3 ms·
Interesting that the case they were using was the Nuclear Threat Initiative and FP uncertainties, I've audited some, ah, nuclear-physics-related code that had
by pseudohadamard 4mo ago
Interesting that the case they were using was the Nuclear Threat Initiative and FP uncertainties, I've audited some, ah, nuclear-physics-related code that had an issue due to FP uncertainties...
- ForOldHack 4mo agoWhere you can encode an entire Command and Control server within rounding errors! You sneaky skunk!
- aa-jv 4mo agoAll software is, eventually, rounding error.
- pseudohadamard 4mo agoIt was actually due to the original developer assuming FP compares were exact. It was never an issue, the code went through huge amounts of auditing and testing, just an interesting coincidence that the underhanded contest used something so similar to its underhandedness.