, Ivan Bratko (2007). Factors affecting diminishing returns for searching deeper. CGW 2007 » Crafty, Rybka, Shredder, Diminishing Returns
  • Matej Guid, Ivan Bratko (2007). Factors affecting diminishing returns for searching deeper. ICGA Journal, Vol. 30, No. 2, pdf

1996 ..

  • Information about Shredder by Stefan Meyer-Kahlen, rgcc, October 22, 1996
  • Shredder2 by Mike Cooter, rgcc, July 26, 1998
  • What's wrong with Shredder? by Stefan Meyer-Kahlen, CCC, February 19, 1999
  • Re: Chess Genius 6.5 by pc..@delthis.co.uk, rgcc, November 24, 1999 » Chess Genius, Millennium Chess System

2000 ..

  • Shredder in the SSDF list by Stefan Meyer-Kahlen, CCC, April 10, 2000 » SSDF
  • Shredder wins the WMCCC 2000 by Enrique Irazoqui, CCC, August 25, 2000 » WMCCC 2000
  • Shredder 5 by Stefan Meyer-Kahlen, CCC, November 17, 2000
  • Shredder5 patch by Stefan Meyer-Kahlen, CCC, December 06, 2000
  • Wb2Uci from Odd Gunnar Malin is working well on Shredder5 GUI by Carlos E.A. Drake, Winboard Forum, March 06, 2002 » Wb2UCI
  • Shredder and Engine Book Learning by Stephen Ham, CCC, April 24, 2003 » Book Learning
  • Shredder wins in Graz after controversy by Darse Billings, CCC, December 09, 2003
  • Shredder 8 secret: search depth? by Jouni Uski, CCC, March 23, 2004
Re: Shredder 8 secret: search depth? by Vasik Rajlich, CCC, March 23, 2004 » Depth, Junior, Fritz

2005 ..

  • Shredder 9 improvements over Shredder 8 by Sandro Necchi, CCC, January 28, 2005
  • Shredder 9 by Alex Shalamanov, CCC, December 26, 2005
  • Non-deterministic behaviour of Deep Shredder - REALLY interesting by George Sobala, CCC, December 31, 2005
  • Shredder Classic 1.3 by phili_ppe, CCC, January 09, 2006
  • Shredder 10 by John Jack, CCC, February 17, 2006
  • Shredder 10.1 now released by AGove, CCC, April 11, 2007
  • Shredder for iPhonet by Stefan Meyer-Kahlen, CCC, July 06, 2007
  • Deep Shredder 11 by AGove, CCC, October 11, 2007
  • Shredder 11 Linux by Kurt Utzinger, CCC, June 08, 2008
  • UEP Shredder WC Edition by Ingo Bauer, CCC, October 18, 2008
  • Shredder 12 infos and release date by Ingo Bauer, CCC, October 03, 2009
  • Deep Shredder 12 Linux by Kurt Utzinger, CCC, October 11, 2009
  • Differences between Shredder12 and Shredder Classic 4? by Volker Pittlik, CCC, October 14, 2009
  • Shredder for Android by Stefan Meyer-Kahlen, CCC, November 17, 2009

2010 ..

  • Shredder 12 by George Bodkin, CCC, April 14, 2010
  • Which engine version is Shredder Classic 3? by Luca Hemmerich, CCC, May 09, 2010
  • Shredder for Android .. by Ingo Bauer, CCC, September 08, 2010
  • Shredder gets Gold by Peter Martan, CCC, October 02, 2010 » WCSC 2010
  • Shredder 4 classic with S12 SE engine for free by Volker Pittlik, CCC, November 06, 2011

2015 ..

  • Shredder-Chess-Engine will be updated by JoeBoden, CCC, February 09, 2015
  • Shredder is the new World Chess Software Champion by Mehmet Karaman, CCC, July 05, 2015 » WCSC 2015
  • New Shredder information? by Robert Flesher, CCC, December 27, 2015
  • Deep Shredder 13 Windows out by Stefan Meyer-Kahlen, CCC, October 30, 2016
  • Shredder 13 for Mac and Linux out by Stefan Meyer-Kahlen, CCC, November 22, 2016

