This is a complete rewrite (moving from Pascal to C/C++) utilizing bitboards. It took me quite a time, but I am very satisfied. NPS almost doubled(!). Search algorithm is almost the same with only a few bugfixes. Static evaluation is a bit different due to bitboards, but all major terms are pretty close maybe except king safety.
32 & 64bit binaries + opening book:
http://www.mediafire.com/?wzmmxzg5djj
Richard
Critter 0.52 available
Moderators: hgm, Rebel, chrisw
-
- Posts: 481
- Joined: Thu Apr 16, 2009 12:00 pm
- Location: Slovakia, EU
-
- Posts: 9773
- Joined: Wed Mar 08, 2006 8:44 pm
- Location: Amman,Jordan
Re: Critter 0.52 available
Thanks,much appreciated Richard
One of the most promising rising stars in the computer chess world
Cheers,
Dr.D
P.S.Will be tested with pleasure
One of the most promising rising stars in the computer chess world
Cheers,
Dr.D
P.S.Will be tested with pleasure
_No one can hit as hard as life.But it ain’t about how hard you can hit.It’s about how hard you can get hit and keep moving forward.How much you can take and keep moving forward….
-
- Posts: 200
- Joined: Tue Apr 01, 2008 11:53 pm
Re: Critter 0.52 available
Thanks richard - looking forward to test the new version.rvida wrote:This is a complete rewrite (moving from Pascal to C/C++) utilizing bitboards. It took me quite a time, but I am very satisfied. NPS almost doubled(!). Search algorithm is almost the same with only a few bugfixes. Static evaluation is a bit different due to bitboards, but all major terms are pretty close maybe except king safety.
32 & 64bit binaries + opening book:
http://www.mediafire.com/?wzmmxzg5djj
Richard
I cant believe you achieved such a rating with a pascal engine - now its in c++ i cant wait to see what critter delivers
Regards Mark
Quad Q6600 3.0 Ghz, Vista 64bit, 4gb Ram
Quad Q6600 3.0 Ghz, Vista 64bit, 4gb Ram
-
- Posts: 6640
- Joined: Thu Mar 09, 2006 4:21 am
Re: Critter 0.52 available
Hopefully, I think we should expect the 100 more elo improvement. C++ I heard is obviously orders of magnitude efficient.meschle wrote:Thanks richard - looking forward to test the new version.rvida wrote:This is a complete rewrite (moving from Pascal to C/C++) utilizing bitboards. It took me quite a time, but I am very satisfied. NPS almost doubled(!). Search algorithm is almost the same with only a few bugfixes. Static evaluation is a bit different due to bitboards, but all major terms are pretty close maybe except king safety.
32 & 64bit binaries + opening book:
http://www.mediafire.com/?wzmmxzg5djj
Richard
I cant believe you achieved such a rating with a pascal engine - now its in c++ i cant wait to see what critter delivers
Thanks, Richard, Testing it now.
-
- Posts: 1222
- Joined: Wed Mar 08, 2006 8:28 pm
- Location: Florida, USA
Re: Critter 0.52 available
I'm experience in both Delphi / Pascal and C - the difference is *maybe* 30% at most in favor of C / C++. Certainly not an order of magnitude different.swami wrote:Hopefully, I think we should expect the 100 more elo improvement. C++ I heard is obviously orders of magnitude efficient.
Cheers,
Steve
-
- Posts: 41520
- Joined: Sun Feb 26, 2006 10:52 am
- Location: Auckland, NZ
Re: Critter 0.52 available
Thanks Richard. The Critter is slowly becoming a monster.
