I got the permission by Steen Surrballe (Gandalf) ...

[Damir]

It is pity he no longer works on Gandalf, cause I was really expecting Gandalf 8 by now, but will have to be happy with 2-cores version of gandalf 7. Can not he at least add unlimited number of Threads for Gandalf 7 to use ? 2-cores is like a single core, does not make that much of a difference.

[AdminX]

Here is a long time control game Gandalf 7 using 1 core.

[pgn]
[Event "60 Moves in 120 min; Rest in 30 min"]
[Site "Engine Match"]
[Date "2017.04.27"]
[Round "3"]
[White "Deep Gandalf 7.0-64"]
[Black "Chess Tiger 2007"]
[Result "1-0"]
[ECO "D43"]
[Annotator "Komodo 10.4 CB14 Tactical Analysis"]
[PlyCount "165"]
[EventDate "2017.??.??"]

1. d4 {book 0s} d5 {book 0s} 2. c4 {book 0s} c6 {book 0s} 3. Nf3 {book 0s} Nf6
{book 0s} 4. Nc3 {book 0s} e6 {book 0s} 5. Bg5 {book 0s} Bb4 {book 0s} 6. Bxf6
{0.48/17 373} Qxf6 {0.08/23 139 (gxf6)} 7. e3 {0.51/16 113 D43: Semi-Slav: 5
Bg5 h6} O-O {0.04/22 55 (c5)} 8. a3 {0.28/17 539 (Qb3)} (8. Be2 c5 9. cxd5 exd5
10. O-O Rd8 11. Rc1 Nc6 12. a3 Bxc3 13. bxc3 Qe7 14. a4 c4 {1/2-1/2 (55)
Epishin,V (2675)-Mitkov,N (2470) France 1994}) 8... Be7 $146 {0.49/24 book 0s
(Bxc3+)} (8... Bxc3+ $11 {-0.14/27} 9. bxc3 b6) (8... Ba5 9. b4 Bc7 10. Bd3 Nd7
11. O-O b6 12. Qc2 g6 13. h3 Bb7 14. cxd5 cxd5 15. Nxd5 Bxd5 16. Qxc7 Bxf3 17.
gxf3 {0-1 (29) Kosov,I (1752)-Glazov,A (1978) St Petersburg 2009}) 9. Be2 {
0.32/16 90 White is slightly better.} b6 {-0.04/22 132 (dxc4)} 10. O-O {
0.38/16 86} Qh6 {0.02/22 334} 11. Ne5 {0.70/18 0 (Bd3)} g6 {0.10/20 158 (Bb7)}
12. cxd5 {0.73/14 122 (Qc2)} exd5 {0.15/21 120} 13. Qa4 {0.76/16 0 (Bg4)} Bd6 {
1.05/25 127} ({Better is} 13... Bb7 $14 {0.41/26}) 14. f4 $16 {0.67/17 90} b5 {
0.22/22 166} 15. Qc2 {0.67/16 0 (Qb3)} a6 {0.12/22 203 (Be6)} 16. e4 {0.96/15
128 (Rf3)} f6 {0.22/21 147} 17. Nd3 {1.02/16 0} (17. exd5 $6 fxe5 18. fxe5 Bf5
$11) 17... Be6 {1.65/27 132 (dxe4)} (17... dxe4 $16 {0.98/25} 18. Nxe4 Bc7) 18.
Rae1 {1.08/15 131 (exd5)} dxe4 {0.15/22 116 (Kh8)} 19. Nxe4 {1.28/18 98} Bc7 {
0.40/24 109} 20. Nec5 {1.28/16 19} Bf7 {0.42/22 111 (Bc4)} 21. Bg4 {1.69/17
110 (Rf3)} f5 {0.84/21 164} 22. Bd1 {1.69/17 0} (22. Re7 Bd6 23. Rb7 (23. Rxf7
Rxf7 24. Bd1 Re7 $11) 23... fxg4 $16) 22... a5 {2.29/24 133} (22... Bb6 $16 {
1.58/26} 23. Ne5 Qg7) 23. Ne5 $1 $18 {1.79/17 0} Qg7 {1.30/22 189 (Bd6)} 24.