Chess Engine

Shredder Computer Chess | News
  • Shredder 5.0 - das Weltmeister-Programm! by Peter Schreiner, Schachclub Leinzell Informationen zu aktueller Schachsoftware, Dezember 2000 (German)
  • Shredder 9 on top of the world - A short introduction to Shredder 9 by Peter Schreiner, ChessBase News, March 01, 2005 [17]
  • Programs and essentials by Steve Lopez, ChessBase News, March 13, 2005 » Junior 9, Shredder 9
  • Shredder 10 now available - Interview with Shredder author Stefan Meyer-Kahlen, ChessBase News, May 26, 2006
  • ChessBase DeepShredder 12 by Frank Quisinsky, Schachwelt - Computerschach, January 30, 2010 (German)
  • First win against Shredder 12! by Dana Mackenzie, Dana Blogs Chess, July 03, 2011

Shredderbases

  • Shredder 10 and the Shredderbases, ChessBase News, July 07, 2006
  • Shredderbases by Steve Lopez, ChessBase News, November 11, 2006

Misc

  1. ↑Shredder 13 Windows Screenshots from Shredder Chess
  2. ↑Image captured from Paderborn 1999 1.mp4 by Thorsten Czub hosted by Ed Schröder
  3. ↑Paderborn 1999 - Chess - Round 6 - Game 3 (ICGA Tournaments)
  4. ↑Shredder Computer Chess, 19th title added by editor
  5. ↑Shredder's ICGA Tournaments
  6. ↑Shredder Computer Chess Download - Linux
  7. ↑Shredder Computer Chess Download - Macintosh
  8. ↑Shredder 5.0 - das Weltmeister-Programm! - Deep Shredder verfügbar! - April 2001 by Peter Schreiner, Schachclub Leinzell Informationen zu aktueller Schachsoftware, Dezember 2000 (German)
  9. ↑Shredder Computer Chess Download - Deep Shredder 13
  10. ↑Deep Shredder 13 Windows out by Stefan Meyer-Kahlen, CCC, October 30, 2016
  11. ↑Re: Deep Shredder 13 Windows out by Ingo Bauer, CCC, October 30, 2016
  12. ↑Shredder Classic 3 user interface
  13. ↑Pocket Shredder: Chess for Windows Mobile
  14. ↑Shredder Mobile
  15. ↑hredder Computer Chess Download - Shredder for Android
  16. ↑Shredderbases
  17. ↑Ten years of Computer Chess revisited by Kai Laskos, CCC, March 13, 2015
Retrieved from 'https://www.chessprogramming.org/index.php?title=Shredder&oldid=16610'
TopicThe Rybka Lounge / Computer Chess / Deep Shredder 12 info and some results
Hello all
Deep Shredder 12 for WINDOWS, MAC and LINUX will be available around midnight at [url]www.shredderchess.com[/url] or [url]www.shredderchess.de[/url]!
Here I have some more details about the engine:
1 THREAD
PONDER ON
256 MB Hash
6 min + 3 sec Time control
(3GHz Core2 speed)
only 4pc Shredderbases, Tablebases, Bitbases
Starting position, changing color
NO lerning (of course)
NO book
64 bit engines default, if 32 bit was used it is mentioned in the rating list
Engine:
Program Elo + - Games Score Av.Op. Draws
1 Rybka 3 : 2827 10 10 3500 76.3 % 2624 31.5 %
2 Rybka 3 1T 32-bit : 2766 14 14 1600 68.1 % 2634 35.1 %
3 Deep Shredder 12 : 2722 14 14 1600 64.1 % 2622 36.5 %
4 Rybka 2.3.2a : 2716 6 6 8800 62.6 % 2627 39.3 %
5 Naum 4 : 2695 7 7 6000 55.0 % 2660 40.6 %
6 Rybka 2.2n2 : 2681 10 10 2700 58.8 % 2620 41.3 %
7 Rybka 1.2f : 2668 6 6 8000 64.1 % 2568 33.8 %
8 Fritz 12 32-bit : 2645 17 17 1000 52.2 % 2630 41.0 %
9 Deep Fritz 11 32-bit : 2636 14 14 1500 55.9 % 2595 38.6 %
10 Stockfish 1.4 JA : 2624 13 13 1700 49.9 % 2625 38.4 %
12 Shredder Bonn 1T 32-bit : 2621 10 10 3000 50.8 % 2616 40.0 %
13 Zappa Mexico II : 2619 6 6 8700 46.3 % 2645 40.3 %
14 Deep Shredder 11 : 2600 9 9 3700 52.1 % 2585 39.4 %
15 Onno-1-1-1 : 2598 13 13 1600 47.5 % 2615 41.2 %
16 Rybka 1.0 : 2595 11 11 2500 53.2 % 2572 36.3 %
17 Zappa Mexico I : 2590 9 9 3600 53.2 % 2568 40.4 %
18 Naum 3.1 1T : 2589 7 7 5600 47.8 % 2604 38.9 %
19 Onno-1-0-0 : 2584 13 13 1600 44.9 % 2620 39.9 %
20 DSjeng WC2008 : 2576 7 7 5800 38.4 % 2658 35.6 %
