Calcudoku puzzle forum
https://www.calcudoku.org/forum/

Today's (12/31/2012) 7x7 difficult
https://www.calcudoku.org/forum/viewtopic.php?f=18&t=392		 Page 1 of 1
Author:  mparisi  [ Mon Dec 31, 2012 2:29 am ]
Post subject:  Today's (12/31/2012) 7x7 difficult
I believe I have solved this puzzle, but the 1^ cage in the middle is being flagged as incorrect with 1,5,7,7 as its contents -- integer overflow perhaps?
Author:  bram  [ Mon Dec 31, 2012 2:40 am ]
Post subject:  Re: Today's (12/31/2012) 7x7 difficult
Same. And while all other puzzles today have been solved a number of times, the counter to the left of this one reads "(solved 0 time(s) so far)".
Author:  clm  [ Mon Dec 31, 2012 3:00 am ]
Post subject:  Re: Today's (12/31/2012) 7x7 difficult
Same problem. Possible overflow in the calculation of powers depending on the sequence followed by the software, since (5 ^ 7) ^ 7 = 1,776... e+34. Consequently, the correct combination c3 = 7, d3 = 1, c4 = 5, d4 = 7 is not accepted as valid giving the red indication in cage 1^. At this time still 0 solutions as observed by bram. Suggestion: To modify the operation to 20+, for instance.
Author:  wizardofhoz  [ Mon Dec 31, 2012 4:43 am ]
Post subject:  Re: Today's (12/31/2012) 7x7 difficult
Glad to see not just me - 1^ last cage I completed and was going crazy going over rest to make sure hadn't made any errors [confused]
Author:  pnm  [ Mon Dec 31, 2012 10:10 am ]
Post subject:  Re: Today's (12/31/2012) 7x7 difficult
Yes, this was reported by many people, seeing it only now (8.09am here),
my first guess is also no overflow in the generator/solver, but overflow
in the code that's running in the browser.

Edit: this turns out to have been the case.

The problem has been "solved" now by a hack: if a cage has the ^ operator,
and 1 as the result, and there's a 1 in there somewhere, it is approved..
(should work :-) )
Author:  bram  [ Mon Dec 31, 2012 1:41 pm ]
Post subject:  Re: Today's (12/31/2012) 7x7 difficult
pnm wrote:
The problem has been "solved" now by a hack: if a cage has the ^ operator,
and 1 as the result, and there's a 1 in there somewhere, it is approved..
(should work :-) )

That hack has a very human, reasoning-rather-than-just-calculating feel to it. Your seem to be on your way to developing strong AI here [smile]

Happy new year – hopefully with lots of new subscribers!
Author:  pnm  [ Mon Dec 31, 2012 1:45 pm ]
Post subject:  Re: Today's (12/31/2012) 7x7 difficult
bram wrote:
That hack has a very human, reasoning-rather-than-just-calculating feel to it. Your seem to be on your way to developing strong AI here [smile]

Yes, I'm waiting for the site to become self-aware [scared]
Author:  nylimb  [ Tue Jan 01, 2013 5:06 am ]
Post subject:  Re: Today's (12/31/2012) 7x7 difficult
pnm wrote:
The problem has been "solved" now by a hack: if a cage has the ^ operator,
and 1 as the result, and there's a 1 in there somewhere, it is approved.


There are other cases that could still cause trouble. E.g. for an "8^" cage with 6 elements, you could have 8 = 2^3^1^7^7^7, but the 7^7^7 would cause an overflow.
Author:  pnm  [ Tue Jan 01, 2013 3:01 pm ]
Post subject:  Re: Today's (12/31/2012) 7x7 difficult
nylimb wrote:
There are other cases that could still cause trouble. E.g. for an "8^" cage with 6 elements, you could have 8 = 2^3^1^7^7^7, but the 7^7^7 would cause an overflow.
Absolutely. That shows the drawback of putting in a hack.. [blink]
Page 1 of 1 All times are UTC + 1 hour [ DST ]
Powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group
http://www.phpbb.com/