Bf3 {1.35/24 122 (a4)} (24. Nxf7 $18 {2.28/25} Qxf7 25. Ne6) 24... Rc8 {
2.32/25 136 (Bxe5)} (24... Bxe5 $16 {1.35/24} 25. fxe5 (25. dxe5 Bc4 $16) 25...
Bc4) 25. Rd1 {1.24/24 143 (a4)} (25. a4 $18 {2.32/25}) 25... a4 {2.62/25 110
(Bd6)} (25... Bd6 $16 {1.24/24} 26. g4 Na6) 26. Qf2 {1.27/25 173 (Nxf7)} (26.
Nxf7 $18 {2.62/25 has better winning chances.} Qxf7 27. d5) 26... Bb6 {1.85/27
126 (Rd8)} (26... Kh8 $16 {1.27/25} 27. Nxc6 Nxc6 28. Bxc6 Ra5) 27. Rfe1 {
1.50/15 144 (Qh4)} Ra7 {0.92/20 205 (Qf6)} 28. Rc1 {0.70/25 142 (g4)} ({
White should play} 28. g4 $16 {1.58/24}) 28... Re7 {0.74/21 137 (Re8)} 29. Qh4
{1.34/15 317 (g3)} Rce8 {0.73/21 213 (Ba2)} 30. Kh1 {1.18/15 273} Bd8 {0.80/22
0 (Bc4)} 31. Qf2 {1.40/15 132 (Qh3)} Bb6 {0.86/21 121 (Ba5)} 32. Qd2 {1.40/15
132 (Red1)} Qf6 {0.74/21 166 (Qh6)} 33. Red1 {1.34/15 131 (Kg1)} Kg7 {0.82/21
130 (Bc7)} 34. Qb4 {1.34/14 141 (g3)} Bc7 {0.83/21 116} 35. Re1 {1.34/15 14
(g3)} Qd6 {0.74/20 115} 36. Qd2 {1.34/16 0 (Qc3)} Bb6 {0.75/21 125 (Bd5)} 37.
Qc3 {1.31/15 176 (Qf2)} Qf6 {0.76/20 128 (Bd5)} 38. Red1 {1.37/14 121} Rc8 {
1.87/24 58 (Qd6)} (38... Qd6 $14 {0.69/28}) 39. Qb4 {0.63/27 141 (g3)} (39. g4
$1 $18 {1.87/24}) 39... Ra7 $14 {0.82/21 124 [#] (Rcc7). Threatens to win with
...Ba5.} 40. Qe1 {1.24/16 266 (Qd2)} Re7 {0.82/21 155} 41. Qe3 {1.21/15 0 (g3)}
Rd8 {0.78/20 110} 42. Qd3 {1.12/16 31 (g3)} Bc7 {0.77/22 467 (Bd5)} 43. Qc3 {
1.21/14 148 (Qf1)} Bd6 {1.39/25 109 (Bb6)} (43... Bb6 $14 {0.51/26}) 44. Qe1 {
0.72/25 114 (Re1)} (44. g4 $142 {1.39/25} Kg8 45. g5) 44... h6 {0.69/20 144
(Bc4)} 45. Qf2 {1.47/14 134} Rc8 {1.00/21 0 (Kh7)} 46. Re1 {1.56/15 141 (Ncd3)}
Kh7 {1.06/21 114 (Bxc5)} 47. Rc3 {1.47/15 252 (Ncd3)} Bc7 {0.84/22 102 (Rce8).
intending ...Ba5.} 48. Rce3 {1.44/15 133 (Ncd3)} Ba5 {1.56/25 120} (48... Rce8
$142 {1.00/26}) 49. R1e2 $1 $16 {1.44/15 25 (Rc1)} Rd8 {0.66/22 103 (Bb6)} 50.
Rd3 {0.54/25 576} ({White should try} 50. Nxf7 $18 {1.67/27} Rxf7 51. b4 axb3
52. Nxb3) 50... Ba2 {0.81/23 29 (Bb6)} 51. Ree3 {1.37/15 93 (Rd1)} Bb6 {
0.66/20 98 (Bf7)} 52. Qe1 {1.12/16 353 (g3). Black must now prevent Ng4.} Bf7 {
0.66/20 112 (Rxd4)} 53. Qb4 {0.96/15 176 (g3)} Rde8 {0.62/21 121 (Ra7)} 54. Rd1
{1.44/15 56 (g3)} Ra7 {1.03/26 129 (Bc7)} (54... Rd8 $14 {0.44/25}) 55. Qd2 {
1.12/16 157 (Qc3)} Rd8 {0.76/20 103 [#] Strongly threatening ...Bxc5.} 56. Qe1
{1.02/14 2 (Qc3)} Bg8 {0.66/21 205} 57. Qg1 {1.02/14 56 (Qc3)} Re7 {0.67/20
115 (Bd5)} 58. Rc3 {1.05/15 86} Ba5 {0.78/20 27 (Bd5)} 59. Rc2 {1.15/16 86} Bb6
{0.82/20 6 (Bc7)} 60. Qe1 {1.08/14 49 (Qf2)} g5 {0.60/22 151 (Bd5)} 61. g3 {
0.76/15 239} Bc7 {0.84/25 73 (Bc4)} (61... Qd6 $11 {0.12/27 remains equal.})
62. Re2 {1.34/14 55 (h3). aiming for Ng4.} Bd6 {0.67/20 55} 63. Ncd3 {1.15/13
53 (Rc1)} Rde8 {0.72/20 39} 64. Rc1 {1.12/12 13} g4 {0.90/21 153 (Bc4)} 65. Bg2
{1.40/16 51} h5 {1.08/24 213 (Bc4)} 66. Re3 {1.56/14 73} Bc7 {1.11/21 0 (Rd8)}
67. Nc5 {1.63/14 34 (Nb4)} Kg7 {1.50/22 35 (Bxe5)} (67... Bb6 $142 {0.87/23})
68. Rc2 {1.63/13 35 (Qf2)} h4 {2.12/24 32 (Kh6)} (68... Qd6 $16 {1.11/25 might
work better.}) 69. gxh4 $18 {1.79/14 103} Qd6 {1.07/21 0} 70. Qd2 {1.82/13 42}
Bf7 {1.24/20 21 (Qh6)} 71. Qd3 {1.95/14 36 (h3)} Qf6 {1.10/21 59} 72. Rce2 {
1.76/14 0} Bd6 {1.02/19 29} 73. Qc2 {1.56/12 14} (73. Nxf7 $6 Qxf7 74. Rxe7
Rxe7 $14) 73... Bg6 {2.20/24 51 (Bc4)} (73... Bb3 $16 {1.42/23 is a better
defense.} 74. Nxb3 axb3 75. Rxb3 Rf8) 74. Qc3 {2.01/14 38} Bh5 {1.21/19 0 (Bf7)
} (74... Qxh4 $2 75. d5 (75. Nxg6 Kxg6 76. Ne6 Qf6 $14) 75... Qf6 76. dxc6) 75.
Kg1 {1.69/13 119 (Qe1)} Bf7 $2 {3.70/24 18} (75... Kh7 $16 {1.35/23 was called
for.} 76. Ncd3 Rd8) 76. h3 {2.08/12 29} (76. Nxf7 Kxf7 77. Rxe7+ Rxe7 $14)
76... Bc4 {1.70/19 33} (76... Qxh4 77. hxg4 fxg4 78. d5 Bxc5 79. Qxc5 (79.
Nxf7+ Kxf7 80. Qxc5 Rxe3 81. Rxe3 Nd7 $14)) 77. Rf2 {2.08/12 52 (Nxc4)} ({
Less strong is} 77. Nxc4 bxc4 78. Rxe7+ (78. Nxa4 Bxf4 79. Rxe7+ Rxe7 80. Rxe7+
Qxe7 $14) 78... Rxe7 79. Rxe7+ Bxe7 80. hxg4 (80. Qxc4 Qxh4 81. hxg4 Qxg4 82.
Qxa4 Bd6 $16) (80. Nxa4 Nd7 $16) 80... fxg4 81. Qxc4 (81. Nxa4 Bd6 82. Nb6 Bxf4
83. Nxc4 Nd7 $18) 81... Qxh4 $18) (77. hxg4 Bxe2 78. Rxe2 (78. g5 Qg6 $16)
78... fxg4 $18) 77... Qxh4 {1.38/19 34} 78. hxg4 {2.17/11 0} (78. Nxc4 bxc4 79.
Rfe2 Rxe3 80. Rxe3 Rxe3 81. Qxe3 Bxc5 82. dxc5 Qf6 $16) 78... fxg4 {1.20/18 21
(Bxe5)} 79. Nxg4 $1 {2.40/11 29} ({Weaker is} 79. Nxc4 bxc4 80. Rxe7+ Rxe7 $18)
79... Kg8 $2 {8.35/26 50 (Rxe3)} (79... Rxe3 {3.76/24} 80. Nxe3 Kg8 81. Nxc4
bxc4 82. Qxc4+ Kh8) (79... Qxg4 80. Rg3) 80. Ne4 {4.41/13 22 (Rh3). White is
clearly winning.} Rxe4 {4.54/19 98} 81. Bxe4 {4.64/14 0} Rxe4 {5.10/19 63} 82.
Rxe4 {4.64/15 0} Bd5 {6.50/19 29} 83. Qc2 {4.99/14 0 (Qe3)} 1-0

[/pgn]

[Peter Berger]

Gandalf 7 does not work on my computer (Windows 10 64bit) with the ChessBase GUI - it terminates instantly with an exception.

Did anyone experience the same and was able to solve the problem?

Peter

[AdminX]

Gandalf 7 does not work on my computer (Windows 10 64bit) with the ChessBase GUI - it terminates instantly with an exception.

Did anyone experience the same and was able to solve the problem?

Peter

Yes, it works over here.

Windows 10 Pro 64bit
i7 4790 3.6Ghz
Chessbase 14
Fritz 15

Check to make sure OS is not blocking it.

[Robert Flesher]

Many thanks Frank!

[Rowen]

Hi there,

I got the permission from Steen Suurballe (mail from today by Steen itself) to release his older Gandalf versions. I have here a large collection by him. In the past I release Gandalf in cooperation with Gambit-Soft. Gandalf was a very important program in times commercials have no interest to support free protocols. Millennium 2000 have a bad own "MCS" protocol and Winboard aren't very populare for commercials. In combination with the new UCI protocol Stefan is working with Dieter Bürssner and Rudolf Huber (powered at first by Millennium, Gambit-Soft and myself over Frank's Chess Page ...) I try to power UCI with Capture, LambChop, Gandalf and much others. Send Freeware programmers the important Millennium 2000 paket for free (200 EUR software I pay by myself and so one). Same as I do for Winboard in the past before I gave my site to Leo Dijksman.

Again, with Gandalf I had a programmer on "our" site which gave UCI a great support and more and more other commercials follow in the years after. Gandalf itself is a very interesting program with a speculative attacking style. After the older information I have in my brain Steen is working with two eval search options. One of them very speculative and able to find attacking moves in the middlegame with full pieces on board. The reason that Gandalf is one of the strongest in the time version 4.32f-h released by Gambit-Soft ans Winboard / UCI engine.

Unfortunatley, to complicated for much others to find the right configuration. So I produced thousands of games on very strong hardware Gandalf was the number 2 in World here ... not in other rating systems, because the configuration seems to be a problem or the others have problems with it that the engine from Denmark is so strong. Means a bigger problem for commercials that a Winboard engine is to high and different people like to support the commercials only. In official tourneys people give not the right settings we have seen after the tourneys. Much of the tournament Directory sent an sorry after the tounreys ... Note: They have the right configuration before the tourneys started.

A long and bad topic, a lot of such topics again and again at this time with people try to make all to block winboard and UCI ... same with Arena GUI in the beginning. I will not thinking an all the mails by Computer chess Reporters, working for newspaper or other troublemakers. People like to used today Arena made some troubles and me a lot of work. Sorry, but this one is computer chess history too. Not all was nice in the past. And I have written over 200.000 winboard Support mails in these years.

Today not longer important ... UCI is standard and the topic is over, after a hard fight.

But Steen Suurballe is one of the persons we all have to say thanks because Steen gave his support from the first hour ... to Winboard and to UCI. The reason that the older development is very important and I am happy that Steen gave his permission for an release for all of his older versions.

---

I asked Steen for Gandalf 7. Only a hand full testers have it and I made a chess break at this time. My latest versions are Gandalf 6 versions. At the moment I am waiting for an answere. Today in the afternoon I will public the Gandalf versions on my site in download area. Maybe Gandalf 7 will be follow later.

Have fun with it and thanks Steen!
Steen ... one of commercial supporters for Winboard and UCI since the beginning.

Best
Frank

PS: I have no problems with it if others added Gandalf on his own site. Same for the older Zarkov versions I public on my site.

