Congratulations again to MellowMelon for winning TVC XI by a huge margin. xevs and deu take 2nd and 3rd position respectively. EKBM also played really well, but had a mistake in one puzzle. Out of 126 players, 109 players got non-zero scores. The table on the right shows cumulative normalized scores of top 10 players.

TVC IX TVC X TVC XI Total MellowMelon 748.2 1000 1000 2748.2 motris 1000 651.5 840.1 2491.6 deu 857.9 643.1 848.1 2349.1 Para 857.9 721 693.8 2272.8 Kota 726.2 836.9 635.5 2198.6 tomek_s 720 586.9 826.9 2133.8 nyuta 681.3 622.4 717 2020.7 xevs 791 313.2 912.1 2016.3 misko 742.2 586.9 684 2013.1 anderson 711.1 589.1 686.7 1986.9

With Serkan not around, I'll put some remarks like he does every time.

There were lot of discomfort among players when the IB was released, must be because of so many new types and some unfamiliar rules. MellowMelon was one of the first players to participate, he had a great performance, but he also posted in forum immediately that the puzzles were nice. Thanks and congratulations for yet another superb performance.

- The puzzle solved by least competitors was Power of Tapa. It was solved by 20 competitors.
- The puzzle solved by most competitors was Fractional Tapa. It was solved by 96 competitors.
- The puzzle with the highest rating was Visionary Tapa (9.28) and the puzzle with the lowest rating was Mad Max Tapa (Top) (6.16).
- The puzzle submitted with the highest accuracy was Dissected Tapa (100%), 64 competitors), lowest was Mad Max Tapa (Bottom) (48.89%, 22/45 competitors)

Here is the link to Cumulative results table - http://logicmastersindia.com/TVC/2012.asp

See you all in the next and final round.

So as I said a few days ago, the test did turn out to be really great. One suggestion I have that would scare everyone less is to not complicate the IB more than it needs to be. Specifically, say up front that the test's Meiosis Tapa would have only even numbers and no multi-number clues. I guess it's possible the IB was released before the puzzle was ready and/or finalized, but that variation as it was in the IB was just plain unwieldy. It's intimidating as hell to have a clue like 1 2 that can function as 1 2, 1 1 1, 1 1, or 1 5 (this possibility eliminates a lot of connectivity logic the clue might have otherwise). The test puzzle itself was great though.

Still not a fan of Mad Max. One good thing about logic puzzles like these is that as hard as solving them can be, checking them is usually easy and straightforward. But for a type like Mad Max, it's almost impossible to be totally sure you have the right answer; computationally, verifying a solution is as hard as solving the puzzle. I think the low success rate on that second puzzle is both not surprising and not a good thing.

I wasn't really enthusiastic at first when I saw the IB, but found enough interesting variants to fill my 75+ minutes after taking a closer look to all the puzzles.
Unfortunately I spent a lot of time on the meiosis tapa, running into contradictions at the same place all the time. I thought to have understand the rules properly, but obviously it's not. So I am very much interested in the solution. If someone can post it here in the forum I would be happy.

Because of the time spent on the meiosis, I just missed the Visionary tapa by a minute. It was a wonderful puzzle, with a brilliant start. Thanks Serkan!

Hi,
Even I have confused myself during the test, but this is a simple puzzle, after the test it took around 8 minutes for me.
I hope the solution is correct.

Regards,
Ravi





Attachments
----------------
Meiosis Tapa.PNG (13KB - 0 downloads)

Thanks!
My problem was the 6 in the middle. If you divide 6 by 2 yo get 3, not 3/3. So I assumed I did something wrong but didn't know what.
As far as I can see, the IB doesn't contain an example where an even digit gives a two-clue digit after dividing.

Did anyone else have problems with this or is it just me who can't read?

Hi,
Para has explained by taking 2 2 as an example in this thread, probably you might have missed it.
http://logicmastersindia.com/forum/forums/thread-view.asp?tid=458&start=23

Regards,
Ravi

Thanks again for the many fun puzzles, Serkan!

I'd be interested in seeing the solution to the Mad Max Tapa. I wasn't confident in my solution being optimal and didn't want to risk losing points submitting after the 75 minutes expired (even though it was only 5 seconds), which is why I only submitted for 10 puzzles and not 11. If there's some logic involved in confirming the maximum that would also be nice to know. I was hoping for some kind of logic that I had in mine, but I couldn't really find any.

I just solved this one after the fact. There was far more pure Tapa solving to do than I could see during the contest… The picture I intend to attach shows how far I could get just with Tapa deductions. At that point, you can shade in all cells that don't neighbour clues and complete this to a solution that also maximizes the clued cells (around the 6,3,23,11 chain).






Attachments
----------------
madmax.png (19KB - 3 downloads)

I also solved it in the aftermath. The factor which makes the solution unique is column 1.
You have to connect the upper part with the bottom part. You can't do it on the right hand side because of the pait (11). If you do it in column 1 if you blacken R7 and R8 you lose the opportunity to blacken R10 and R11.
If you blacken R7, R10 and R11 you blacken 3 cells insted of 2, and you can still connect top and bottom through R8C7 and R6C10, and this also gives a unique solution for the (11) pair in R3C11

For Mad Max, I did not get as much in the top as rob has in that image, but I had the rest. When I was stuck there I decided I would try to shade in all the unclued squares, save for the top left where at least one had to be left white, and see if that worked. Wasn't 100% sure though when I was done, since it could have been possible to squeeze out more by leaving more squares at the intersections of clues unshaded. I eventually convinced myself that there was no way a puzzle in a timed contest could be that cruel to have the valid and fairly well-constrained solution I had found and not have it be right, but it was a doubt I had even as I went to claim time bonus...

By the way rob, why didn't you shade R10C6 and R10C11? =P

where can i see the cumulative scores of all 3 tapas held till now.

Thanks rob, I hadn't noticed the Tapa bit was that much constrained. I was expecting more of maximise coloured unclued cells vs minimise double clued cells. But it seems the optimise bit was much smaller in this puzzle.



http://logicmastersindia.com/TVC/2012.asp

Yes, I really should have that pattern down by now.