Thread - TVC XII — 31st March-2nd April

@ 2012-03-25 11:02 PM (#7009 - in reply to #7008) (#7009) Top

Country : United States

MellowMelon posted @ 2012-03-25 11:02 PM

Your understanding of Roman Tapa is correct assuming it works the same way as it did in a previous TVC (7 or 8?).

Compass Tapa: Every cell with one or more arrows has a path to the cell with the star. If we don't allow any paths that use a cell twice, the arrows tell all possible directions of the first direction the path goes in.

Here's a quick example of a wall where a cell might have more than one arrow.

_XXX
XX_A
_XXX
XB_*

The _ are white cells and everything else is part of the Tapa wall, with the * in R4C4 being the starred cell. If there were arrows in cell A (R2C4), there would be one pointing south (R2C4-R3C4-R4C4) and north (R2C4-R1C4-R1C3-R1C2-R2C2-R3C2-R3C3-R3C4-R4C4). If there were arrows in cell B (R4C2), there would only be one pointing north (R4C2-R3C2-R3C3-R3C4-R4C4). There would NOT be one pointing west because a path that goes in that direction first would have to retrace its steps to reach the *.

Some more examples of this type here, where it originated from: http://buyaketa.blogspot.com/2012/02/compass-tapa.html

Finally, after all the anxiety over the last IB, I should say I am really excited for this one. Visionary, Make Room, Compass, Borders (

)... so much to look forward to.

@ 2012-03-26 1:20 AM (#7011 - in reply to #7007) (#7011) Top

prasanna16391

Posts: 1801

Country : India

prasanna16391 posted @ 2012-03-26 1:20 AM

^ Thanks for the link.

I'm looking forward to this one too. Quite a nice IB. Hopefully I stop my little run of goof-ups in the last two TVCs.

@ 2012-03-26 1:42 AM (#7012 - in reply to #7007) (#7012) Top

MellowMelon

Country : United States

MellowMelon posted @ 2012-03-26 1:42 AM

Some questions:

- For Wired Tapa, does the 2x2 replacement rule mean that a grid cell can't be surrounded by four edges (i.e. if there were slitherlink clues no 4s would be allowed), or that a dot can't be joined to four edges?
- Is the roman example in the IB wrong? The IIII on bottom is a single 4, which is not proper roman numerals. On the other hand if you let the IIIII be a 1-4 clue there are many other solutions.

@ 2012-03-26 2:10 AM (#7013 - in reply to #7012) (#7013) Top

suboree

Posts: 9

Country : Turkey

suboree posted @ 2012-03-26 2:10 AM

MellowMelon
- For Wired Tapa, does the 2x2 replacement rule mean that a grid cell can't be surrounded by four edges (i.e. if there were slitherlink clues no 4s would be allowed), or that a dot can't be joined to four edges?

Yes, there cannot be wire segments in all of the four edges of any cell.

MellowMelon
- Is the roman example in the IB wrong? The IIII on bottom is a single 4, which is not proper roman numerals. On the other hand if you let the IIIII be a 1-4 clue there are many other solutions.

Sorry, bad example. This has been asked before in TVC VIII, we took the example from the list that contains all Tapa variations so far, and unfortunately have forgotten to correct the example puzzle. 4 is definitely IV, as discussed here:

http://logicmastersindia.com/forum/forums/thread-view.asp?tid=285&p...

@ 2012-03-26 3:10 AM (#7014 - in reply to #7007) (#7014) Top

zachpuzzle

Posts: 14

Country : United States

zachpuzzle posted @ 2012-03-26 3:10 AM

can paths pass through arrows? so, for exapmle, would the following picture be valid?

(untitled.PNG)

Attachments
----------------

untitled.PNG (2KB - 2 downloads)

@ 2012-03-26 3:41 AM (#7015 - in reply to #7014) (#7015) Top

MellowMelon

Country : United States

MellowMelon posted @ 2012-03-26 3:41 AM

Yes, that is allowed, assuming we're using the same rules as the original I linked to above.

Edited by MellowMelon 2012-03-26 3:42 AM

@ 2012-03-26 4:17 AM (#7016 - in reply to #7007) (#7016) Top

MellowMelon

Country : United States

MellowMelon posted @ 2012-03-26 4:17 AM

To give Para either some rest or at least some puzzles for him to practice on that were not made by him, I went ahead and drew up my own practice puzzles for this test. All of the variations save for 1, 5, 10 (Previously seen, Broken, TAPA TAPA) are covered in this set. Go here to find them.

@ 2012-03-26 10:41 PM (#7017 - in reply to #7009) (#7017) Top

swaroop2011

Posts: 668

Country : India

swaroop2011 posted @ 2012-03-26 10:41 PM

MellowMelon - 2012-03-25 11:02 PM

Your understanding of Roman Tapa is correct assuming it works the same way as it did in a previous TVC (7 or 8?).

Compass Tapa: Every cell with one or more arrows has a path to the cell with the star. If we don't allow any paths that use a cell twice, the arrows tell all possible directions of the first direction the path goes in.

Here's a quick example of a wall where a cell might have more than one arrow.