21 Toga II 1.4 beta5c BB : 2571 6 6 8200 39.1 % 2648 37.8 %
22 H12.1 MP 32-bit : 2566 6 6 9300 38.3 % 2649 37.0 %
23 DSjeng 3.0 : 2562 10 10 3200 41.0 % 2626 33.8 %
24 Glaurung 2.2 JA : 2547 8 8 5400 34.1 % 2661 33.6 %
25 Naum 2.2 : 2542 10 10 2700 43.5 % 2587 42.9 %
26 Rybka 1.0 32-bit : 2540 8 8 4400 50.1 % 2540 33.2 %
27 HIARCS 11.2 32.bit : 2534 7 7 6600 45.5 % 2566 36.3 %
28 Fruit 05/11/03 32-bit : 2531 5 5 10200 36.0 % 2631 36.8 %
29 DS 10 Balmung : 2529 10 10 2600 49.9 % 2530 42.5 %
30 LoopMP 12.32 32-bit : 2523 7 7 5600 44.7 % 2560 34.7 %
31 Loop 13.5 32-bit : 2522 8 8 4700 43.4 % 2568 37.0 %
32 Toga II 1.2.1a : 2518 7 7 6600 46.6 % 2542 35.2 %
33 ListMP 11.64b : 2515 12 12 2200 44.0 % 2558 35.9 %
34 Glaurung 2.1 : 2508 19 19 900 42.7 % 2559 33.9 %
35 Deep Shredder 10 : 2505 5 5 12100 35.3 % 2610 33.9 %
36 HIARCS 11 32-bit : 2504 19 19 800 44.9 % 2540 36.4 %
37 Naum 2.1 : 2503 9 9 3700 44.6 % 2541 36.4 %
38 Toga II 1.3x4 : 2496 19 19 800 44.6 % 2534 38.9 %
39 Hiarcs X54 : 2494 22 22 600 43.9 % 2537 36.5 %
40 Spike 1.2 Turin 32-bit : 2479 5 5 13600 32.1 % 2609 33.1 %
41 DS 9.02 1T : 2464 16 16 1400 37.9 % 2551 29.1 %
42 Deep Sjeng 2.7 : 2453 13 13 2000 31.2 % 2590 33.1 %
43 Glaurung 2-epsilon/5 : 2444 16 16 1300 34.8 % 2553 30.9 %
44 Deep Sjeng 2.5 : 2385 20 20 900 29.9 % 2532 30.6 %
and the individual results for are:
Engine:
3 Deep Shredder 12 x64 1T : 2722 1600 (+734,=584,-282), 64.1 %
Zappa Mexico II x64 : 100 (+ 47,= 37,- 16), 65.5 %
Deep Shredder 10 x64 : 100 (+ 67,= 28,- 5), 81.0 %
Spike 1.2 Turin : 100 (+ 66,= 26,- 8), 79.0 %
Fruit 05/11/03 : 100 (+ 60,= 29,- 11), 74.5 %
Rybka 2.3.2a mp : 100 (+ 33,= 45,- 22), 55.5 %
H12.1 MP : 100 (+ 58,= 35,- 7), 75.5 %
Toga II 1.4 beta5c BB : 100 (+ 64,= 30,- 6), 79.0 %
Rybka 3 : 100 (+ 15,= 40,- 45), 35.0 %
Rybka 2.2n2 mp : 100 (+ 34,= 39,- 27), 53.5 %
DSjeng WC2008 x64 : 100 (+ 41,= 42,- 17), 62.0 %
Naum 4 : 100 (+ 38,= 33,- 29), 54.5 %
Glaurung 2.2 JA : 100 (+ 50,= 40,- 10), 70.0 %
Rybka 3 1T 32-bit : 100 (+ 25,= 44,- 31), 47.0 %
Stockfish 1.4 JA : 100 (+ 41,= 43,- 16), 62.5 %
Fritz 12 : 100 (+ 41,= 40,- 19), 61.0 %
Onno-1-1-1 : 100 (+ 54,= 33,- 13), 70.5 %
Enjoy it
Ingo
Excellent !
I am one of them !
Looking forward for this programm.
According the testing, seems very strong and will be a good alternative to Rybka to have an additionnal analysis tool.
regards
Will there be a ChessBase DVD version of 'Deep Shredder 12' ? or an UCI version on DVD ?
regards.
Unfortunately, the 'NO BOOK' part tells us that this only gives a primary impression of how well Shredder 12 plays in the opening phase. Well, we already knew that Shredder plays openings well. Also, how do you get that many games with 'no book'? It seems that many games are repeated, so we actually have a statistically small sample here if you're looking at general playing strength.
Hi,
No, read carefully!
Maybe there are close to no repeated games as I use opening positions with changing colors.
Regards
Ingo
It says, 'starting position, changing color', and 'NO book'. This seems pretty clear. If you have 44 opponents, a double-round robin tournament will have 44*43 = 1892 games. All engines having more games than this have repeated games if it's a single processor.
Also, again, this only tests the programs' abilities in playing the opening, something that many chess programs are NOT programmed to do well, as there is no need to do this.
ByIngoDate 2009-10-03 21:25Edited 2009-10-03 21:34
Hello
There where no 44 Opponents for Shredder, I played 1600 games as you can see in both lists.
But your opinion is fine, just wait for some more results later.
Bye
Ingo
