4 ms·
You didn't read the article (which incidentally isn't worth reading). Basically the final point after lots of text is that you want all the precision because s
by ars 17y ago
You didn't read the article (which incidentally isn't worth reading).
Basically the final point after lots of text is that you want all the precision because sometimes you want to use a number that can be represented exactly in binary float.
Doesn't really address the point that it still makes no difference to the final value.