The New Sudoku Players' Forum

[mario133]

Code: Select all

....8....
..39.56..
.7.....1.
7.......5
.1.....9.
..65.32..
.6..4..7.
9...1...3
.........


....5....
..71.25..
.8.....3.
5.......6
.3.....7.
..19.32..
.4..6..9.
6...8...2
.........


....9....
..38.24..
.1.....7.
8.......9
.6.....5.
..45.82..
.9..3..6.
6...2...7
.........


....1....
..84.97..
.3.....6.
6.......1
.7.....9.
..25.74..
.1..9..4.
5...3...8
.........





I wanted to play the pattern game thing with the puzzle above but have no clue how to get the ranking format to look like this ED=x.x/x.x/x.x

Can someone give me easy to understand step by step processes to get it to look like that please and thank you:)

Edit: Figured I would put all my findings here till I figure out how to put the ranking system in that format.

[m_b_metcalf]

I wanted to play the pattern game thing with the puzzle above but have no clue how to get the ranking format to look like this ED=x.x/x.x/x.x

Can someone give me easy to understand step by step processes to get it to look like that please and thank you:)

Welcome. Go to Page 1 of the Patterns Game thread for full instructions.

Regards,

Mike Metcalf

[mario133]

I still do not get it I read it over many times and tried many things to try to figure it out but not resulted to a solution. Can you type me up somthing or have any easier explantionary guides on how to figure it out in a format that I can easly understand?

[gsf]

I still do not get it I read it over many times and tried many things to try to figure it out but not resulted to a solution. Can you type me up somthing or have any easier explantionary guides on how to figure it out in a format that I can easly understand?

post what you attempted and we may be able to help

[mario133]

Well I first went through the files in the sudoku explainer but I found no serate this was the .jar one I used 7zip to check through the arichives. I then atempted to check the sudoku explainer it self by seeing all the options it has but nothing leading to serate. After that I pretty much just rechecked through the same stuff over again. Can you tell me were to find this serate thing thats my main goal as of now I may be able to figure the rest out from there but who knows.

Edit:Never mind I figured out how to do it manualy with out the program doing it for me but if you guys could still tell me were the serate is it would be well aprciated.

Edit2:I just posted one of my puzzles and I noticed the system gave me a -50 and my sudoku was not added to the list why is that?

[gsf]

Edit:Never mind I figured out how to do it manualy with out the program doing it for me but if you guys could still tell me were the serate is it would be well aprciated.

the first page of the patterns game has links to two sudoku explainer jars
the first one is the original
the second one has the modifications for the patterns game

a java jar file supports multiple entry points
the modified jar file has an entry point for serate
at the bottom of the serate(1) link is an example java invocation for the serate entry point

getting java running the first time can be an adventure, and the
patterns game instructions will be no help there

Edit2:I just posted one of my puzzles and I noticed the system gave me a -50 and my sudoku was not added to the list why is that?

now the patterns game rules come into play
one of the rules is that successive entries be better than their predecessors
the ratings are ER/EP/ED
a new entry can have the same ER as a previous entry, but
either its EP must be greater than the previous entry, or
its EP must be the same and the ED must be greater

you can keep track of the game entries by monitoring the last post in
the results thread
the last post is edited in place and contains the 81 char grids for each entry
in that list you'll see an earlier entry 2.6/2.6/2.6 -- a diamond that can't be bettered (trumped)

it takes a few rounds to get in the flow

[Pat]

I figured out how to do it manualy with out the program doing it for me
but if you guys could still tell me were the serate is it would be well aprciated.

this is the modified vesion
of Sudoku Explainer 1.2.1 --

http://www.research.att.com/~gsf/sudoku/SudokuExplainer.jar
( be sure to save it as a JAR, not a ZIP )

and this is how you use it ( all on one long line ) --

java -Xrs -Xmx500m -cp SudokuExplainer.jar diuf.sudoku.test.serate --format=%r/%p/%d --input=puzzles.txt --output=ratings.txt

-- giving your puzzles in puzzles.txt, one line for each puzzle

[mario133]

I still do not get this I just posted another puzzle that was max 1.5/1.5/1.5 for its grade but it was still not added. Are ties not excepted eather what I read above I though they were

[m_b_metcalf]

I still do not get this I just posted another puzzle that was max 1.5/1.5/1.5 for its grade but it was still not added. Are ties not excepted eather what I read above I though they were

No, see gsf's post on the Patterns Games thread. To make the point again:

A posted puzzle must have an ER rating that is different from all others (in the current game, 9.5/1.2/1.2 would qualify), or it must have a higher EP/ED rating than any other puzzle already submitted (in the curent game 1.7/1.7/1.7 would qualify). It's not much different from many card games. A tie is permitted only within a short time window of about 2 minutes.

Wait for the next game to start and follow its development, joining in wherever possible.

