Recommended Posts

Is it possible to disable the "Hide http" feature?

There's an about:config entry to disable it. I read it from mozillazine. I will post it if I find it again.

To disable "hiding of http" feature , use this about config value and toggle it to false

browser.urlbar.trimURLs

There is a new landing in the UX branch

Expect TI in mc by 1st week of july

Firefox 6 to beta and Firefox 7 to aurora branch would happen on 5th July

I have updated the first page with many links to track the development , also , now that firefox has opted for faster release cycle , i dont think we would be creating more of these threads every quarter! Just like opera and chrome threads are there , i think firefox should follow the same. So henceforth all firefox discussion about PRERELEASE versions would go here, as it was decided at advent of the release mechanism

Azure (Implement Direct2D Azure Backend) landed in mozilla-inbound

https://bugzilla.mozilla.org/show_bug.cgi?id=651858#c101

Soon in mozilla-central.........Yuppeee!!!!!!!

very good! (Y)

i'm really amazed with nightly memory usage; once when i randomly opened the task manager to check the usage, it usually showed about 250 - 300 mb usage; but now it's usually around 140 - 170 mb, great work mozilla! :laugh:

So Azure has landed, and is enabled by default. On some things it gives massive gains, while on other things it barely helps. But more performance is always welcome.

Edit: Well, D2D backed <canvas>, there's still work to be done for general page compositing and cross-process stuff.

Starting with Today's Nightly build , when I hover over the icon in the taskbar, I can't see a preview. All I see is a busy Aero mouse pointer. Anyone else see this? Using Win 7 x 64

working fine here, also Win 7 x64

Switch on the setting of "Show tab preview in windows taskbar"(in tab options) and then look.

Edit: There's more people experiencing the same problem over here. http://forums.mozillazine.org/viewtopic.php?f=23&t=2236367&start=30

Edit: There's more people experiencing the same problem over here. http://forums.mozillazine.org/viewtopic.php?f=23&t=2236367&start=30

That was for Brando212, as he said he does not see the busy aero mouse pointer and you won't if that is NOT checked.

I hope electrolysis project will help Firefox in performance.

I am not super-optimistic though.

Currently there are a few major issues:

1. Firefox tab animations are very laggy compared to Chrome.

Loading new tab pages in a separate process (or at least separating the FF GUI process...) will help here.

Although, part of the issue seems just to stem from expensive repainting in general (new tab animation laggs opening a blank).

From visual analysis, Chrome tabs seem more simple.

2. Firefox expensively loads background tabs.

Firefox repaints the content in background tabs (correct me if I am wrong).

This is painfully seen loading a 2-3MB image in a background tab (while foreground tab is idle),

simply downloading something is not CPU intensive but Firefox uses 30% to load an img in background tab.

3. All loading tabs and foreground tab are required to share on virtual core.

The main component! The slower each virtual core is, the worse Firefox performance is.

This will not benefit single core systems, but will HT and better CPUs.

I have been using Firefox 7a for some time now, but it is painful because of the above.

Azure and TI are not as important. They are simple improvements to something that "works well", whereas process separation is an improvement to something that "works poorly"...

Edited by Udedenkz

i'm really amazed with nightly memory usage; once when i randomly opened the task manager to check the usage, it usually showed about 250 - 300 mb usage; but now it's usually around 140 - 170 mb, great work mozilla! :laugh:

Same here! So far ok but not yet convinced. Need to give it a good week before I am happy maybe two.

I hope electrolysis project will help Firefox in performance.

I am not super-optimistic though.

Currently there are a few major issues:

1. Firefox tab animations are very laggy compared to Chrome.

Loading new tab pages in a separate process (or at least separating the FF GUI process...) will help here.

Although, part of the issue seems just to stem from expensive repainting in general (new tab animation laggs opening a blank).

From visual analysis, Chrome tabs seem more simple.

2. Firefox expensively loads background tabs.

Firefox repaints the content in background tabs (correct me if I am wrong).

This is painfully seen loading a 2-3MB image in a background tab (while foreground tab is idle),

simply downloading something is not CPU intensive but Firefox uses 30% to load an img in background tab.

3. All loading tabs and foreground tab are required to share on virtual core.

The main component! The slower each virtual core is, the worse Firefox performance is.

This will not benefit single core systems, but will HT and better CPUs.

I have been using Firefox 7a for some time now, but it is painful because of the above.

Azure and TI are not as important. They are simple improvements to something that "works well", whereas process separation is an improvement to something that "works poorly"...

I highly disagree.

Though the improvements are fine by me but I am in no rush. Runs smooth as butter for me.

I hope electrolysis project will help Firefox in performance.

I am not super-optimistic though.

Currently there are a few major issues:

1. Firefox tab animations are very laggy compared to Chrome.

Loading new tab pages in a separate process (or at least separating the FF GUI process...) will help here.

Although, part of the issue seems just to stem from expensive repainting in general (new tab animation laggs opening a blank).

From visual analysis, Chrome tabs seem more simple.

2. Firefox expensively loads background tabs.

Firefox repaints the content in background tabs (correct me if I am wrong).

This is painfully seen loading a 2-3MB image in a background tab (while foreground tab is idle),

simply downloading something is not CPU intensive but Firefox uses 30% to load an img in background tab.

3. All loading tabs and foreground tab are required to share on virtual core.

The main component! The slower each virtual core is, the worse Firefox performance is.

This will not benefit single core systems, but will HT and better CPUs.