Hello Mr Quisinsky

Thank you for making these Gandalf versions available on your web site.
I notice that there are many versions of Gandalf 4.32 UCI on the download link , many development versions, I am not sure which one is the actual 4.32 UCI version, would it be possible to have a specific link to version 4.32uci , or could you indicate to me which is the actual 4.32 UCI version?

Once again, thank you for making there versions available.

[JohnW]

Hi there,

what I have from Gandalf is now available in my download area.

A lot of Winboard versions, dev. UCI versions and Version 6 of Gandalf, MSDOS v2.1 & v3.0 ... the readme files in different languages from the Gambit-Soft CD with configuration examples!

Note 1:
Gandalf 3 DOS and the first Winboard versions used older tablebases. The file to the older 4-pieces database I added too.

Note 2:
My personal favorite all the time was Gandalf 4.32d (most speculative). I am using the older versions today for games I am playing. This version is most speculative in my opinion. The latest 4.32g and 4.32h versions of Gandalf are stronger as 4.32f and my favorite 4.32d but most interesting playing style will have 4.32f. Newer Gandalf 5 versions are weaker in my opinion and lost a lot of the powerful style. Gandalf 6 was stronger as Gandalf 4.32f, more powerful in playing style as 5er versions but for self playing games ... please used 4.32f or 4.32d for a nice day on your chess board. Maybe with the last version of Winboard GUI?

Download around 112Mb.

Again ...
If Steen send me Gandalf 7 I will added it in a new *.zip file to my download area. So you haven't download again the bigger 112Mb file.

Best
Frank

Where is your download area??

[Guenther]

Hi there,

I got the permission from Steen Suurballe (mail from today by Steen itself) to release his older Gandalf versions. I have here a large collection by him. In the past I release Gandalf in cooperation with Gambit-Soft. Gandalf was a very important program in times commercials have no interest to support free protocols. Millennium 2000 have a bad own "MCS" protocol and Winboard aren't very populare for commercials. In combination with the new UCI protocol Stefan is working with Dieter Bürssner and Rudolf Huber (powered at first by Millennium, Gambit-Soft and myself over Frank's Chess Page ...) I try to power UCI with Capture, LambChop, Gandalf and much others. Send Freeware programmers the important Millennium 2000 paket for free (200 EUR software I pay by myself and so one). Same as I do for Winboard in the past before I gave my site to Leo Dijksman.

...

Have fun with it and thanks Steen!
Steen ... one of commercial supporters for Winboard and UCI since the beginning.

Best
Frank

PS: I have no problems with it if others added Gandalf on his own site. Same for the older Zarkov versions I public on my site.

Hello Mr Quisinsky

Thank you for making these Gandalf versions available on your web site.
I notice that there are many versions of Gandalf 4.32 UCI on the download link , many development versions, I am not sure which one is the actual 4.32 UCI version, would it be possible to have a specific link to version 4.32uci , or could you indicate to me which is the actual 4.32 UCI version?

Once again, thank you for making there versions available.

Actually a good question, but as you can see the original dates are preserved, so I would always use the latest one.
There are 12 versions of Gandalf 4.32UCI in the original download collection.

The date diff between uci_gandalf_01.rar and uci_gandalf_12.rar is just two months.
I checked the output between the oldest and newest one (01 vs. 12) and it seems the changes mostly concerned
better uci support itself. You can see that Steen Surballe added a few more uci options and changed default hash.
The PV and eval remain exactly the same until depth 10 from startposition. The difference in nodes could be explained
with different default hash - I was too lazy to try with same hash.

Hope this helps.

