I've received a lot of interesting comments and questions from Sudoku fans over the last few years and this page
is where I try to answer them. I'm also directing Str8ts feedback here. Please feel free to drop me a note on the side of the page. Or you can email me directly at andrew@str8ts.com.
Post a Comment or Question here...
Thursday 1-Apr-2021
... by: Peter Foster, United Kingdom
Hi I have a Killer Sudoku from Puzzler #181 that the solver can't crack.
I'm not sure if it is because the strategy required hasn't been programmed or whether the puzzle is flawed.
If you are interested, is there a way I can send it to you ?
Many thanks
Pete
Andrew Stuart writes:
Yes the Killer solver does lag behind the Sudoku in terms of solving power. Mainly because there are zillions more candidates and the chaining strategies can't bite I have example that don’t complete, so no sweat
Saturday 20-Feb-2021
... by: Richard Pasma, United States
Thank you so much for all the work you have done and are doing in providing an excellent and probably the best resource there is for people interested in puzzles. Since the web site has been recovered I have not been able to get the Sudoku Solver -Solution Count button to work. It returns a SudokuWiki.org/404 on the address line and an "Article Not Found" as a result. It was working well for me before the crash. I am using Windows 10 and tried getting it to work with IE, Fire Fox and Chrome to no avail. Thanks again, Rich
Andrew Stuart writes:
Restored it this morning. Quite difficult to remember how to install a DLL in the web server environment, but dredged it out of memory somehow
Friday 19-Feb-2021
... by: Tom, Connecticut USA
So sorry about the vicious hack. I'm saddened that our world has people so evil as to destroy Andrew's site that has given pleasure to so many people the world over. When I was working I used to purchase monthly puzzle packs both as a way to support the site, and give me diversions on long flights. I especially enjoy the symmetrical patterns of the daily puzzles. I'm going back to purchasing monthly puzzle packs as a way to support the site financially.
Andrew Stuart writes:
Thanks you very much! Really appreciated and very heartening to hear My pleasure
Tuesday 16-Feb-2021
... by: Dikla, Israel
Why the daily jigsaw does not display on the web page? I have enabled Javascript.
Andrew Stuart writes:
Jigsaw player is back online. Still have to work out why the state is not saving in cookies
Sunday 14-Feb-2021
... by: Mark Boyd, Canada
Welcome back. You were very much missed.
Andrew Stuart writes:
Thank you !
Sunday 14-Feb-2021
... by: Philip Trueman, England
Has the definition of 'midnight' changed as a result of the hack/restore? Not complaining at all, just asking.
Andrew Stuart writes:
It has. The old server was set to US Eastern Standard Time and is now on GMT This was just how it came and it was too late to change it It's in my job queue to go through the player feed scripts and add some kind of adjustment EST was fair to the US and Europe
Saturday 13-Feb-2021
... by: Eugene, US
I made a small donation...hopefully it helps a little with the recent hack. Hope it gets sorted out!
Andrew Stuart writes:
Thank you! Going towards a better backup system, for sure. Really appreciated All the best Andrew Stuart
Saturday 13-Feb-2021
... by: Charles R Cochems, United States
Well, seems there's a strategy that's not known by your solver. Cracking the cryptic found it. I call it odd/even. it's very usable by humans, and not any sort of bifurcation.
Divide the puzzle into odd/even columns (or rows).
In the odd columns (or rows) there must be a total of 20 evens, minus any that are already placed (4 per column).
now for each row, count the number of known evens. subtract that number from four. color that many squares that are on an odd columns. doesn't matter which ones.
now count the number of colored squares on the odd rows. if this is the same as the maximum left to place on those columns, then we can make some deductions.
Specifically if any row has all of it's cells in the odd columns marked as potential evens at this point, then you can eliminate odds from those squares. This is because placing an odd on one of them would mean there aren't enough spaces left to put all the evens.
On the rows you couldn't eliminate the odds from the odd columns, you can then eliminate the evens from the even columns on those rows, since all the evens MUST be in the odd columns. uncolor the odd cells, and color the empty even cells a different color.
This can be done with row and column swapped. Since there are less even numbers than odd, this technique usually works better with evens in the odd rows, or evens in the odd columns. smaller number, greater area.
Usually, this will get you nothing. But if you notice very few evens in the odd rows, or very few evens in the odd columns, this can give a useful result.
Andrew Stuart writes:
I will have to look that one up on Cracking the Cryptic then, thanks for a first look
Saturday 13-Feb-2021
... by: Klaus Stärk, Switzerland
numbers 1 to 9 on the right side of sudoku?
Andrew Stuart writes:
Some people use an alternative coordinate system row num + col num
Thursday 1-Apr-2021
... by: Peter Foster, United Kingdom
I'm not sure if it is because the strategy required hasn't been programmed or whether the puzzle is flawed.
If you are interested, is there a way I can send it to you ?
Many thanks
Pete
I have example that don’t complete, so no sweat
Saturday 20-Feb-2021
... by: Richard Pasma, United States
Friday 19-Feb-2021
... by: Tom, Connecticut USA
My pleasure
Tuesday 16-Feb-2021
... by: Dikla, Israel
I have enabled Javascript.
Sunday 14-Feb-2021
... by: Mark Boyd, Canada
Sunday 14-Feb-2021
... by: Philip Trueman, England
This was just how it came and it was too late to change it
It's in my job queue to go through the player feed scripts and add some kind of adjustment
EST was fair to the US and Europe
Saturday 13-Feb-2021
... by: Eugene, US
Really appreciated
All the best
Andrew Stuart
Saturday 13-Feb-2021
... by: Charles R Cochems, United States
Divide the puzzle into odd/even columns (or rows).
In the odd columns (or rows) there must be a total of 20 evens, minus any that are already placed (4 per column).
now for each row, count the number of known evens. subtract that number from four. color that many squares that are on an odd columns. doesn't matter which ones.
now count the number of colored squares on the odd rows. if this is the same as the maximum left to place on those columns, then we can make some deductions.
Specifically if any row has all of it's cells in the odd columns marked as potential evens at this point, then you can eliminate odds from those squares. This is because placing an odd on one of them would mean there aren't enough spaces left to put all the evens.
On the rows you couldn't eliminate the odds from the odd columns, you can then eliminate the evens from the even columns on those rows, since all the evens MUST be in the odd columns. uncolor the odd cells, and color the empty even cells a different color.
This can be done with row and column swapped. Since there are less even numbers than odd, this technique usually works better with evens in the odd rows, or evens in the odd columns. smaller number, greater area.
Usually, this will get you nothing. But if you notice very few evens in the odd rows, or very few evens in the odd columns, this can give a useful result.
Saturday 13-Feb-2021
... by: Klaus Stärk, Switzerland