Perft(13), final fifth

Discussion of chess software programming and technical issues.

Moderator: Ras

User avatar
sje
Posts: 4675
Joined: Mon Mar 13, 2006 7:43 pm

Re: Perft(13), final fifth

Post by sje »

The operating system on the 3.4 GHz 16 GB RAM machine doing the perft(13) run has been changed from Ubuntu to Debian 6.0.5. The perft(13) run itself has been restarted and at present has completed 14 of the 20 draft 12 calculations and 340 of the 400 draft 11 calculations.

The only downside so far is that the vanilla Debian installation did not come up with a driver for the WiFi card. For the moment, I'm connecting a cat 5 cable every so often to move data. What I really need is a pointer to a site with known good Debian WiFi drivers along with instructions.
Top
User avatar
sje
Posts: 4675
Joined: Mon Mar 13, 2006 7:43 pm

After 55 weeks

Post by sje »

After 55 weeks:

The run has produced 777,905 draft 8 records, about 95% of the total needed. The average draft 8 record generation time is about 42.8 seconds.

There are now 341 draft 11 records, 85.25% of the 400 needed. The average draft 11 record generation time is about 27.1 hours.
http://dl.dropbox.com/u/31633927/Perft/Perft13/draft11

There are now 14 draft 12 records, 70% of the 20 needed. The average draft 12 record generation time is about 27.5 days.
http://dl.dropbox.com/u/31633927/Perft/Perft13/draft12

Recent output:

Code: Select all

rnbqkbnr/ppppp1p1/7p/5p2/3P4/7P/PPPQPPP1/RNB1KBNR b KQkq - 1 3 8 231684261618
rnbqkbnr/pppp1ppp/8/8/3P1p2/8/PPPNPPPP/R2QKBNR b KQkq - 1 3 8 458763829780
rnbqkbn1/ppppppp1/7r/7p/8/3PQ3/PPP1PPPP/RNB1KBNR b KQq - 3 3 8 894259412750
rnbqkbnr/pppp1ppp/B7/8/4p3/4PQ2/PPPP1PPP/RNB1K1NR b KQkq - 1 3 8 1035887488155
rnbqk1nr/pppp1ppp/8/4p3/3P3P/b7/PPP1PPPR/RNBQKBN1 b Qkq - 0 3 8 998661829313
rnbqkbnr/ppppp1p1/7p/5p2/3P4/3Q3P/PPP1PPP1/RNB1KBNR b KQkq - 1 3 8 453829702362
rnbqk1nr/pppp1ppp/B3p3/2b5/4P3/8/PPPPKPPP/RNBQ2NR b kq - 3 3 8 925843320800
rnbqkbnr/ppppp1p1/7p/5p2/3P4/7P/PPPKPPP1/RNBQ1BNR b kq - 1 3 8 106278113645
rnbqkbnr/ppppp1p1/7p/5p2/3P4/4P2P/PPP2PP1/RNBQKBNR b KQkq - 0 3 8 286202075632
rnbqkbnr/pppp1ppp/8/8/3P1p2/N7/PPP1PPPP/R2QKBNR b KQkq - 1 3 8 442653128413
Top
User avatar
sje
Posts: 4675
Joined: Mon Mar 13, 2006 7:43 pm

Re: Perft(13), final fifth

Post by sje »

After three hours of working on getting my Asus RT2860 WiFi chipset to talk to my router, I gave up. I'm sure it's possible to get it to work, but in the end I just moved the machine to within cabling distance of the router. Perhaps the next Debian version will have the needed WiFi support more easily available.

On the other hand, I did get Debian to handle WiFi correctly on two old notebooks, each with different WiFi chipsets.
Top
User avatar
sje
Posts: 4675
Joined: Mon Mar 13, 2006 7:43 pm

The French

Post by sje »

For perft(11) after 1 e4 e6: 28,917,008,057,410,835.

The run is now working on perft(11) for 1 e4 e5 which is expected to be 22,653,878,987,990,237.

