The Real Test for Komodo 4

[A Distel]

Bxh6!!!, Sting SF 1.0 Finds it![d]r5k1/1qrnb1p1/3p2Pp/3Pp3/pp2N1R1/3QBP2/PPP5/1K1R4 w - - 0 1

Analysis by Sting SF JA 111210 64bit:

3.Bxh6 b3 4.cxb3 axb3 5.a3 Rc2 6.Bg5 Bxg5 7.Nxg5 Qa7 8.Qxb3 Rh2 9.Ne6 Nc5 10.Nxc5 Qxc5 11.Rc1 Qa7 12.Rc6 Qe7 13.Rgc4 Qd7
+- (1.73) Depth: 21/40 00:06:46 2167mN
(28.12.2011)

Thanks Alain! It means that this position too :

[d]r5k1/1qrnbppp/3p4/3Pp1PP/pp2N1R1/3QBP2/PPP5/1K1R4 w - - 0 1

I believe that my idea works though . I have to slow computer to prove it.

[Event "Blitz 5m"]

[Date "2011.12.28"]
[Round "1"]
[White "Sting SF JA 111210 64bit]
[Black "Komodo64 SSE Version 4 - 64bit]

[ECO "C68"]



1. e4 {0.28/18 5} e5 {0.19/17 4} 2. Nf3 {0.36/18 5 (Nc3)} Nc6 {0.14/17 5} 3.
Bb5 {0.36/18 7 (Bc4)} a6 {0.08/17 6} 4. Bxc6 {0.36/19 7} dxc6 {0.06/19 6} 5.
O-O {0.48/20 11} Qd6 {-0.01/19 19 (Bg4)} 6. d3 {0.52/18 6 (b3)} f6 {-0.04/18 3
(Nf6)} 7. Be3 {0.52/19 8} Be6 {-0.12/17 2} 8. Nbd2 {0.40/18 4} O-O-O {0.01/16 2
} 9. Nb3 {0.32/19 5} Qd7 {-0.06/17 6 (Ne7)} 10. Qe2 {0.32/18 5 (Nfd2)} Qf7 {-0.
07/16 7} 11. Nfd2 {0.40/18 7} g5 {-0.01/18 6 (Ne7)} 12. Nc4 {0.52/18 5 (d4)}
Ne7 {-0.21/18 7 (Bxc4)} 13. Nca5 {0.16/17 8 (Bc5)} Ng6 {-0.23/18 6} 14. f3 {0.
00/17 5} Bb4 {-0.22/17 22 (Nf4)} 15. Rfd1 {0.04/18 17 (a3)} h5 {-0.18/15 7
(Nf4)} 16. a3 {0.40/17 5} Bd6 {-0.08/16 14 (Nf4)} 17. Qf2 {0.40/16 5 (Nxb7)}
Bxb3 {-0.29/16 8 (Rde8)} 18. cxb3 {0.40/18 9 (Nxb3)} Kb8 {-0.43/16 7} 19. b4 {
0.40/18 7 (Rac1)} Nf4 {-0.41/15 1} 20. Rac1 {0.32/17 7} g4 {-0.43/13 1 (Ne6)}
21. Kh1 {0.52/14 3 (Ba7+)} f5 {-0.50/14 10 (Ne6)} 22. b3 {0.36/15 13 (d4)} h4 {
-1.02/14 1 (gxf3)} 23. fxg4 {-0.12/14 3} fxg4 {-0.91/14 1} 24. Qc2 {-0.16/16 3
(Qd2)} h3 {-1.03/16 1} 25. g3 {0.00/19 4} Ng2 {-1.03/17 2} 26. Qf2 {0.00/20 3}
Qg7 {-0.93/18 4 (Qxf2)} 27. Bd2 {0.00/18 3 (Ba7+)} Rhf8 {-0.97/18 3} 28. Qe2 {
0.08/20 5} Rf3 {-0.86/16 1} 29. Rf1 {0.08/19 5} Rdf8 {-0.85/18 4 (Qf6)} 30. Nc4
{0.24/19 4} b6 {-0.73/16 6 (c5)} 31. Rxf3 {0.36/19 4 (Bc3)} gxf3 {-0.71/15 1}
