Floating Point to Unsigned?

The old fault was to truncate, like, round down ito the nearest whole number. The new thing is saturating conversions. Floating point signd has always been centris conforment. That is not changing. O and justthat the main thing that are actually changing is if you're doing like a nan onversion to int or ifits like a large floating point, number larger than youow the size of the to your using. I knew there was a little detail in here that i had forgotten.