HTH

Mike Metcalf

[mario133]

Thanks I think I understand how this works now I'll wait for the next game to start doing. Also when will the next game be up?

[gsf]

Thanks I think I understand how this works now I'll wait for the next game to start doing. Also when will the next game be up?

the referee is a prgram running on a machine that crashed
friday and won't be completely recovered until monday
the current game will resume but will probably be over by wednesday
the next game will start a day or so after that

[champagne]

This seem to be an unkhown bug in serate??

during game 112, at a moment gremlin was down, I submitted that puzzle

Code: Select all

2..3....5.....473.....9..1......1.9.....7.4...6.8....34.5......6..5......79.....8 ed=9.7 9.7 9.3

when gremlin came back, he took the puzzles properly, but found a rating 9.8 9.8 9.3 and applied a penalty for "SE mismatch".

after discussions with glenn, we came to that key post

:::comment:::

my version of se (1.2.1.3) rates this 9.8/9.8/9.3
can we get corroboration from the other players?

I am using version 1.2.1 (help file)
here the three puzzles I have in cache (inclusding that one) rated 9.7 9.7 9.3

Code: Select all

6..4....2.....573.....9..1......1.9.....7.5...4.8....34.5......8..2......63.....8 9.7 9.7 9.3
2..3....5.....473.....9..1......1.9.....7.4...6.8....34.5......6..5......79.....8 9.7 9.7 9.3
2..3....6.....473.....9..1......1.4.....7.9...6.2....55.9......8..5......32.....8 9.7 9.7 9.3

I get 9.8/9.8/9.3 for the middle one

This happenned late wednesday, so I first thought I had a release problem and answered in that way to glenn.

I am afraid we are using the same release and we found here a "volatility" problem in SE Rating.

A) Next day, I made the same run as glenn and I got the same result

Code: Select all

6..4....2.....573.....9..1......1.9.....7.5...4.8....34.5......8..2......63.....8 9.7 9.7 9.3ED=9.7/9.7/9.3
2..3....5.....473.....9..1......1.9.....7.4...6.8....34.5......6..5......79.....8 9.7 9.7 9.3ED=9.8/9.8/9.3
2..3....6.....473.....9..1......1.4.....7.9...6.2....55.9......8..5......32.....8 9.7 9.7 9.3ED=9.7/9.7/9.3

B) I then added to the batch as first puzzle the puzzles submitted and got

Code: Select all

2..3....5.....473.....9..1......1.9.....7.4...6.8....34.5......6..5......79.....8 9.7/9.7/9.3ED=9.7/9.7/9.3
6..4....2.....573.....9..1......1.9.....7.5...4.8....34.5......8..2......63.....8 9.7 9.7 9.3ED=9.7/9.7/9.3
2..3....5.....473.....9..1......1.9.....7.4...6.8....34.5......6..5......79.....8 9.7 9.7 9.3ED=9.7/9.7/9.3
2..3....6.....473.....9..1......1.4.....7.9...6.2....55.9......8..5......32.....8 9.7 9.7 9.3ED=9.7/9.7/9.3

C) with no surprise, the suspect submitted puzzles rated alone gave in all trials

2..3....5.....473.....9..1......1.9.....7.4...6.8....34.5......6..5......79.....8 9.7/9.7/9.3ED=9.7/9.7/9.3



I don't know in which conditions gremlin run the rating for that puzzle,
but this reflects undoubtly a small bug in SE RATING.

this gives a king of humanity to gremlin, our referee, but this is not exactly what we are looking for.

champagne

[Pat]

the suspect puzzle rated alone
gave in all trials
2..3....5.....473.....9..1......1.9.....7.4...6.8....34.5......6..5......79.....8
9.7/9.7/9.3

not for me -- i rated it alone and got 9.8


2..3....5.....473.....9..1......1.9.....7.4...6.8....34.5......6..5......79.....8
9.8/9.8/9.3

[champagne]

the suspect puzzle rated alone
gave in all trials
2..3....5.....473.....9..1......1.9.....7.4...6.8....34.5......6..5......79.....8
9.7/9.7/9.3

not for me -- i rated it alone and got 9.8


2..3....5.....473.....9..1......1.9.....7.4...6.8....34.5......6..5......79.....8
9.8/9.8/9.3

thanks pat for having tried it.
It looks lke this would come from an uninitialized area acting in very specific conditions.

very small chances in that case to put the finger on the bug.

If this remains an isolated case, we can accept it. No chance anyway to have it fixed.


champagne

[gsf]

It looks lke this would come from an uninitialized area acting in very specific conditions.

very small chances in that case to put the finger on the bug.

If this remains an isolated case, we can accept it. No chance anyway to have it fixed.

good idea about uninitialized data
I also wonder about 32 vs 64 bit arch (I run on 64 bit x86)

we have the java source
and we know its failing in the 9.7-9.8 code
and the source has already been modified for the patterns game
so the chances to fix it are > 0