32. Qf2 {0.12/19 8} Kb7 {-0.52/16 4} 33. Bc3 {0.36/19 9 (Rf1)} Qg5 {-0.76/15 1}
34. Rf1 {0.48/18 3} c5 {-0.58/14 1} 35. Kg1 {0.56/19 10 (a4)} Qe7 {-0.39/13 3
(Rf7)} 36. Qd2 {0.56/18 3} Ka7 {-0.26/15 10 (Qf6)} 37. Rf2 {0.92/17 5 (Qb2)}
Rf7 {-0.07/15 4 (Qf6)} 38. Qb2 {1.33/16 4 (Qc1)} Qe6 {0.05/16 5 (cxb4)} 39. Qa1
{1.09/16 3 (Qd2)} cxb4 {-0.13/16 1} 40. axb4 {1.01/17 2} Kb7 {-0.13/14 0} 41.
Bxe5 {1.17/18 2 (Qd1)} Bxb4 {-0.17/16 2} 42. Bc3 {1.09/17 3} Bc5 {-0.06/18 3
(Bd6)} 43. d4 {1.21/18 2} Bd6 {-0.01/17 1} 44. e5 {0.80/18 2} Be7 {-0.01/16 0}
45. Qb1 {1.17/17 2} Qd5 {0.05/15 1 (Qf5)} 46. Qg6 {0.52/15 2 (Qd3)} a5 {0.05/
16 3} 47. Qd3 {0.48/16 2} Bg5 {0.05/17 4 (Rf8)} 48. Bd2 {0.48/16 1} Be7 {0.05/
19 3} 49. Kf1 {0.48/17 4 (Bc3)} Kb8 {0.05/17 3} 50. Nb2 {0.02/17 5 (Kg1)} Rf8 {
0.05/15 2} 51. Nc4 {0.00/16 1 (Qc4)} Rd8 {-0.34/16 1 (Rf7)} 52. Rxf3 {0.20/16
4 (Bc3)} a4 {-0.32/14 2} 53. Ke2 {0.24/16 1 (Bc3)} axb3 {-0.21/13 2} 54. Bc3 {
0.00/16 1} Bg5 {-0.21/15 2 (Bf8)} 55. Bb2 {0.00/16 2} Re8 {-0.15/14 1 (Bh6)}
56. Nd2 {0.72/14 1 (Rf5)} Bxd2 {0.05/14 2 (Qa5)} 57. Kxd2 {0.84/16 1} Rd8 {0.
05/16 1} 58. Qe2 {1.31/17 3 (Ke2)} Ka7 {-0.09/13 2 (Kb7)} 59. Rf6 {1.25/14 1
(Kc1)} Qc5 {0.05/15 2 (Kb8)} 60. Qe4 {1.13/15 1} Kb8 {0.05/15 2} 61. Rf2 {1.21/
16 1 (Rf3)} Qa5+ {-0.20/14 1 (Qb4+)} 62. Kd1 {0.92/17 2 (Kc1)} Qb4 {0.01/15 2}
63. Rf3 {1.09/17 2} Rxd4+ {0.05/15 2} 64. Bxd4 {1.21/11 0} b2 {0.05/15 0} 65.
Rf8+ {1.01/16 1} Qxf8 {0.05/16 0} 66. Bxb2 {0.80/16 1} Qf1+ {0.05/16 3} 67. Kd2
{0.60/16 1} Qf2+ {0.05/17 1} 68. Qe2 {0.56/15 0} Qc5 {0.02/18 2 (Qf5)} 69. Bc3
{0.72/13 1 (Qe4)} Qd5+ {0.01/15 1} 70. Qd3 {0.72/16 0} Qa2+ {0.05/16 1 (Qe6)}
71. Kc1 {0.76/14 0} c5 {0.05/15 1 (Qb3)} 72. Qd8+ {1.01/11 0 (Bb2)} Kb7 {0.06/
15 1 (Ka7)} 73. Qd7+ {1.21/12 0} Ka6 {0.05/16 1} 74. Qd3+ {1.33/14 1} Ka7 {0.
05/16 1} 75. Qe4 {1.33/13 0} Qa3+ {0.05/15 1 (Qb3)} 76. Kc2 {1.25/13 1} Qa2+ {
0.05/17 1} 77. Bb2 {1.41/13 0} c4 {0.05/17 1} 78. Kc1 {0.88/15 1} Qb3 {0.05/16
0} 79. Bd4 {0.88/14 1} Qb4 {0.05/16 1} 80. e6 {0.80/13 0} Ne1 {0.05/16 1} 81.
