dailysudoku.com

George Woods · Joined: 28 Mar 2006 Posts: 304 Location: Dorset UK

Today's Mon 8th puzzle Hard starts with 8 fairly simple moves - Fill up row 5 3 numbers into box 1 a 4 into box3 and 2 into box6

We now find box2 is "attacked" by (35) combos that lead to a "hidden pair 35" into row3

Now an analysis of col5 shows the only place for 2 is at row 4... and so on

Curiously Draw finds the first 8 moves, but fails to spot the 2 in r4c5. Asking for hint, expecting this 2 to be revealed, leads to "Too difficult for me" type comment.

Interestingly the puzzle is graded hard after the first 8 moves, and Easy once the 2 is inserted into r4c5!
So perhaps the coder of Draw could investigate

Chuck Boyer · Guest

I found it even more bizarre.

'Grade' yields difficulty "too hard" after these initial
move(s) : '3' in row 1, col 2 and/or '7' in row 5, col 2.

It later reverts to 'hard', but the 'hint' function fails.

Strange!

Chuck

Guest · Posted: Mon May 08, 2006 4:48 pm Post subject: Re: May 8 puzzle anomaly

ravel · Joined: 21 Apr 2006 Posts: 536

I dont know this program, but there was a strange hint mentioned elsewhere. Alternatives are SimpleSudoku, susser, intosudoku and SudoCue, all free.

David Bryant · Joined: 29 Jul 2005 Posts: 559 Location: Denver, Colorado

I'm not sure if others have thought of this yet, or not.

We just recently (May 2) saw the very first DailySudoku classic puzzle that _required_ the use of an X-Wing pattern to reach the solution. Most likely, Samgj had to modify the logic in the "Draw" program so that it would recognize this pattern.

I have quite a bit of experience (over 20 years' worth) with computer programming. New bugs are often introduced into existing code -- unintentionally, of course -- when one attempts to add new functionality to a program that already works.

So my guess is that the "strange" behavior of the "Draw" program is associated, somehow, with the recent appearance of the X-Wing, and that Samgj will get it straightened out as soon as he has the time and the energy to do so. dcb

Guest · Posted: Tue May 09, 2006 6:47 pm Post subject:

Excellent point, David!

I cut my programming teeth about 40 years ago on Algol, btw... :wink:

Cheers,

Chuck

Clement N. · Guest

After the initial moves i.e 3 in r1c2; 7 in r5c2; 4 in r2c9(lone number) we see that row r5 can be completed as 1 & 9 remain to complete the row. 9 in r9c5 cannot occupy the empty cell in r5c5, therefore 1 should fit that cell and 9 completes the row in r5c8. Now look at BOX1 the numbers to complete the Box are 1,5,6,7 but, 1&6cannot occupy the two empty cells in r2c2 and r3c3. They are left for 5&7; but, 7 is already in r5c2 in BOX4; therefore 5 goes inr2c2 and 7 in r3c3. From there the puzzle becomes manageable somehow.
The grading was proper it was Hard.

Clement. Very Happy

George Woods · Joined: 28 Mar 2006 Posts: 304 Location: Dorset UK

just for the hell of it I went to draw, today's program (hard) put in the v obvious 2 asked for grade and it became v Hard!!!

samgj

Guest · Posted: Thu May 11, 2006 1:31 pm Post subject: May 8th puzzle

A friend of mine and I, who are both new to this game by the way, have been trying to solve the May 8th puzzle. We found matches to the first 8 hints but at that point we get stuck. The next hit puts a 2 in R4C5. Can anyone explain why a 2 goes there for us newbees?

Also, I would like to make a suggestion for a possible enhancement to the "hint" option. I would like to be able to see, in addition to the hint, the reason for the hint. Maybe just the name of the pattern used to arrive at that conclusion would be sufficient.

Thanks.
Brad

Guest · Posted: Thu May 11, 2006 6:28 pm Post subject:

Nevermind. I (actually, my friend) just found it!

PeteGrant · Posted: Sat May 13, 2006 6:06 pm Post subject:

ravel · Joined: 21 Apr 2006 Posts: 536

PeteGrant · Posted: Sat May 13, 2006 9:10 pm Post subject:

Chuck · Guest