Version 12

Code: Select all

uci

id name Gandalf 4.32UCI
id author Steen Suurballe
option name Hash type spin default 4 min 4 max 104
option name MultiPV type spin default 1 min 1 max 20
option name NalimovPath type string default c:\
option name NalimovCache type spin default 16 min 8 max 32
option name Ponder type check default true
option name Use TB type check default true
option name Contempt factor type spin default 0 min -50 max 50
option name Clear Hash type button
option name Clear Learning type button
option name Style type combo default Normal var Solid var Normal var Wild
uciok
ucinewgame
isready
readyok
position startpos
go infinite
info depth 1 seldepth 0
info depth 2 seldepth 2
info depth 3 seldepth 5
info depth 4 seldepth 6
info depth 5 seldepth 13
info depth 6 seldepth 19
info depth 6 seldepth 25 score cp 12 time 100 nodes 33228 nps 166140 pv  e2e4 e7e5 g1f3 g8f6 d2d4 d7d5 f1b5 c7c6
info depth 7 seldepth 27 score cp 36 time 200 nodes 112988 nps 376626 pv  e2e4 d7d5 e4d5 d8d5 g1f3 b8c6 b1c3 d5d8
info depth 8 seldepth 32 score cp 6 time 600 nodes 355315 nps 507592 pv  e2e4 e7e5 b1c3 b8c6 g1f3 g8f6 d2d4 f8b4 f3e5 c6e5 d4e5 b4c3 b2c3 f6e4
info depth 9 seldepth 36 score cp 44 time 2100 nodes 1303529 nps 592513 pv  e2e4 d7d5 e4d5 g8f6 d2d4 c7c6 d5c6 b8c6 g1f3 c8f5
info depth 10 seldepth 40 score cp 20 time 5100 nodes 3295984 nps 633843 pv  e2e4 e7e5 b1c3 b8c6 g1f3 g8f6 f1b5 f8b4 b5c6 d7c6 e1g1

Version 01

Code: Select all

White(1): uci
id name Gandalf 4.32UCI
id author Steen Suurballe
option name Hash type spin default 56 min 4 max 200
option name NalimovPath type string default c:\
option name NalimovCache type spin default 16 min 8 max 32
option name Ponder type check default true
option name OwnBook type check default true
option name Ponder check default true
uciok
ucinewgame
isready
readyok
position startpos
go infinite
info depth 6 seldepth 25 score cp 12 time 100 nodes 33118 nps 165590 pv  e2e4 e7e5 g1f3 g8f6 d2d4 d7d5 f1b5 c7c6
info depth 7 seldepth 27 score cp 36 time 200 nodes 111106 nps 370353 pv  e2e4 d7d5 e4d5 d8d5 g1f3 b8c6 b1c3 d5d8
info depth 8 seldepth 32 score cp 6 time 600 nodes 332753 nps 475361 pv  e2e4 e7e5 b1c3 b8c6 g1f3 g8f6 d2d4 f8b4 f3e5 c6e5 d4e5 b4c3 b2c3 f6e4
info depth 9 seldepth 36 score cp 44 time 1900 nodes 1187582 nps 593791 pv  e2e4 d7d5 e4d5 g8f6 d2d4 c7c6 d5c6 b8c6 g1f3 c8f5
info depth 10 seldepth 38 score cp 20 time 4300 nodes 2846184 nps 646860 pv  e2e4 e7e5 b1c3 b8c6 g1f3 g8f6 f1b5 f8b4 b5c6 d7c6 e1g1
info hashfull 658 nps 638888

[carldaman]

https://www.amateurschach.de/

The 'downloads' link, which will take you to the downloads page, should be on the green-colored part of the home page (on the left side).

[Rowen]

Hi there,

I got the permission from Steen Suurballe (mail from today by Steen itself) to release his older Gandalf versions. I have here a large collection by him. In the past I release Gandalf in cooperation with Gambit-Soft. Gandalf was a very important program in times commercials have no interest to support free protocols. Millennium 2000 have a bad own "MCS" protocol and Winboard aren't very populare for commercials. In combination with the new UCI protocol Stefan is working with Dieter Bürssner and Rudolf Huber (powered at first by Millennium, Gambit-Soft and myself over Frank's Chess Page ...) I try to power UCI with Capture, LambChop, Gandalf and much others. Send Freeware programmers the important Millennium 2000 paket for free (200 EUR software I pay by myself and so one). Same as I do for Winboard in the past before I gave my site to Leo Dijksman.

...

Have fun with it and thanks Steen!
Steen ... one of commercial supporters for Winboard and UCI since the beginning.

Best
Frank

PS: I have no problems with it if others added Gandalf on his own site. Same for the older Zarkov versions I public on my site.

Hello Mr Quisinsky

Thank you for making these Gandalf versions available on your web site.
I notice that there are many versions of Gandalf 4.32 UCI on the download link , many development versions, I am not sure which one is the actual 4.32 UCI version, would it be possible to have a specific link to version 4.32uci , or could you indicate to me which is the actual 4.32 UCI version?

Once again, thank you for making there versions available.

