Glaurung 0.2.1: Faster Windows executable

[Tord Romstad]

Hi everybody,

Dann has now compiled a faster and Cygwin-free executable of Glaurung 0.2.1. It can be found on Dann's FTP site or on my Glaurung page.

Tord

[Michael Yee]

I noticed something strange while trying to benchmark/compare your old cygwin build and Dann's new build.

It appears that Dann's windows build doesn't respond to "stop" during infinite analysis when run from the commandline, even though it seems to work fine under arena. In fact, in arena's engine debug, you can clearly see glaurung responding to "stop".

Here are the steps I used:

uci
position startpos
go infinite
(wait until depth 12 or so)
stop

But ultimately I just had to kill the program since it wouldn't stop.

Interestingly, I observed similar behavior when compiling fruit with a couple windows compilers (mingw, vctoolkit). Fruit also wouldn't respond to "stop" from the commandline.

Any thoughts? Maybe I'm just misunderstanding the UCI protocol?

Michael

[Tord Romstad]

Hi Michael!

I noticed something strange while trying to benchmark/compare your old cygwin build and Dann's new build.

Do you have any numbers? It would be interesting to see how much faster Dann's build is.

It appears that Dann's windows build doesn't respond to "stop" during infinite analysis when run from the commandline, even though it seems to work fine under arena. In fact, in arena's engine debug, you can clearly see glaurung responding to "stop".

Here are the steps I used:

uci
position startpos
go infinite
(wait until depth 12 or so)
stop

But ultimately I just had to kill the program since it wouldn't stop.

Interestingly, I observed similar behavior when compiling fruit with a couple windows compilers (mingw, vctoolkit). Fruit also wouldn't respond to "stop" from the commandline.

Any thoughts? Maybe I'm just misunderstanding the UCI protocol?

Very strange. You are not misunderstanding the UCI protocol, the engine is really supposed to stop thinking immediately when receiving the 'stop' command. The Linux and Mac OS X versions of Glaurung do respond to the 'stop' command when running from the command line, and if I understand you correctly the Cygwin executable for Windows also does. This makes me believe that the behaviour you observed is caused by the different code used for input polling during the search in Dann's compile. The OS X, Linux and Cygwin executables uses the select() function for polling, and this function does not exist in Windows (IIRC). Dann has therefore replaced my polling code with some Windows-specific code which I don't understand at all in his Glaurung binary. Perhaps this Windows-specific code for some reason behaves differently when running from the command line than when running from a GUI like Arena.

This is nothing more than a guess, however, and it is very possible that I am entirely wrong. I think we need an explanation from someone with greater Windows expertize before we can say something definite about the strange behaviour you observe.

Tord

[Michael Yee]

Hi Tord,

Here are some lines with nps info from both executables:

Code: Select all


(cygwin) C:\games\chess\glaurung021>glaurung

info depth 2 score cp 0 time 1 nodes 43 nps 43000 pv b1c3 g8f6
info depth 3 score cp 20 time 2 nodes 149 nps 74500 pv b1c3 g8f6 g1f3
info depth 4 score cp 0 time 3 nodes 310 nps 103333 pv b1c3 g8f6 g1f3 b8c6
info depth 5 score cp 14 time 10 nodes 1278 nps 127800 pv b1c3 g8f6 g1f3 b8c6 e2e3
info depth 6 score cp 0 time 19 nodes 2515 nps 132368 pv b1c3 g8f6 g1f3 b8c6 e2e3 d7d6
info depth 7 score cp 15 time 83 nodes 11044 nps 133060 pv b1c3 b8c6 g1f3 d7d5 d2d4 g8f6 c1g5
info depth 8 score cp 1 time 288 nodes 53952 nps 187333 pv d2d4 g8f6 g1f3 b8c6 d4d5 c6b4 b1c3 d7d6
info depth 9 score cp 10 time 413 nodes 101375 nps 245460 pv d2d4 g8f6 g1f3 d7d6 c1f4 b8c6 b1d2 f6d5 f4g5
info depth 10 score cp 6 time 977 nodes 311130 nps 318454 pv d2d4 d7d5 g1f3 b8c6 b1c3 c8g4 c1g5 g4f3 g2f3 g8f6
info nodes 320000 nps 319042 time 1003
info depth 11 score cp 10 time 1707 nodes 580715 nps 340196 pv d2d4 d7d5 g1f3 b8c6 b1c3 g8f6 c1g5 c8g4 f3e5 c6e5 d4e5
info nodes 700000 nps 342801 time 2042
info nodes 1060000 nps 345276 time 3070
info depth 11 score cp 16 time 3380 nodes 1165565 nps 344841 pv e2e4 b8c6 b1c3 g8f6 d2d4 d7d5 e4e5 f6e4 c3e4 d5e4 f1b5
info nodes 1420000 nps 347613 time 4085
info nodes 1780000 nps 347996 time 5115
info depth 12 score cp 11 time 5123 nodes 1782473 nps 347935 pv e2e4 b8c6 g1f3 g8f6 e4e5 f6g4 d2d4 d7d6 h2h3 g4h6 f1b5 c8d7
info nodes 2140000 nps 349387 time 6125
info nodes 2500000 nps 349113 time 7161
info nodes 2860000 nps 350232 time 8166
info nodes 3200000 nps 348356 time 9186
info nodes 3560000 nps 347384 time 10248
info nodes 3900000 nps 346205 time 11265
info nodes 4260000 nps 345582 time 12327
info depth 13 score cp 9 time 12837 nodes 4432699 nps 345306 pv e2e4 e7e5 b1c3 b8c6 g1f3 g8f6 f1b5 f8b4 e1g1 d7d6 d2d3 e8g8 c1g5 b4c3 b2c3
info nodes 4620000 nps 345653 time 13366
info nodes 5000000 nps 346836 time 14416
info nodes 5360000 nps 347645 time 15418

