I Can imagine in the past, this was “faster”, yet clang/gcc can emit the same by just writing a basic A/B function.
Seems the win goes to readability by reducing some of these old school hacks.
What say you, greybeards ?
Similarly, a fast divisibility test (we’ll assume we’re dividing n by some odd prime p):
1. Shift the bits of p right so that there is a 1 in the last position.
2. If n = p then p∣n, if n < p then p∤n, otherwise continue.
3. Subtract p and go back to step 1.
(One of my ADD habits during meetings is to find the prime factors phone numbers or anything else very long. I do something similar with the numbers in decimal, but for this, I’ll subtract multiples of the potential divisor to get 0s at the end of the number in decimal. I remember co-workers puzzling over a piece of paper with my notes I left behind in a conference room trying to figure out what the numbers represented and how the process worked.)
For example, see the article and discussion on Bitwise Division from two days ago: https://news.ycombinator.com/item?id=34981027
Of course that tiny bit of extra work is usually negligible, but might explain why the idiom has stuck around longer than you might otherwise expect.
But yes, fast inverse sqrt is obsolete.
Makes me wonder who pays attention to this sort of thing these days :)
Usually, it can't -- but sometimes...
Oh, yes. I used to do that sort of thing frequently because the time savings was significant enough. As you say, though, compilers have improved a great deal since then, so it's not generally needed anymore.
If stupid bit tricks like that aren't necessary, they shouldn't be used. They do bring a readability/mental load cost with them.
With -O1 it performed the optimisation.