3 ms·

Clever code that's actually significantly faster is fine as long as you document every single thing that's clever about it, even if that means writing 20 lines

by xmprt 6y ago

Clever code that's actually significantly faster is fine as long as you document every single thing that's clever about it, even if that means writing 20 lines of comments for a 2 line function. The worst thing is someone who writes a clever one liner without any comments because that would "ruin" the elegance of the one liner and because their code is "self-documenting"

woutr_be 6y ago
> The worst thing is someone who writes a clever one liner without any comments because that would "ruin" the elegance of the one liner and because their code is "self-documenting" Sadly this was some of my colleagues, no documentation and clever one liners without comments and often obscure naming conventions. And again, we didn't need the performance benefits for our use case. I think in the end it was just a play to keep some people employed, because without them we wouldn't be able to work on this part of our application. Sadly for them the business decided to outsource and they ended up being forced to switch teams.
njharman 6y ago
Faster code is not fine. Faster code should only be written after finished product is shown to be too slow for purpose and after profiler has shown which parts of code are worth optimizing.
- username90 6y ago
  You should always have a good idea of what part of the codebase will be the bottleneck when you write it. If you have that just optimize the things it spends time on. The only time you really need a profiler before optimizing is when you are micro optimizing to try to handle cpu ordering etc, other than that performance is mostly predictable.
- meheleventyone 6y ago
  This can be too late for the product though. If you have real performance requirements for example you’re making a game and it needs to render at 60Hz optimising at the end can be an impossible task. This is also further complicated when data dependencies also impact performance. For example level designers need some understanding of the complexity of the scenes the game will be able to support.

by xmprt 6y ago

woutr_be 6y ago

> The worst thing is someone who writes a clever one liner without any comments because that would "ruin" the elegance of the one liner and because their code is "self-documenting" Sadly this was some of my colleagues, no documentation and clever one liners without comments and often obscure naming conventions. And again, we didn't need the performance benefits for our use case. I think in the end it was just a play to keep some people employed, because without them we wouldn't be able to work on this part of our application. Sadly for them the business decided to outsource and they ended up being forced to switch teams.

njharman 6y ago

Faster code is not fine. Faster code should only be written after finished product is shown to be too slow for purpose and after profiler has shown which parts of code are worth optimizing.

username90 6y ago
You should always have a good idea of what part of the codebase will be the bottleneck when you write it. If you have that just optimize the things it spends time on. The only time you really need a profiler before optimizing is when you are micro optimizing to try to handle cpu ordering etc, other than that performance is mostly predictable.
meheleventyone 6y ago
This can be too late for the product though. If you have real performance requirements for example you’re making a game and it needs to render at 60Hz optimising at the end can be an impossible task. This is also further complicated when data dependencies also impact performance. For example level designers need some understanding of the complexity of the scenes the game will be able to support.