Maybe this is interesting for someone.

It's a distributed attack on SHA-1. They try to find collision with a interesting approach.

http://boinc.iaik.tugraz.at/sha1_coll_search/

OHPen

This is the same technique used for MD5 (sequential multi-message modification via tunneling), though in the case of MD5 it takes only a few seconds, while for SHA-1 the search space is much larger.
What.

To overheat, not burn to ashes :P

@LLXX: I think the remark refers to a 100 % cpu efficiency. I think we shouldn't take it literally.

OHPen

Getting even closer...

Differential Path for SHA-1 with complexity O(2^52):
http://eprint.iacr.org/2009/259.pdf