Mozilla GFX Blog

I led the Stylo project to integrate Servo's CSS engine into Firefox as part of project Quantum [1]. I have the utmost respect for the engineers on the Servo team, and am sad to see them go (though I am certain they will have no shortage of opportunities as to what to work on next).

Servo had two major roles within Mozilla - as an incubator for novel browser technology we wanted to ship in Firefox, and more recently, as a lighter-weight vehicle for Mixed Reality products. The latter has been the focus for the last three years, and those products now appear to be winding down. But the former was a huge success - both Servo's parallel CSS engine and its GPU-based graphics layer are now shipping in Firefox.

While it seems unlikely that Mozilla will continue to prototype things in Servo, we're still building lots of innovative technology (and writing lots of Rust code) directly in Firefox. A few of the teams have blogged recently about the work they're doing [2] [3], and I'd encourage anyone interested to check it out.

The Servo team accomplished a ton and left its ongoing mark on the Web. These changes are tough for everyone within Mozilla, but are not indicative of any change in strategy for Firefox. Gecko is alive and well, and there are no plans to switch to Blink.

[1] https://bholley.net/blog/2017/stylo.html
[2] https://mozilla-spidermonkey.github.io/blog/
[3] https://mozillagfx.wordpress.com/

You may be thinking of Core Animation usage, which is really good to see. As I understand it this is also a prerequisite for WebRender.

The bug in question: https://bugzilla.mozilla.org/show_bug.cgi?id=1429522

A blog post breakdown: https://mozillagfx.wordpress.com/2019/10/22/dramatically-reduced-power-usage-in-firefox-70-on-macos-with-core-animation/

This does make a difference, but macOS is in many respects a "death by a thousand cuts" platform due to the UI/UX bar being fairly high. It notably clips in many areas where Chrome and Safari simply don't, which leads to a jarring experience. Scrolling and resizing is still not smooth, even after tweaking it behind the scenes - although to be fair, the post above calls out that they're aware they can make this smoother with CA.

Firefox on macOS also still has issues with hardware acceleration. Comparing the following URL across browsers I can achieve 60fps, no drops, on Safari. Firefox on my Mac averages 45fps.

https://webglsamples.org/aquarium/aquarium.html

I love Firefox (and the associated lineage) - I really do. I've kept it installed since the Mozilla Suite days. I still check every damn release to see if I could switch back. I'm sad every time I find I can't.

We're currently rewriting most of the graphics rendering of Firefox (with what is called the WebRender project). This is the first time Linux users will have hardware accelerated rendering by default, and is a pre-requisite to take real advantage of hardware-accelerated video decoding. It's enabled on some configuration in Nightly builds [0], and it's being worked on actively. If you want to know more, the graphics teams has a blog where updates are periodically posted about what is happening, often with interesting insight about the innards of a web browser rendering stack [1].

Once rendering is hardware accelerated, it makes sense to look into hardware video decoding. Doing hardware video decoding without hardware rendering roughly means we need to read back from GPU memory into main memory, composite the image in software, and then upload back to the GPU to display, which is super super inefficient.

[0]: https://wiki.mozilla.org/Platform/GFX/WebRender_Where
[1]: https://mozillagfx.wordpress.com/

Firefox and Gecko has a lot of tech debt, and it's big. It'd be too risky to just try to throw big chunks of it out and hope that the replacement will be great. Though the momentum seems lost, the WebRender integration is still ongoing. ( https://mozillagfx.wordpress.com/2018/11/29/webrender-newsletter-32/ ) And it's big, it's complex, it's hairy, the web is a mess. So, it's slow and full of terrors. (Shaders, caching, multi-threading, mobile, multi-core, C++ and Rust interfacing, etc.)

The next thing seems to be Fission, which is the continuation of the e10s (electrolysis) content process model into full/more Site Isolation. ( https://wiki.mozilla.org/Project_Fission - but it seems hard as fuck: https://mail.mozilla.org/pipermail/firefox-dev/2018-July/006628.html nowadays they seem to be stripping out the old Gecko XUL stuff and replacing it with something newer and better: https://mail.mozilla.org/pipermail/firefox-dev/2018-November/006903.html )

One thought, cant one reason to the difference when you checked be that Firefox is the active application and have focus? And do you maybe have any addons in Firefox? Extensions can give bigger energy impact, you can check that under about:performance. Anyways I think both have low energy impact even if safari happens to be abit lower.
Personally I use Firefox on mac and can many times have 100+ tabs and it works without any problems.

It could also be mention that Firefox now use macos coreanimation that drastically improved Firefox battery performance [1]. So give it a real test again, Firefox is really awesome and needs all support :) We need bigger diversity among webbrowsers, not just chromium based.

[1]
https://mozillagfx.wordpress.com/2019/10/22/dramatically-reduced-power-usage-in-firefox-70-on-macos-with-core-animation/

It's hilarious how a macOS-lookalike desktop environment is using Linux-like technologies like Openbox and KWin, rather than building something in Cocoa. I suspect it will cause trouble trying to bodge KWin to have Mac-like window management, or support apps that integrate with Apple's graphical acceleration and compositing frameworks (Core Animation? https://mozillagfx.wordpress.com/2019/10/22/dramatically-reduced-power-usage-in-firefox-70-on-macos-with-core-animation/).