(dann) C:\games\chess\glaurung021dc>glaurung

info depth 2 score cp 0 time 0 nodes 43 pv b1c3 g8f6
info depth 3 score cp 20 time 0 nodes 149 pv b1c3 g8f6 g1f3
info depth 4 score cp 0 time 10 nodes 310 nps 31000 pv b1c3 g8f6 g1f3 b8c6
info depth 5 score cp 14 time 20 nodes 1278 nps 63900 pv b1c3 g8f6 g1f3 b8c6 e2e3
info depth 6 score cp 0 time 30 nodes 2515 nps 83833 pv b1c3 g8f6 g1f3 b8c6 e2e3 d7d6
info depth 7 score cp 15 time 90 nodes 11134 nps 123711 pv b1c3 b8c6 g1f3 d7d5 d2d4 g8f6 c1f4
info depth 8 score cp 0 time 180 nodes 25083 nps 139350 pv b1c3 b8c6 g1f3 g8f6 e2e3 e7e6 f1d3 f8d6
info depth 8 score cp 1 time 300 nodes 51186 nps 170620 pv d2d4 g8f6 g1f3 b8c6 d4d5 c6b4 b1c3 d7d6
info depth 9 score cp 11 time 481 nodes 121722 nps 253060 pv d2d4 g8f6 b1c3 d7d5 e2e3 b8c6 g1f3 c8g4 f1d3
info depth 10 score cp 6 time 891 nodes 289047 nps 324407 pv d2d4 d7d5 b1c3 b8c6 g1f3 c8g4 c1g5 g4f3 g2f3 g8f6
info nodes 340000 nps 333006 time 1021
info depth 11 score cp 10 time 1752 nodes 628502 nps 358734 pv d2d4 d7d5 b1c3 b8c6 g1f3 g8f6 c1g5 c8g4 f3e5 c6e5 d4e5
info nodes 740000 nps 360448 time 2053
info depth 11 score cp 14 time 2974 nodes 1095143 nps 368239 pv e2e4 b8c6 b1c3 e7e5 g1f3 f8c5 f1b5 d7d6 e1g1 d8f6 d2d3
info nodes 1140000 nps 368455 time 3094
info nodes 1520000 nps 369290 time 4116
info depth 12 score cp 11 time 4516 nodes 1674211 nps 370728 pv e2e4 b8c6 g1f3 g8f6 e4e5 f6g4 d2d4 d7d6 h2h3 g4h6 f1b5 c8d7
info nodes 1920000 nps 373032 time 5147
info nodes 2300000 nps 373437 time 6159
info nodes 2680000 nps 373779 time 7170
info nodes 3060000 nps 372172 time 8222
info nodes 3420000 nps 370812 time 9223
info nodes 3800000 nps 369829 time 10275
info nodes 4180000 nps 370370 time 11286
info depth 13 score cp 12 time 12278 nodes 4554644 nps 370959 pv e2e4 e7e5 g1f3 g8f6 b1c3 f8b4 f1c4 d7d6 e1g1 e8g8 d2d3 c8e6 c4e6 f7e6
info nodes 4560000 nps 371093 time 12288
info nodes 4960000 nps 372960 time 13299



So it seems that Dann's is about 7-8% faster (372k vs 347).

I agree that the i/o handling must be different when executing in a command shell (vs having standard input/output mapped to something else by arena). I'm not a windows programmer either, so my only hunch is that it has something to do with different blocking or buffering or something else.

Michael

[Michael Yee]

... and tests were run on a Pentium M 1.7 GHz laptop

[Jim Ablett]

Hello Michael,

Would you mind testing my Intel P4 optimized
compile against the others?

http://homepages.tesco.net/henry.ablett ... g2.1ja.zip

Thanks,
Jim.

[Michael Yee]

Hi Jim,

No problem:

Code: Select all

C:\games\chess\glaurung021ja>glaurung2.1ja.exe

position startpos
go infinite
info depth 2 score cp 0 time 0 nodes 43 pv b1c3 g8f6
info depth 3 score cp 20 time 0 nodes 149 pv b1c3 g8f6 g1f3
info depth 4 score cp 0 time 0 nodes 310 pv b1c3 g8f6 g1f3 b8c6
info depth 5 score cp 14 time 10 nodes 1278 nps 127800 pv b1c3 g8f6 g1f3 b8c6 e2e3
info depth 6 score cp 0 time 20 nodes 2515 nps 125750 pv b1c3 g8f6 g1f3 b8c6 e2e3 d7d6
info depth 7 score cp 15 time 90 nodes 11134 nps 123711 pv b1c3 b8c6 g1f3 d7d5 d2d4 g8f6 c1f4

info depth 8 score cp 0 time 161 nodes 25083 nps 155795 pv b1c3 b8c6 g1f3 g8f6 e2e3 e7e6 f1d3 f8d6
info depth 8 score cp 1 time 231 nodes 51253 nps 221874 pv d2d4 g8f6 g1f3 b8c6 d4d5 c6b4 b1c3 d7d6
info depth 9 score cp 11 time 411 nodes 119891 nps 291705 pv d2d4 g8f6 b1c3 d7d5 e2e3 b8c6 g1f3 c8g4 f1d3
info depth 10 score cp 6 time 811 nodes 275600 nps 339827 pv d2d4 d7d5 b1c3 b8c6 g1f3 c8g4 c1g5 g4f3 g2f3 g8f6
info nodes 360000 nps 345489 time 1042
info depth 11 score cp 10 time 1683 nodes 603031 nps 358307 pv d2d4 d7d5 b1c3 g8f6 g1f3 b8c6 c1g5 c8g4 f3e5 c6e5 d4e5
info nodes 740000 nps 356970 time 2073
info depth 11 score cp 16 time 2975 nodes 1066053 nps 358337 pv e2e4 b8c6 b1c3 g8f6 d2d4 d7d5 e4e5 f6e4 c3e4 d5e4 f1b5
info nodes 1100000 nps 357723 time 3075
info nodes 1480000 nps 359572 time 4116
info nodes 1840000 nps 356037 time 5168
info depth 12 score cp 9 time 5578 nodes 1994730 nps 357606 pv e2e4 e7e5 b1c3 g8f6 g1f3 b8c6 f1b5 f8b4 d2d3 d7d6 b5c6 b7c6 c1d2
info nodes 2220000 nps 358700 time 6189
info nodes 2600000 nps 359066 time 7241
info nodes 2960000 nps 357401 time 8282
info nodes 3320000 nps 356835 time 9304
info depth 13 score cp 9 time 9664 nodes 3448535 nps 356843 pv e2e4 e7e5 b1c3 g8f6 g1f3 b8c6 f1b5 f8b4 d2d3 d7d6 e1g1 e8g8 c1g5 b4c3 b2c3
info nodes 3700000 nps 357660 time 10345
info nodes 4080000 nps 358933 time 11367


The 358k seems to fall right between Tord's and Dann's nps. (As a side note, the uci "stop" command doesn't seem to work from the commandline with this build either.)

Michael

[Jim Ablett]

Thanks Michael.
I thought the Intel compiler was
supposed to be the fastest out there.
Oh well.

Regards,
Jim.

[Tord Romstad]

Hi Michael,

Thanks for the benchmarks! The difference between the Cygwin and MSVC executables is smaller than I expected. Evidently gcc is quite good!

Unfortunately, there is one thing which worries me. The node count is different in all three executables, and the MSVC executable even has a different score and PV than the two others. Were all three tests done with the same hash table size? If they were, I'm afraid I have a nasty bug to hunt down.

Tord

[Peter Fendrich]

Unfortunately, there is one thing which worries me. The node count is different in all three executables, and the MSVC executable even has a different score and PV than the two others. Were all three tests done with the same hash table size? If they were, I'm afraid I have a nasty bug to hunt down.
Tord

Maybe your random code behaves differently in different environments? I don't now what you're using but just a hint
/Peter

[Bryan Hofmann]

Try this compile in your bench test, I am sure you will find this to be the fastest.

http://www.freewebs.com/bkhofmann/glaurung.zip


Bryan

[Norm Pollock]

Try this compile in your bench test, I am sure you will find this to be the fastest.

http://www.freewebs.com/bkhofmann/glaurung.zip


Bryan

something strange with that link. I hand-typed it and then it worked.

[Volker Pittlik]

...

something strange with that link. I hand-typed it and then it worked.

Yes. Strange "feature" of the forum software. Please copy and paste it.

Volker

[Tord Romstad]

Try this compile in your bench test, I am sure you will find this to be the fastest.

http://www.freewebs.com/bkhofmann/glaurung.zip

Thanks, Bryan!

What compiler did you use?

If somebody can test Bryan's executable and confirm that it is indeed faster than Dann's, I will replace the Windows executable on my home page.