I have been using Firefox 7a for some time now, but it is painful because of the above.

Azure and TI are not as important. They are simple improvements to something that "works well", whereas process separation is an improvement to something that "works poorly"...

True , or just taking chrome out as a separate process would do enough to keep the UI fast

I'd definitely agree that the UI is sluggish. Hell, even running your cursor up and down the Firefox menu isn't as responsive as it should be.

Boot times are still pretty trashy for me, even on an SSD. :/

The UI on firefox 7.0a is fine for me but firefox still crashes sort of often for me. Maybe once a day, chrome just seems to glide. I use both of them anyway but mostly chrome. The tests are catching up with Chrome.

unled2nt.png

Palemoon hmm, they should make nightly builds too :p

I'd definitely agree that the UI is sluggish. Hell, even running your cursor up and down the Firefox menu isn't as responsive as it should be.

Boot times are still pretty trashy for me, even on an SSD. :/

Opens instantly on my ssd. I do agree the UI is sluggish compared to other browsers though.

I'd definitely agree that the UI is sluggish. Hell, even running your cursor up and down the Firefox menu isn't as responsive as it should be.

Boot times are still pretty trashy for me, even on an SSD. :/

What are your boot times ?

The UI on firefox 7.0a is fine for me but firefox still crashes sort of often for me. Maybe once a day, chrome just seems to glide. I use both of them anyway but mostly chrome. The tests are catching up with Chrome.

unled2nt.png

I never had Firefox 7.0a crash on me ever. It has had the occasional annoying bug or problem but no crashes. That is to be expected with a alpha build of a program though. It glides for me just like my friend's Chrome on his pc. Actually Firefox feels faster than Chrome now.

Never had 7.0a crash on me, either. In fact, I just checked and I've only had 7 crashes this year and I've been using Nightly/Minefield as my main browser the whole time (except for a week of trying Aurora).

I don't understand how it crashes so much on other people's systems.

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
  • Posts

    • Unsurprisingly, there's what the law says and what the old white wealthy males legally enforce...
    • Or anything online that requires an anti-cheat
    • Gf needed a new Surface and was looking at a Surface Laptop because of the Snapdragon. Seeing as it was a two year old chip she just decided to get a Lenovo Yoga 2 in 1 instead. Personally this Surface Ultra Cassis reminds me a bit of Razor. It would be interesting if it could handle proper gaming and be 17 inch.
    • No idea, frankly, I'm not into minimum requirements gaming, but it would be an interesting test to find out. Also, I just have to point out that it wasn't my intention to downplay the performance of DXVK on Linux or Linux gaming in general (despite my own experience being a bit of a mixed bag). I just thought it would be good to point out that DXVK is not Linux exclusive and that you can benefit from using it even in Windows.
    • Fastfetch 2.64 released bringing new logos and other improvements by David Uzondu Fastfetch, the popular command-line system information tool that developers created as a fast alternative to the classic Neofetch utility, has updated its codebase to version 2.64, bringing experimental scripting power, streamlined compilation options, a smarter logo renderer, and Codec module support. As noted earlier, Fastfetch can now detect hardware-accelerated video codecs across Windows, macOS, Linux, and Android through this new Codec module. On Linux and BSD, the utility uses VA-API by default, with a fallback to VDPAU on Nvidia hardware if compiled with libva and libvdpau. Windows users get D3D12VA on Windows 11 or D3D11VA with Media Foundation Transforms on older systems, while macOS relies on VideoToolbox and Android utilizes AMediaCodec. You can manually toggle Vulkan Video via the config file, and the program will report both encoders and decoders unless configured otherwise. Logo support for Quasar, Origami, Origami_small, NixOS2, and BerserkArch also landed in this release. BerserkArch, if you have never heard of it, is a specialized Arch Linux derivative that targets security researchers and power users. This distro comes with an offensive security tool manager, simply called berserk, which allows users to install complex hacking toolkits with single terminal commands. Moving on, Fastfetch now has experimental scripting options for custom formats using Lua or QuickJS. The Lua integration supports versions 5.3 through 5.5, sharing a single interpreter instance across all modules so you can store variables globally. T Alternatively, if you prefer JavaScript, you can use QuickJS-ng version 0.15.0 or newer to evaluate your custom formats with the qjs: prefix. Other changes that version 2.64 brings include native CMake compilation flags to disable specific modules to shrink the final binary size. Users can delete unwanted ASCII logo files directly from the source directory before building to save additional space. The format engine now boasts ANSI-escape awareness, meaning you can center text with the new vertical bar specifier without breaking colored outputs. Haiku users receive preliminary support for boot manager, window manager theme, screen brightness, and other basic properties. Finally, the Linux edition now extracts desktop wallpaper and theme details from the modern COSMIC desktop environment.
  • Recent Achievements

    • Apprentice
      fernan99 went up a rank
      Apprentice
    • One Month Later
      nothanks earned a badge
      One Month Later
    • One Month Later
      B2Proxy earned a badge
      One Month Later
    • One Year In
      MadMung0 earned a badge
      One Year In
    • Week One Done
      jefred earned a badge
      Week One Done
  • Popular Contributors

    1. 1
      +primortal
      474
    2. 2
      PsYcHoKiLLa
      247
    3. 3
      Skyfrog
      79
    4. 4
      FloatingFatMan
      78
    5. 5
      Michael Scrip
      59
  • Tell a friend

    Love Neowin? Tell a friend!