If I run perft(6) in the position after 1. e4 e5 what should be the correct result ?
[Skipper crashes because of a short castling move after:
1 g4 e5 2 Nf3 Nf6 3 Bg2 Ng4 4 0-0
]
Or is there a website where I can run perft myself.
perft(6) after e4 e5
Moderators: hgm, Rebel, chrisw
-
- Posts: 6442
- Joined: Tue Jan 09, 2007 12:31 am
- Location: PA USA
- Full name: Louis Zulli
Re: perft(6) after e4 e5
My move generator saysHenk wrote:If I run perft(6) in the position after 1. e4 e5 what should be the correct result ?
Leaf nodes = 673070116
Time taken = 15043 ms
-
- Posts: 91
- Joined: Wed Mar 26, 2014 4:29 pm
- Location: Buettelborn/Hessen/Germany
Re: perft(6) after e4 e5
Code: Select all
XFEN 00: rnbqkbnr/pppp1ppp/8/4p3/4P3/8/PPPP1PPP/RNBQKBNR w KQkq - 0 2
+-*--b--c--d--*--f--g--*-+
8 |[r][n][b][q][k][b][n][r]| Compiled on Feb 26 2015
7 |[p][p][p][p]:::[p][p][p]| MS Vis.Studio C/C++ 64-Bit Vers. 18.0
6 | ::: ::: ::: :::|
5 |::: ::: [p] ::: | Break, when time >= 5.0 sec
4 | ::: :::<P>::: :::|
3 |::: ::: ::: ::: | (run with 4-fold TT hashing)
2 |<P><P><P><P> <P><P><P>|
1 |<R><N><B><Q><K><B><N><R>|
(w)+-*--b--c--d--*--f--g--*-+
Ply Moves all [x] [ep] all [+] [++] Prom Cstl Time
1: 29 0 0 0 0 0 0 0.00
2: 835 8 0 4 0 0 0 0.00
3: 24825 576 0 284 0 0 0 0.00
4: 728887 23907 2 10306 1 0 0 0.02
5: 22273312 998407 420 430046 52 0 19682 0.21
6: 673070116 36172846 14340 14028981 3418 0 598518 2.36
7: 21181685531 1359626731 924634 528409498 134777 0 36210043 32.84
-
- Posts: 7218
- Joined: Mon May 27, 2013 10:31 am
Re: perft(6) after e4 e5
zullil wrote:My move generator saysHenk wrote:If I run perft(6) in the position after 1. e4 e5 what should be the correct result ?
Leaf nodes = 673070116
Time taken = 15043 ms
ok. Thank you. Only 15 seconds ? Perft of skipper uses 13 minutes. But I have to rerun it and compare for I did not add an assertion in the test.
-
- Posts: 7218
- Joined: Mon May 27, 2013 10:31 am
Re: perft(6) after e4 e5
Enough castling moves so it might be a good test.SMIRF wrote:Code: Select all
XFEN 00: rnbqkbnr/pppp1ppp/8/4p3/4P3/8/PPPP1PPP/RNBQKBNR w KQkq - 0 2 +-*--b--c--d--*--f--g--*-+ 8 |[r][n][b][q][k][b][n][r]| Compiled on Feb 26 2015 7 |[p][p][p][p]:::[p][p][p]| MS Vis.Studio C/C++ 64-Bit Vers. 18.0 6 | ::: ::: ::: :::| 5 |::: ::: [p] ::: | Break, when time >= 5.0 sec 4 | ::: :::<P>::: :::| 3 |::: ::: ::: ::: | (run with 4-fold TT hashing) 2 |<P><P><P><P> <P><P><P>| 1 |<R><N><B><Q><K><B><N><R>| (w)+-*--b--c--d--*--f--g--*-+ Ply Moves all [x] [ep] all [+] [++] Prom Cstl Time 1: 29 0 0 0 0 0 0 0.00 2: 835 8 0 4 0 0 0 0.00 3: 24825 576 0 284 0 0 0 0.00 4: 728887 23907 2 10306 1 0 0 0.02 5: 22273312 998407 420 430046 52 0 19682 0.21 6: 673070116 36172846 14340 14028981 3418 0 598518 2.36 7: 21181685531 1359626731 924634 528409498 134777 0 36210043 32.84
-
- Posts: 6442
- Joined: Tue Jan 09, 2007 12:31 am
- Location: PA USA
- Full name: Louis Zulli
Re: perft(6) after e4 e5
Is likely irrelevant, but Reinhard's FEN is incorrect. Should beHenk wrote:Enough castling moves so it might be a good test.SMIRF wrote:Code: Select all
