Thread Subject:

Spring 2012 MATLAB Contest, April 4 - 11

Hi MATLAB Central Community -

The Spring MATLAB Contest will start at noon, Boston time/16:00 UTC today, Wednesday April 4and will run through noon, Boston time Wednesday, April 11.

Please post any comments or questions about this contest as replies to this thread so that we can track the conversations.

If you have not participated in any of our past contests, you can learn more by going to the contest website at http://www.mathworks.com/matlabcentral/contest/

We look forward to seeing you online!
 
Helen Chen and the MATLAB Contest Team

The Tiles Contest is officially launched. Get started by reading the rules at http://www.mathworks.com/matlabcentral/contest/contests/36/rules .

Good luck everybody!
Helen

"Helen Chen" <helen.chen@mathworks.com> wrote in message <jlhlej$bkp$1@newscl01ah.mathworks.com>...
> Hi MATLAB Central Community -
>
> The Spring MATLAB Contest will start at noon, Boston time/16:00 UTC today, Wednesday April 4and will run through noon, Boston time Wednesday, April 11.
>
> Please post any comments or questions about this contest as replies to this thread so that we can track the conversations.
>
> If you have not participated in any of our past contests, you can learn more by going to the contest website at http://www.mathworks.com/matlabcentral/contest/
>
> We look forward to seeing you online!
>
> Helen Chen and the MATLAB Contest Team

http://www.mathworks.com/matlabcentral/contest/
FYI the Contest Main page does not show the Tiles contest website.

- Amitabh


"Helen Chen" <helen.chen@mathworks.com> wrote in message <jlht2h$cdc$1@newscl01ah.mathworks.com>...
> The Tiles Contest is officially launched. Get started by reading the rules at http://www.mathworks.com/matlabcentral/contest/contests/36/rules .
>
> Good luck everybody!
> Helen
>
> "Helen Chen" <helen.chen@mathworks.com> wrote in message <jlhlej$bkp$1@newscl01ah.mathworks.com>...
> > Hi MATLAB Central Community -
> >
> > The Spring MATLAB Contest will start at noon, Boston time/16:00 UTC today, Wednesday April 4and will run through noon, Boston time Wednesday, April 11.
> >
> > Please post any comments or questions about this contest as replies to this thread so that we can track the conversations.
> >
> > If you have not participated in any of our past contests, you can learn more by going to the contest website at http://www.mathworks.com/matlabcentral/contest/
> >
> > We look forward to seeing you online!
> >
> > Helen Chen and the MATLAB Contest Team

Hey, I'm bit confused about the contest rules. It says;
You don't have to use all tiles.
Tiles you play don't need to fill your board.
You may receive fewer tiles than your board can contain.

Is it possible to have an empty table?

Nuwan Ganganath wrote
>
> Is it possible to have an empty table?

Yes. Depending on the number of tiles provided, you can use all, some, or none of them on the board. If you have too many tiles to use them all, you get to pick which ones go on the board and which ones don't.

The edges values may be negative integers or floating point also? Thanks!

> The edges values may be negative integers or floating point also? Thanks!

Only positive integers!

Thanks,
Doug

> The edges values may be negative integers or floating point also? Thanks!

Only positive integers!

Thanks,
Doug

Score any empty squares on board as if filled with tiles [0 0 0 0], right? Thanks!

> Score any empty squares on board as if filled with tiles [0 0 0 0], right? Thanks!

That is one way to think about it, so yes.

Doug

Amitava Biswas wrote
> Score any empty squares on board as if filled with tiles [0 0 0 0], right?

Right.

"Doug Hull" <hull@mathworks.SPAMPROOFcom> wrote in message <jli713$nma$1@newscl01ah.mathworks.com>...
>
> > The edges values may be negative integers or floating point also? Thanks!
>
> Only positive integers!
>
> Thanks,
> Doug

zero or any positive (not necessarily below 10 or 1000, but including the 0) right? Thanks!

The newsreader link on the bottom of the contest rules page still links to the Fall 2011 Contest thread.

Andreas Bonelli wrote
> The newsreader link on the bottom of the rules still links to the Fall 2011 Contest thread.

Thanks for the note! Fixed it.

"Doug Hull" wrote:
> > The edges values may be negative integers or floating point also? Thanks!
>
> Only positive integers!
>
> Thanks,
> Doug

Test 3 of the test suite contains a tile with a negative edge!

"Oliver Woodford" wrote in message <jlk42s$pjj$1@newscl01ah.mathworks.com>...
> "Doug Hull" wrote:
> > > The edges values may be negative integers or floating point also? Thanks!
> >
> > Only positive integers!
> >
> > Thanks,
> > Doug
>
> Test 3 of the test suite contains a tile with a negative edge!

