Hacker News Reader

For floating point math is this why I see "/2" written as "*0.5" all the time, and is that really necessary and can't compilers figure that out?

View on HN

They don't necessarily produce the same result and strict IEEE conformance prohibits such transformations. With GCC, --ffast-math enables optimizations like this.

That makes sense. I typically have to worry about larger relative tolerances anyway so I think I can start using that trick.

In this specific example, they do produce exactly the same result:

`a / b` is completely equivalent to `a * (1 / b)` when both b and 1/b are perfectly representable in floating point without rounding, which is the case for 2 and 0.5

Reply on news.ycombinator.com