![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
According to my little benchmark at http://bolson.org/flops_js/, on my MBP13, recent Safari gets about 200 megaflops, Chrome(beta) is getting 150, and the latest Firefox 3.6 gets an astounding 700 (there's actually a range of numbers depending on mix of operations, I'm eyeballing a representative number). Java on the same machine gets around 1600 megaflops.
Now the super amazing thing that happens when I try it just now is that running virtually the same source, transliterated from C to Java and Javascript, I run `gcc -O2 -m64` and run the binary and get results slower than Java. I'll eyeball it as C being 80% the speed of Java. Of course, I'm sure this will vary from problem to problem and some things may optimize better on one system or the other. On the plus side, I also ran a binary from 2006 based an an unknown gcc version of that era. It was slower than the latest gcc. At least that's getting better.
Now the super amazing thing that happens when I try it just now is that running virtually the same source, transliterated from C to Java and Javascript, I run `gcc -O2 -m64` and run the binary and get results slower than Java. I'll eyeball it as C being 80% the speed of Java. Of course, I'm sure this will vary from problem to problem and some things may optimize better on one system or the other. On the plus side, I also ran a binary from 2006 based an an unknown gcc version of that era. It was slower than the latest gcc. At least that's getting better.
