3 ms·
That’s weird, this article about performance didn’t begin with the design of the experiment and benchmark. In fact, unless I missed something, it was completel
by binary132 2y ago
That’s weird, this article about performance didn’t begin with the design of the experiment and benchmark. In fact, unless I missed something, it was completely omitted. It must not be an article about performance at all, then.
- ameliaquining 2y agoYou're being unnecessarily dismissive, but you've got a point. I do think it would have been good to include actual numbers.
- binary132 2y agoIn my opinion discussions on performance theory are so subtle and can be so misleading that you really need to model and demonstrate e.g. reduced instruction count, memory reads, or other metrics in order to prove out your claims. Otherwise it’s nearly useless. The Go blog post about performance tuning versus C++ is a great model to follow. The claim in this article is not 2x, it’s “50% of CPU was consumed by this loop, now it’s ~0%” I’d love to see a series of improvements and how they impact the machine behavior / speedup....