View unanswered posts | View active topics It is currently Sat Dec 16, 2017 10:44 pm



← Back to the Calcudoku puzzle page




Reply to topic  [ 5 posts ] 
 Step by step analytical solution 9x9 “very dif” 2012-05-15 
Author Message

Posted on: Wed May 16, 2012 10:02 pm




Posts: 694
Joined: Fri May 13, 2011 6:51 pm
Post Step by step analytical solution 9x9 “very dif” 2012-05-15
Step by step analytical solution 9x9 “very dif” 2012-05-15.

The tuesday’s 9x9 is usually the most difficult calcudoku during the week. They come with different shapes requiring different techniques. This is the 9x9 difficult (tuesday May 15, 2012), Puzzle id: 461715, solving rate 132.4. The difficulty level is theorically a little higher than the equivalent 9x9 on tuesday Apr 17, 2012 (solved step by step in a different thread). Our purpose here is to use only the analysis (no “guessing” or many “trial and error” will be necessary, everything will come straight forward) (in this post I will mainly use conventional language) (graphic 1 shows the problem):

Image

First (graphic 2) we write (in blue colour) all known numbers and candidates; in steps 1 and 2 additionally we will place some 5’s, 7’s and 9's.

Step 1.
In row 4, a 5 can only go to f4, consequently d4 + e4 = 16, then d4 = 9 (due to e7 = 9), e4 = 7.
The cage “64x” (in two lines) has initially three possibilities: [1,1,8,8], [1,2,4,8] and [2,2,4,4] but due to the L-shape the [1,1,8,8] and [2,2,4,4] are not valid then “64x” = [1,2,4,8].

Step 2.
(In violet colour) Row 5: the 7 cann’t go inside any of the cages “11+” (due to i5 = 4) so now the only valid place is d5 = 7 with e5 + f5 = 10 >>> [1,9], since [2,8], [3,7] and [4,6] are not possible now and, consequently, e5 = 1, f5 = 9 (due again to e7 = 9).
Also we observe that since 7 is not a divisor of “162x” or “432x” and b5 <> 7, the 7 of column b must go to b9 and after this the 5 of column b must go to b5 >>> c5 = 6 (obviously the 5 is not a divisor of those multiplication cages). The cage “11+” (g5-h5) = [3,8].
The 3-cell cage “144x”, initially with these four combinations: [2,8,9], [3,6,8], [4,4,9] and [4,6,6], has been left to [3,6,8] since the 9’s are not allowed (already in use in columns d, e and f) and the 6 is being repeated in [4,6,6]. Since the 3 of column e cann’t go to the cage “2-” (due to e3 = 5 and e5 = 1) it must be e6 = 3 and e1-e2 = [6,8]. Also d6-f6 = [6,8].

Image

Step 3.
We will now obtain the possible combinations for the other products (just in case we need it, the “addition value”, that is, the sum of each combination, is shown in parentheses):

“640x” = [1,2,5,8,8] (24), [1,4,4,5,8] (22), [2,2,4,5,8] (21);
“2268x” = [1,4,7,9,9] (30), [1,6,6,7,9] (29), [2,2,7,9,9] (29), [2,3,6,7,9] (27), [3,3,4,7,9] (26), [3,3,6,6,7] (25);
“162x” = [1,2,9,9] (21), [1,3,6,9] (19), [2,3,3,9] (17);
“432x” = [1,6,8,9] (24), [2,3,8,9] (22), [2,4,6,9] (21), [3,3,6,8] (20), [3,4,4,9] (20), [3,4,6,6] (19);
“24x” = [1,3,8] (12), [1,4,6] (11), [2,2,6] (10), [2,3,4] (9).

Step 4.
Let’s reduce a little bit these combinations. For the cage “432x”, the [3,3,6,8] and the [3,4,6,6] are inmediately cancelled (due to the 3 and the 6 in “144x” in such a way that a 3 or a 6 can not be repeated). The other four combinations of “432x” have a 9.
For the cage “162x” the [1,2,9,9] is not valid (d4 = 9) but the other two have a 9 which being in b2 or b3 force the required 9 of cage “432x” to be in c6 (we write it in green colour, graphic 3). Now the cage “12+” = [5,7] and then the 7 of cage “2268x” goes to g2.