Qe3 {0.84/14 0 (e7)} Nd3+ {-0.11/10 0} 82. Kd1 {0.80/13 0 (Kc2)} Qb1+ {0.05/13
1} 83. Ke2 {0.56/14 0} Qh1 {0.05/14 1 (Nc1+)} 84. e7 {0.92/10 0 (Bxb6+)} Qg2+ {
0.05/13 1} 85. Kd1 {0.48/9 0} Qh1+ {0.05/14 1} 86. Kc2 {0.56/13 1 (Qg1)} Qxh2+
{0.05/13 1} 87. Qd2 {0.36/13 0} Nb4+ {0.05/14 1 (Qxd2+)} 88. Kc1 {0.36/13 0}
Na2+ {0.05/15 1 (Nd3+)} 89. Kd1 {0.04/12 0 (Kc2)} Qh1+ {-0.03/12 0} 90. Ke2 {
0.08/13 0} Nc1+ {0.05/12 1} 91. Ke3 {0.08/13 0} Qc6 {0.05/13 1} 92. Qa5+ {0.00/
13 0 (Qxc1)} Kb7 {0.05/12 0} 93. Qe5 {0.00/13 0 (Qh5)} Nd3 {0.05/12 1 (Qh6+)}
94. Qg7 {0.24/11 0 (e8Q)} Qe6+ {0.05/13 1 (Ka6)} 95. Be5 {0.00/12 0} Ka6 {0.05/
16 1} 96. Qf6 {0.00/13 0} Qxf6 {0.05/13 0} 97. Bxf6 {0.00/13 0} h2 {0.05/13 0}
98. e8=Q {0.00/14 0} h1=Q {0.05/14 0} 99. Qa4+ {0.00/15 0} Kb7 {0.05/15 0} 100.
Qxc4 {0.00/15 0 (Qd7+)} Qe1+ {-0.04/12 1} 101. Kxd3 {0.00/16 0} Qf1+ {-0.05/13
1 (Qxg3+)} 102. Kd4 {0.00/13 0} Qxf6+ {-0.02/12 0} 103. Ke3 {0.00/14 0} Qg5+ {
-0.03/12 0} 104. Qf4 {0.00/14 0 (Kf3)} Qc5+ {0.03/12 1} 105. Ke4 {0.00/12 0
(Qd4)} Qc6+ {-0.04/11 1 (Qc2+)} 106. Kd3 {0.00/14 0} Qd5+ {0.00/13 1} 107. Kc3
{0.00/15 0 (Ke3)} Qc5+ {0.00/11 1} 108. Kb3 {0.00/15 0 (Kd2)} Qb5+ {0.00/12 0}
109. Kc2 {0.00/14 0 (Kc3)} Qe2+ {0.00/13 1} 110. Kc3 {0.00/15 0} Ka7 {0.05/13
1 (Qe1+)} 111. g4 {0.76/8 0} Qe1+ {0.05/13 1} 112. Kd3 {0.76/10 0} Qb1+ {0.05/
14 1} 113. Ke2 {0.76/10 0} Qc2+ {0.05/14 0} 114. Kf3 {0.80/12 0} Qc6+ {0.05/14
1 (Qc3+)} 115. Qe4 {0.80/13 0} Qf6+ {0.05/14 0} 116. Ke3 {0.80/13 0 (Kg2)} Qc3+
{0.05/14 1} 117. Kf4 {0.80/11 0} Qc7+ {0.07/14 0 (Qf6+)} 118. Kf3 {0.72/13 0
(Qe5)} Qc3+ {0.05/14 0} 119. Qe3 {0.48/13 0 (Kg2)} Qc6+ {0.05/16 1} 120. Kf4 {
1.13/1 0 (Kg3)} Qc4+ {0.05/15 0} 121. Kf5 {0.76/4 0 (Kg3)} Qf7+ {0.05/14 0}
122. Ke5 {0.76/5 0 (Kg5)} Qc7+ {0.05/13 0 (Qe8+)} 123. Kf6 {1.17/7 0} Qd8+ {0.
05/14 0 (Qc6+)} 124. Kf5 {0.68/9 0 (Kg6)} Qf8+ {0.05/14 0} 125. Ke4 {0.72/8 0
(Kg5)} Qe8+ {0.05/14 0 (Qe7+)} 126. Kf4 {0.72/8 0} Qf7+ {0.05/15 0 (Qf8+)} 127.
