[oldmanjay]

I get that this is a translation but it's very poorly written even considering that. lots of repetitive conclusion jumping in what I can only assume is an attempt to baffle the reader into uncritically accepting what appears to be a largely unsupported opinion.

the only evidence the post seems to offer is that rust is slow because it showed up 5th in a set of microbenchmarks. everything else is rhetorical garbage, frankly.

[jcranmer]

Correction: a set of microbenchmarks whose own documentation emphatically states the difficulty of drawing any conclusions from those results.

Comparing different programming languages is notoriously difficult, particularly since the metric is underdefined. The best programmers might write faster code in language A compared to language B, but if average programmers write code that is no faster, does that mean language A is better than language B?

[igouy]

> the difficulty of drawing any conclusions from those results <

The difficulty of drawing general conclusions from program performance measurements -- not just "those results".

[plosi]

Additionally, even the microbenchmark barely supports it based upon the size of the error bars

[igouy]

Incidentally, those are the upper and lower quartiles not error bars.

[EugeneOZ]

This article is difficult to read in original (for native speaker) too. Just raw flow of minds, without any shape.

[igouy]

At this point, the only response needed to "rust is slow" claims is - rustc is beta.

Come back when it's cooked :)

[Manishearth]

Technically Rust will be 1.0 by the end of the week; but that won't magically make it faster.

The real point to be made here is that Rust can be optimized without dropping any design principles. It's a solvable problem. On the other hand, Java and Go's design puts a limit on the optimization you can use.

(Also, those benchmarks are a year old IIRC, Rust might actually have sped up in that time!)

[karmakaze]

Indeed for x64-one-core[1] the geometric mean is now closer to Fortran and Ada.

[1] http://benchmarksgame.alioth.debian.org/u64/which-programs-a...

[igouy]

> Also, those benchmarks are a year old IIRC <

Why didn't you check the commit age in the repo?