I bet another application version is just what you've all been waiting for, and here it is! (Just kidding). I've put v0.15 on the server for OSX and Linux, and as usual, Windows will be on the way when Dave gets to it. Anyways, the update
to v0.14 reduced the number of bad results from a couple in a hundred to a couple in a thousand. I'm hoping v0.15 will be the last of them. I've turned off the code that is denying credits for these bad results, because you guys are still doing
the work and it's a problem on our end. In fact if you get a bad result it helps us figure out whats going wrong.

The code is in the code release
directory as usual. I'd really like to thank our users (and the people making their own compiled code) for putting up with the changes this last week. Once this last issue is sorted out I think the code should remain stable for awhile as Nate
will be doing a lot of optimizations across different stripes.