We should forget about small efficiencies, say about 97% of the time: premature optimization is the root of all evil. Yet we should not pass up our opportunities in that critical 3%.Famous computer scientist Donald Knuth
His quote is often clipped down to just “premature optimization is the root of all evil”, but that is NOT TRUE.
Yes, don’t micro-benchmark everything when you begin a project, but please do think ahead. And once you’ve got the program working as intended, then make it faster by profiling and looking for the bottlenecks.
Choose the proper data structures. Array? List? Hashset? Queue? Dictionary? Caching?
Take a few moments and think it over.
How often will this line of code be ran? Do I need to inline it, or pull it out into a separate method?