Tord

[Michael Yee]

Here's a run of Bryan Hofmann's compile:

Code: Select all

C:\games\chess\glaurung021bh>glaurung.exe

position startpos
go infinite

info depth 2 score cp 0 time 0 nodes 43 pv b1c3 g8f6
info depth 3 score cp 20 time 0 nodes 149 pv b1c3 g8f6 g1f3
info depth 4 score cp 0 time 0 nodes 310 pv b1c3 g8f6 g1f3 b8c6
info depth 5 score cp 14 time 10 nodes 1278 nps 127800 pv b1c3 g8f6 g1f3 b8c6 e2e3
info depth 6 score cp 0 time 20 nodes 2515 nps 125750 pv b1c3 g8f6 g1f3 b8c6 e2e3 d7d6
info depth 7 score cp 15 time 80 nodes 11134 nps 139175 pv b1c3 b8c6 g1f3 d7d5 d2d4 g8f6 c1f4
info depth 8 score cp 0 time 180 nodes 25083 nps 139350 pv b1c3 b8c6 g1f3 g8f6 e2e3 e7e6 f1d3 f8d6
info depth 8 score cp 1 time 330 nodes 51186 nps 155109 pv d2d4 g8f6 g1f3 b8c6 d4d5 c6b4 b1c3 d7d6
info depth 9 score cp 11 time 501 nodes 121722 nps 242958 pv d2d4 g8f6 b1c3 d7d5 e2e3 b8c6 g1f3 c8g4 f1d3
info depth 10 score cp 6 time 901 nodes 289047 nps 320806 pv d2d4 d7d5 b1c3 b8c6 g1f3 c8g4 c1g5 g4f3 g2f3 g8f6
info nodes 340000 nps 329776 time 1031
info depth 11 score cp 10 time 1742 nodes 628502 nps 360793 pv d2d4 d7d5 b1c3 b8c6 g1f3 g8f6 c1g5 c8g4 f3e5 c6e5 d4e5
info nodes 760000 nps 366618 time 2073
info depth 11 score cp 14 time 2914 nodes 1095143 nps 375821 pv e2e4 b8c6 b1c3 e7e5 g1f3 f8c5 f1b5 d7d6 e1g1 d8f6 d2d3
info nodes 1160000 nps 374919 time 3094
info nodes 1560000 nps 379008 time 4116
info depth 12 score cp 11 time 4396 nodes 1674211 nps 380848 pv e2e4 b8c6 g1f3 g8f6 e4e5 f6g4 d2d4 d7d6 h2h3 g4h6 f1b5 c8d7
info nodes 1980000 nps 383944 time 5157
info nodes 2360000 nps 383179 time 6159
info nodes 2760000 nps 384401 time 7180
info nodes 3140000 nps 382367 time 8212
info nodes 3520000 nps 381241 time 9233
info nodes 3900000 nps 380673 time 10245
info nodes 4300000 nps 381003 time 11286
info depth 13 score cp 12 time 11927 nodes 4554638 nps 381876 pv e2e4 e7e5 g1f3 g8f6 b1c3 f8b4 f1c4 d7d6 e1g1 e8g8 d2d3 c8e6 c4e6 f7e6
info nodes 4700000 nps 382486 time 12288
info nodes 5120000 nps 384413 time 13319

Some notes:

(1) definitely seems fastest--382k (bryan) vs 372k (dann) vs 247k (cygwin)
(2) doesn't respond to "stop" in console mode

Finally, a disclaimer: these tests are just tests from the start position, so they might not representative of the speed over all positions encountered over a game.

Michael

[Michael Yee]

Hi Tord,

Regarding your question about hash size and different node counts...

Each run uses glaurung's defaults. But I noticed an interesting thing. All the node counts for the early iterations are identical. And Dann and Bryan only diverge after depth 12. Could it be that some of your info strings are sent based on elapsed time? Or are they only sent when the pv changes?

Michael

[Tord Romstad]

Hi Michael,

First of all, thanks for repeating your benchmark with Bryan's executable! I will replace the Windows download on my home page tonight or tomorrow.

Regarding your question about hash size and different node counts...

Each run uses glaurung's defaults. But I noticed an interesting thing. All the node counts for the early iterations are identical. And Dann and Bryan only diverge after depth 12. Could it be that some of your info strings are sent based on elapsed time? Or are they only sent when the pv changes?

The info strings with a PV are always displayed at the exact moment when the PV changes. The elapsed time should not have any influence at all. Peter's guess that it has something to do with random numbers also cannot be correct. It looks to me like the only possible explanation is a bug.

Tord

[Peter Eizenhammer]

Hi Tord,

it may not be of any use any longer, but I have some outputs
for BH and DC versions nevertheless. (Athlon XP 1,67GHz, 128MB hash for each engine,
Arena1.1, four positions tested, engines got restarted after every position, hash was cleared.).
Bryans version seems to be about 3%-5% faster, at least on my AMD.
(Btw: Many thanks to those who compile Crafties, Glaurungs, Fruits... .!)

Peter

Code: Select all



FEN: r5k1/Rb4p1/2q2pBp/1pp5/1b4QN/1P2P2P/5PP1/6K1 w - - 0 30

Glaurung 0.2.1 BH :
     9   00:01        263.812   179.586   +0,78   30.Txa8+ Lxa8 31.Ld3 Lb7 32.Sf5 Dc7 33.Lxb5 Kh8 34.Dg6 Le1    

                  35.De8+ Kh7
  10   00:03        560.392   166.832   +0,81   30.Txa8+ Lxa8 31.Ld3 g5 32.Sf5 Kh8 33.Sxh6 La5 34.h4 Dd5    

                  35.Dc8+ Ld8
  11   00:07      1.108.569   163.868   +0,79   30.Txa8+ Lxa8 31.Ld3 g5 32.Sf5 Kh8 33.Sxh6 Kg7 34.Sf5+ Kf8    

                  35.Dg3 Db6 36.Sd6
  12   00:20      3.200.404   160.646   +0,91   30.Txa8+ Lxa8 31.Ld3 Kf8 32.Sf5 g5 33.Sxh6 Kg7 34.Sf5+ Kf8    

                  35.h4 Dd7 36.Le4 Lxe4 37.Dxe4 gxh4 38.Sxh4
  13   00:41      6.676.505   160.338   +0,91   30.Txa8+ Lxa8 31.Ld3 Kf8 32.Sf5 g5 33.Sxh6 Kg7 34.Sf5+ Kf8    

                  35.h4 Dd7 36.Le4 Lxe4 37.Dxe4 gxh4 38.Sxh4
  13   01:14     11.602.308   156.161   +4,52   30.Txb7 Dxb7 31.De6+ Kh8 32.Le4 Da7 33.Sg6+ Kh7 34.Se7+ Kh8    

                  35.Df5 Da1+ 36.Lb1 Dxb1+ 37.Dxb1 Lc3 38.Sg6+ Kg8 39.Df5 Ta3
  14   01:57     18.167.801   155.011   +4,74   30.Txb7 Dxb7 31.De6+ Kh8 32.Le4 Da7 33.Sg6+ Kh7 34.Se7+ Kh8    

                  35.Df5 Da1+ 36.Lb1 Dxb1+ 37.Dxb1 La3 38.Dd3 Tb8 39.Sg6+ Kh7    

                  40.Se5+ Kg8
  15   03:44     34.633.091   154.579   +5,03   30.Txb7 Dxb7 31.De6+ Kh8 32.Le4 Da7 33.Sg6+ Kh7 34.Se5+ g6    

