The New Sudoku Players' Forum

[koushanejad74]

Blocku is quite similar to classic Sudoku,

For a 6 by 6 puzzle:

-There's a block in each row and column
-Each row and each column must include numbers from 1 to 5 (similar to classic Sudoku)
-For each row and column: the numbers on each side of the block must be consecutive

Here's a small sample:

Blocku_6_1_0_Problem.png (10.36 KiB) Viewed 809 times

And here's the solution:

Blocku_6_1_0_Solution.png (22.18 KiB) Viewed 809 times

And here's a challenging one for you to solve, enjoy

Blocku_10_1_3_Problem.png (28.59 KiB) Viewed 809 times

[Leren]

This looks to be the same as Str8ts to me. Andrew Stuart has a site for the 9 x 9 version here. You can also put clues in the black cells if you want. A Wikipedia article about Str8ts is here.

You can try solving the following clueless 9 x 9 Str8ts puzzle shown here. You are only given 1 hint : r6c3 (or F3 in Andrew's cell ID convention) is not 3. It's not all that hard.

Leren

[koushanejad74]

Hi Leren,

Thanks for your reply, It's similar to Str8ts, but not the same as I have only one blocker in each row and column; and also every row and column must include all the numbers;
By the way, did you solve the one I posted,

Regards,

-Kousha

[Leren]

Yes, I finally solved it via paper and pencil. Here is my solution

Code: Select all

4569378X21
567418923X
678549X312
78965X2143
89X7213456
9X38621574
312X745968
X251934687
13428567X9
2413X67895


Technically, it was very easy, requiring nothing more than singles, and a few naked pairs and triples. Also my Str8ts experience definitely helped with some easy Hi/LO decisions early on.

In practice, as it was the first time I've actually manually solved any puzzle in years, I kept making stupid mistakes, so it took me quite some time to solve.

A pleasant variant. No doubt the smaller puzzles will keep many people glued to their phones, if they aren't already.

Leren

[SpAce]

I couldn't get very far without pencil marks, but after that it was trivial even without any Str8ts experience. I guess poker experience helps too Besides, my p&p system kills pretty much any problem that can be expressed with candidates, so I didn't expect much trouble (and there wasn't). Only the lightest level of my system was required anyway since it was just basics.

Would be fun to see something like this with a bit more advanced methods required. At one point I considered using a UR but it was never necessary. The hardest technique I used was a naked triple, and even that was obvious (all cells next to each other). (Added. Actually there were probably more triples, maybe even quads, if you consider the blocked regions affecting each other. I didn't really pay attention because it was all basics and quite automatic.)

In practice, as it was the first time I've actually manually solved any puzzle in years, I kept making stupid mistakes,

It's probably not that. I bet it's because you've never developed a manual system that really works and prevents making mistakes. I'm almost certain that I could quit sudoku for many years, and come back at almost the same level and with almost the same error rate (close to zero). Routine like that just doesn't go away if you've put some thought into creating a system that works. It becomes part of you. (These days I use my p&p system only a few times a year because I've become lazy, but it seems to make no difference -- even with a new puzzle type. It's like riding a bicycle.)

[tarek]

The 10x10 version is probably closest to what we all know. I can see very little difference between this variant and Vanilla sudoku IMO. (so adding the consecutive twist is good but then you can remove that or add any similar constraint as with vanilla). Looking at the solution for the 1st sample puzzle, I may have missed how the numbers on either side of the block are consecutive.

Well done nevertheless. Developing a variant that actually works and allows people to solve and enjoy it must come with degree of satisfaction to the developer.

tarek

[Leren]

Hi Tarek, I think there may be just a linguistic quirk about the consecutive rule description. What is meant is that the set of numbers between a black cell (called a block here) and the edge of the puzzle have to form a contiguous set, but their physical positions can be in any order. There is no requirement for numbers touching opposite sides of a block to differ by one.

Leren

[koushanejad74]

Thanks for you comments

Added a more challenging sample here: blocku-part-ii-t36323.html

[tarek]

I see.

Code: Select all

In a set of X continuous white cells A-B >= X is forbidden

tarek

[Wecoc]

It would be more challenging if not all blocks were placed from the start, that way the logic would be more like Between 1&9 Sudoku (sometimes called Sandwich sudoku).
The consecutive rule in both 'sides' is VERY restrictive, there aren't many cases for the 6x6 grid.

Another way to approch this would be considering the block is like a "9" in a default sudoku grid (9x9).
That way you can also include the 3x3 blocks as an extra constrain. That being said, I'm not sure a sudoku with those restrictions is even possible.

For example: 7, 6, 4, 5 is valid, but 1, 3, 8, 7 is not.

4567 isn't valid because it breaks the other 'side'. It can only be 1234 or 6789.