As a xoogler, I was thinking the same thing. I assume they knew from the get-go, because well, I can't see how they wouldn't. But I guess it doesn't matter all that much, anyway.
>The most relevant web benchmarks today are Speedometer, MotionMark, and Jetstream.
Being a Mid-Tier JIT, am I correct to assume ( I may be wrong ) it brings very little benefits to my day to day browsing if most of my usage are Web page only? My only JS heavy site is Feedly, second being Twitter.
I am thinking if there are any benchmarks for testing JPEG, Photos, HTML and CSS Rendering responsiveness and speed.
Actually this concern is exactly why we built Maglev: A very fast JIT that can generate decent enough code very early so it can help even if you're not running tight benchmarks. In fact, the JIT by itself isn't particularly good at the traditional benchmarks yet. It's just very fast, and the code is most often good enough for real browsing usecases.
Lots of performance benchmarks but no measurements for power efficiency. Granted this is not easy, but it's more of a requirement than developing more performance benchmarks IMHO. My guess is that most people would rather Chrome improve power efficiency as a priority, which is why many use Safari on the Mac instead of Chrome.
They do mention reduced power consumption in the opening paragraph. Usually executing the code you want to execute faster also translates to the CPU sleeping earlier again and this increased battery life.
19 comments
[ 3.2 ms ] story [ 48.2 ms ] threadFunny name clash: https://research.google/pubs/pub44824/
[1] https://maglev.github.io
Being a Mid-Tier JIT, am I correct to assume ( I may be wrong ) it brings very little benefits to my day to day browsing if most of my usage are Web page only? My only JS heavy site is Feedly, second being Twitter.
I am thinking if there are any benchmarks for testing JPEG, Photos, HTML and CSS Rendering responsiveness and speed.
https://chromium.googlesource.com/v8/v8/+/refs/heads/main/sr...