                  35.Sd7 Da1+ 36.Kh2 Kh8 37.Lxa8 Dxa8 38.Dxf6+ Kg8 39.Dxg6+ Kh8    

                  40.Dxh6+ Kg8 41.Dg6+ Kh8 42.Dh5+ Kg7 43.Dg5+ Kf7 44.Df5+ Kg7    

                  45.Sxc5

Glaurung 0.2.1 DC :
   9   00:02        263.812   154.910   +0,78   30.Txa8+ Lxa8 31.Ld3 Lb7 32.Sf5 Dc7 33.Lxb5 Kh8 34.Dg6 Le1    

                  35.De8+ Kh7
  10   00:04        560.392   153.911   +0,81   30.Txa8+ Lxa8 31.Ld3 g5 32.Sf5 Kh8 33.Sxh6 La5 34.h4 Dd5    

                  35.Dc8+ Ld8
  11   00:07      1.108.569   155.916   +0,79   30.Txa8+ Lxa8 31.Ld3 g5 32.Sf5 Kh8 33.Sxh6 Kg7 34.Sf5+ Kf8    

                  35.Dg3 Db6 36.Sd6
  12   00:21      3.200.404   155.404   +0,91   30.Txa8+ Lxa8 31.Ld3 Kf8 32.Sf5 g5 33.Sxh6 Kg7 34.Sf5+ Kf8    

                  35.h4 Dd7 36.Le4 Lxe4 37.Dxe4 gxh4 38.Sxh4
  13   00:43      6.676.505   155.491   +0,91   30.Txa8+ Lxa8 31.Ld3 Kf8 32.Sf5 g5 33.Sxh6 Kg7 34.Sf5+ Kf8    

                  35.h4 Dd7 36.Le4 Lxe4 37.Dxe4 gxh4 38.Sxh4
  13   01:16     11.602.308   152.756   +4,52   30.Txb7 Dxb7 31.De6+ Kh8 32.Le4 Da7 33.Sg6+ Kh7 34.Se7+ Kh8    

                  35.Df5 Da1+ 36.Lb1 Dxb1+ 37.Dxb1 Lc3 38.Sg6+ Kg8 39.Df5 Ta3
  14   02:00     18.167.801   151.654   +4,74   30.Txb7 Dxb7 31.De6+ Kh8 32.Le4 Da7 33.Sg6+ Kh7 34.Se7+ Kh8    

                  35.Df5 Da1+ 36.Lb1 Dxb1+ 37.Dxb1 La3 38.Dd3 Tb8 39.Sg6+ Kh7    

                  40.Se5+ Kg8
  15   03:50     34.633.091   150.967   +5,03   30.Txb7 Dxb7 31.De6+ Kh8 32.Le4 Da7 33.Sg6+ Kh7 34.Se5+ g6    

                  35.Sd7 Da1+ 36.Kh2 Kh8 37.Lxa8 Dxa8 38.Dxf6+ Kg8 39.Dxg6+ Kh8    

                  40.Dxh6+ Kg8 41.Dg6+ Kh8 42.Dh5+ Kg7 43.Dg5+ Kf7 44.Df5+ Kg7    

                  45.Sxc5


FEN: r2qk2r/1p1bbp2/1P2p3/p2pPp2/n2N1N1p/3PB3/5QPP/R4RK1 w kq - 0 24

Glaurung 0.2.1 BH :
   9   00:03        461.087   156.141   -1,58   24.Sb3 Lb4 25.Ld4 Tg8 26.Tac1 Tc8 27.Tc7 Txc7 28.bxc7 Dxc7    

               29.Dxh4
   9   00:06        859.385   149.458   -1,53   24.Sf3 Tg8 25.Tfc1 Tc8 26.Txc8 Dxc8 27.De1 Dc2 28.Dxa5
  10   00:08      1.282.797   150.087   -1,56   24.Sf3 Tc8 25.Ta2 Tg8 26.Taa1 Tc3 27.Ld4 Tb3 28.Sd2 Ta3
  10   00:09      1.449.827   149.899   -1,53   24.Sb3 Lb4 25.Ld4 Tg8 26.Tac1 Tc8 27.Sh5 Txc1 28.Sf6+ Kf8    

               29.Sxc1
  11   00:22      3.248.917   147.678   -1,47   24.Sb3 Ta6 25.Sc5 Lxc5 26.Lxc5 Sxc5 27.Dxc5 Dxb6 28.Dxb6 Txb6    

               29.Txa5
  12   00:51      7.369.272   144.762   -1,43   24.Sb3 Lb4 25.Ld4 Th7 26.Ta2 Tc8 27.Tfa1 Sc3 28.Txa5 Lxa5    

               29.Sxa5 Kf8 30.Sxb7
  12   01:18     11.157.090   142.668    0,00   24.Txa4 Lxa4 25.Sdxe6 fxe6 26.Sxe6 Dc8 27.Dxf5 Lc6 28.Df7+ Kd7    

               29.Sc5+ Kd8 30.Se6+ Kd7
  13   01:38     13.908.144   141.401   +0,57   24.Txa4 Lxa4 25.Sdxe6 fxe6 26.Sxe6 Dc8 27.Dxf5 Kd7 28.Tc1 Dxc1+

                  29.Lxc1 Kc6 30.Le3 The8
  14   02:06     17.849.394   141.416   +0,81   24.Txa4 Lxa4 25.Sdxe6 fxe6 26.Sxe6 Dc8 27.Dxf5 Kd7 28.Tc1 Dxc1+

                  29.Lxc1 Kc6 30.Dg4 Lb4 31.Dd4
  15   03:26     28.842.137   139.957   +1,17   24.Txa4 Lxa4 25.Sdxe6 fxe6 26.Sxe6 Dc8 27.Dxf5 Dc6 28.Tc1 Dxc1+

                  29.Lxc1 Tc8 30.Sc7+ Kd8 31.De6 Ld7 32.Dxd5

Glaurung 0.2.1 DC :
   9   00:03        461.087   148.259   -1,58   24.Sb3 Lb4 25.Ld4 Tg8 26.Tac1 Tc8 27.Tc7 Txc7 28.bxc7 Dxc7    

               29.Dxh4
   9   00:06        859.385   143.589   -1,53   24.Sf3 Tg8 25.Tfc1 Tc8 26.Txc8 Dxc8 27.De1 Dc2 28.Dxa5
  10   00:09      1.282.797   140.580   -1,56   24.Sf3 Tc8 25.Ta2 Tg8 26.Taa1 Tc3 27.Ld4 Tb3 28.Sd2 Ta3
  10   00:10      1.449.827   141.653   -1,53   24.Sb3 Lb4 25.Ld4 Tg8 26.Tac1 Tc8 27.Sh5 Txc1 28.Sf6+ Kf8    

               29.Sxc1
  11   00:23      3.248.917   142.709   -1,47   24.Sb3 Ta6 25.Sc5 Lxc5 26.Lxc5 Sxc5 27.Dxc5 Dxb6 28.Dxb6 Txb6    

               29.Txa5
  12   00:52      7.369.272   141.333   -1,43   24.Sb3 Lb4 25.Ld4 Th7 26.Ta2 Tc8 27.Tfa1 Sc3 28.Txa5 Lxa5    

               29.Sxa5 Kf8 30.Sxb7
  12   01:21     11.157.090   138.463    0,00   24.Txa4 Lxa4 25.Sdxe6 fxe6 26.Sxe6 Dc8 27.Dxf5 Lc6 28.Df7+ Kd7    