Kg3 {0.56/9 0} Qc7+ {0.05/14 0} 128. Kf2 {0.60/2 0 (Kh3)} Qc2+ {0.05/15 0} 129.
Qe2 {0.56/7 0 (Kf3)} Qc5+ {0.05/15 0} 130. Kf3 {0.56/8 0 (Qe3)} Qc6+ {0.05/13
0 (Qf8+)} 131. Kf4 {0.52/9 0 (Kf2)} Qf6+ {0.05/13 0} 132. Kg3 {0.60/8 0} Qd6+ {
0.05/14 0} 133. Kh3 {0.52/10 0} Qh6+ {0.05/14 0} 134. Kg2 {0.24/8 0} Qc6+ {0.
05/14 0 (Qg5)} 135. Qf3 {0.44/8 0} Qc2+ {0.05/14 0} 136. Qf2 {0.36/8 0 (Kg3)}
Qe4+ {0.05/14 0 (Qc6+)} 137. Kg3 {0.80/4 0 (Qf3)} Qe5+ {0.05/13 0 (Qd3+)} 138.
Kf3 {0.36/5 0 (Kg2)} Qd5+ {0.05/13 0 (Qf6+)} 139. Kf4 {0.36/6 0 (Kg3)} Qc4+ {
0.05/12 0} 140. Kg3 {0.36/8 0} Qd3+ {0.05/14 0 (Qc7+)} 141. Qf3 {0.76/3 0} Qd6+
{0.05/14 0} 142. Qf4 {0.76/2 0 (Kg2)} Qd3+ {0.05/14 0} 143. Kf2 {0.96/3 0} Qc2+
{0.05/15 0 (b5)} 144. Kf1 {0.48/8 0 (Kg3)} Qd3+ {0.05/13 0 (b5)} 145. Kg1 {0.
00/9 0} Qd1+ {0.05/14 0} 146. Kh2 {0.00/11 0 (Kf2)} Qe2+ {0.05/13 0 (Qc2+)}
147. Kh3 {0.24/6 0 (Kg3)} Qd3+ {0.05/14 0} 148. Kh4 {0.46/9 0} Qh7+ {0.05/17 0}
149. Kg5 {0.56/7 0 (Kg3)} Qg7+ {0.05/14 0 (b5)} 150. Kh5 {0.72/4 0 (Kh4)} Qh8+
{0.05/13 0 (Qh7+)} 151. Qh6 {0.00/8 0 (Kg5)} Qe5+ {0.05/12 0 (Qe8+)} 152. g5 {
0.64/7 0 (Qg5)} Qe8+ {0.21/12 0} 153. Qg6 {1.10/9 0} Qh8+ {0.24/12 0} 154. Kg4
{1.41/7 0} Qd4+ {0.25/12 0} 155. Kf5 {1.37/9 0} Qd5+ {0.24/11 0} 156. Kf6 {1.
37/9 0} Qd4+ {0.26/10 0 (b5)} 157. Kf7 {1.61/8 0} Qd5+ {0.27/10 0} 158. Kf8 {
1.57/8 0} Ka6 {0.27/10 0 (Qd8+)} 159. Qf6 {1.57/8 0} Qc5+ {0.32/10 0} 160. Kg7
{1.49/8 0 (Kf7)} Qc7+ {0.33/10 0} 161. Kg8 {1.45/10 0 (Kg6)} Qc8+ {0.33/10 0
(Qc4+)} 162. Kf7 {1.45/8 0 (Qf8)} Qd7+ {0.34/11 0 (Qb7+)} 163. Qe7 {1.45/8 0
(Kg6)} Qd5+ {0.46/11 0} 164. Kg7 {1.45/10 0 (Kf8)} Qd4+ {0.33/10 0} 165. Qf6 {
1.45/8 0} Qd7+ {0.36/11 0} 166. Kh6 {1.49/10 0 (Qf7)} Qh3+ {0.29/10 0} 167. Kg6
{1.41/8 0} Qd3+ {0.33/11 0} 168. Qf5 {1.49/9 0} Qd6+ {0.32/11 0} 169. Kf7 {1.
53/7 0 (Kg7)} Qc7+ {0.29/10 0} 170. Kf6 {1.61/8 0} Qd8+ {0.32/11 0} 171. Ke5 {
1.57/8 0 (Kg7)} Qc7+ {0.30/8 0 (Qe7+)} 172. Kd4 {2.02/1 0} b5 {0.32/9 0 (Qd6+)}
173. Qf6+ {2.46/2 0 (g6)} Kb7 {0.35/7 0} 174. g6 {1.85/6 0 Clear win for Sting 1.0