_XXX
XX_A
_XXX
XB_*

The _ are white cells and everything else is part of the Tapa wall, with the * in R4C4 being the starred cell. If there were arrows in cell A (R2C4), there would be one pointing south (R2C4-R3C4-R4C4) and north (R2C4-R1C4-R1C3-R1C2-R2C2-R3C2-R3C3-R3C4-R4C4). If there were arrows in cell B (R4C2), there would only be one pointing north (R4C2-R3C2-R3C3-R3C4-R4C4). There would NOT be one pointing west because a path that goes in that direction first would have to retrace its steps to reach the *.

Some more examples of this type here, where it originated from: http://buyaketa.blogspot.com/2012/02/compass-tapa.html

Finally, after all the anxiety over the last IB, I should say I am really excited for this one. Visionary, Make Room, Compass, Borders (

)... so much to look forward to.

Thank you for the explanation.
So in your example for the arrows in cell B.
Only one pointing towards North exist ok.
So it means that arrow has two paths to reach star right.
one mentioned by you and the other as (r4c2-r3c2-r2c2-r1c2-r1c3-r1c4-r2c4-r3c4-r4c4)

@ 2012-03-27 12:45 AM (#7018 - in reply to #7008) (#7018) Top

Para

Posts: 315

Country : The Netherlands

Para posted @ 2012-03-27 12:45 AM

swaroop2011 - 2012-03-25 10:36 PM

In roman Tapa is it that
III may be 1,1,1 or 2,1 or 3 itself
and IIIII may be 2,1,2 or 2,3 or 5 itself..etc.

If I understood it right. IIIII can't be 5. 5 would have to be a V, 5 I's can't be a 5.

@ 2012-03-27 2:29 AM (#7019 - in reply to #7018) (#7019) Top

prasanna16391

Posts: 1801

Country : India

prasanna16391 posted @ 2012-03-27 2:29 AM

Para - 2012-03-27 12:45 AM

swaroop2011 - 2012-03-25 10:36 PM

In roman Tapa is it that
III may be 1,1,1 or 2,1 or 3 itself
and IIIII may be 2,1,2 or 2,3 or 5 itself..etc.

If I understood it right. IIIII can't be 5. 5 would have to be a V, 5 I's can't be a 5.

Yes, on all the examples I've solved so far, that was the way I interpreted and they solved fine. You can't have a 4-1 either for the IIIII.

@ 2012-03-28 7:06 PM (#7032 - in reply to #7007) (#7032) Top

anurag

Posts: 136

Country : India

anurag posted @ 2012-03-28 7:06 PM

Compass tapa:For the puzzle to be elegant,every arrow should find a unique path.This is not mentioned in the rules.I believe Takeya's third puzzle has unique paths,otherwise the puzzle would have three solutions.How is the test puzzle gonna be like?

@ 2012-03-28 9:51 PM (#7035 - in reply to #7007) (#7035) Top

rob

Posts: 170

Country : Germany

rob posted @ 2012-03-28 9:51 PM

Neither of Takeya's first and third puzzle has unique paths, and both are uniquely solvable. Also they're both elegant in my opinion…

@ 2012-03-29 1:28 AM (#7041 - in reply to #7007) (#7041) Top

RALehrer

Posts: 31

Country : United States

RALehrer posted @ 2012-03-29 1:28 AM

For Tapa Tapa:

The directions say that the cities CAN have Tapa clues. Does this mean they also have the option of not having clues? (The example shows only length 1 segments adjacent to cities, but without clues there could be longer segments. Can a city be shaded?

@ 2012-03-29 1:44 AM (#7042 - in reply to #7041) (#7042) Top

prasanna16391

Posts: 1801

Country : India

prasanna16391 posted @ 2012-03-29 1:44 AM

RALehrer - 2012-03-29 1:28 AM

For Tapa Tapa:

The directions say that the cities CAN have Tapa clues. Does this mean they also have the option of not having clues? (The example shows only length 1 segments adjacent to cities, but without clues there could be longer segments. Can a city be shaded?

Cities can have Tapa clues, using ONLY the digit 1. So 1, 1-1, 1-1-1, 1-1-1-1 are all possible around the train, and it has to be 1 of these 4 possibilities since otherwise a connective path between two trains wouldn't exist.

@ 2012-03-29 8:50 AM (#7043 - in reply to #7042) (#7043) Top

RALehrer

Posts: 31

Country : United States

RALehrer posted @ 2012-03-29 8:50 AM

If I read you right, then, cities MUST have tapa clues - as a square with no tapa clues can have any number of segments around it.

@ 2012-03-29 9:17 AM (#7044 - in reply to #7043) (#7044) Top

prasanna16391

Posts: 1801

Country : India

prasanna16391 posted @ 2012-03-29 9:17 AM

RALehrer - 2012-03-29 8:50 AM

If I read you right, then, cities MUST have tapa clues - as a square with no tapa clues can have any number of segments around it.

Well true, I guess the cities are clue cells with at least a 1.

