1) I'm running a MacBook Pro (Retina, 15-inch, Mid 2015)

2) My display options are set to "more space"

3) I loaded this hackernews thread.

4) Firefox and it's child processes are using <1% cpu.

If I aggressively scroll this page I can push the cpu usage up. We have a variety of fixes in the pipeline that will help with this (https://bugzilla.mozilla.org/show_bug.cgi?id=1429522, https://bugzilla.mozilla.org/show_bug.cgi?id=1265824). That being said, only 4.7% of our users are on MacOS so it's more difficult to justify prioritizing work specific to that platform.

4.7% of your users, but what percentage of web developers?

If that second number is substantially higher (and I personally suspect it is, but I don't have data backing that up) I think it would be very reasonable for Mozilla to prioritize MacOS. Getting web developers to use firefox makes firefox work better for everyone because it makes more sites work well with firefox.

Yeah, we do prioritize Mac disproportionately to its market share for that reason. It does sometimes happen that we have to make hard calls though, and that's what Jeff is referring to.

On that note, one of the key motivators behind WebRender (our new graphics backend) is that it provides hardware acceleration across all platforms, including mac (whereas our current Direct2D acceleration works only on Windows). So Mac graphics performance should generally improve once we get that shipped.

I build internal tools in my organization. Most of our developers are on MacOS and use Chrome day to day. Most of our users are on Windows. Whenever a browser specific issue comes up, we do our best to address it, but our first response is typically "go install Chrome". Chrome adoption is 97% internally, 2.5% Safari, and the remaining 0.5% is Edge. We have a 0% internal Firefox usage rate.

Draw from that what you may, but I bet that dynamic is happening in other places.

4.7% of your users... and how much of that is because you are finding it difficult to prioritise work specific to that platform? It sounds like a chicken-and-egg problem to me.

First off, thank you for all of your hardwork. I'm grateful for the existence of Mozilla.

However, I would like to ask... 4.7% translates to how many millions of people?

Now don't get me wrong: I'm a big fan of Firefox, and have used the dev edition as my main work browser since Quantum came out. However, that's a terrible attitude to take. You have half a billion dollars per year of revenue. The second-largest desktop platform should be a major priority. Certainly above all of the other non-core-Firefox projects you're working on, however cool they may be.

I also have this problem, and it's a reason not to use Firefox for me. Maybe you only have 4.7% of users on MacOS precisely because of problems like these?

I spent a few minutes trying to replicate on a MacBook Pro (Retina, 13-inch, Late 2013) with Firefox 62.0.2 and could not replicate, browsing a number of resource intense sites.

Interesting, as this is the same MBP I have, and same Firefox version. I now have a perf file showing the issue. Can't upload it (it gives an error), but would be happy to send it to anyone who thinks they can help.

You can send it to me at [email protected]

I'm on Mac OS 10.11.6 (15G22010). The regression might be due to some interaction between Firefox and changes in the OS, if you're on a more recent version.

I've been doing this since the first-gen rMBP, and I've never had this issue. No matter my scaling, I idle at about 5% CPU with about 80 tabs and 10 windows actively open.

I'm not sure what it is with your MBP but my Firefox does the same thing anytime I open a phys.org page -- one Firefox at work using Ubuntu 16.04 and the other at home using Windows 7 both exhibit it.

Edit: Actually I just checked and the Windows 7 machine doesn't do it any more. I'm not sure if maybe a Firefox update fixed that here. I'll double check at work tomorrow too

Can't reproduce that here (on either of two different machines)-- something else may be going on on your machine.

I don't have a rMBP, and others who may be curious to help may not either. Uploading performance profiles from various websites to the mentioned URL will give everyone something to poke at.