[lech]

Alain, I represent mostly the opposite culture.
[url] http://www.talkchess.com/forum/viewtopi ... ight=sting [/ url]

[A Distel]

Alain, I represent mostly the opposite culture.
[url] http://www.talkchess.com/forum/viewtopi ... ight=sting [/ url]

I know but i just like the engine, and when Sting 2.0 is ready let me
know!

All the best,

Alain.

[lech]

Alain, I represent mostly the opposite culture.
[url] http://www.talkchess.com/forum/viewtopi ... ight=sting [/ url]

I know but i just like the engine, and when Sting 2.0 is ready let me
know!

All the best,

Alain.

Thanks Alain,

However, there is a significant problem. I changed my idea seriously. I wanted to make it faster and stronger. But not so that Sting could be stronger than Stockfish.
In this case, my idea should be incorporated into Stockfish rather than being a separate entity. But this is subject to a separate topic, in fact.

best for you too,
Marek

[fern]

Point is, Don, I am not interested in my "development" as chess player. My development as a writer, yes. Ches is just fun. In any case, at my age, i am more in the path of underdevelopment.
Playing at full force, my goal is to survive the more I can. If I get to move 40, I feel i am a genius. If 50 or more, then I organize a party. And all this NOT playing anti computer chess..


Fern

[MM]

Point is, Don, I am not interested in my "development" as chess player. My development as a writer, yes. Ches is just fun. In any case, at my age, i am more in the path of underdevelopment.
Playing at full force, my goal is to survive the more I can. If I get to move 40, I feel i am a genius. If 50 or more, then I organize a party. And all this NOT playing anti computer chess..


Fern

+1

[IGarcia]

Point is, Don, I am not interested in my "development" as chess player. My development as a writer, yes. Ches is just fun. In any case, at my age, i am more in the path of underdevelopment.
Playing at full force, my goal is to survive the more I can. If I get to move 40, I feel i am a genius. If 50 or more, then I organize a party. And all this NOT playing anti computer chess..


Fern

That remembers me to GM Symslov:

"I will play 40 good moves. If you play 40 good moves, we will draw”

[Don]

Point is, Don, I am not interested in my "development" as chess player. My development as a writer, yes. Ches is just fun. In any case, at my age, i am more in the path of underdevelopment.
Playing at full force, my goal is to survive the more I can. If I get to move 40, I feel i am a genius. If 50 or more, then I organize a party. And all this NOT playing anti computer chess..

Fern

Hi Fern,

YOU are big reason I put in the strength feature. It came out of talkchess discussions you and I had about being able to realistically evaluate your improvements and skill level. This was a first small step in that direction.

You were definitely interested in some kind of strength feature, here is a quote from one of these past discussion from you:

What is preposterous in any and every case is that an expert level player that anyway will lose almost every game to a 3000 engine be considered a 800 player due to that reason. It is not fair. same for club players, etc.

I like your idea of an engine going up or down in his capabilities according to results. that would be a "camarade" engine, or "Partner".

My original intent was to make the strength feature one that you specify by ELO, and that will probably be in the next version but I want some feedback on how the current stuff is working before proceeding. That was a feature in Rexchess many many years ago. There would be a spin button where you specify the level of play. Didn't have time to implement this properly because it requires some work calibrating the levels properly and determining which rating system to use (FIDE? USCF? Other?) I'm not sure FIDE has much meaning when you are talking about 1200 rated players, but USCF is probably not appropriate either since the U.S. is more into passively being entertained than doing anything intellectual and chess is just not big here. Faked up wrestling and Jerry Springer is the kind of thing that goes over big here.

It's also very difficult getting below the beginner level because even a 1 ply search doesn't miss hanging pieces and plays much stronger positionally than say a 1000-1200 player and would require more work getting this all working nicely.

[Uri Blass]

Point is, Don, I am not interested in my "development" as chess player. My development as a writer, yes. Ches is just fun. In any case, at my age, i am more in the path of underdevelopment.
Playing at full force, my goal is to survive the more I can. If I get to move 40, I feel i am a genius. If 50 or more, then I organize a party. And all this NOT playing anti computer chess..