               29.Sc5+ Kd8 30.Se6+ Kd7
  13   01:41     13.908.144   137.470   +0,57   24.Txa4 Lxa4 25.Sdxe6 fxe6 26.Sxe6 Dc8 27.Dxf5 Kd7 28.Tc1 Dxc1+

                  29.Lxc1 Kc6 30.Le3 The8
  14   02:11     17.849.394   136.727   +0,81   24.Txa4 Lxa4 25.Sdxe6 fxe6 26.Sxe6 Dc8 27.Dxf5 Kd7 28.Tc1 Dxc1+

                  29.Lxc1 Kc6 30.Dg4 Lb4 31.Dd4
  15   03:32     28.842.137   136.128   +1,17   24.Txa4 Lxa4 25.Sdxe6 fxe6 26.Sxe6 Dc8 27.Dxf5 Dc6 28.Tc1 Dxc1+

                  29.Lxc1 Tc8 30.Sc7+ Kd8 31.De6 Ld7 32.Dxd5


FEN: r3rnk1/2qn1pb1/1p1p2pp/pPp5/2PPb3/P1Q2NP1/1B3PBP/R2R1NK1 w - - 1 19

Glaurung 0.2.1 BH :
   9   00:02        462.287   176.109   +0,14   19.Dc1 Se6 20.dxc5 Lxb2 21.Dxb2 dxc5 22.S1d2 Lb7 23.a4
  10   00:05        842.721   170.694   +0,07   19.Dc1 Se6 20.dxc5 dxc5 21.Lxg7 Kxg7 22.Se3 Lxf3 23.Lxf3 Tad8
  11   00:14      2.308.722   160.428   +0,14   19.Dc1 Se6 20.dxc5 dxc5 21.Lxg7 Kxg7 22.Se3 Kg8 23.Se5 Sxe5    

                  24.Lxe4
  12   00:26      4.101.178   158.978   +0,10   19.Dc1 Se6 20.dxc5 Sdxc5 21.Lxg7 Sxg7 22.Ta2 Db7 23.Dc3 Sf5    

                  24.Tad2 f6
  13   00:55      8.691.956   157.588   +0,14   19.Dc1 Se6 20.dxc5 Sdxc5 21.Lxg7 Sxg7 22.Ta2 Sf5 23.g4 Sd3    

                  24.Dc3 Lxf3 25.Lxf3
  14   01:56     17.991.413   155.265   +0,13   19.Dc1 Sf6 20.S1d2 Lb7 21.d5 S8d7 22.Dc2 Te2 23.Te1 Tae8    

                  24.Txe2 Txe2 25.Te1 Txe1+ 26.Sxe1
  15   04:55     45.890.068   155.263   +0,14   19.Dc1 Se6 20.dxc5 dxc5 21.Lxg7 Kxg7 22.Db2+ f6 23.Se3 Tad8    

                  24.Sd5 Lxd5 25.Txd5 Kh7 26.Tad1 Te7

Glaurung 0.2.1 DC :
   9   00:03        462.287   154.870   +0,14   19.Dc1 Se6 20.dxc5 Lxb2 21.Dxb2 dxc5 22.S1d2 Lb7 23.a4
  10   00:06        842.721   155.857   +0,07   19.Dc1 Se6 20.dxc5 dxc5 21.Lxg7 Kxg7 22.Se3 Lxf3 23.Lxf3 Tad8
  11   00:15      2.308.722   151.699   +0,14   19.Dc1 Se6 20.dxc5 dxc5 21.Lxg7 Kxg7 22.Se3 Kg8 23.Se5 Sxe5    

                  24.Lxe4
  12   00:27      4.101.178   152.778   +0,10   19.Dc1 Se6 20.dxc5 Sdxc5 21.Lxg7 Sxg7 22.Ta2 Db7 23.Dc3 Sf5    

                  24.Tad2 f6
  13   00:58      8.691.956   149.980   +0,14   19.Dc1 Se6 20.dxc5 Sdxc5 21.Lxg7 Sxg7 22.Ta2 Sf5 23.g4 Sd3    

                  24.Dc3 Lxf3 25.Lxf3
  14   02:00     17.991.413   149.499   +0,13   19.Dc1 Sf6 20.S1d2 Lb7 21.d5 S8d7 22.Dc2 Te2 23.Te1 Tae8    

                  24.Txe2 Txe2 25.Te1 Txe1+ 26.Sxe1
  15   05:05     45.890.068   150.351   +0,14   19.Dc1 Se6 20.dxc5 dxc5 21.Lxg7 Kxg7 22.Db2+ f6 23.Se3 Tad8    

                  24.Sd5 Lxd5 25.Txd5 Kh7 26.Tad1 Te7


FEN: 4k3/rp3ppb/1np1p2p/N7/1PP5/6P1/4PPBP/3R2K1 w - - 4 35

Glaurung 0.2.1 BH :
   9   00:00        135.466   377.342   +0,76   35.Sxb7 Sxc4 36.Tc1 Se5 37.Sa5 Kd7 38.Sxc6 Sxc6 39.Txc6
  10   00:01        290.889   273.906   +0,66   35.Sxb7 Sxc4 36.Tc1 Se5 37.Sa5 Kd7 38.Sxc6 Sxc6 39.Lxc6+ Kd6    

                  40.Kg2
  11   00:03        710.102   244.357   +0,72   35.Sxb7 Txb7 36.Lxc6+ Td7 37.Lxd7+ Sxd7 38.c5 Le4 39.Td4 Lc6    

                  40.Tg4 Kf8 41.Tc4
  12   00:06      1.526.399   233.715   +0,68   35.Sxb7 Sxc4 36.Tc1 Ke7 37.Txc4 Txb7 38.Lxc6 Tc7 39.b5 Kd6    

                  40.Td4+ Kc5 41.Td8
  13   00:14      3.167.707   227.794   +0,71   35.Sxb7 Sxc4 36.Tc1 Ke7 37.Txc4 Txb7 38.Lxc6 Ta7 39.Kg2 Ta2    

                  40.Kf3 Kd6 41.b5
  14   00:46     10.353.412   223.032   +0,75   35.Sxb7 Sxc4 36.Lxc6+ Ke7 37.f4 Se3 38.Td7+ Kf6 39.b5 Lg6    

                  40.Sc5 Ta2 41.b6 Txe2 42.b7 Tb2 43.Td8
  15   01:34     20.514.488   218.895   +0,69   35.Sxb7 Sxc4 36.Lxc6+ Ke7 37.f4 Lf5 38.b5 Sb6 39.Td8 Ta2 40.e4    

                  Lg4 41.e5 Ta1+ 42.Kg2 Ta2+ 43.Kf1 Le2+ 44.Kf2 Lc4+ 45.Kg1
  16   03:06     40.705.712   218.885   +0,71   35.Sxb7 Sxc4 36.Lxc6+ Ke7 37.f4 Lf5 38.Kf2 Ta2 39.Td4 Sb6    

                  40.Sd6 Lg4 41.Lf3 Lxf3 42.Kxf3 Kf6 43.b5

Glaurung 0.2.1 DC :
  9   00:01        135.466   228.057   +0,76   35.Sxb7 Sxc4 36.Tc1 Se5 37.Sa5 Kd7 38.Sxc6 Sxc6 39.Txc6
  10   00:01        290.889   216.435   +0,66   35.Sxb7 Sxc4 36.Tc1 Se5 37.Sa5 Kd7 38.Sxc6 Sxc6 39.Lxc6+ Kd6    