gbanksnz at gmail.com
-
- Posts: 6640
- Joined: Thu Mar 09, 2006 4:21 am
Re: Critter 0.52 available
Code: Select all
----------------------------------------------------------------------------------------------------------------------------------
| STS1 | STS2 | STS3 | STS4 | STS5 | STS6 | STS7 | STS8 |
----------------------------------------------------------------------------------------------------------------------------------
Critter_0.52_32bit | 80 | 73 | 73 | 75 | 77 | 75 | 61 | 63 |
----------------------------------------------------------------------------------------------------------------------------------
Grade | A+ | A- | A- | A | A | A | B | B |
Score / 1000 | 864 | 759 | 809 | 834 | 821 | 872 | 693 | 724 |
Grade | S | A | A+ | A+ | A+ | S | B+ | A- |
Best Move : 577 / 800 72.12 % Grade : A-
Total : 6376 / 8000 79.7 % Grade : A
----------------------------------------------------------------------------------------------------------------------------------
-
- Posts: 2873
- Joined: Wed Mar 08, 2006 10:09 pm
- Location: Germany
- Full name: Werner Schüle
Some problems with Critter 0.52
Hi,
I had some problems with Critter running inside Shredder 12 GUI. First I saw very bad results. Then I had a look at the games (played on different pcs). All look the same:
Critter always displays a 0.00 eval. I had this problem not in Arena 2.01. Any ideas what happens??
Games look like these:
Werner
[Event "40 Züge in 10 min"]
[Site "Engine Match"]
[Date "2010.01.30"]
[Round "4"]
[White "Toga II 1.2.1a"]
[Black "Critter 0.52 x64"]
[Result "1-0"]
[PlyCount "95"]
[EventDate "2010.??.??"]
1. d4 {Buch 0s} d5 {Buch 0s} 2. Nf3 {Buch 0s} Nf6 {Buch 0s} 3. c4 {Buch 0s} c6
{Buch 0s} 4. Nc3 {Buch 0s} e6 {Buch 0s} 5. Bg5 {Buch 0s} h6 {Buch 0s} 6. Bh4 {B
uch 0s} dxc4 {Buch 0s} 7. e4 {Buch 0s} g5 {Buch 0s} 8. Bg3 {Buch 0s} b5 9. Ne5
{+0.22/14 19s} h5 {0.00/17 37s (Lg7)} 10. h4 {+0.02/14 17s} g4 {0.00/17 1:25m}
11. a3 {-0.07/14 17s (Lf4)} Bh6 {0.00/17 31s (Lb7)} 12. Be2 {+0.17/14 16s} Qb6
{0.00/18 30s (Lb7)} 13. a4 {+0.19/14 14s} O-O {0.00/18 29s} 14. a5 {+0.04/13
11s (axb5)} Qb7 {0.00/18 28s} 15. f3 {+0.20/14 18s (Dc2)} Qe7 {0.00/16 32s
(Td8)} 16. O-O {+0.48/14 16s (Dc2)} Be3+ {0.00/15 26s (b4)} 17. Kh1 {+0.89/14
18s} Nfd7 {0.00/17 24s} 18. fxg4 {+1.22/14 18s (Sxd7)} Nxe5 {0.00/17 33s} 19.
dxe5 {+1.20/15 14s} Qd8 {0.00/17 22s (Td8)} 20. Qe1 {+1.43/14 11s (gxh5)} hxg4
{0.00/18 20s} 21. Bh2 {+1.42/15 16s (Lf4)} Nd7 {0.00/16 27s (Lh6)} 22. Qg3 {
+1.26/1513s} Bh6 {0.00/18 27s} 23. Qxg4+ {+1.60/16 19s} Bg7 {0.00/19 15s} 24.
Rad1 {+1.81/14 19s (Tf3)} b4 {0.00/17 14s (De8)} 25. Bf4 {+2.45/13 19s (h5)}
Qc7 {0.00/15 19s (Kh7)} 26. Rxd7 {+5.17/14 19s (Td6)} Qxd7 {0.00/17 14s} 27.
