3 ms·
That's partially true but remember that you're generalizing from micro-benchmarks – which did not universally tilt Chrome – to the complex behavior of real site
by acdha 9y ago
That's partially true but remember that you're generalizing from micro-benchmarks – which did not universally tilt Chrome – to the complex behavior of real sites, where networking, memory usage, and general de-optimization tended to cancel out many of those small wins.
That's not saying that the browser teams haven't made big strides, only that it's not such a simple story. Chrome running a certain JavaScript function 15% faster isn't a game changer if a side is limited by layout or poorly structured DOM interactions.
- twobyfour 9y agoAren't micro-benchmarks kind of pointless if they don't reflect real-world behavior? The fact is that javascript execution (not networking behavior but javascript performance) on many web apps differs by orders of magnitude between Firefox and Chrome. There are a handful of sites that I find literally unusable in Firefox on a new Macbook Pro because even after loading, sitting in the background doing nothing, they make the entire computer unresponsive due to high CPU usage. These same sites work just fine in Chrome and Safari.