The New Sudoku Players' Forum

[gsf]

besides, its an SE bug that rates some automorphisms differently
surely the ones rated harder aren't

Then, by your own logic, a puzzle with a lower rating should preempt/eject any currently posted automorphism with a higher rating. Of course, after the higher rated puzzle is ejected, you may then have a problem if another puzzle already exists at the lower rating. This could introduce a whole new level of strategy into the game!!!

Bottom Line: It's not fair that the higher rated puzzle gets credit for a rating that it doesn't deserve!!!

I'd object to adding exceptions to 2 rules based on a bug
the real solution is to get SE fixed, but that may be problematic
for now the SE automorphism bug is like the luck of the draw in a card game
and luck isn't fair over small samples

[papy999]

::: comment :::

About rarity champagne wrote

the only change I could propose would be to increase the rarity bonus due to the fact that we have more and more puzzles in the data base

We could keep the actual bonuses but consider to evaluate the rarity of a rating only the puzzles that were posted in the last 50 or 80 games for instance...

What about using a version of SE with less bugs ? is it possible ?
The rarity rule could be though as a problem to change the rating engine... previous posted puzzles could have a different rating ... but in fact it is not a problem if the rating engine is near enough. We don't need to rerate the previous game... we can take the submited rating as they are.

I don't remember exactly but, IKsudoku has done some bug fixes ?

[m_b_metcalf]

for now the SE automorphism bug is like the luck of the draw in a card game

Exactly. And, as we don't play for money, even less important.

Shouldn't g.r.emlin have announced the start of a new game about 6 hours ago?

Regards,

Mike Metcalf

[gsf]

Shouldn't g.r.emlin have announced the start of a new game about 6 hours ago?

g.r.emlin's host machine is two hops away and the intermediate machine rebooted over the weekend
it may not be until friday until I get physical access reconnect it
time for some recoding ... and I'll check into adding the referee to crontab so this won't happen again

[Patrice]

For those who are not so good in English as myself :

To have the 'Luck of the draw' is to win something in a competition where the winner is chosen purely by chance.

Glenn,

Can you give us an indication 48h before the starting of the next game ?... just to organize the week-end
Thanks

[gsf]

Glenn,
Can you give us an indication 48h before the starting of the next game ?... just to organize the week-end

sure

[champagne]

::: more :::

Code: Select all

.....987.6...8....5..4.....3..8.24...6.....2...17.6..3.....5..1....3...9.539.....  ED=9.00/9.00/9.00       
.....987.9...8....6..5.....8..4.75...6.....4...36.5..8.....4..2....5...1.123.....  ED=8.50/8.50/8.30 


champagne

[champagne]

::: comment :::

I got these puzzles in a run I made after game closure to try to test and start taking benefit of the "SE clone" work in progress.
Out of more than 3 million non isomorphic puzzles, less than 9000 have been rated using serate, but I can have missed some interesting ones.
Unhappily, I can not make the test in live for the next game.

The final cache is very similar to what I got in other games, between 50% and 60% of the final entries. Here below, the final entries adjusted of the "more". Are marked 'x' the entries I have in cache.

