Jeffrey Phillips Freeman

@CC007 I really like this idea. Would you be interested in doing a pull request for this if I offered up a bounty as reward?

@CC007 Added a $50 bounty on this if your interested.

IT should be noted this is not simply an issue with rolling over. if I change the calculate line to `return (long) tc + 212600881053l;` instead it should produce the...

@CC007 The issue seems to occur on linux as well and on AMD App SDK. It appears this bug is not platform specific. IF you runt he code in the...

@savaskoc Not yet but i will give it a go this evening. If your saying it produces the same incorrect results however then I expect you to be correct that...

@savaskoc You realize the bug is producing the result but clipping it to 32 bits? To use the test I posted above as the example (the same is true for...

@savaskoc I think there is more to it than that. I think when i was testing i tried some variants that produced some very odd results. I need to test...

@savaskoc also as I stated in the OP if you change the operation to addition but change the value of the operand to make it mathematically equivalent, it magically works....

I've added @CoreRasurae as a reviewer so he can also contribute any comments as this PR develops.

@yana2301 I agree with @CoreRasurae . Several tests were removed that were working well.