Largest perft(11) results so far:

Code: Select all

rnbqkbnr/pppp1ppp/4p3/8/8/4P3/PPPP1PPP/RNBQKBNR w KQkq - 0 2 11 29547429295793347
rnbqkbnr/pppp1ppp/4p3/8/4P3/8/PPPP1PPP/RNBQKBNR w KQkq - 0 2 11 28917008057410835
rnbqkbnr/ppp1pppp/8/3p4/4P3/8/PPPP1PPP/RNBQKBNR w KQkq - 0 2 11 28808031383517106
rnbqkbnr/pppp1ppp/4p3/8/3P4/8/PPP1PPPP/RNBQKBNR w KQkq - 0 2 11 22835016719850486
rnbqkbnr/ppp1pppp/8/3p4/8/4P3/PPPP1PPP/RNBQKBNR w KQkq - 0 2 11 22334345588975690
rnbqkbnr/ppp1pppp/3p4/8/4P3/8/PPPP1PPP/RNBQKBNR w KQkq - 0 2 11 17344682849928475
rnbqkbnr/ppp1pppp/8/3p4/3P4/8/PPP1PPPP/RNBQKBNR w KQkq - 0 2 11 17022349302490272
rnbqkbnr/ppp1pppp/3p4/8/8/4P3/PPPP1PPP/RNBQKBNR w KQkq - 0 2 11 16408433376099590
rnbqkbnr/ppp1pppp/3p4/8/3P4/8/PPP1PPPP/RNBQKBNR w KQkq - 0 2 11 15948394992914745
rnbqkbnr/pppp1ppp/8/4p3/8/3P4/PPP1PPPP/RNBQKBNR w KQkq - 0 2 11 15651641608357144
Top
User avatar
sje
Posts: 4675
Joined: Mon Mar 13, 2006 7:43 pm

After 56 weeks

Post by sje »

After 56 weeks:

The run has produced 785,195 draft 8 records, about 95% of the total needed. The average draft 8 record generation time is about 43.1 seconds.

There are now 350 draft 11 records, 87.5% of the 400 needed. The average draft 11 record generation time is about 26.9 hours.
http://dl.dropbox.com/u/31633927/Perft/Perft13/draft11

There are now 14 draft 12 records, 70% of the 20 needed. The average draft 12 record generation time is about 28.0 days.
http://dl.dropbox.com/u/31633927/Perft/Perft13/draft12

Recent output:

Code: Select all

rnbqkb1r/pppppppp/8/6B1/4n3/3P4/PPPNPPPP/R2QKBNR b KQkq - 4 3 8 513437970639
r1bqkb1r/pppppppp/n4n2/8/6Q1/4P2P/PPPP1PP1/RNB1KBNR b KQkq - 2 3 8 749272271126
rnbqkbnr/pppp1p1p/6p1/4p2P/4P3/8/PPPP1PP1/RNBQKBNR b KQkq - 0 3 8 889451923346
rnbqkbnr/pppp1p1p/6p1/4p3/4P2P/8/PPPP1PP1/RNBQKBNR w KQkq - 0 3 9 26855174565098
rnbqkbnr/pppp1p1p/7p/4p3/3P4/2N5/PPP1PPPP/R2QKBNR b KQkq - 1 3 8 522604483843
rnb1kbnr/pppp1ppp/B7/4p3/6Qq/4P3/PPPP1PPP/RNB1K1NR b KQkq - 3 3 8 3454243157375
rnbqkbnr/pppp1p1p/6p1/4p2Q/2B1P3/8/PPPP1PPP/RNB1K1NR b KQkq - 1 3 8 1665450938742
r1bqkb1r/pppppppp/n4n2/7Q/8/4P2P/PPPP1PP1/RNB1KBNR b KQkq - 2 3 8 621688761967
r1bqkb1r/pppppppp/n4n2/8/8/4P2P/PPPPKPP1/RNBQ1BNR b kq - 2 3 8 120296200855
r1bqkb1r/pppppppp/n4n2/8/8/4P2P/PPPPBPP1/RNBQK1NR b KQkq - 2 3 8 227994593299
Top
User avatar
sje
Posts: 4675
Joined: Mon Mar 13, 2006 7:43 pm

