Last-Minute Delays

As you may have noticed, testing of Coldest hasn't actually started yet. This is due to a number of different things, including but not limited to:
[list]
[*]The death of my home webserver.
[*]More Windows STL problems.
[*]A crash (possibly particle-related) that I discovered during testing
[/list]
The first two are mostly dealt with and aren't still blocking testing, but the last one is shaping up to be a big headache because the location of the crash doesn't seem to be where the problem actually is. It appears to be a memory corruption issue, but checked iterators aren't turning up anything and I can't reproduce the problem under Valgrind because it slows everything down too much. It's also possible that it's a threading issue, but at the time of the crashes the other threads aren't doing anything related to the crash so I'm not sure. It could be that a threading issue is causing memory corruption which eventually leads to the crash. Who knows?

Anyway, the important point you should take from this post is that I no longer have any idea when testing will actually start because in the past this type of problem has sometimes taken an extremely long time to fix. Watch this space for updates because I could be ready for testing anywhere from tonight to next month.