Bh6 {+5.22/16 20s} f5 {0.00/17 13s} 28. exf6 {+6.04/13 11s} Rf7 {0.00/17 15s}
29. Na4 {+6.10/14 20s} c3 {0.00/16 11s (Dc7)} 30. bxc3 {+6.80/14 20s} bxc3 {
0.00/15 8s (Dc7)} 31. Nxc3 {+7.66/12 20s} Rb8 {0.00/15 6s (Dc7)} 32. Rd1 {
+8.62/1321s} Qc7 {0.00/15 5s} 33. Bxg7 {+14.24/12 18s} Ba6 {0.00/14 3s (Ld7)}
34. Bxa6 {+16.29/11 20s} Rxg7 {0.00/0 1s} 35. fxg7 {+18.68/12 22s} Qxg7 {
0.00/0 1s (Df7)} 36. Qxe6+ {+22.41/13 22s} Qf7 {0.00/0 1s} 37. Qxc6 {+23.50/12
22s} Qh7 {0.00/0 1s (Tf8)} 38. Qe6+ {+21.16/12 23s (Lc4+)} Qf7 {0.00/0 1s} 39.
Qg4+ {+M11/12 23s} Kh8 {0.00/15 1s (Kh7)} 40. Rd6 {+M9/11 22s (Td5)} Kh7 {0.00/
14 1s} 41. e5 {+M8/11 19s (Dg5)} Qg7 {0.00/20 30s} 42. Qh5+ {+M7/12 18s} Kg8 {
0.00/2329s} 43. Bc4+ {+M6/14 5s} Kf8 {0.00/26 28s} 44. Rf6+ {+M5/33 3s} Ke7 {
0.00/5625s} 45. Nd5+ {+M4/63 0s} Kd8 {0.00/0 0s (Kd7)} 46. Rd6+ {+M3/63 0s} Kc8
{0.00/0 0s} 47. Qe8+ {+M2/63 0s} Kb7 {0.00/0 0s} 48. Qc6# {+M1/63 0s} 1-0
edited: just tried to repeat: inside engine window I see the correct evaluation but in moves window it looks like 0.00 /17 35 sec
I had some problems with Critter running inside Shredder 12 GUI. First I saw very bad results. Then I had a look at the games (played on different pcs). All look the same:
Critter always displays a 0.00 eval. I had this problem not in Arena 2.01. Any ideas what happens??
Games look like these:
Werner
[Event "40 Züge in 10 min"]
[Site "Engine Match"]
[Date "2010.01.30"]
[Round "4"]
[White "Toga II 1.2.1a"]
[Black "Critter 0.52 x64"]
[Result "1-0"]
[PlyCount "95"]
[EventDate "2010.??.??"]
1. d4 {Buch 0s} d5 {Buch 0s} 2. Nf3 {Buch 0s} Nf6 {Buch 0s} 3. c4 {Buch 0s} c6
{Buch 0s} 4. Nc3 {Buch 0s} e6 {Buch 0s} 5. Bg5 {Buch 0s} h6 {Buch 0s} 6. Bh4 {B
uch 0s} dxc4 {Buch 0s} 7. e4 {Buch 0s} g5 {Buch 0s} 8. Bg3 {Buch 0s} b5 9. Ne5
{+0.22/14 19s} h5 {0.00/17 37s (Lg7)} 10. h4 {+0.02/14 17s} g4 {0.00/17 1:25m}
11. a3 {-0.07/14 17s (Lf4)} Bh6 {0.00/17 31s (Lb7)} 12. Be2 {+0.17/14 16s} Qb6
{0.00/18 30s (Lb7)} 13. a4 {+0.19/14 14s} O-O {0.00/18 29s} 14. a5 {+0.04/13
11s (axb5)} Qb7 {0.00/18 28s} 15. f3 {+0.20/14 18s (Dc2)} Qe7 {0.00/16 32s
(Td8)} 16. O-O {+0.48/14 16s (Dc2)} Be3+ {0.00/15 26s (b4)} 17. Kh1 {+0.89/14
18s} Nfd7 {0.00/17 24s} 18. fxg4 {+1.22/14 18s (Sxd7)} Nxe5 {0.00/17 33s} 19.
dxe5 {+1.20/15 14s} Qd8 {0.00/17 22s (Td8)} 20. Qe1 {+1.43/14 11s (gxh5)} hxg4
{0.00/18 20s} 21. Bh2 {+1.42/15 16s (Lf4)} Nd7 {0.00/16 27s (Lh6)} 22. Qg3 {
+1.26/1513s} Bh6 {0.00/18 27s} 23. Qxg4+ {+1.60/16 19s} Bg7 {0.00/19 15s} 24.