Then you need to be more clear: did you use an opening book or not? You stated in the initial post that you did not--if this is the case, then most of the games you played should be repeats. The statistics imply that there are not many repeats--so did you use a test suite or something? When you say that they were played from the 'starting position', that implies that the games were played from the initial board setup. Or are you implying that the starting positions were varied, as in a test suite, such that they're different from the initial board setup? If so, you should state this, and state which test suite you used. This will help buyers make an informed decision early about Shredder 12. First impressions are everything, and if these test results are actually legitimate, i.e. if they were performed with a test suite or something similar, then these are actually very strong results.
ByIngoDate 2009-10-03 22:17Edited 2009-10-03 22:22
I think you have a problem with 'starting position' - I have to admit I missed an 's' but hey .. would it make ANY sense without it?
No book, and no repeats!
I have 50 DIFFERENT opening positions
I change colors after each game
50 * 2 = 100 games vs each opponent
16 oponents
16 * 100 = 1600 games
NO repeats AND NO book - clear.
Thats it from my side - I really hope that it is enough now!
CU
Ingo
PS: I posted my list about a hundred times in the past (of course without S12) and NEVER did anyone have a problem understanding the principle. I will try to be even more precise in the future.
Thank you; this is clear now. I guess the reason there wasn't much trouble before is that there wasn't such a major situation: we have apparently a HUGE improvement by Shredder 12 over the previous version, leaping it into the second strongest engine available behind Rybka 3. With such an unexpectedly large improvement, everything had to be examined more closely.
Now that this has been done and everything has been made clear, there should be no question which engine to buy if people are only going to buy one of the engines between Fritz 12 and Shredder 12. Shredder 12 is leagues ahead of Fritz 12; indeed, it's surprising that they even bothered releasing Fritz 12 at this point with such a puny amount of improvement. I guess they just wanted to try to see if they could snag some sales before Rybka 4 comes out.
TJ lets not go all OCD on him.
what are the new features of the GUI?
Hi Felix,
Just wait for the release, there will be a brief deiscription on the web site.
Bye
Ingo
Can't wait ;)
I´m sure, I clicked 10x on the refresh button on Shredders home page in the last hour! :-)
Same here :)
there's an excellent movie on wdr at the moment (lady vengeance), so waiting is no problem ;)
Now maintenance on Shredder home page! :-)
Nice! We have a new #2! And for the 32-bit users: There are only 44 Elo-points to #1!
And for the rest (doubts, accusations, etc.): Let us see! :-)
PS: I´m most interested in playing style and rekursive analysis and off course analysis strength!
Well, as I already wrote here, SMK told me in Mainz that Shredder 12 is stronger than Naum. I'll wait for the independent tests before saying more, but I guess Ingo's tests should be about ok.
As for the 44 Elo gap: Vas is working on a good answer to that :) But it's good to see that Shredder makes progress, competition is welcome :)
I for one am finding the wait very difficult. I was never good at resisting temptations ;)
Would be awesome when R4 is out!
The posts from Lukas feel like someone taking a kid to do window-shopping at a candy store :) No doubt Vas and Lukas are creating a super 'build up' for the eventual release.

