The New Sudoku Players' Forum

by **m_b_metcalf** » Wed Dec 19, 2007 2:03 pm

TTHsieh wrote:Are the following 13 puzzles pearls
wrote: wrote:000012300000400000403005600307000080500090007010000506004300209000006000008240000 # 7.1
000012300000400000503006700304000080600050007010000609009300205000007000006240000 # 7.1
000012300000400000506007200301000040800090003090000602003600105000008000007150000 # 8.4

I think the three left in your quote are not Pearls.

Regards,

Mike Metcalf

by **gsf** » Wed Dec 19, 2007 3:10 pm

tarek wrote:
gsf wrote:
gsf wrote:
Code: Select all
ER=4.69

I've never seen a 4.699999999999999
can someone corroborate with their copy of SE

I just checked....
The cmd batch solver reports 4.699999999999999(Unique Loop type 3), the GUI reports 4.7 (Unique Loop type 3)

The GUI also reports a 6.7 for the 6.699999999999999

aha
so the .699999999999999s seem to be an artifact of the implementation
should we then always do the roundup to the next tenth?

by **JPF** » Wed Dec 19, 2007 3:14 pm

m_b_metcalf wrote:I think the three left in your quote are not Pearls.

IMO, there are weak pearls.

gsf,
Could you give the command lines of your solver to filter weak and strong pearls ?

btw, I personnaly think that SCORE_PEARL_STRONG=100 is too high.

JPF

by **gsf** » Wed Dec 19, 2007 3:22 pm

JPF wrote:
m_b_metcalf wrote:I think the three left in your quote are not Pearls.
IMO, there are weak pearls.

right, they require naked tuples in ss constraint order

gsf,
Could you give the command lines of your solver to filter weak and strong pearls ?

solver version 2007-12-18

Code: Select all: sudoku -S -qss+G -e '(%#Pq)>2' -f'%#0v # %-2#pq %2#Pq'

-S for stepwise constraint application
-qss+G is SS (simple sudoku) constraint order (or close to it) plus guessing
%#Pq is the pearl ordinal -- the highest SS constraint ordinal up to the first placement
%#pq is the shorthand name of the highest SS constraint applied up to the first placement
the -e (%...) allows any -f format numeric quantity to be a filter expression operand

btw, I personnaly think that SCORE_PEARL_STRONG=100 is too high.

ok, but I need a suggested value ...

by **TTHsieh** » Wed Dec 19, 2007 3:56 pm

gsf wrote:... I need a suggested value ...

IMO,
SCORE_PEARL_WEAK = 25
SCORE_PEARL_STRONG = 50

TTHsieh

by **m_b_metcalf** » Wed Dec 19, 2007 4:04 pm

gsf wrote:
JPF wrote:
m_b_metcalf wrote:I think the three left in your quote are not Pearls.
IMO, there are weak pearls.

right, they require naked tuples in ss constraint order

The first two require a naked triple (3.6) to solve the first cell with SE, the third a hidden pair (3.4).

btw, I personnaly think that SCORE_PEARL_STRONG=100 is too high.

ok, but I need a suggested value ...

I think that, until we can agree on a methodology, it should be a token value, like 10.

As far as I understand gemstones, they are classified and valued using a number of criteria, not only weight. In an ideal world we could classify our pearls by the SE ratings for first step/first cell/whole puzzle but, since only the last of these three values can be obtained in batch mode (AFAIK), I admit that this isn't a very practical suggestion.

[Edit: I do think we should bear in mind that, the more the rules become involved and even esoteric, the higher the barrier to new players rises.]

Regards,

Mike Metcalf

by **TTHsieh** » Wed Dec 19, 2007 5:11 pm

m_b_metcalf wrote:... the more the rules become involved and even esoteric, the higher the barrier rises to new players.

Right !

But, if these pearl's rules only are bonus rules, then it does not constitute any barrier.

TTHsieh

by **wintder** » Wed Dec 19, 2007 11:35 pm

gsf wrote:
tarek wrote:
gsf wrote:
gsf wrote:
Code: Select all
ER=4.69

I've never seen a 4.699999999999999
can someone corroborate with their copy of SE

I just checked....
The cmd batch solver reports 4.699999999999999(Unique Loop type 3), the GUI reports 4.7 (Unique Loop type 3)

The GUI also reports a 6.7 for the 6.699999999999999

aha
so the .699999999999999s seem to be an artifact of the implementation
should we then always do the roundup to the next tenth?

The 6.6999... lists a different hardest step than the 6.7 does.
It should not be rounded up.

The 4.6999.. indicates the identical hardest step to the 4.7, it should not be considered distinct, IMHO.

by **gsf** » Thu Dec 20, 2007 3:43 am

::: pearl :::

Code: Select all: 000028500000100000908005700403000070700090008010000603006200904000006000004830000 4.4 T2/3 000017500000600000902004600507000020300090005020000803005800907000001000008430000 8.3 Y/11 000012500000400000103008700806000040200070009030000105001500408000001000008390000 8.9 G/99 000028500000700000902005100403000070200010008060000405006200304000006000008390000 8.5 T4/6 000028500000700000902005700403000070200090008050000604006200803000006000004150000 7.3 Y/11 000085300000900000904006700807000010500030007040000205002700106000004000009160000 3.0 T2/3 000092300000400000406007100301000020800060007090000508008100706000008000004750000 2.0 T2/3 000012700000900000104007900607000080500060007030000604002500109000004000008320000 7.1 Y/11 000061700000500000408007200306000050500080001070000902009400608000002000003670000 7.2 B2/4 000021800000600000301008700204000030100050006070000908009700604000009000002410000 9.0 G/99

by **gsf** » Thu Dec 20, 2007 3:45 am

TTHsieh wrote:
m_b_metcalf wrote:... the more the rules become involved and even esoteric, the higher the barrier rises to new players.

