Given the interest raised by this contest, a second edition has been launched. This year, the goal is to still to evaluate the strength of attacks against an unprotected implementation of a block cipher (AES). Of course, most commercial smartcards embed advanced countermeasures that make such side-channel attacks difficult. Thus, we view this 2nd contest more as a way to learn how to compare attacks ; for this purpose, we will rate them according to metrics such as partial/global success rates or guessing entropy, and also according to the computational power of the attacker. We hope this will improve our understanding of the real threats on secure hardware…

More information there: http://www.dpacontest.org/v2/

Thanks for posting about the results of this first DPA contest. I agree with the most parts of you text. I would like to give a small comment though.

This is exact that the partipants to the DPA contest knew the secret key, but this was only for verification purpose. It is not true that their methods have been tailored for that key value. At least this was not the case for the different methods I proposed. I am quite confident that my methods would have given very similar figures whatever the key.