champagne

Code: Select all

    1.2/1.2/1.2 joel64        x4.4/4.4/2.6 joel64         6.7/6.7/6.6 gsf            8.2/8.2/7.1 gsf
    1.5/1.5/1.5 gsf           x4.5/4.5/2.6 joel64        x6.8/6.8/2.6 joel64        x8.3/8.3/8.2 m_b_metcalf
   x1.7/1.7/1.7 gsf            4.6/4.6/2.6 joel64         6.9/6.9/2.6 ronk           8.4/8.4/8.3 papy999
   x2.0/2.0/2.0 joel64         4.7/4.7/2.6 joel64        x7.0/7.0/7.0 m_b_metcalf   x8.5/8.5/8.2 ronk
   x2.3/2.3/2.3 gsf            4.8/1.7/1.7 joel64        x7.1/7.1/7.1 m_b_metcalf    8.6/8.6/7.2 m_b_metcalf
   x2.5/2.5/2.5 joel64         5.0/1.7/1.7 joel64        x7.2/7.2/7.1 m_b_metcalf   x8.7/8.7/2.6 ronk
   x2.6/2.6/2.6 ronk           5.2/3.0/2.6 joel64        x7.3/7.3/7.1 ronk          x8.8/8.8/7.1 m_b_metcalf
   x2.8/2.8/2.6 joel64        x5.6/4.4/2.6 ronk           7.4/7.4/7.1 m_b_metcalf    8.9/8.9/8.2 m_b_metcalf
   x3.0/3.0/3.0 ronk          x5.7/4.0/2.6 ronk           7.5/7.5/7.1 papy999       x9.0/9.0/8.8 ronk   
   x3.2/3.2/2.6 joel64        x5.8/3.4/2.6 joel64         7.6/7.6/6.7 ronk           9.1/9.1/7.2 joel64
   x3.4/3.4/3.4 joel64        x5.9/3.4/2.6 ronk           7.7/7.7/7.1 m_b_metcalf   x9.2/9.2/8.9 joel64 
   x3.6/3.6/3.4 ronk          x6.0/3.2/2.6 joel64        x7.8/7.8/2.6 joel64        x9.3/9.3/2.6 joel64
    3.8/3.6/2.6 ronk          x6.2/6.2/2.6 ronk          x7.9/7.9/7.2 papy999        9.4/9.4/2.6 joel64
   x4.0/4.0/2.6 ronk          x6.5/3.4/2.6 joel64         8.0/8.0/2.6 ronk
    4.2/4.2/3.0 joel64        x6.6/6.6/6.6 m_b_metcalf    8.1/2.0/2.0 ronk


[g.r.emlin]

::: game 0137 start at 2011-03-22+02:00:00-0000 :::

Code: Select all

dealer status

candidate patterns (via private message to g.r.emlin)

   1 coloin            2 Mauricio          3 m_b_metcalf       6 JPF              87 g.r.emlin
   1 Pat               3 champagne         3 ronk             31 tarek

past patterns

   1 claudiarabia      1 ravel             3 ronk             17 coloin           27 JPF
   1 draco             1 wintder           9 gsf              19 TTHsieh
   1 Eioru             3 champagne        15 Pat              23 Mauricio
   1 pseudocoup        3 g.r.emlin        16 m_b_metcalf      26 tarek


[g.r.emlin]

::: game 0137 open 2011-03-25+19:29:39-0000 dealer champagne 24 clues full rotational (90 180 270 degrees) symmetry :::

Code: Select all

. 9 . . 8 . . 7 .
6 . . . . 9 . . 5
. . 4 . . . 3 . .
. 3 . . 7 . . . .
2 . . 9 . 4 . . 3
. . . . 5 . . 8 .
. . 5 . . . 7 . .
1 . . 8 . . . . 6
. 8 . . 6 . . 2 .  ED=2.3/1.2/1.2

[gsf]

::: comment :::
well we didn't even get a 48 second notice

the referee machine was hosed
the referee script is in crontab (scheduled to start when the machine restarts)
so it restarted when the machine rebooted
the referee doesn't know it was offline and so jumped in with I'm late for a very important date
another item for the todo list ...

[ronk]

Code: Select all

 . 5 . . 3 . . 4 .
 8 . . . . 5 . . 2
 . . 7 . . . 9 . .
 . 7 . . 4 . . . .
 9 . . 2 . 6 . . 8
 . . . . 7 . . 6 .
 . . 4 . . . 5 . .
 1 . . 3 . . . . 4
 . 3 . . 2 . . 1 .    ED=7.2/7.2/7.2

[gsf]

Code: Select all

. 7 . . 2 . . 5 .
8 . . . . 1 . . 9
. . 4 . . . 3 . .
. 3 . . 7 . . . .
2 . . 9 . 4 . . 3
. . . . 5 . . 1 .
. . 5 . . . 7 . .
1 . . 8 . . . . 6
. 8 . . 6 . . 2 .  ED=3.4/3.4/3.4

[ronk]

Code: Select all

 . 6 . . 5 . . 8 .
 2 . . . . 9 . . 7
 . . 3 . . . 9 . .
 . 9 . . 8 . . . .
 7 . . 3 . 1 . . 2
 . . . . 6 . . 4 .
 . . 5 . . . 4 . .
 1 . . 4 . . . . 3
 . 3 . . 2 . . 1 .    ED=8.5/8.5/8.5

[g.r.emlin]

::: adjustment 2011-03-25+20:37:00-0000 ronk premature-entry -50 :::