                  40.Kg2
  11   00:03        710.102   221.699   +0,72   35.Sxb7 Txb7 36.Lxc6+ Td7 37.Lxd7+ Sxd7 38.c5 Le4 39.Td4 Lc6    

                  40.Tg4 Kf8 41.Tc4
  12   00:07      1.526.399   220.005   +0,68   35.Sxb7 Sxc4 36.Tc1 Ke7 37.Txc4 Txb7 38.Lxc6 Tc7 39.b5 Kd6    

                  40.Td4+ Kc5 41.Td8
  13   00:15      3.167.707   218.221   +0,71   35.Sxb7 Sxc4 36.Tc1 Ke7 37.Txc4 Txb7 38.Lxc6 Ta7 39.Kg2 Ta2    

                  40.Kf3 Kd6 41.b5
  14   00:50     10.353.412   208.502   +0,75   35.Sxb7 Sxc4 36.Lxc6+ Ke7 37.f4 Se3 38.Td7+ Kf6 39.b5 Lg6    

                  40.Sc5 Ta2 41.b6 Txe2 42.b7 Tb2 43.Td8
  15   01:38     20.514.488   209.800   +0,69   35.Sxb7 Sxc4 36.Lxc6+ Ke7 37.f4 Lf5 38.b5 Sb6 39.Td8 Ta2 40.e4    

                  Lg4 41.e5 Ta1+ 42.Kg2 Ta2+ 43.Kf1 Le2+ 44.Kf2 Lc4+ 45.Kg1
  16   03:15     40.705.712   209.401   +0,71   35.Sxb7 Sxc4 36.Lxc6+ Ke7 37.f4 Lf5 38.Kf2 Ta2 39.Td4 Sb6    

                  40.Sd6 Lg4 41.Lf3 Lxf3 42.Kxf3 Kf6 43.b5



[Peter Eizenhammer]

Hi Tord,

it may not be of any use any longer, but I have some outputs
for BH and DC versions nevertheless. (Athlon XP 1,67GHz, 128MB hash for each engine,
Arena1.1, four positions tested, engines got restarted after every position, hash was cleared.).
Bryans version seems to be about 3%-5% faster, at least on my AMD.
(Btw: Many thanks to those who compile Crafties, Glaurungs, Fruits... .!)

Peter

Code: Select all



FEN: r5k1/Rb4p1/2q2pBp/1pp5/1b4QN/1P2P2P/5PP1/6K1 w - - 0 30

Glaurung 0.2.1 BH :
     9   00:01        263.812   179.586   +0,78   30.Txa8+ Lxa8 31.Ld3 Lb7 32.Sf5 Dc7 33.Lxb5 Kh8 34.Dg6 Le1    

                  35.De8+ Kh7
  10   00:03        560.392   166.832   +0,81   30.Txa8+ Lxa8 31.Ld3 g5 32.Sf5 Kh8 33.Sxh6 La5 34.h4 Dd5    

                  35.Dc8+ Ld8
  11   00:07      1.108.569   163.868   +0,79   30.Txa8+ Lxa8 31.Ld3 g5 32.Sf5 Kh8 33.Sxh6 Kg7 34.Sf5+ Kf8    

                  35.Dg3 Db6 36.Sd6
  12   00:20      3.200.404   160.646   +0,91   30.Txa8+ Lxa8 31.Ld3 Kf8 32.Sf5 g5 33.Sxh6 Kg7 34.Sf5+ Kf8    

                  35.h4 Dd7 36.Le4 Lxe4 37.Dxe4 gxh4 38.Sxh4
  13   00:41      6.676.505   160.338   +0,91   30.Txa8+ Lxa8 31.Ld3 Kf8 32.Sf5 g5 33.Sxh6 Kg7 34.Sf5+ Kf8    

                  35.h4 Dd7 36.Le4 Lxe4 37.Dxe4 gxh4 38.Sxh4
  13   01:14     11.602.308   156.161   +4,52   30.Txb7 Dxb7 31.De6+ Kh8 32.Le4 Da7 33.Sg6+ Kh7 34.Se7+ Kh8    

                  35.Df5 Da1+ 36.Lb1 Dxb1+ 37.Dxb1 Lc3 38.Sg6+ Kg8 39.Df5 Ta3
  14   01:57     18.167.801   155.011   +4,74   30.Txb7 Dxb7 31.De6+ Kh8 32.Le4 Da7 33.Sg6+ Kh7 34.Se7+ Kh8    

                  35.Df5 Da1+ 36.Lb1 Dxb1+ 37.Dxb1 La3 38.Dd3 Tb8 39.Sg6+ Kh7    

                  40.Se5+ Kg8
  15   03:44     34.633.091   154.579   +5,03   30.Txb7 Dxb7 31.De6+ Kh8 32.Le4 Da7 33.Sg6+ Kh7 34.Se5+ g6    

                  35.Sd7 Da1+ 36.Kh2 Kh8 37.Lxa8 Dxa8 38.Dxf6+ Kg8 39.Dxg6+ Kh8    

                  40.Dxh6+ Kg8 41.Dg6+ Kh8 42.Dh5+ Kg7 43.Dg5+ Kf7 44.Df5+ Kg7    

                  45.Sxc5

Glaurung 0.2.1 DC :
   9   00:02        263.812   154.910   +0,78   30.Txa8+ Lxa8 31.Ld3 Lb7 32.Sf5 Dc7 33.Lxb5 Kh8 34.Dg6 Le1    

                  35.De8+ Kh7
  10   00:04        560.392   153.911   +0,81   30.Txa8+ Lxa8 31.Ld3 g5 32.Sf5 Kh8 33.Sxh6 La5 34.h4 Dd5    

                  35.Dc8+ Ld8
  11   00:07      1.108.569   155.916   +0,79   30.Txa8+ Lxa8 31.Ld3 g5 32.Sf5 Kh8 33.Sxh6 Kg7 34.Sf5+ Kf8    

                  35.Dg3 Db6 36.Sd6
  12   00:21      3.200.404   155.404   +0,91   30.Txa8+ Lxa8 31.Ld3 Kf8 32.Sf5 g5 33.Sxh6 Kg7 34.Sf5+ Kf8    

                  35.h4 Dd7 36.Le4 Lxe4 37.Dxe4 gxh4 38.Sxh4
  13   00:43      6.676.505   155.491   +0,91   30.Txa8+ Lxa8 31.Ld3 Kf8 32.Sf5 g5 33.Sxh6 Kg7 34.Sf5+ Kf8    

                  35.h4 Dd7 36.Le4 Lxe4 37.Dxe4 gxh4 38.Sxh4
  13   01:16     11.602.308   152.756   +4,52   30.Txb7 Dxb7 31.De6+ Kh8 32.Le4 Da7 33.Sg6+ Kh7 34.Se7+ Kh8    

                  35.Df5 Da1+ 36.Lb1 Dxb1+ 37.Dxb1 Lc3 38.Sg6+ Kg8 39.Df5 Ta3
  14   02:00     18.167.801   151.654   +4,74   30.Txb7 Dxb7 31.De6+ Kh8 32.Le4 Da7 33.Sg6+ Kh7 34.Se7+ Kh8    