Deep Shredder For Mac Pro

ByRoland RöslerDate 2009-10-03 23:39Edited 2009-10-03 23:43
I have no doubts; Ingo is a serious tester!
So S12 is 120 Elo better than S11, 100 Elo better than S11 Bonn and 80 Elo better than F12! And we can buy it with mp-Version! :-)

> rekursive analysis Network card for macbook pro.


Shredder is already the best out there for position learning, it'll be good to see this applied to a much stronger program.
I love the current Shredder GUI. It is my favorite GUI hands down. And I love doing analysis with Shredder and Rybka. They make a great 1-2 punch. I hope that the Shredder team doesn't try to keep up with the Jones by making a gimmicky GUI to 'keep up' with Aquarium. (as Fritz 12 apparently did judging by the screen shots)
This may not be true, as the CEGT seem to indicate that Shredder 64 bit is stronger than 32 bit. That means the gap on 32 bit is bigger than in Ingo's test. As I said, I would be careful with not independend results :)
I´m rather sure there isn´t any gap in playing strength between S12 w32 and x64. If you can see a gap now in CEGT, it´s only because of few matches!
well see, of course not many games were played yet
And have in mind: Ingo´s test with 6'+3' and ponder=on on 3 GHz machine is much more similar to CEGT 40/20 (in reality only 40/11 or 40/10 on 3 GHz C2) than to 40/4 (in reality only 40/3)!
but with 64 bit- my question is if there is any 32/64 bit performence change
Ingos Kommentar.
nothing new then
on Shredder 11,i found no diff between w32 and w64,i imagine it is the same with S12.
Deep Shredder 12 w32 1CPU ELO 2972 out of 800 games
Deep Shredder 12 x64 1CPU ELO 3020 out of 700 games
It's getting unlikely that there is no difference between 32 and 64 bit. As I said, be careful with unindependend test results :)

>It's getting unlikely that there is no difference between 32 and 64 bit.


Deep Shredder 12 w32 1CPU ELO 2980 out of 1100 games
Deep Shredder 12 x64 1CPU ELO 3030 out of 1200 games
What´s your conclusion, when you know! there is no difference in playing strength?
I don't know that :) Why do you think you know?
No excuse! I´m your teacher now! The (statistic) exercise: Why we see 50 Elo difference after so many games although the two engines are equal in playing strength?
I want to see a paper (post) with at least two good eloborated reasons in two days (without asks; all the information you can get at CEGT home page and computerschach.de)! :-)
Because they're not of the same strength?
Really, we don't need any games, just 5 minutes of infinite analysis on the opening position would resolve it.
We do not know that the 2 engines are equal in the playing strength.
It is also not easy to prove it without games.
analyzing single position with no hash to get knodes per second is not enough to prove no difference.
How do you prove that the only difference between 32 bits and 64 bits is speed?
Analyzing from fixed positions is not a proof because it is not a game situation and in a game situation shredder may be using hash information from previous positions and you may get moves that you cannot reproduce from clear hash.
In theory it is possible that there is a bug that cause 32 bits to crash in games when the 64 bits does not crash and it is possible that there is a different bug that cause the 32 bits only to play weaker moves in part of the games.
Uri
I'm only looking at the 97% likelyhood, that if we get the exact same output from 64bit, just faster, it's that it's the same just faster, like Rybka. Maybe we get different output, then that'd make things different, but meanwhile there's no reason to be obsessive about the other 3% of cases (where everything is possible.)
Okay Uri, maybe good arguments that equal analyse strength isn´t equal playing strength in the same (?) program. I don´t know, if S12 has only one code and two compilations (x64 and w32). There can be differences (bugs) in tc, EGTB or EGBB usage, and many oher things (I know there is a difference in hashtable usage; with x64 you can use more than 4 GB). I see that, but I don´t believe it, because it´s SMK who writes the program! :-)
PS: I´m eased, you don´t give Felix a hint for his exercise! :-)
It is also not easy to prove it without games.
But we can prove it with games! :-) All sp engines are deterministic (?)! C.K. has to play all his games he played with S12 x64 with the same openings against the same opponents on the same machine with the same parameters again with S12 w32! If we see different results (more than 1 Elo), I have to think! :-) Maybe it´s enough he proves it only for one opponent (100 games)! If we see exactly the same games (with x64 and w32), we have a proof, I think!
Okay, you can prove it with one opening, one opponent and S12 x64 and S12 w32! Only three games are needed! It´s over after less than 20 minutes (40/3 games)!
I can´t prove it, because I have 32-bit OS only! :-(
Hello Roland
1 thread, same hash, same opponent, same time, no learning (all the same condition - you name it) and you do not get 1000 times the same game! Wanna bet?
Bye
Ingo
ByFelix KlingDate 2009-10-10 09:47Edited 2009-10-10 09:55
What I'd like to see is analysis output of the start position of both engines on the same hardware. This could at least indicate if there's a fundamental difference or not.
You could also make your own test with Shredder 12 32 bit and look what happens in your list.
Btw., I only trust Stefan, he's the only one who could convince me that the engines are equally good even if some rating list indicates something else. :)
OKOK
This is starting position with Shredder 12 UCI 32 bit:

rnbqkbnr/pppppppp/8/8/8/8/PPPPPPPP/RNBQKBNR w KQkq -

Engine: DS 12 UCI 1T (1024 MB)
von Stefan Meyer-Kahlen
13/41 0:01 +0.37 1.e2-e4 e7-e5 2.Sg1-f3 Sb8-c6
3.Lf1-b5 a7-a6 4.Lb5xc6 d7xc6
5.d2-d3 Sg8-f6 6.O-O Lf8-d6 7.d3-d4 Lc8-g4
8.d4xe5 Lg4xf3 9.Dd1xf3 Ld6xe5 (1.249.344) 661
14/31 0:02 +0.36 1.e2-e4 e7-e5 2.Sg1-f3 Sb8-c6
3.Lf1-b5 Lf8-c5 4.O-O Sg8-f6
5.Sb1-c3 Dd8-e7 6.d2-d3 a7-a6
7.Lb5-c4 b7-b5 8.Sc3-d5 De7-d8
9.b2-b4 Sc6xb4 (1.969.035) 663
15/34 0:04 +0.41 1.e2-e4 e7-e5 2.Sg1-f3 Sb8-c6
3.Lf1-b5 Lf8-c5 4.O-O Sg8-f6
5.Sb1-c3 Dd8-e7 6.d2-d3 a7-a6
7.Lb5xc6 d7xc6 8.Lc1-e3 Lc5xe3
9.f2xe3 O-O 10.h2-h3 h7-h6 11.Sf3-d2 Lc8-d7 (3.188.050) 669
16/41 0:11 +0.42 1.e2-e4 e7-e5 2.Sg1-f3 Sb8-c6
3.Lf1-b5 a7-a6 4.Lb5xc6 d7xc6 5.O-O Dd8-d6
6.Sb1-c3 Lc8-g4 7.h2-h3 Lg4xf3
8.Dd1xf3 Sg8-f6 9.d2-d3 Lf8-e7
10.Lc1-e3 O-O 11.a2-a3 (8.017.928) 671
17/34 0:16 +0.40 1.e2-e4 e7-e5 2.Sg1-f3 Sb8-c6
3.Lf1-b5 a7-a6 4.Lb5xc6 d7xc6 5.O-O Dd8-d6
6.Sb1-c3 Lc8-g4 7.h2-h3 Lg4xf3
8.Dd1xf3 Sg8-f6 9.d2-d3 Lf8-e7
10.Lc1-e3 O-O 11.a2-a3 Dd6-e6
12.b2-b4 b7-b5 (11.259.774) 675
18/41 0:38 +0.29 1.e2-e4 e7-e5 2.Sg1-f3 Sg8-f6

