now that was interesting

That was quite a competition at the end :)
Grats to the cyclist.

The entry which is puzzling me most is the algorithm from Hannes
Naudé & Cobus Potgieter, which shows some strange runtime behaviour
:) (especially Probulator VI). On my laptop this algorithm is even
slower than the original one from markus but on the test machine it
seems to be blazingly fast :) (this may depend on the actual data in
the testsuites) The funny thing is I tried out their entry and
disregarded it as to slow to become a competition (oh how wrong I?ve
been :) ). Otherwise I would have tried to make a last minute merger
of my changes with theirs.

The human readable version of my entries came in at place 20 <http://www.mathworks.com/contest/blockbuster.cgi/view_submission.html?id=31882>
and is almost as fast as the obfuscated ones (or am I to stupid to
obfuscate properly for matlab? :) ).

Another interesting thing is the chaotic reaction of all leading
entries towards small changes of the parameters, something that has
to be kept in mind for real world applications.

Last but not least there are some things I would like to see in
future matlab contests:

1. longer twilight phase, since this is usually the phase where the
major algorithmic ideas are developed.

2. a limit of entries one person can post in a certain timeframe
(lets say 10/hour), which may be lifted on the last day to give the
tweakers some time to play around :)

3. clear rules on obfuscation or at least a default obfuscator
.

Relevant Pages

  • Re: Algorithms to generate permutations
    ... >>The position on algorithm design, ... > I claim that my government should not insist ... Would this have forced a US national competition? ...
    (sci.crypt)
  • SUMMARY: changing password on NIS problems
    ... My problem was that I was trying to use the C1Crypt (The entry ... u_newcrypt set to 3 in the /etc/auth/default file), this encryption ... algorithm as a bug that led me to the problem. ... Is this a bug from NIS or passwd, ...
    (Tru64-UNIX-Managers)
  • Re: Fasctode - Sort estimating complexity and B&V
    ... > Estimating complexity and needed time for N elements according ... > since any algorithm is allowed to challange. ... from the competition. ... a person could always send a lot of ill-behaved entries just to ...
    (borland.public.delphi.language.basm)
  • Re: Difference between StoreIDs in Outlook 2007 and 2003
    ... The reason for the difference is the new algorithm to encode the Entry ID ... (See LPWSTR EncodeID(ULONG cbEID, LPENTRYID rgbID) ...
    (microsoft.public.office.developer.com.add_ins)
  • Re: Prime counting algorithms, speculation
    ... > in fact, I never entered any competition, and actually my fastest ... the fastest algorithm you ever published, ...
    (sci.math)