                  35.Df5 Da1+ 36.Lb1 Dxb1+ 37.Dxb1 La3 38.Dd3 Tb8 39.Sg6+ Kh7    

                  40.Se5+ Kg8
  15   03:50     34.633.091   150.967   +5,03   30.Txb7 Dxb7 31.De6+ Kh8 32.Le4 Da7 33.Sg6+ Kh7 34.Se5+ g6    

                  35.Sd7 Da1+ 36.Kh2 Kh8 37.Lxa8 Dxa8 38.Dxf6+ Kg8 39.Dxg6+ Kh8    

                  40.Dxh6+ Kg8 41.Dg6+ Kh8 42.Dh5+ Kg7 43.Dg5+ Kf7 44.Df5+ Kg7    

                  45.Sxc5


FEN: r2qk2r/1p1bbp2/1P2p3/p2pPp2/n2N1N1p/3PB3/5QPP/R4RK1 w kq - 0 24

Glaurung 0.2.1 BH :
   9   00:03        461.087   156.141   -1,58   24.Sb3 Lb4 25.Ld4 Tg8 26.Tac1 Tc8 27.Tc7 Txc7 28.bxc7 Dxc7    

               29.Dxh4
   9   00:06        859.385   149.458   -1,53   24.Sf3 Tg8 25.Tfc1 Tc8 26.Txc8 Dxc8 27.De1 Dc2 28.Dxa5
  10   00:08      1.282.797   150.087   -1,56   24.Sf3 Tc8 25.Ta2 Tg8 26.Taa1 Tc3 27.Ld4 Tb3 28.Sd2 Ta3
  10   00:09      1.449.827   149.899   -1,53   24.Sb3 Lb4 25.Ld4 Tg8 26.Tac1 Tc8 27.Sh5 Txc1 28.Sf6+ Kf8    

               29.Sxc1
  11   00:22      3.248.917   147.678   -1,47   24.Sb3 Ta6 25.Sc5 Lxc5 26.Lxc5 Sxc5 27.Dxc5 Dxb6 28.Dxb6 Txb6    

               29.Txa5
  12   00:51      7.369.272   144.762   -1,43   24.Sb3 Lb4 25.Ld4 Th7 26.Ta2 Tc8 27.Tfa1 Sc3 28.Txa5 Lxa5    

               29.Sxa5 Kf8 30.Sxb7
  12   01:18     11.157.090   142.668    0,00   24.Txa4 Lxa4 25.Sdxe6 fxe6 26.Sxe6 Dc8 27.Dxf5 Lc6 28.Df7+ Kd7    

               29.Sc5+ Kd8 30.Se6+ Kd7
  13   01:38     13.908.144   141.401   +0,57   24.Txa4 Lxa4 25.Sdxe6 fxe6 26.Sxe6 Dc8 27.Dxf5 Kd7 28.Tc1 Dxc1+

                  29.Lxc1 Kc6 30.Le3 The8
  14   02:06     17.849.394   141.416   +0,81   24.Txa4 Lxa4 25.Sdxe6 fxe6 26.Sxe6 Dc8 27.Dxf5 Kd7 28.Tc1 Dxc1+

                  29.Lxc1 Kc6 30.Dg4 Lb4 31.Dd4
  15   03:26     28.842.137   139.957   +1,17   24.Txa4 Lxa4 25.Sdxe6 fxe6 26.Sxe6 Dc8 27.Dxf5 Dc6 28.Tc1 Dxc1+

                  29.Lxc1 Tc8 30.Sc7+ Kd8 31.De6 Ld7 32.Dxd5

Glaurung 0.2.1 DC :
   9   00:03        461.087   148.259   -1,58   24.Sb3 Lb4 25.Ld4 Tg8 26.Tac1 Tc8 27.Tc7 Txc7 28.bxc7 Dxc7    

               29.Dxh4
   9   00:06        859.385   143.589   -1,53   24.Sf3 Tg8 25.Tfc1 Tc8 26.Txc8 Dxc8 27.De1 Dc2 28.Dxa5
  10   00:09      1.282.797   140.580   -1,56   24.Sf3 Tc8 25.Ta2 Tg8 26.Taa1 Tc3 27.Ld4 Tb3 28.Sd2 Ta3
  10   00:10      1.449.827   141.653   -1,53   24.Sb3 Lb4 25.Ld4 Tg8 26.Tac1 Tc8 27.Sh5 Txc1 28.Sf6+ Kf8    

               29.Sxc1
  11   00:23      3.248.917   142.709   -1,47   24.Sb3 Ta6 25.Sc5 Lxc5 26.Lxc5 Sxc5 27.Dxc5 Dxb6 28.Dxb6 Txb6    

               29.Txa5
  12   00:52      7.369.272   141.333   -1,43   24.Sb3 Lb4 25.Ld4 Th7 26.Ta2 Tc8 27.Tfa1 Sc3 28.Txa5 Lxa5    

               29.Sxa5 Kf8 30.Sxb7
  12   01:21     11.157.090   138.463    0,00   24.Txa4 Lxa4 25.Sdxe6 fxe6 26.Sxe6 Dc8 27.Dxf5 Lc6 28.Df7+ Kd7    

               29.Sc5+ Kd8 30.Se6+ Kd7
  13   01:41     13.908.144   137.470   +0,57   24.Txa4 Lxa4 25.Sdxe6 fxe6 26.Sxe6 Dc8 27.Dxf5 Kd7 28.Tc1 Dxc1+

                  29.Lxc1 Kc6 30.Le3 The8
  14   02:11     17.849.394   136.727   +0,81   24.Txa4 Lxa4 25.Sdxe6 fxe6 26.Sxe6 Dc8 27.Dxf5 Kd7 28.Tc1 Dxc1+

                  29.Lxc1 Kc6 30.Dg4 Lb4 31.Dd4
  15   03:32     28.842.137   136.128   +1,17   24.Txa4 Lxa4 25.Sdxe6 fxe6 26.Sxe6 Dc8 27.Dxf5 Dc6 28.Tc1 Dxc1+

                  29.Lxc1 Tc8 30.Sc7+ Kd8 31.De6 Ld7 32.Dxd5


FEN: r3rnk1/2qn1pb1/1p1p2pp/pPp5/2PPb3/P1Q2NP1/1B3PBP/R2R1NK1 w - - 1 19

Glaurung 0.2.1 BH :
   9   00:02        462.287   176.109   +0,14   19.Dc1 Se6 20.dxc5 Lxb2 21.Dxb2 dxc5 22.S1d2 Lb7 23.a4
  10   00:05        842.721   170.694   +0,07   19.Dc1 Se6 20.dxc5 dxc5 21.Lxg7 Kxg7 22.Se3 Lxf3 23.Lxf3 Tad8
  11   00:14      2.308.722   160.428   +0,14   19.Dc1 Se6 20.dxc5 dxc5 21.Lxg7 Kxg7 22.Se3 Kg8 23.Se5 Sxe5    

                  24.Lxe4
  12   00:26      4.101.178   158.978   +0,10   19.Dc1 Se6 20.dxc5 Sdxc5 21.Lxg7 Sxg7 22.Ta2 Db7 23.Dc3 Sf5    

                  24.Tad2 f6
  13   00:55      8.691.956   157.588   +0,14   19.Dc1 Se6 20.dxc5 Sdxc5 21.Lxg7 Sxg7 22.Ta2 Sf5 23.g4 Sd3    