Oops. We had intended to make 0 be the lowest value, but it looks like a few -1s slipped through the cracks. So all tile edges are one of the following:
   -1
   0
   A positive integer

You may notice an interesting twist that this adds: Any '-1' not played on the board will subtract 1 from your total score rather than add 1. This may help your score, but you may get a better result by playing the tile on the board and matching it to another -1.

Does anyone know why cpu time shown on the contest webpage is so different from what we can get from our own computer using the test case? It seems using bsxfun is much faster on the contest server than on our compters. Why?

Yi

"Yi Cao" <y.cao@cranfield.ac.uk> wrote in message <jlvqk6$q45$1@newscl01ah.mathworks.com>...
> Does anyone know why cpu time shown on the contest webpage is so different from what we can get from our own computer using the test case? It seems using bsxfun is much faster on the contest server than on our compters. Why?
>
> Yi

Yi, the contest machine is running MATLAB R2012a on 64-bit Linux. It is not unreasonable to observe differences in performance between MATLAB versions and machine architectures. Also, keep in mind that the actual testsuite used to score the entries is different than the sample testsuite provided on the File Exchange. Thus, the exact amount of work the code has to do might be a bit different.

Hope this helps,
David

> Yi, the contest machine is running MATLAB R2012a on 64-bit Linux. It is not unreasonable to observe differences in performance between MATLAB versions and machine architectures. Also, keep in mind that the actual testsuite used to score the entries is different than the sample testsuite provided on the File Exchange. Thus, the exact amount of work the code has to do might be a bit different.
>
> Hope this helps,
> David

I notice that the current leader contains a parfor loop. Can you give us any information about whether the contest machine has multiple parallel threads available to it?
-Nick

"Nicholas Howe" wrote in message <jm446c$7eq$1@newscl01ah.mathworks.com>...
> > Yi, the contest machine is running MATLAB R2012a on 64-bit Linux. It is not unreasonable to observe differences in performance between MATLAB versions and machine architectures. Also, keep in mind that the actual testsuite used to score the entries is different than the sample testsuite provided on the File Exchange. Thus, the exact amount of work the code has to do might be a bit different.
> >
> > Hope this helps,
> > David
>
> I notice that the current leader contains a parfor loop. Can you give us any information about whether the contest machine has multiple parallel threads available to it?
> -Nick

Nick,

We do not open a MATLAB pool, so PARFOR will act the same as FOR wherever it's used. As always, only base MATLAB functions are available to be used in the contest. While the official time we measure won’t see any benefit from using PARFOR, it might help a user develop/test code more quickly if they have the Parallel Computing Toolbox available.

Good luck on the home stretch!

- Dan

Another fun contest has wrapped up. Congratulations to Sergey and his cats on the big win this afternoon - that was a job well done! Read more about it in Ned's blog posts http://blogs.mathworks.com/contest/ .

It was so much fun seeing both old friends and new playing in this season's contest!. Thank you to everyone for participating. We look forward to seeing you again next season. :-)

Helen

"Helen Chen" <helen.chen@mathworks.com> wrote in message <jlhlej$bkp$1@newscl01ah.mathworks.com>...
> Hi MATLAB Central Community -
>
> The Spring MATLAB Contest will start at noon, Boston time/16:00 UTC today, Wednesday April 4and will run through noon, Boston time Wednesday, April 11.
>
> Please post any comments or questions about this contest as replies to this thread so that we can track the conversations.
>
> If you have not participated in any of our past contests, you can learn more by going to the contest website at http://www.mathworks.com/matlabcentral/contest/
>
> We look forward to seeing you online!
>
> Helen Chen and the MATLAB Contest Team

Congratulations, Sergey. Thank you to the contest team for coming up with yet another interesting problem for us to try to solve - I often wonder if these aren't customer optimization problems, just rephrased in a fun way!
Oliver

Since the Tiles Contest is over, it would be very appreciated if the actual test suite is posted and/or a link is posted here. Thanks, Amitava Biswas.

"Amitava Biswas" <abiswas@utep.edu> wrote in message <jmhh36$hvf$1@newscl01ah.mathworks.com>...
> Since the Tiles Contest is over, it would be very appreciated if the actual test suite is posted and/or a link is posted here. Thanks, Amitava Biswas.
Hi Amitava -

So sorry on the delay in getting this up for you! We've updated the download with the test suite as of this afternoon.

Thank you for the reminder on this.

Best,
Helen