The Winners Page, Contest Evolution, and Final Analysis are now on the site. Thanks to Hannes and Cobus for writing a detailed analysis of their code that was part of the winning entry. Do check out these pages to hear more from your fellow contestants!
Don’t miss hearing about the next MATLAB Contest. Sign up for our MATLAB Contest mailing list. Send e-mail to lists@mathworks.com with subscribe contest-announce in the e-mail body.
Thanks for playing and we look forward to seeing you all again in the Fall.
Thanks for the writeups. I’m still curious what the cause of the speedup for obfuscated code was. You said you figured it out midway through the contest and would tell us about it later.
The explained CalculateMoves looks a bit similar to my early SHsolver-calculations. I expect that was too slow, because I wanted to hold too much intermediate results for faster later use. I also did it on an enlarged board, I only didn’t think of the unnecessary bottom row.
I have just read the nice contest analysis of Hannes and Cobus. But there is actually one mistake in it: It was not me who introduced the nthroot-computation. It was not included in my twilight winning entry.
Leave a Reply
About
The MATLAB Programming Contest is a semi-annual competition where contestants submit MATLAB code to try to solve a challenge. For more information, see the overview.
Thanks for the writeups. I’m still curious what the cause of the speedup for obfuscated code was. You said you figured it out midway through the contest and would tell us about it later.
The explained CalculateMoves looks a bit similar to my early SHsolver-calculations. I expect that was too slow, because I wanted to hold too much intermediate results for faster later use. I also did it on an enlarged board, I only didn’t think of the unnecessary bottom row.
I have just read the nice contest analysis of Hannes and Cobus. But there is actually one mistake in it: It was not me who introduced the nthroot-computation. It was not included in my twilight winning entry.