Deep Shredder 13


3.Sf3xe5 d7-d6 4.Se5-f3 Sf6xe4
5.Dd1-e2 Dd8-e7 6.d2-d3 Se4-f6
7.Sb1-c3 Sb8-c6 8.Lc1-g5 De7xe2+
9.Lf1xe2 Lc8-e6 10.O-O-O O-O-O
11.d3-d4 Td8-e8 12.Le2-b5 Lf8-e7
13.Th1-e1 (26.018.677) 678
19/49 1:07 +0.31 1.e2-e4 e7-e5 2.Sg1-f3 Sg8-f6
3.d2-d4 Sf6xe4 4.Lf1-d3 d7-d5
5.Sf3xe5 Lf8-d6 6.Sb1-d2 Ld6xe5
7.d4xe5 Se4-c5 8.Sd2-f3 Lc8-g4
9.Lc1-e3 Sc5xd3+ 10.Dd1xd3 c7-c6
11.O-O-O Sb8-d7 12.Dd3-d4 Lg4xf3
13.g2xf3 Dd8-e7 14.f3-f4 (46.087.995) 682 20/48 1:50 +0.33 1.e2-e4 e7-e5 2.Sg1-f3 Sg8-f6
3.d2-d4 Sf6xe4 4.Lf1-d3 d7-d5
5.Sf3xe5 Lf8-d6 6.Sb1-d2 Ld6xe5
7.d4xe5 Se4-c5 8.Sd2-f3 Lc8-g4
9.Lc1-e3 Sc5xd3+ 10.Dd1xd3 c7-c6
11.O-O-O Sb8-d7 12.Dd3-d4 Lg4xf3
13.g2xf3 Dd8-e7 14.f3-f4 (75.617.608) 686 21/53 3:17 +0.34 1.e2-e4 e7-e5 2.Sg1-f3 Sg8-f6
3.d2-d4 Sf6xe4 4.Lf1-d3 d7-d5
5.Sf3xe5 Lf8-d6 6.Sb1-d2 Ld6xe5
7.d4xe5 Se4-c5 8.Sd2-f3 Lc8-g4
9.Lc1-e3 Lg4xf3 10.g2xf3 Sb8-d7
11.Ld3-f5 c7-c6 12.O-O g7-g6
13.Lf5-g4 f7-f5 14.e5xf6 (136.040.761) 687
22/56 6:03 +0.27 1.e2-e4 e7-e5 2.Sg1-f3 Sg8-f6
3.d2-d4 Sf6xe4 4.Lf1-d3 d7-d5
5.Sf3xe5 Lf8-d6 6.Sb1-d2 Ld6xe5
7.d4xe5 Se4-c5 8.Sd2-f3 Lc8-g4
9.Lc1-e3 Lg4xf3 10.g2xf3 Sb8-d7
11.Ld3-f5 c7-c6 12.f3-f4 g7-g6 (250.517.081) 688
Bester Zug: e2-e4 Zeit: 7:00.610 min K/s: 688.831 CPU 99.8% K/s(1CPU): 690.211 Knoten: 289.707.522

and this is Shredder 12 x64 1 Thread:

rnbqkbnr/pppppppp/8/8/8/8/PPPPPPPP/RNBQKBNR w KQkq -

Engine: DS 12 x64 1T (1024 MB)
von Stefan Meyer-Kahlen
13/41 0:01 +0.37 1.e2-e4 e7-e5 2.Sg1-f3 Sb8-c6
3.Lf1-b5 a7-a6 4.Lb5xc6 d7xc6
5.d2-d3 Sg8-f6 6.O-O Lf8-d6 7.d3-d4 Lc8-g4
8.d4xe5 Lg4xf3 9.Dd1xf3 Ld6xe5 (1.249.344) 660
14/31 0:02 +0.36 1.e2-e4 e7-e5 2.Sg1-f3 Sb8-c6
3.Lf1-b5 Lf8-c5 4.O-O Sg8-f6
5.Sb1-c3 Dd8-e7 6.d2-d3 a7-a6
7.Lb5-c4 b7-b5 8.Sc3-d5 De7-d8
9.b2-b4 Sc6xb4 (1.969.035) 663
15/34 0:04 +0.41 1.e2-e4 e7-e5 2.Sg1-f3 Sb8-c6
3.Lf1-b5 Lf8-c5 4.O-O Sg8-f6
5.Sb1-c3 Dd8-e7 6.d2-d3 a7-a6
7.Lb5xc6 d7xc6 8.Lc1-e3 Lc5xe3
9.f2xe3 O-O 10.h2-h3 h7-h6 11.Sf3-d2 Lc8-d7 (3.188.050) 666
16/41 0:11 +0.42 1.e2-e4 e7-e5 2.Sg1-f3 Sb8-c6
3.Lf1-b5 a7-a6 4.Lb5xc6 d7xc6 5.O-O Dd8-d6
6.Sb1-c3 Lc8-g4 7.h2-h3 Lg4xf3
8.Dd1xf3 Sg8-f6 9.d2-d3 Lf8-e7 For
10.Lc1-e3 O-O 11.a2-a3 (8.017.928) 668
17/34 0:16 +0.40 1.e2-e4 e7-e5 2.Sg1-f3 Sb8-c6
3.Lf1-b5 a7-a6 4.Lb5xc6 d7xc6 5.O-O Dd8-d6
6.Sb1-c3 Lc8-g4 7.h2-h3 Lg4xf3
8.Dd1xf3 Sg8-f6 9.d2-d3 Lf8-e7
10.Lc1-e3 O-O 11.a2-a3 Dd6-e6
12.b2-b4 b7-b5 (11.259.774) 672
18/41 0:38 +0.29 1.e2-e4 e7-e5 2.Sg1-f3 Sg8-f6
3.Sf3xe5 d7-d6 4.Se5-f3 Sf6xe4
5.Dd1-e2 Dd8-e7 6.d2-d3 Se4-f6
7.Sb1-c3 Sb8-c6 8.Lc1-g5 De7xe2+
9.Lf1xe2 Lc8-e6 10.O-O-O O-O-O
11.d3-d4 Td8-e8 12.Le2-b5 Lf8-e7
13.Th1-e1 (26.018.677) 675
19/49 1:07 +0.31 1.e2-e4 e7-e5 2.Sg1-f3 Sg8-f6
3.d2-d4 Sf6xe4 4.Lf1-d3 d7-d5
5.Sf3xe5 Lf8-d6 6.Sb1-d2 Ld6xe5
7.d4xe5 Se4-c5 8.Sd2-f3 Lc8-g4
9.Lc1-e3 Sc5xd3+ 10.Dd1xd3 c7-c6
11.O-O-O Sb8-d7 12.Dd3-d4 Lg4xf3
13.g2xf3 Dd8-e7 14.f3-f4 (46.087.995) 679
20/48 1:50 +0.33 1.e2-e4 e7-e5 2.Sg1-f3 Sg8-f6
3.d2-d4 Sf6xe4 4.Lf1-d3 d7-d5
5.Sf3xe5 Lf8-d6 6.Sb1-d2 Ld6xe5
7.d4xe5 Se4-c5 8.Sd2-f3 Lc8-g4 Mac
9.Lc1-e3 Sc5xd3+ 10.Dd1xd3 c7-c6
11.O-O-O Sb8-d7 12.Dd3-d4 Lg4xf3
13.g2xf3 Dd8-e7 14.f3-f4 (75.617.608) 683
21/53 3:18 +0.34 1.e2-e4 e7-e5 2.Sg1-f3 Sg8-f6
3.d2-d4 Sf6xe4 4.Lf1-d3 d7-d5
5.Sf3xe5 Lf8-d6 6.Sb1-d2 Ld6xe5
7.d4xe5 Se4-c5 8.Sd2-f3 Lc8-g4
9.Lc1-e3 Lg4xf3 10.g2xf3 Sb8-d7
11.Ld3-f5 c7-c6 12.O-O g7-g6
13.Lf5-g4 f7-f5 14.e5xf6 (136.040.761) 684
22/56 6:05 +0.27 1.e2-e4 e7-e5 2.Sg1-f3 Sg8-f6
3.d2-d4 Sf6xe4 4.Lf1-d3 d7-d5
5.Sf3xe5 Lf8-d6 6.Sb1-d2 Ld6xe5
7.d4xe5 Se4-c5 8.Sd2-f3 Lc8-g4
9.Lc1-e3 Lg4xf3 10.g2xf3 Sb8-d7
11.Ld3-f5 c7-c6 12.f3-f4 g7-g6 (250.517.081) 686
Bester Zug: e2-e4 Zeit: 7:00.406 min K/s: 686.484 CPU 99.8% K/s(1CPU): 687.859 Knoten: 288.580.807

Virtually NO difference.
BTW: There ARE differences when learning is on, but of course there is no serious tester who does this!
Deep Shredder For MacBye
Ingo
ok :) it looks like there is some systematic error in the cegt testing.
And Roland Rösler was right all along!
In this case it's clear that 1 analysis was worth more than 1000 games ;)
'Roland Rösler was right'
Statistical noise :)
Buuuuuhhh! :-)
ok, but what is the expected value for this result (>=50 Elo difference after 2300 (1200+1100) games), if there is no (systematic) error in testing and both engines have same playing strength? Is it <1% or >3%? :-)

File Shredder Mac

TopicThe Rybka Lounge / Computer Chess / Deep Shredder 12 info and some results

Deep Shredder For Mac Computers

Powered by mwForum 2.27.4 © 1999-2012 Markus Wichitill