Perft(12) for 1 h3

Post by sje »

Perft(12) for 1 h3 is 53,765,991,918,774,270 and this is the 15th draft 12 calculation made by the run so far; five more to go. There are now five hyperthreads (1 d3, 1 d4, 1 e3, 1 e4, 1 h4) running on four 3.4 GHz Intel i7 physical cores. The thread for 1 d3 should finish very soon and the thread for 1 h4 soon after that.
Top
User avatar
sje
Posts: 4675
Joined: Mon Mar 13, 2006 7:43 pm

Perft(12) for 1 d3

Post by sje »

Perft(12) for 1 d3 is 141,516,447,571,471,634 and this is the 16th draft 12 calculation made by the run so far; four more to go. There are now four hyperthreads (1 d4, 1 e3, 1 e4, 1 h4) running on four 3.4 GHz Intel i7 physical cores.
Top
User avatar
sje
Posts: 4675
Joined: Mon Mar 13, 2006 7:43 pm

Remaining work

Post by sje »

Remaining work in terms of per thread draft 11 results:

For 1 d4: 11 to go

For 1 e3: 11 to go

For 1 e4: 11 to go

For 1 h4: 8 to go

Total: 41 to go with 4 in progress

Time remaining first order estimate: 45 days
Top
User avatar
sje
Posts: 4675
Joined: Mon Mar 13, 2006 7:43 pm

After 57 weeks

Post by sje »

After 57 weeks:

There are now 360 draft 11 records, 90% of the 400 needed. The average draft 11 record generation time is about 26.6 hours.
http://dl.dropbox.com/u/31633927/Perft/Perft13/draft11

There are now 16 draft 12 records, 80% of the 20 needed. The average draft 12 record generation time is about 24.9 days.
http://dl.dropbox.com/u/31633927/Perft/Perft13/draft12

Recent output:

Code: Select all

rnbqkbnr/pppp1p1p/B5p1/4p3/2P1P3/8/PP1P1PPP/RNBQK1NR b KQkq - 0 3 8 627957077127
rnbqkb1r/pppppp1p/7n/6p1/7P/4P3/PPPPNPP1/RNBQKB1R b KQkq - 2 3 8 207032192546
rnbqkb1r/pppppp1p/7n/6pP/8/4P3/PPPP1PP1/RNBQKBNR b KQkq - 0 3 8 279691586899
rnbqk1nr/pppp1ppB/8/4p3/8/b3P3/PPPP1PPP/RNBQK1NR b KQkq - 0 3 8 1024638287196
rnbqk1nr/pppp1ppp/8/4p3/8/b2BP3/PPPP1PPP/RNBQK1NR w KQkq - 2 3 9 36008819426161
rnbqkb1r/pppppp1p/7n/6P1/8/4P3/PPPP1PP1/RNBQKBNR b KQkq - 0 3 8 438376383220
rnbqkb1r/pppppp1p/7n/6p1/7P/4P3/PPPP1PP1/RNBQKBNR w KQkq - 1 3 9 12660141313607
rnbqkbnr/pppppp1p/8/6p1/7P/4P3/PPPP1PP1/RNBQKBNR b KQkq - 0 2 10 366761025081859
rnbqkbnr/pppp1p1p/B5p1/4p3/4P3/3P4/PPP2PPP/RNBQK1NR b KQkq - 0 3 8 1038896409581
rnbqkbnr/pppppp1p/8/8/4P1pP/8/PPPPQPP1/RNB1KBNR b KQkq - 1 3 8 313876698517
Top
User avatar
sje
Posts: 4675
Joined: Mon Mar 13, 2006 7:43 pm

Re: Interlude 2

Post by sje »

The earlier calculation of perft(11) for 1 e4 e5 using 56 bits of hash was 22,653,878,987,990,237 and this is apparently INCORRECT.

The main run, using a 120 bit hash, gives the most likely by far correct number of 22,653,878,988,011,879.

Close, but close isn't good enough in this game.
Top

Return to “Computer Chess Club: Programming and Technical Discussions”