Right !
But, if these pearl's rules only are bonus rules, then it does not constitute any barrier.

the bonus only counts for submitted puzzles
this adds a bit of strategy
you have a non-pearl 9.1 in hand
should you submit it, or wait to see if you can coax a pearl from it
wait too long and another player may grab the 9.1 slot
and you are left with nothing

by **gsf** » Thu Dec 20, 2007 3:50 am

wintder wrote:The 6.6999... lists a different hardest step than the 6.7 does.
It should not be rounded up.

The 4.6999.. indicates the identical hardest step to the 4.7, it should not be considered distinct, IMHO.

thanks for checking this
but I'm not going to keep an exception list for SE's bugs
it will be either one way or the other
right now x.6999.. will be truncated to x.69 which will be distinct from x.6 and x.7,
whether or not SE calls them distinct by other means

by **gsf** » Thu Dec 20, 2007 3:54 am

::: comment :::

gsf wrote:::: pearl :::
Code: Select all
000028500000100000908005700403000070700090008010000603006200904000006000004830000 4.4 T2/3 000017500000600000902004600507000020300090005020000803005800907000001000008430000 8.3 Y/11 000012500000400000103008700806000040200070009030000105001500408000001000008390000 8.9 G/99 ...

the third column is the rightmost SS technique required up to and including the first placement
the number is the postition of the technique in the SS technique order, from the left
G means "guess" -- no SS technique will result in a placement -- and has a position value 99

by **g.r.emlin** » Thu Dec 20, 2007 4:30 am

::: game 0003 idle -- with no further entries the game will close at 2007-12-20+06:00:00-0000 :::

Code: Select all: game 0003 2007-12-15+18:17:00-0000 duration 3d11h dealer JPF index 150 (open) 6 players 2 resigned 1.2 1.5 1.7 2.0 2.3 2.5 2.6 2.8 3.0 3.2 3.4 3.6 3.8 4.0 4.2 4.4 4.5 4.6 4.69 4.7 5.0 5.2 5.6 5.7 5.8 5.9 6.2 6.5 6.6 6.69 6.7 6.8 6.9 7.0 7.1 7.2 7.3 7.4 7.5 7.6 7.7 7.8 7.9 8.0 8.1 8.2 8.3 8.4 8.5 8.6 8.7 8.8 8.9 9.0 9.1 9.2 9.3 9.4 9.5 submitter entries score response notes ----------- ------- ----- -------- ----- wintder 0 -50 0 dup-rating tarek 6 132 7m19s m_b_metcalf 7 214 9m39s resigned-entry,resigned gsf 12 250 21m39s last-entry,bad-pattern,consecutive-entry JPF 9 455 18m59s highest-rating,dealer TTHsieh 25 665 4m45s resigned ----------- ------- 6 59

by **g.r.emlin** » Thu Dec 20, 2007 7:15 am

::: game 0003 closed (idle for ) -- preliminary results :::

Code: Select all: game 0003 2007-12-15+18:17:00-0000 duration 3d11h dealer JPF index 150 (idle) 6 players 2 resigned 1.2 1.5 1.7 2.0 2.3 2.5 2.6 2.8 3.0 3.2 3.4 3.6 3.8 4.0 4.2 4.4 4.5 4.6 4.69 4.7 5.0 5.2 5.6 5.7 5.8 5.9 6.2 6.5 6.6 6.69 6.7 6.8 6.9 7.0 7.1 7.2 7.3 7.4 7.5 7.6 7.7 7.8 7.9 8.0 8.1 8.2 8.3 8.4 8.5 8.6 8.7 8.8 8.9 9.0 9.1 9.2 9.3 9.4 9.5 submitter entries score response notes ----------- ------- ----- -------- ----- wintder 0 -50 0 dup-rating tarek 6 132 7m19s m_b_metcalf 7 214 9m39s resigned-entry,resigned gsf 12 250 21m39s last-entry,bad-pattern,consecutive-entry JPF 9 455 18m59s highest-rating,dealer TTHsieh 25 665 4m45s resigned ----------- ------- 6 59

by **TTHsieh** » Thu Dec 20, 2007 8:32 am

::: comment :::

g.r.emlin

Code: Select all: game 0003 idle 2007-12-15+18:17:00-0000 duration 3d11h dealer JPF index 150 6 players 2 resigned 1.2 1.5 1.7 2.0 2.3 2.5 2.6 2.8 3.0 3.2 3.4 3.6 3.8 4.0 4.2 4.4 4.5 4.6 4.69 4.7 5.0 5.2 5.6 5.7 5.8 5.9 6.2 6.5 6.6 6.69 6.7 6.8 6.9 7.0 7.1 7.2 7.3 7.4 7.5 7.6 7.7 7.8 7.9 8.0 8.1 8.2 8.3 8.4 8.5 8.6 8.7 8.8 8.9 9.0 9.1 9.2 9.3 9.4 9.5 submitter entries score response notes ----------- ------- ----- -------- ----------------- wintder 0 -50 0 dup-rating tarek 6 132 7m19s m_b_metcalf 7 214 9m39s resigned-entry resigned gsf 12 250 21m39s last-entry bad-pattern consecutive-entry JPF 9 455 18m59s highest-rating dealer TTHsieh 25 665 4m45s resigned ----------- ------- 6 59

TTHsieh

The New Sudoku Players' Forum

Patterns Game 1.5