XFEN 00: rnbqkbnr/pppp1ppp/8/4p3/4P3/8/PPPP1PPP/RNBQKBNR w KQkq - 0 2 +-*--b--c--d--*--f--g--*-+ 8 |[r][n][b][q][k][b][n][r]| Compiled on Feb 26 2015 7 |[p][p][p][p]:::[p][p][p]| MS Vis.Studio C/C++ 64-Bit Vers. 18.0 6 | ::: ::: ::: :::| 5 |::: ::: [p] ::: | Break, when time >= 5.0 sec 4 | ::: :::<P>::: :::| 3 |::: ::: ::: ::: | (run with 4-fold TT hashing) 2 |<P><P><P><P> <P><P><P>| 1 |<R><N><B><Q><K><B><N><R>| (w)+-*--b--c--d--*--f--g--*-+ Ply Moves all [x] [ep] all [+] [++] Prom Cstl Time 1: 29 0 0 0 0 0 0 0.00 2: 835 8 0 4 0 0 0 0.00 3: 24825 576 0 284 0 0 0 0.00 4: 728887 23907 2 10306 1 0 0 0.02 5: 22273312 998407 420 430046 52 0 19682 0.21 6: 673070116 36172846 14340 14028981 3418 0 598518 2.36 7: 21181685531 1359626731 924634 528409498 134777 0 36210043 32.84
rnbqkbnr/pppp1ppp/8/4p3/4P3/8/PPPP1PPP/RNBQKBNR w KQkq e6 0 2
-
- Posts: 91
- Joined: Wed Mar 26, 2014 4:29 pm
- Location: Buettelborn/Hessen/Germany
Re: perft(6) after e4 e5
As you will have noticed: it is compatible ! XFEN. There only then an e.p. coordinate is set, when an opposite pawn is standing at the side of the double stepped pawn.zullil wrote:... Is likely irrelevant, but Reinhard's FEN is incorrect. Should be
rnbqkbnr/pppp1ppp/8/4p3/4P3/8/PPPP1PPP/RNBQKBNR w KQkq e6 0 2
-
- Posts: 6442
- Joined: Tue Jan 09, 2007 12:31 am
- Location: PA USA
- Full name: Louis Zulli
Re: perft(6) after e4 e5
No, I failed to notice that. Sorry.SMIRF wrote:As you will have noticed: it is compatible ! XFEN. There only then an e.p. coordinate is set, when an opposite pawn is standing at the side of the double stepped pawn.zullil wrote:... Is likely irrelevant, but Reinhard's FEN is incorrect. Should be
rnbqkbnr/pppp1ppp/8/4p3/4P3/8/PPPP1PPP/RNBQKBNR w KQkq e6 0 2
http://en.wikipedia.org/wiki/X-FEN#Encoding_en-passant
-
- Posts: 358
- Joined: Thu Jan 22, 2015 3:21 pm
- Location: Zurich, Switzerland
- Full name: Jonathan Rosenthal
Re: perft(6) after e4 e5
If you suspect you have a bug in move generation, I can recommend downloading the latest stockfish version and starting it in a terminal (not in a GUI). Then manually enter the position via UCI protocol and call it's perft function, which is extremely fast. The reason I suggest this is that SF actually offers the functionality of a divide function which means it gives a breakdown of perft (n-1) for each legal move when you call perft (n).
This is what I get for perft 6 after e2e4 e7e5:
This is what I get for perft 6 after e2e4 e7e5:
Code: Select all
position startpos moves e2e4 e7e5
perft 6
Position: 1/1
a2a3: 21100966
b2b3: 23596663
c2c3: 23416950
d2d3: 25540907
f2f3: 15742641
g2g3: 23364786
h2h3: 20760629
a2a4: 24042446
b2b4: 22001165
c2c4: 20896102
d2d4: 38353660
f2f4: 24044398
g2g4: 20005227
h2h4: 24059746
b1a3: 22001586
b1c3: 24912209
g1e2: 14479984
g1f3: 20419850
g1h3: 21454237
f1e2: 19399143
f1d3: 20878923
f1c4: 26828547
f1b5: 20591524
f1a6: 22471973
d1e2: 22164322
d1f3: 34059395
d1g4: 33583821
d1h5: 27807644
e1e2: 15090672
===========================
Total time (ms) : 4158
Nodes searched : 673070116
Nodes/second : 161873524
-
- Posts: 7218
- Joined: Mon May 27, 2013 10:31 am
Re: perft(6) after e4 e5
Perft 5 and 6 both did not fail. So I have to look further. Skipper crashed when taking back rook on f1 and it said that the square is not occupied.