                  24.Dc3 Lxf3 25.Lxf3
  14   01:56     17.991.413   155.265   +0,13   19.Dc1 Sf6 20.S1d2 Lb7 21.d5 S8d7 22.Dc2 Te2 23.Te1 Tae8    

                  24.Txe2 Txe2 25.Te1 Txe1+ 26.Sxe1
  15   04:55     45.890.068   155.263   +0,14   19.Dc1 Se6 20.dxc5 dxc5 21.Lxg7 Kxg7 22.Db2+ f6 23.Se3 Tad8    

                  24.Sd5 Lxd5 25.Txd5 Kh7 26.Tad1 Te7

Glaurung 0.2.1 DC :
   9   00:03        462.287   154.870   +0,14   19.Dc1 Se6 20.dxc5 Lxb2 21.Dxb2 dxc5 22.S1d2 Lb7 23.a4
  10   00:06        842.721   155.857   +0,07   19.Dc1 Se6 20.dxc5 dxc5 21.Lxg7 Kxg7 22.Se3 Lxf3 23.Lxf3 Tad8
  11   00:15      2.308.722   151.699   +0,14   19.Dc1 Se6 20.dxc5 dxc5 21.Lxg7 Kxg7 22.Se3 Kg8 23.Se5 Sxe5    

                  24.Lxe4
  12   00:27      4.101.178   152.778   +0,10   19.Dc1 Se6 20.dxc5 Sdxc5 21.Lxg7 Sxg7 22.Ta2 Db7 23.Dc3 Sf5    

                  24.Tad2 f6
  13   00:58      8.691.956   149.980   +0,14   19.Dc1 Se6 20.dxc5 Sdxc5 21.Lxg7 Sxg7 22.Ta2 Sf5 23.g4 Sd3    

                  24.Dc3 Lxf3 25.Lxf3
  14   02:00     17.991.413   149.499   +0,13   19.Dc1 Sf6 20.S1d2 Lb7 21.d5 S8d7 22.Dc2 Te2 23.Te1 Tae8    

                  24.Txe2 Txe2 25.Te1 Txe1+ 26.Sxe1
  15   05:05     45.890.068   150.351   +0,14   19.Dc1 Se6 20.dxc5 dxc5 21.Lxg7 Kxg7 22.Db2+ f6 23.Se3 Tad8    

                  24.Sd5 Lxd5 25.Txd5 Kh7 26.Tad1 Te7


FEN: 4k3/rp3ppb/1np1p2p/N7/1PP5/6P1/4PPBP/3R2K1 w - - 4 35

Glaurung 0.2.1 BH :
   9   00:00        135.466   377.342   +0,76   35.Sxb7 Sxc4 36.Tc1 Se5 37.Sa5 Kd7 38.Sxc6 Sxc6 39.Txc6
  10   00:01        290.889   273.906   +0,66   35.Sxb7 Sxc4 36.Tc1 Se5 37.Sa5 Kd7 38.Sxc6 Sxc6 39.Lxc6+ Kd6    

                  40.Kg2
  11   00:03        710.102   244.357   +0,72   35.Sxb7 Txb7 36.Lxc6+ Td7 37.Lxd7+ Sxd7 38.c5 Le4 39.Td4 Lc6    

                  40.Tg4 Kf8 41.Tc4
  12   00:06      1.526.399   233.715   +0,68   35.Sxb7 Sxc4 36.Tc1 Ke7 37.Txc4 Txb7 38.Lxc6 Tc7 39.b5 Kd6    

                  40.Td4+ Kc5 41.Td8
  13   00:14      3.167.707   227.794   +0,71   35.Sxb7 Sxc4 36.Tc1 Ke7 37.Txc4 Txb7 38.Lxc6 Ta7 39.Kg2 Ta2    

                  40.Kf3 Kd6 41.b5
  14   00:46     10.353.412   223.032   +0,75   35.Sxb7 Sxc4 36.Lxc6+ Ke7 37.f4 Se3 38.Td7+ Kf6 39.b5 Lg6    

                  40.Sc5 Ta2 41.b6 Txe2 42.b7 Tb2 43.Td8
  15   01:34     20.514.488   218.895   +0,69   35.Sxb7 Sxc4 36.Lxc6+ Ke7 37.f4 Lf5 38.b5 Sb6 39.Td8 Ta2 40.e4    

                  Lg4 41.e5 Ta1+ 42.Kg2 Ta2+ 43.Kf1 Le2+ 44.Kf2 Lc4+ 45.Kg1
  16   03:06     40.705.712   218.885   +0,71   35.Sxb7 Sxc4 36.Lxc6+ Ke7 37.f4 Lf5 38.Kf2 Ta2 39.Td4 Sb6    

                  40.Sd6 Lg4 41.Lf3 Lxf3 42.Kxf3 Kf6 43.b5

Glaurung 0.2.1 DC :
  9   00:01        135.466   228.057   +0,76   35.Sxb7 Sxc4 36.Tc1 Se5 37.Sa5 Kd7 38.Sxc6 Sxc6 39.Txc6
  10   00:01        290.889   216.435   +0,66   35.Sxb7 Sxc4 36.Tc1 Se5 37.Sa5 Kd7 38.Sxc6 Sxc6 39.Lxc6+ Kd6    

                  40.Kg2
  11   00:03        710.102   221.699   +0,72   35.Sxb7 Txb7 36.Lxc6+ Td7 37.Lxd7+ Sxd7 38.c5 Le4 39.Td4 Lc6    

                  40.Tg4 Kf8 41.Tc4
  12   00:07      1.526.399   220.005   +0,68   35.Sxb7 Sxc4 36.Tc1 Ke7 37.Txc4 Txb7 38.Lxc6 Tc7 39.b5 Kd6    

                  40.Td4+ Kc5 41.Td8
  13   00:15      3.167.707   218.221   +0,71   35.Sxb7 Sxc4 36.Tc1 Ke7 37.Txc4 Txb7 38.Lxc6 Ta7 39.Kg2 Ta2    

                  40.Kf3 Kd6 41.b5
  14   00:50     10.353.412   208.502   +0,75   35.Sxb7 Sxc4 36.Lxc6+ Ke7 37.f4 Se3 38.Td7+ Kf6 39.b5 Lg6    

                  40.Sc5 Ta2 41.b6 Txe2 42.b7 Tb2 43.Td8
  15   01:38     20.514.488   209.800   +0,69   35.Sxb7 Sxc4 36.Lxc6+ Ke7 37.f4 Lf5 38.b5 Sb6 39.Td8 Ta2 40.e4    

                  Lg4 41.e5 Ta1+ 42.Kg2 Ta2+ 43.Kf1 Le2+ 44.Kf2 Lc4+ 45.Kg1
  16   03:15     40.705.712   209.401   +0,71   35.Sxb7 Sxc4 36.Lxc6+ Ke7 37.f4 Lf5 38.Kf2 Ta2 39.Td4 Sb6    

                  40.Sd6 Lg4 41.Lf3 Lxf3 42.Kxf3 Kf6 43.b5



[Bryan Hofmann]

Try this compile in your bench test, I am sure you will find this to be the fastest.

http://www.freewebs.com/bkhofmann/glaurung.zip

Thanks, Bryan!

What compiler did you use?

If somebody can test Bryan's executable and confirm that it is indeed faster than Dann's, I will replace the Windows executable on my home page.

Tord

Microsoft Compiler, No code changes nor special inlining (Like I do with Crafty) was used. Let me know when you have an new version I would be more than happy to compile it for you.

Bryan