Jean-Baptiste Kempf

Once in a while, it never gets old, the speed improvements to show 62x.

There are months, on a personal level, I run our internal test suite at work and just see.

I'm still in awe of the gains we have.

Yeah, but ARM in some ways is more... x86 with out-of-order execution, it's not so bad.

ARM, you really need to understand all the different generations of ARM processor because they're all different.

There's...

a72 etc etc and there's the apple variant there's this variant there's that and you need to write code that works efficiently on all of x86 well broadly speaking you have intel amd you have sub variants but generally speaking there's something fast is going to remain fast on all of the variants whereas an arm is a completely much more complicated ball game

12, 14, something like that.

Yeah, and then they switched back.

To be fair, on Apple, when you type GCC, you get CLang.

They did something like that as well.

All the crazy edge cases of XML under ridiculous circumstances, and they get attacked by security researchers because there's one other crazy edge case that they haven't thought of.

I was like, yeah, but...

The body of knowledge to actually resolve that is massive.

In the security community, they certainly did.

That was one of the things I think that argument came out is,

There was a portion of the security community that said, no, these guys write crap code.

They need to fix their crap code.

I'm like, no, no, no, no.

This is a guy's hobby project.