@ 2012-03-29 10:23 AM (#7045 - in reply to #7043) (#7045) Top

Para

Posts: 315

Country : The Netherlands

Para posted @ 2012-03-29 10:23 AM

RALehrer - 2012-03-29 8:50 AM

If I read you right, then, cities MUST have tapa clues - as a square with no tapa clues can have any number of segments around it.

The rule is that any segment of the wall touching a city can only be 1 cell long.
From this my question comes. Do all cities have to be connected to eachother? For example, would the following puzzle be allowed?
The question of course can become irrelevant quickly if each city has at least one distance given.

(Valid Tapa Tapa.png)

Attachments
----------------

Valid Tapa Tapa.png (12KB - 3 downloads)

@ 2012-03-29 10:46 AM (#7046 - in reply to #7007) (#7046) Top

anurag

Posts: 136

Country : India

anurag posted @ 2012-03-29 10:46 AM

Interesting.Such a city can be used as a trick.But I think Serkan wouldnt, in all probability.The test puzzle would most likely show distances involving all cities
(certainly not all pairs though),and also likely have all cities connected.

@ 2012-03-29 11:07 AM (#7048 - in reply to #7045) (#7048) Top

yureklis

Posts: 183

Country : Turkey

yureklis posted @ 2012-03-29 11:07 AM

Para - 2012-03-29 10:23 AM

RALehrer - 2012-03-29 8:50 AM

If I read you right, then, cities MUST have tapa clues - as a square with no tapa clues can have any number of segments around it.

So far i've solved only Rauno's example. I had some questions about the type too before solving. But, after solving the example i thought every city should have at least a 1 clue. I suppose that Rauno thought same way. So all cities have to be connected like in the example.

Edited by yureklis 2012-03-29 11:10 AM

@ 2012-03-29 4:16 PM (#7050 - in reply to #7007) (#7050) Top

anurag

Posts: 136

Country : India

anurag posted @ 2012-03-29 4:16 PM

Wired tapa: Can the wire loop in to itself? This is different from using all four nodes of a cell. I know this is possible,but a bit useful to know in advance so we know that such loops do not exist.

@ 2012-03-30 11:55 PM (#7051 - in reply to #7007) (#7051) Top

Administrator

Country : India

Administrator posted @ 2012-03-30 11:55 PM

Puzzle	Points	Group
1a. Visionary Tapa	59	Puzzles 1-2
1b. Make Room For Tapa	104	Puzzles 1-2
2a. Twilight Tapa	25	Puzzles 1-2
2b. Twilight Tapa	83	Puzzles 1-2
3a. Compass Tapa	14	Puzzles 3-4
3b. Compass Tapa	92	Puzzles 3-4
4a. Wired Tapa	33	Puzzles 3-4
4b. Wired Tapa	94	Puzzles 3-4
5a. Broken Tapa	40	Puzzles 5-6
5b. Broken Tapa	71	Puzzles 5-6
6a. Roman Tapa	19	Puzzles 5-6
6b. Roman Tapa	63	Puzzles 5-6
7a. Sweeper Tapa	42	Puzzles 7-8
7b. Sweeper Tapa	163	Puzzles 7-8
8a. Tapa - Like Loop	14	Puzzles 7-8
8b. Tapa - Like Loop	90	Puzzles 7-8
9a. Tapa [Borders]	35	Puzzles 9-10
9b. Tapa [Borders]	144	Puzzles 9-10
10. TAPA TAPA	62	Puzzles 9-10

Announcements

Password protected puzzle has 12 pages (including a cover page)
Time bonus will be computed after you click "Claim Bonus"
Penalty will be given irrespective of correctness of submission

Contest Length : 75 minutes
Total Points : 1247
Bonus : 13 points per minute saved
Grace Period : 5 minutes
Penalty : 9 points per minute used (Computed upto seconds)

@ 2012-03-31 12:43 AM (#7055 - in reply to #7007) (#7055) Top

Para

Posts: 315

Country : The Netherlands

Para posted @ 2012-03-31 12:43 AM

19 puzzles in 75 minutes. That's going to be challenging. Although I guess I did 50 in an hour at the WPC.

@ 2012-03-31 7:24 AM (#7058 - in reply to #7007) (#7058) Top

Administrator

Country : India

Administrator posted @ 2012-03-31 7:24 AM

Link to score page - http://logicmastersindia.com/TVC/XII/score.asp

@ 2012-03-31 11:41 PM (#7060 - in reply to #7058) (#7060) Top

prasanna16391

Posts: 1801

Country : India

prasanna16391 posted @ 2012-03-31 11:41 PM

Excellent puzzles. Absolutely loved some of them (Yet to do 2). Thanks for a great 2 months of puzzles Serkan :)

@ 2012-04-01 9:06 PM (#7063 - in reply to #7007) (#7063) Top

swaroop2011

Posts: 668

Country : India

swaroop2011 posted @ 2012-04-01 9:06 PM

Nice Puzzles. Njoyed a lot. I only knew 6 types to solve . So Kept a target to solve 12 puzzles at least. But was able to do only 10.
But even then i guess it was better than my first two tvcs and worst than 3rd tvc.
My favorite was Compass Tapa and Tapa Borders.