Comment by pojntfx
1 day ago
Don't see why it wouldn't - as long as pstate etc. works it should be the same. I'd argue it's probably better given that modern desktops use far less resources in the background compared to Windows
I bet they don't publish Linux numbers because it depends on which desktop you use etc.
A lot of office workers these days spend a lot of time in video calls.
So to get the best battery life you need, for example, your browser to use GPU-accelerated video encoding and decoding.
Linux is something of a second-class citizen for both GPU vendors and browser vendors. So for example if you're using Firefox and an nvidia GPU on Linux? No video encode/decode acceleration for you. The browser will silently switch to CPU decoding.
This translates into worse battery life.
HW video decoding is now available and by default on in Chrome on at least Ubuntu with my Intel iGPU. I was also surprised when they turned it on under the radar. I saw this the other day debugging a problem and saw others see it too: https://www.reddit.com/r/Ubuntu/comments/1ojydv9/comment/nm8...
HW decoding works fine. But some distros (looking at you Fedora) have legal issues around providing it out of the box.
Call me crazy, but most people working typically leave their laptops wired in to either a charger or a hub so they can have more monitors. I know some people will go through the effort of charging and pulling the cord, and charging later, but most people don't want to micromanage something they can forget about while working. If you're living on battery life for a work call, it would not matter if you're on Windows, changes are high your batterly life will self-terminate quicker than you realize.
Among Linux users, long battery life is for in-office workers (who leave their desk to attend meetings) in hybrid companies (where no meetings are laptop-free) in roles that sometimes involve back-to-back meetings.
4 replies →
CPU decoding/encoding for video means warm chassis + spinning fans. Fan noise is very annoying with video calls.
Firefox has had GPU video decoding in Linux on by default since 2023 for Intel and 2025 for AMD from what I've read
I don't understand why. From what I can tell, some of this is remedied just by changing feature flags in Firefox, which (if correct) would mean it ships with the capability, but decides not to.
There's other software that can do GPU acceleration in the repos, and there are plenty of distros that enable closed-source software. It's shocking to me how difficult it seems to be to get GPU acceleration working in Linux.
> I bet they don't publish Linux numbers because it depends on which desktop you use etc.
They ship with Ubuntu on it, which would be quite natural choice for such benchmark. Also they do do the standby test on Ubuntu for some reason.
Can't help but suspect there's a reason why Linux numbers are not given. :(
There have been fairly recent changes to the linux kernel to better support panther lake in terms of power performance. I'd suspect a major reason for holding back is because ubuntu 26.04 has not been released yet and it is using kernel 7.0 which includes these power improvements. 24.04 does not.
By the time these laptops start shipping, 26.04 should be released and testing should be easy. I suspect no major differences from it vs windows.
7.1 includes even more performance improvements for panther lake. [1]
[1] https://www.phoronix.com/news/Linux-7.1-Enabled-Intel-FRED
If I was releasing a laptop with Linux support as a key selling point, and the battery life was bad on Ubuntu 24.04 but good on the pre-release 26.04, then I'd advertise the good figures and write "tested on Ubuntu 26.04 beta, requires Linux 7.0 or later" in the footnotes.
I definitely /wouldn't/ rely on just Windows figures for a machine that's otherwise advertised as "Linux first". If the battery life was the same on both, I'd prominently mention that.
2 replies →