Fern

Hi Fern,

YOU are big reason I put in the strength feature. It came out of talkchess discussions you and I had about being able to realistically evaluate your improvements and skill level. This was a first small step in that direction.

You were definitely interested in some kind of strength feature, here is a quote from one of these past discussion from you:

What is preposterous in any and every case is that an expert level player that anyway will lose almost every game to a 3000 engine be considered a 800 player due to that reason. It is not fair. same for club players, etc.

I like your idea of an engine going up or down in his capabilities according to results. that would be a "camarade" engine, or "Partner".

My original intent was to make the strength feature one that you specify by ELO, and that will probably be in the next version but I want some feedback on how the current stuff is working before proceeding. That was a feature in Rexchess many many years ago. There would be a spin button where you specify the level of play. Didn't have time to implement this properly because it requires some work calibrating the levels properly and determining which rating system to use (FIDE? USCF? Other?) I'm not sure FIDE has much meaning when you are talking about 1200 rated players, but USCF is probably not appropriate either since the U.S. is more into passively being entertained than doing anything intellectual and chess is just not big here. Faked up wrestling and Jerry Springer is the kind of thing that goes over big here.

It's also very difficult getting below the beginner level because even a 1 ply search doesn't miss hanging pieces and plays much stronger positionally than say a 1000-1200 player and would require more work getting this all working nicely.

Going below the beginner level is easy because you can tell the program to play random moves with some positive probability when the rating is function of the probability for a random move.

Trying to play similiar to human beginners is harder but the question is why do people need a program to do it when they can find players in the internet who do it in chess sites.

[Don]

Point is, Don, I am not interested in my "development" as chess player. My development as a writer, yes. Ches is just fun. In any case, at my age, i am more in the path of underdevelopment.
Playing at full force, my goal is to survive the more I can. If I get to move 40, I feel i am a genius. If 50 or more, then I organize a party. And all this NOT playing anti computer chess..

Fern

Hi Fern,

YOU are big reason I put in the strength feature. It came out of talkchess discussions you and I had about being able to realistically evaluate your improvements and skill level. This was a first small step in that direction.

You were definitely interested in some kind of strength feature, here is a quote from one of these past discussion from you:

What is preposterous in any and every case is that an expert level player that anyway will lose almost every game to a 3000 engine be considered a 800 player due to that reason. It is not fair. same for club players, etc.

I like your idea of an engine going up or down in his capabilities according to results. that would be a "camarade" engine, or "Partner".

My original intent was to make the strength feature one that you specify by ELO, and that will probably be in the next version but I want some feedback on how the current stuff is working before proceeding. That was a feature in Rexchess many many years ago. There would be a spin button where you specify the level of play. Didn't have time to implement this properly because it requires some work calibrating the levels properly and determining which rating system to use (FIDE? USCF? Other?) I'm not sure FIDE has much meaning when you are talking about 1200 rated players, but USCF is probably not appropriate either since the U.S. is more into passively being entertained than doing anything intellectual and chess is just not big here. Faked up wrestling and Jerry Springer is the kind of thing that goes over big here.

It's also very difficult getting below the beginner level because even a 1 ply search doesn't miss hanging pieces and plays much stronger positionally than say a 1000-1200 player and would require more work getting this all working nicely.

Going below the beginner level is easy because you can tell the program to play random moves with some positive probability when the rating is function of the probability for a random move.

Trying to play similiar to human beginners is harder but the question is why do people need a program to do it when they can find players in the internet who do it in chess sites.

It's not as easy as you think, because the desired behavior is to be able to gradually "dial in" any level. I would like Komodo to be able to set a beginner level of 600 ELO for example and have the 600 player want to move up to 700 next time. But I also want to be able to gradually transition from there to ANY level up to perhaps 2400 ELO or so.

The way it's implemented in Komodo is that a delay is introduced to each node and the delay is the same on any computer so if you set it to level 6 it should play the same speed on any computer. For instance on my computer level 6 make it play about 3500 nodes per second from the opening position. It should be the same on every computer.

As you say, it's easy to make it play really weak by just playing random moves. I could make it play weaker than random by doing a search and forbidding it to play any of the good moves, but I need something that can be flexibly scaled in up to senior master level. I don't want to suddenly switch away from random moves for example.