2 ms·

Sounds like a good call to me. If you have something mission critical that's tried and tested, and reliably does 99% of what you need it to, you don't mess with

by taneq 5y ago

Sounds like a good call to me. If you have something mission critical that's tried and tested, and reliably does 99% of what you need it to, you don't mess with it. It's very easy in software dev to get caught up in software optimizations, and it's a lot harder to quantify the risk of continuing to mess with something the closer you get to ship date. Of course it "shouldn't make any difference" but who among us hasn't at some point been bitten by a "that couldn't possibly affect... ohhh. Ohhhhhhh." style bug? When you're dealing not just with a single piece of software, but an entire hardware+software ecosystem, you don't risk jeopardizing the whole stack for an extra few FPS in some back corner.

bombcar 5y ago
Especially in an era where a buggy release could require an expensive return/reprinting step. Back in those days it had to ship as close to perfect as possible because you couldn’t just send out a patch layer.
- anthk 5y ago
  -O2 was considered "OK-ish". With -O3 often you get a can of worms.
  - bombcar 5y ago
    Yeah, and most development houses would test and develop with -O2 on. The problems came if you ended up thinking you might need to switch near the end.