Actually a good question, but as you can see the original dates are preserved, so I would always use the latest one.
There are 12 versions of Gandalf 4.32UCI in the original download collection.

The date diff between uci_gandalf_01.rar and uci_gandalf_12.rar is just two months.
I checked the output between the oldest and newest one (01 vs. 12) and it seems the changes mostly concerned
better uci support itself. You can see that Steen Surballe added a few more uci options and changed default hash.
The PV and eval remain exactly the same until depth 10 from startposition. The difference in nodes could be explained
with different default hash - I was too lazy to try with same hash.

Hope this helps.

Version 12

Code: Select all

uci

id name Gandalf 4.32UCI
id author Steen Suurballe
option name Hash type spin default 4 min 4 max 104
option name MultiPV type spin default 1 min 1 max 20
option name NalimovPath type string default c:\
option name NalimovCache type spin default 16 min 8 max 32
option name Ponder type check default true
option name Use TB type check default true
option name Contempt factor type spin default 0 min -50 max 50
option name Clear Hash type button
option name Clear Learning type button
option name Style type combo default Normal var Solid var Normal var Wild
uciok
ucinewgame
isready
readyok
position startpos
go infinite
info depth 1 seldepth 0
info depth 2 seldepth 2
info depth 3 seldepth 5
info depth 4 seldepth 6
info depth 5 seldepth 13
info depth 6 seldepth 19
info depth 6 seldepth 25 score cp 12 time 100 nodes 33228 nps 166140 pv  e2e4 e7e5 g1f3 g8f6 d2d4 d7d5 f1b5 c7c6
info depth 7 seldepth 27 score cp 36 time 200 nodes 112988 nps 376626 pv  e2e4 d7d5 e4d5 d8d5 g1f3 b8c6 b1c3 d5d8
info depth 8 seldepth 32 score cp 6 time 600 nodes 355315 nps 507592 pv  e2e4 e7e5 b1c3 b8c6 g1f3 g8f6 d2d4 f8b4 f3e5 c6e5 d4e5 b4c3 b2c3 f6e4
info depth 9 seldepth 36 score cp 44 time 2100 nodes 1303529 nps 592513 pv  e2e4 d7d5 e4d5 g8f6 d2d4 c7c6 d5c6 b8c6 g1f3 c8f5
info depth 10 seldepth 40 score cp 20 time 5100 nodes 3295984 nps 633843 pv  e2e4 e7e5 b1c3 b8c6 g1f3 g8f6 f1b5 f8b4 b5c6 d7c6 e1g1

Version 01

Code: Select all

White(1): uci
id name Gandalf 4.32UCI
id author Steen Suurballe
option name Hash type spin default 56 min 4 max 200
option name NalimovPath type string default c:\
option name NalimovCache type spin default 16 min 8 max 32
option name Ponder type check default true
option name OwnBook type check default true
option name Ponder check default true
uciok
ucinewgame
isready
readyok
position startpos
go infinite
info depth 6 seldepth 25 score cp 12 time 100 nodes 33118 nps 165590 pv  e2e4 e7e5 g1f3 g8f6 d2d4 d7d5 f1b5 c7c6
info depth 7 seldepth 27 score cp 36 time 200 nodes 111106 nps 370353 pv  e2e4 d7d5 e4d5 d8d5 g1f3 b8c6 b1c3 d5d8
info depth 8 seldepth 32 score cp 6 time 600 nodes 332753 nps 475361 pv  e2e4 e7e5 b1c3 b8c6 g1f3 g8f6 d2d4 f8b4 f3e5 c6e5 d4e5 b4c3 b2c3 f6e4
info depth 9 seldepth 36 score cp 44 time 1900 nodes 1187582 nps 593791 pv  e2e4 d7d5 e4d5 g8f6 d2d4 c7c6 d5c6 b8c6 g1f3 c8f5
info depth 10 seldepth 38 score cp 20 time 4300 nodes 2846184 nps 646860 pv  e2e4 e7e5 b1c3 b8c6 g1f3 g8f6 f1b5 f8b4 b5c6 d7c6 e1g1
info hashfull 658 nps 638888

Hello
Thanks for the reply, yes I saw that version 12 was the latest and had more uci options, although the oldest, version 1. was the only exe that was labelled 4.32 the others were simply labelled U. But yes it would seem that 12 is the latest and probably the version to use, Thanks for your time.