Announcement

Collapse
No announcement yet.

A Quick Look At The Firefox 66.0 vs. Chrome 73.0 Performance Benchmarks

Collapse
X
  • Filter
  • Time
  • Show
Clear All
new posts

  • A Quick Look At The Firefox 66.0 vs. Chrome 73.0 Performance Benchmarks

    Phoronix: A Quick Look At The Firefox 66.0 vs. Chrome 73.0 Performance Benchmarks

    Given the recent releases of Chrome 73 and Firefox 66, here are some fresh tests of these latest browsers on Linux under a variety of popular browser benchmarks...

    http://www.phoronix.com/scan.php?pag...-73-Benchmarks

  • #2
    Firefox clang vs gcc?

    Comment


    • #3
      Clear linux just re-packages the official firefox release tarball, it doesn't build it from source.
      https://github.com/clearlinux-pkgs/f...r/firefox.spec

      (Firefox PGO requires access to X server (xvfb), dbus and drm devices as firefox has to run during the PGO build. Clear linux's build infrastructure may not be set up for this. If I had to guess It's also way easier to repackage as matching the perf from the official tarball from mozilla is possible but quite difficult.)

      Comment


      • #4
        Wow that's quite a bummer. I'm really looking forward to see if webrender makes a difference there. Although it might still take some time until it's enabled by default on linux

        Comment


        • #5
          Hopefully the performance of Firefox will soon improve by upcoming WebRender.
          It is experimental and can enabled in about:config with the gfx.webrender.all key.

          Comment


          • #6
            Originally posted by xpris View Post
            Firefox clang vs gcc?
            Rather than clang vs gcc, I'd be willing to bet some of Firefox's performance shortcomings are a Windows vs. Linux issue. Go read bugzilla issues around enabling performance related features - multi-threading, webrender, GPU acceleration, etc. and devs are continuously refusing to turn on the Linux defaults. GPU drivers for Linux still get blamed, and while Google is trying to improve the situation with drivers (incentivized by ChromeOS), Mozilla just claims it's not their fault and moves on.

            If I'm sounding overly harsh to Firefox, that's not my intent. I actually prefer Firefox since it uses less memory than Chrome, has been less crash prone, and doesn't ship my data off to Google. Even if it is slower on a few render benchmarks, Firefox still runs great on my slow quad core AMD APU laptop.

            Comment


            • #7
              In real world situations in which specific cases is there a noticeable performances difference?

              Comment


              • #8
                As others said above, it's interesting to identify the reasons for the performance difference. Is it due to Chrome supporting GPU acceleration by default, does it have anything to do with relative JavaScript performance etc.

                Comment


                • #9
                  Originally the intent was to look at the performance of Clear Linux's default Firefox build compared to the upstream Firefox x86_64 Linux binary to see how the performance differed or if any Clear optimizations paid off there. In this particular instance, Clear Linux's Firefox build and that of the upstream/generic Firefox Linux binary basically came down to the same.
                  Of course, that would be the case since the clearlinux team just repackages the upstream pre-built firefox binary.

                  https://github.com/clearlinux-pkgs/f...r/firefox.spec
                  Optimizing Firefox #329

                  Comment


                  • #10
                    Originally posted by xpris View Post
                    Firefox clang vs gcc?
                    Better a comparison between Firefox,Chromium,Opera,Vivaldi
                    So that we can see if any differences between Chromium based Browsers to Firefox..

                    Comment

                    Working...
                    X