Image

The combination [1,2,5,8,8] is not valid for “640x” due to the double 8 (there is an 8 in “2-”) but the other two combinations for this cage also contain an 8. In these conditions there are four 8’s in the four top rows (considering the cages “64x” and “16x”) so “2:” cann’t be [1,2], [2,4] or [4,8] so it must be [3,6] and we write those candidates (in green).

Step 5.
Now let’s see row 3: The only place for a 7 in this row is c3 (in brown colour) and then d2-d3 = [1,3]. Now, since the pairs 1-3, 1-4 or 3-4 are in column d they cann’t go “simultaneously” to the cell c7 so the only valid combination for “24x” is [2,2,6] and we write those numbers. As a consequence d6 = 8 and f6 = 6.

Step 6 (in red colour).
If, for instance, we name “a” the cell g3 (for simplicity) and “b” the cell g7, applying the property of the addition to column f we find:
a + b + 45 (45 is the full column f) = 7 + 15 + 5 + 9 + 6 + 14 + 3 = 59 >>> a + b = 14. This sum can not be obtained with the pair 6-8 because we already have three 8’s in columns g, h and i (in “64x”, “16x” and “11+”) so it must be [5,9] with g3 = 9 and g7 = 5. As a consequence f2-f3 = [2,4] and f7-f8 = [1,8].
And we can also determine the cage “2268x” applying the addition rule to the three rightmost columns:
“2268x” (+) + 9 + 15 + 10 + 11 + 4 + 20 + 12 + 5 + 22 = “2268x” (+) + 108 = 135 (3 x 45, that is, three columns). Then “2268x” (+) = 27 >>> “2268x” = [2,3,6,7,9].

Step 7 (graphic 4).
Let’s continue: i6 = 5, i7 = 7 (in light blue colour). Also, since “162x” contains a 9, b2 = 9 (due to d4 = 9 and g3 = 9). It is clear that [2,3,3,9] is not valid now for the cage “162x” (two 3’s, but there is already a 3 in “2:”) so “162x” = [1,3,6,9] (19).

(In purple colour). The 6 in b3-b4 and the 6 in c5 cancel the combinations [1,6,8,9] and [2,4,6,9] for “432x” and since [3,4,4,9] is impossible (a double 4 would go to b6-b7-b8), then “432x” = [2,3,8,9] (22).
The 3’s in “2:” and “162x” force d3 = 1 and d2 = 3. Now the 1 in d3 makes b4 or c4 to be a 1 and, consequently, the 1 of cage “64x” goes to h2.
Now the 2’s in “64x” and “16x” force f3 = 4 and f2 = 2. This 2 makes the combination [2,2,4,5,8] impossible for “640x” >>> “640x” = [1,4,4,5,8] (22) (the only left combination for this cage).
Applying the addition rule to the three leftmost columns:
22 + 19 + 7 + 9 + 2 + 11 + “3-” (+) + 22 + 2 + 36 = “3-” (+) + 130 = 135 (3 x 45, that is, three columns). Then “3-” (+) = 5 >>> “3-” = [1,4].

Image

Let’s continue with graphic 5.

Step 8 (in green colour).
The 1-4 in a6-a7 produce b1 = 4, c2 = 4, c1 = 1; additionally a2 = 5 and a1 = 8. And then e1 = 6, e2 = 8 and we place those numbers; e1 = 6 forces the 6 of cage “2268x” to i2.
We write the rest of the numbers in green: b9 = 7 >>> a8 = 7, a9 = 9; c7 = 2 and d8 = 2 >>> b6 = 2 and b7-b8 = [3,8]; c8 + c9 = 36 - 7 - 9 - 7 = 13 = [5,8] ([4,9] and [6,7] not valid now in column c), but also the 3 of “162x” must go to c4 (due to the 3 in b7 or b8), then again the only left numbers in column c are the pair 5-8. Besides this we can write the candidates (among 2, 3 and 9) for the cells g1, h1 and i1.

Step 9 (in brown colour).
To place the final numbers in the case of the cages “2:” and “162x” is elementary. Also since b7-b8 = [3,8] and f7-f8 = [1,8] >>> c9 = 8 and c8 = 5. We complete the cells g9, h9 and i9: g7 = 5 and i6 = 5 >>> h9 = 5; i2 = 6 >>> g9 = 6 and then i9 = 1.

