Benchmarking Mozilla's Firefox Performance Over The Past Two Years

[phoronix]

Phoronix: Benchmarking Mozilla's Firefox Performance Over The Past Two Years

With 2019 quickly drawing to an end, I figured it would be interesting to see how the performance of Mozilla Firefox has been trending over the longer term. So for this article today is a look at the Firefox 57 through Firefox 71 stable performance plus tests of Firefox 72 beta and Firefox 73 alpha all from the same system and using a variety of browser benchmarks.

http://www.phoronix.com/vr.php?view=28641

[Cape]

These benches just confirm my feelings.

I wonder what Mozilcuck is doing with all those donation money?!

[rene]

These benches just confirm my feelings.

I wonder what Mozilcuck is doing with all those donation money?!

sync, pocket, "premium", maybe vpn (in testing?) and I guess a party or two, ... :-/

[Charlie68]

I've always had a contrasting relationship with Firefox, as it has always had problems on my PCs. I'm not a performance fanatic, but it has to work fine and it doesn't work well on my PCs. I often see the CPU working abnormally and from a glance of the system monitor, it tells me the "web" item to grind cpu. All this happens without reason, no multimedia content, simple web pages of the forum standard. This never happens with Chromium or very rarely.

[kozman]

Far be it for me to be a Mozilla apologist, but in looking at these bench graphs and coinciding them with what changed after FF60, I'd point you to https://www.phoronix.com/scan.php?pa...-All-Platforms.

I'm not implying it's causal but that's probably the main thing that changed after FF60 that *could* be responsible. All compilers are not created equal regardless of how artificial benchmarks may portray them. Real world use is where it counts.

I've been reading Phoronix for more than a decade and have followed LLVM vs. GCC progress through Michael's benchmarking stories and I can't recall many instances where LLVM outright trounced GCC across the board. I don't know why Mozilla switched from GCC, to (I think) VS and then to LLVM in such a short span of time but as an outside observer, I always thought it was very disruptive. Like stepping over dollars to save pennies. </2 Cents>

[Ray_o]

These benches just confirm my feelings.

I wonder what Mozilcuck is doing with all those donation money?!

donations don't go to Firefox development https://donate.mozilla.org/en-US/faq#item_8.

[NateHubbard]

donations don't go to Firefox development https://donate.mozilla.org/en-US/faq#item_8.

That's really too bad. Other than Firefox, I don't care about whatever else they are doing.

[eltomito]

[...] I don't know why Mozilla switched from GCC, to (I think) VS and then to LLVM in such a short span of time but as an outside observer, I always thought it was very disruptive. Like stepping over dollars to save pennies. </2 Cents>

So, how about building Firefox with GCC and comparing the benchmarks? Michael?

[NateHubbard]

So, how about building Firefox with GCC and comparing the benchmarks? Michael?

That might be easier said than done, since their build page seems to indicate you need llvm/clang. I guess it would depend on how much effort you wanted to put into it.

[Ray_o]

Far be it for me to be a Mozilla apologist, but in looking at these bench graphs and coinciding them with what changed after FF60, I'd point you to https://www.phoronix.com/scan.php?pa...-All-Platforms.

I'm not implying it's causal but that's probably the main thing that changed after FF60 that *could* be responsible. All compilers are not created equal regardless of how artificial benchmarks may portray them. Real world use is where it counts.

I've been reading Phoronix for more than a decade and have followed LLVM vs. GCC progress through Michael's benchmarking stories and I can't recall many instances where LLVM outright trounced GCC across the board. I don't know why Mozilla switched from GCC, to (I think) VS and then to LLVM in such a short span of time but as an outside observer, I always thought it was very disruptive. Like stepping over dollars to save pennies. </2 Cents>

How you came to this conclusion from looking at the graphs ?
They switched to clang in firefox 64, when i compare firefox 64 to 63, i don't see anything that stand.

Another thing, they didn't jump from GCC to MSVC, MSVC doesn't work in mac or linux, they had been using MSVC to build for windows, GCC to build for linux and clang to build for mac before jumping to clang in all platforms.