Very well, now look at the other results. The solution both I and kvale proposed is the most straightforward, arguably most maintainable, and least "creative" one. It comes in third, much faster than the slowest, very "clever" solution, and achieves over 80% of the speed of tye's winning entry. Do you see anything wrong with your picture yet?
The reality is that "boring" solutions chosen with a clear understanding of the task at hand are usually plenty fast enough. In my experience, they tend to achieve close to 85% of the possible performance.
Given the points already made by tye about the reality of the impact of microoptimizations, there is very little worth in evaluating the obscure. Go with the maintainable, clear, concise solution and worry only if performance is actually insufficient.
And even if you have gotten there, it is still more worthwhile to optimize your expensive_functions instead.
Makeshifts last the longest.
In reply to Re^4: Matching First Character of Strings Efficiently (benchmarking)
by Aristotle
in thread Matching First Character of Strings Efficiently
by Limbic~Region
| For: | Use: | ||
| & | & | ||
| < | < | ||
| > | > | ||
| [ | [ | ||
| ] | ] |