Image

Graphic 6.
Step 10 (in red colour).
To finish the puzzle is very easy considering that g8 + i8 = 10 (the rest for the cage “22+”), so g8-i8 = [1,9] due to [2,8], [3,7] and [4,6] not valid for row 8. So, g8 = 1 and i8 = 9.

Image

Now g8 = 1 >>> f8 = 8, f7 = 1. Also f8 = 8 >>> b8 = 3, b7 = 8 and f7 = 1 >>> a7 = 4, a6 = 1. From this we deduce the candidates for the cells g6-h6, that is, [4,7], and, since g2 = 7, we have g6 = 4 and h6 = 7. Additionally i8 = 9 >>> i1 = 3 >>> g1 = 2 >>> h1 = 9.

Step 11 (in blue colour). In row 8 we need a 6: h8 = 6; in row 7 we need a 3: h7 = 3. This 3 implies g5 = 3, h5 = 8; now the 8 of cage “64x” goes to g4 and the rest is self explanatory.

And the official solution (graphic 7):

Image


Profile

Posted on: Sun May 20, 2012 8:04 pm




Posts: 246
Location: Lisbon, Portugal
Joined: Sun Sep 18, 2011 5:40 pm
Post Re: Step by step analytical solution 9x9 “very dif” 2012-05-
I don't understand why this puzzle so difficult becomes easy after the solution of Clm.( [smile] )
There isn't any "trial and error" whether openly or disguised. This is a solution 100% analytical and perfect.

_________________
Visit http://www.calcudoku.org the most interesting and addictive site of puzzles.


Profile
User avatar

Posted on: Sun May 20, 2012 9:14 pm




Posts: 422
Location: Canada
Joined: Fri May 13, 2011 2:43 am
Post Re: Step by step analytical solution 9x9 “very dif” 2012-05-
I'm always impressed by the amount of time clm has to invest in doing all the diagrams and step-by-step solution method, etc. I don't even have the time to read all that he's written and I'm sure it takes much more time to make it than to read it...


Profile

Posted on: Mon May 21, 2012 10:42 am




Posts: 694
Joined: Fri May 13, 2011 6:51 pm
Post Re: Step by step analytical solution 9x9 “very dif” 2012-05-
sneaklyfox wrote:
I'm always impressed by the amount of time clm has to invest in doing all the diagrams and step-by-step solution method, etc. I don't even have the time to read all that he's written and I'm sure it takes much more time to make it than to read it...


Because I'm as fast prepearing the "papers" as you are solving the timed puzzles [biggrin]. Seriously, prepearing the big "papers" requires me 4-5-6 hours, in different sessions, at nights (when everybody is sleeping), sacrificing tv movies, reading, sleep or going out, but it compensates. And I have found with practice a faster method by doing many intermediate graphics so minimizing the time (if a make mistakes during the process). Sending the post is about 20 to 25 minutes. But the priority for me are of course the daily puzzles, I send most on-line but I print the larger or the very difficult ones to analyze them with pencil and rubber, and I do them along the morning here and there, sometimes even having lunch with other persons [biggrin], when I have many complaints I will gave up this bad habit [smile] though smoking a big cigar is worse [biggrin].


Profile

Posted on: Mon May 21, 2012 4:32 pm




Posts: 246
Location: Lisbon, Portugal
Joined: Sun Sep 18, 2011 5:40 pm
Post Re: Step by step analytical solution 9x9 “very dif” 2012-05-
It´s always a pleasure to follow a solution of yours for a puzzle. As Sneaklyfox says where do you find time? We are waiting for the next one. Perhaps your next work will be to build some baby puzzles ( 1x1 and 2x2 ) for her little Daniel. [smile] [smile]

_________________
Visit http://www.calcudoku.org the most interesting and addictive site of puzzles.


Profile
Display posts from previous:  Sort by  
Reply to topic   [ 5 posts ] 

You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum

Search for:
Jump to:  
cron
All forum contents © Patrick Min, and by the post authors.

Forum software phpBB © 2000, 2002, 2005, 2007 phpBB Group.
Designed by STSoftware.