Perft(7) challenge position #3

Discussion of chess software programming and technical issues.

Moderator: Ras

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

Perft(7) challenge position #3

Post by sje »

Perft(7) challenge position #3

From line 94,715 of Perft(14) work unit #415:
[d]
What is the value of perft(7) for the above?

One program says: 4,517,444,758
A second says: 4,517,445,538

Ooooh, so close!
elcabesa
Posts: 855
Joined: Sun May 23, 2010 1:32 pm

Re: Perft(7) challenge position #3

Post by elcabesa »

4,517,445,538
sorry for ther delay... but I had to count them one by one :P
User avatar
sje
Posts: 4675
Joined: Mon Mar 13, 2006 7:43 pm

64 bit false positive rate circa 2.0*10^-6

Post by sje »

64 bit false positive rate circa 2.0*10^-6

Verified work units (15):
400-408 411-412 414 416 419 422

Work units with at least one false positive (3):
409 413 415

Each work unit has 100,000 unique positions.
User avatar
Ajedrecista
Posts: 2102
Joined: Wed Jul 13, 2011 9:04 pm
Location: Madrid, Spain.

Re: Perft(7) challenge position #3.

Post by Ajedrecista »

Hello Steven:

Divided perft(7) by JetChess:

Code: Select all

rnbq1bnr/ppppk1pp/5p2/4p3/1Q6/2PP4/PP2PPPP/RNB1KBNR b KQ - 2 4

  1   c7-c5   707733571
  2   d7-d6  1046824320
  3  ke7-f7  1038401617
  4  ke7-e6   870696961
  5  ke7-e8   853789069

Total:       4517445538
Regards from Spain.

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

Re: 64 bit false positive rate circa 2.0*10^-6

Post by sje »

Err, actually that error rate figure is closer to 1.7*10^-6

In a month or so, there will be over five million positions covered which will give a more confident answer. In about a year, all forty million 64 bit results will be tested against the 128 bit answers.
zullil
Posts: 6442
Joined: Tue Jan 09, 2007 12:31 am
Location: PA USA
Full name: Louis Zulli

Re: Perft(7) challenge position #3

Post by zullil »

sje wrote:Perft(7) challenge position #3

From line 94,715 of Perft(14) work unit #415:
[d]
What is the value of perft(7) for the above?

One program says: 4,517,444,758
A second says: 4,517,445,538

Ooooh, so close!

Code: Select all

FEN string = rnbq1bnr/ppppk1pp/5p2/4p3/1Q6/2PP4/PP2PPPP/RNB1KBNR b KQ -  
Depth = 7
Leaf nodes = 4517445538
Time taken = 69596 ms
StuartRiffle
Posts: 25
Joined: Tue Apr 05, 2016 9:34 pm
Location: Canada

Re: Perft(7) challenge position #3

Post by StuartRiffle »

Pigeon agrees (FWIW):

Code: Select all

divide 7
info string divide 7 e7e8  853789069
info string divide 7 c7c5  707733571
info string divide 7 d7d6  1046824320
info string divide 7 e7f7  1038401617
info string divide 7 e7e6  870696961
info string divide 7 total 4517445538
-Stuart
(Pigeon)
pkumar
Posts: 100
Joined: Tue Oct 15, 2013 5:45 pm

Re: Perft(7) challenge position #3

Post by pkumar »

My engine says
D7 4517445538
User avatar
stegemma
Posts: 859
Joined: Mon Aug 10, 2009 10:05 pm
Location: Italy
Full name: Stefano Gemma

Re: Perft(7) challenge position #3

Post by stegemma »

Satana confirms 4,517,445,538:

Code: Select all

setboard rnbq1bnr/ppppk1pp/5p2/4p3/1Q6/2PP4/PP2PPPP/RNB1KBNR b KQ - 2 4
perft 1
Nodes: 5, Time: 21 ms, Nodes/s: 227
perft 2
Nodes: 200, Time: 21 ms, Nodes/s: 9090
perft 3
Nodes: 4866, Time: 21 ms, Nodes/s: 221181
perft 4
Nodes: 183643, Time: 25 ms, Nodes/s: 7063192
perft 5
Nodes: 4677872, Time: 136 ms, Nodes/s: 34145051
perft 6
Nodes: 170451708, Time: 3525 ms, Nodes/s: 48341380
perft 7
Nodes: 4517445538, Time: 116311 ms, Nodes/s: 38839032
Author of Drago, Raffaela, Freccia, Satana, Sabrina.
http://www.linformatica.com
User avatar
yurikvelo
Posts: 710
Joined: Sat Dec 06, 2014 1:53 pm

Re: Perft(7) challenge position #3

Post by yurikvelo »

What is highest number of legal moves in a position, ever observed?

In perft<14, not in puzzles

https://www.chess.com/forum/view/fun-wi ... ible-moves