Rad1 {+1.81/14 19s (Tf3)} b4 {0.00/17 14s (De8)} 25. Bf4 {+2.45/13 19s (h5)}
Qc7 {0.00/15 19s (Kh7)} 26. Rxd7 {+5.17/14 19s (Td6)} Qxd7 {0.00/17 14s} 27.
Bh6 {+5.22/16 20s} f5 {0.00/17 13s} 28. exf6 {+6.04/13 11s} Rf7 {0.00/17 15s}
29. Na4 {+6.10/14 20s} c3 {0.00/16 11s (Dc7)} 30. bxc3 {+6.80/14 20s} bxc3 {
0.00/15 8s (Dc7)} 31. Nxc3 {+7.66/12 20s} Rb8 {0.00/15 6s (Dc7)} 32. Rd1 {
+8.62/1321s} Qc7 {0.00/15 5s} 33. Bxg7 {+14.24/12 18s} Ba6 {0.00/14 3s (Ld7)}
34. Bxa6 {+16.29/11 20s} Rxg7 {0.00/0 1s} 35. fxg7 {+18.68/12 22s} Qxg7 {
0.00/0 1s (Df7)} 36. Qxe6+ {+22.41/13 22s} Qf7 {0.00/0 1s} 37. Qxc6 {+23.50/12
22s} Qh7 {0.00/0 1s (Tf8)} 38. Qe6+ {+21.16/12 23s (Lc4+)} Qf7 {0.00/0 1s} 39.
Qg4+ {+M11/12 23s} Kh8 {0.00/15 1s (Kh7)} 40. Rd6 {+M9/11 22s (Td5)} Kh7 {0.00/
14 1s} 41. e5 {+M8/11 19s (Dg5)} Qg7 {0.00/20 30s} 42. Qh5+ {+M7/12 18s} Kg8 {
0.00/2329s} 43. Bc4+ {+M6/14 5s} Kf8 {0.00/26 28s} 44. Rf6+ {+M5/33 3s} Ke7 {
0.00/5625s} 45. Nd5+ {+M4/63 0s} Kd8 {0.00/0 0s (Kd7)} 46. Rd6+ {+M3/63 0s} Kc8
{0.00/0 0s} 47. Qe8+ {+M2/63 0s} Kb7 {0.00/0 0s} 48. Qc6# {+M1/63 0s} 1-0
edited: just tried to repeat: inside engine window I see the correct evaluation but in moves window it looks like 0.00 /17 35 sec
-
- Posts: 4186
- Joined: Wed Mar 08, 2006 9:31 pm
- Location: Sulu Sea
Re: Critter 0.52 available
Thanks for 0.52
was using 0.42 vs Crafty,TLogic Fritz5.32 , Albin Opening Theme tournament.
Will change the 0.42 with this new 0.52.
was using 0.42 vs Crafty,TLogic Fritz5.32 , Albin Opening Theme tournament.
Will change the 0.42 with this new 0.52.
rvida wrote:This is a complete rewrite (moving from Pascal to C/C++) utilizing bitboards. It took me quite a time, but I am very satisfied. NPS almost doubled(!). Search algorithm is almost the same with only a few bugfixes. Static evaluation is a bit different due to bitboards, but all major terms are pretty close maybe except king safety.
32 & 64bit binaries + opening book:
http://www.mediafire.com/?wzmmxzg5djj
Richard
.
.
................. Mu Shin ..........................
.
................. Mu Shin ..........................