This article is about a refinement of Google's Octane benchmark four years ago. In the meantime, the benchmark has been abandoned. It has been replaced by (guess what!) Octane 2.0. Also see this interesting article about the transition to Octane 2.0. Also, read the comment: while adding little information to the main article, it provides an insider's view of things.
Google enhances its Octane Benchmark measuring Javascript performance in an interesting way. It adds two benchmarks measuring technologies developed by Google's competitors: Typescript has been developed by Microsoft, while asm.js is a Mozilla project aiming at speeding up a subset of Javascript. Google does not support the asm.js. According to them, it's better to optimize the entire Javascript runtime instead of focusing on a subset of it. In my eyes, this is a convincing argument: asm.js code is designed to be generated by compilers, not by humans. As for now, Javascript compilers like Dart, Ceylon, Kotlin or Typescript haven't had that much of an impact. The majority of Javascript code is written by humans.
Thus it's not surprising Chrome 30 performs the asm.js test worse than Firefox 25 (22561 points vs 40039 points on my machine). Google adding an asm.js test shows the company is aware of the asm.js even if it doesn't support it actively.
Another interesting improvement is a test measuring latencies caused by Javascript compilation and garbage collection.
Further reading
Benchmark results on my PC
For the sake of completeness, I add the benchmark results on my machine. You have to take the results with a grain of salt: they are by no means scientific. I didn't take them under clean-room conditions. Nonetheless, they show the relative performance of the browsers as Google sees them. It goes without saying that the competitors' benchmarks tell a slightly different story.
Internet Explorer 10:
Chrome 30:
Firefox 25:
