this post was submitted on 05 Feb 2024
83 points (88.1% liked)

Open Source

33127 readers
456 users here now

All about open source! Feel free to ask questions, and share news, and interesting stuff!

Useful Links

Rules

Related Communities

Community icon from opensource.org, but we are not affiliated with them.

founded 5 years ago
MODERATORS
 

I was looking for a non-chromium alternative browser to Firefox and found mercury. According to this site it is one of the fastest Firefox forks and also has optimizations from other well known forks like librewolf.

top 33 comments
sorted by: hot top controversial new old
[–] [email protected] 43 points 1 year ago* (last edited 1 year ago) (1 children)

Claims 7-10% performance improvement on an old AMD FX thing. No information about the baseline though; whether it's the terrible Snap or Firefox's official binary package. I suspect it's the former because it has known performance issues IIRC and the latter has quite good compiler optimisations already (LTO+PGO making most of the difference).

When I built Firefox for x86_64-v3, I saw no measurable improvement over x86_64-v1 in speedometer. I didn't dare to build the most security critical application on my system with unsafe compiler optimisations though..

[–] [email protected] 3 points 1 year ago (1 children)
[–] [email protected] 13 points 1 year ago (1 children)

Everything involving -O3. That usually stands for "enable all standards-compliant compiler optimisations, no matter how how little their benefit or their stability".

What would be even worse would be -Ofast which won't even care about strict standards compliance. No sane distributor distributes -Ofast and the only distributor I'd trust to use -O3 correctly is Intel's Clear Linux.

I don't know about OPT_LEVEL but it's likely an abstraction of the build system for this flag.

[–] [email protected] 6 points 1 year ago

Interesting. Thanks!

[–] [email protected] 23 points 1 year ago (2 children)

After the shenanigans with that dev and his other project, Thorium. No fucking thanks.

[–] [email protected] 24 points 1 year ago (1 children)

Out-of-the-loop here, what shenanigans?

[–] [email protected] 30 points 1 year ago (2 children)

The developer included furry porn art as part of the code base of his other browser project, Thorium. He removed it after it was discovered.

[–] [email protected] 8 points 1 year ago

Furries hold up the entire IT field though

[–] [email protected] 6 points 1 year ago (2 children)

Do I really want to know what furry porn is?🤨

[–] [email protected] 11 points 1 year ago (1 children)

Its an art form depicting humanoid animals engaging in sexual activities

[–] [email protected] 4 points 1 year ago (1 children)

Oh well - why should one integrate this in a browser lol? Jesus sometimes I regret sharing something I think is good at first glance - but well I think I will then go back to librewolf

[–] [email protected] 2 points 1 year ago

Mercury is ESR and not privacy focused. The dev found nice build flags though. Using Librewolf for privacy is always advised, if you dont compile FF yourself and use it with arkenfox and policies

[–] [email protected] 6 points 1 year ago (1 children)

Don't ever go to 4chan. Ever. You will lose your innocence.

[–] [email protected] 0 points 1 year ago

Thanks for pointing it out 😂

[–] [email protected] 4 points 1 year ago (1 children)

Could you elaborate? I'm not familiar with the drama.

[–] [email protected] 18 points 1 year ago (1 children)

just make sure to download one of the newer versions without the furry porn

[–] [email protected] 44 points 1 year ago (1 children)

Where can I get the versions WITH the furry porn

[–] pastermil 8 points 1 year ago

Easy, you start with the one without and add it yourself.

[–] [email protected] 15 points 1 year ago (1 children)

It is based off ESR but the flags are really cool!

You can build Firefox yourself, takes like 30min on a modern Laptop.

I did that to remove the jemalloc memory allocator, to make it work on Secureblue

Just havent found how to compile regular Releases yet.

[–] [email protected] 6 points 1 year ago

All hail the mighty AUR ;)

[–] [email protected] 11 points 1 year ago* (last edited 1 year ago) (2 children)

Tried this myself, performance differences are non existent. In fact I noticed more regressions on speedometer than improvements.

Don't bother, use Floorp instead.

[–] [email protected] 2 points 1 year ago

Interesting, because I saw a 20 point increase between vanilla Firefox and Mercury when testing last night.

[–] [email protected] 1 points 1 year ago

Floorp looks nice in theory but in practice it is very outdated...

[–] cyanarchy 9 points 1 year ago (1 children)

Been using it, absolutely cannot recommend.

[–] [email protected] 1 points 1 year ago (1 children)

Any reasons why you can't recommend it?

[–] [email protected] 4 points 1 year ago

I mean, "it's shit" is technically discussion but I was hoping for more too lol

[–] [email protected] 6 points 1 year ago

You won't notice any difference whatsoever.

[–] [email protected] 3 points 1 year ago* (last edited 1 year ago) (1 children)

Does this enable x86-64-v4 (AVX512) optimisations? I've checked the github page but don't see any references to x86-64-v4.

[–] [email protected] 4 points 1 year ago

https://github.com/Alex313031/Mercury/blob/main/mozconfigs/mozconfig

You can likely use Mozillas documentation to optimize your Firefox that way

[–] [email protected] 0 points 1 year ago