Matt BirchlerSep 15, 2024
Everyone says Chrome devastates Mac battery life, but does it? I tested for 36 hours to find out. https://birchtree.me/blog/everyone-says-chrome-devastates-mac-battery-life-but-does-it-i-tested-for-36-hours-to-find-out/
Everyone says Chrome devastates Mac battery life, but does it? I tested for 36 hours to find out.

Long time Birchtree readers know I love data, and love data even more when I can throw it into a graph. I’m also a fan of testing things that everyone generally agrees are true, but no one seems to have any data to back up. That brings us to

Birchtree
Show threadMae, gaeSep 15, 2024
@matt Okay, so you asked for feedback on how you did it wrong, all wrong, so here's some (sorry if that was sarcasm) First off, you're doing battery percentage tests, and as you have seen, those do not correspond neatly to actual charge on the battery (the os fudges them). You could have used the more detailed stats, although idk if mac os has those neatly accessible. But the real problem isn't that; 1/2
Show threadMae, gae

@matt The real problem in my opinion is that you should have used error analysis. That would have compensated for use of battery percentage (most of the time). Just calculating the standard error of your experiment would have a way more significant result. (you can still do that, if you want) A simple t-test would have helped your argument a lot.

But I do really like the idea of actually testing this instead of following people's guts. Thanks for that! 2/2

Sep 15, 2024 at 9:38amWeb
Show threadYury MolodtsovSep 15, 2024
@deciMae @matt Yep, considering the effort involved in this test anyway calculating errors would be just perfect
Show threadYury MolodtsovSep 15, 2024
@deciMae @matt I've also seen this interesting test
https://www.reddit.com/r/macapps/comments/12n7162/part_3_final_browser_energy_efficiency_benchmarks/
Show threadMatt BirchlerSep 15, 2024
@deciMae Interesting, I didn’t know of this concept before now, so thank you. I did note the variability and tried to extensively explain how I attempted to reduce that variability skewing the overall results, but as a real amateur over here, I get that might not stand up to strict scrutiny.