Recommended Posts

i can understand why Australis dont look great in any Windoows Envirement but on Linux an MacOSX it looks good.

I dunno, looks pretty nice on my Windows desktop too, think my only real complaint would be nuking the add-on bar, that was just stupid. Can work around it mostly but there's one addon I use that hasn't been updated it, it only works in the add-on bar. But I'm sticking with beta, not in that build yet, only toying with nightly in a sandbox, hopefully things will even out by release.

 

it would be so much easier if it was in a TAB rather than a pop-up

There's a few addons that do that now -- Download Manager Tweak for example will let you have the downloads show up in a tab, sidebar or separate window.

All the in-content work is eventually going to transition that stuff to tabs anyway.

 

Didn't they make mockups for that as far back as version 4?  I don't know what channels to follow for this kind of info, so I'm curious why the extreme hold up.  16+ versions and one theme change later, and we're still waiting on things like downloads in a tab, etc.

Didn't they make mockups for that as far back as version 4?  I don't know what channels to follow for this kind of info, so I'm curious why the extreme hold up.  16+ versions and one theme change later, and we're still waiting on things like downloads in a tab, etc.

 

If you follow their BugZilla a little bit, you get a good view at how things work (or don't) at Mozilla. It takes them ages to implement something so all they really do is mockups. That is due to their "organizational" structure or however you want to call that. You get the feeling that noone really works there and just does a little work here, a little there, then a student project comes a long, re-does everything someone made but does not have time to finish so it sits there for months etc. It's horrible.

i aint been using Chrome for awhile now, its been unusable for awhile so i removed the package.

 

i can understand why Australis dont look great in any Windoows Envirement but on Linux an MacOSX it looks good. but there's no point whinging about it. its the way of the future. i see Loudnoise has stuck with the ESR, which is funny how these whinging people still use the Australis builds ( Nightly or Aurora ) . .

 

I have choices so I am not whining.. 

New things in Nightly:
 
HTML5 Canvas Hit testing support (only basic support landed - gained few points on html5test.com - although its behind preference "canvas.hitregions.enabled"): https://bugzilla.mozilla.org/show_bug.cgi?id=966591
Improving little performance of DOM: https://bugzilla.mozilla.org/show_bug.cgi?id=976272 (it improved basically performance of this bug: (https://bugzilla.mozilla.org/show_bug.cgi?id=684305)
Moving Moz2D in libxul: https://bugzilla.mozilla.org/show_bug.cgi?id=976392 (If people here remember old PGO linker hitting limits on 32 bit) [Not landed but same: https://bugzilla.mozilla.org/show_bug.cgi?id=977701]
Oldie bug fixed: https://bugzilla.mozilla.org/show_bug.cgi?id=401564
Golden Bug removing Main Thread I/O (quite big contender so now its down UI and Firefox will feel snappier for sure): https://bugzilla.mozilla.org/show_bug.cgi?id=975750
 
 
CSS GRID basic support is also landing soon behind preference.. Above mentioned bugs already landed.

Landed in mozilla-inbound:
Interesting No Audio bug on driver with vendor enhancement (more specifically SoundBlaster sound card): https://bugzilla.mozilla.org/show_bug.cgi?id=971693
CSS Blending bug fix with border-radius and overflow:hidden : https://bugzilla.mozilla.org/show_bug.cgi?id=952051
Native themeing for Help Button on Mac OS X: https://bugzilla.mozilla.org/show_bug.cgi?id=970079 (Image: https://bug970079.bugzilla.mozilla.org/attachment.cgi?id=8380796)
JS Debugger 2 related bug: https://bugzilla.mozilla.org/show_bug.cgi?id=969786
Code cleanup (I don't know how much it affect Desktop but still Desktop share code base with Firefox OS so posting here): https://bugzilla.mozilla.org/show_bug.cgi?id=458300
CSS Transform Perspective + Box Shadow bug fix: https://bugzilla.mozilla.org/show_bug.cgi?id=970294
Rewrite of WebConsole API in C++ to take out about 3 different console API: https://bugzilla.mozilla.org/show_bug.cgi?id=965860
Eviction Algorithm for Media Source Extensions: https://bugzilla.mozilla.org/show_bug.cgi?id=976037

mix-blend-mode is still fairly buggy, I found a bad performance bug with it that is basically going to require it to be re-implemented :laugh:

 

Hahahahaha!!!!  :D  Efforts from start again!

Limited support for Pointer Events on non-metro environments landed in this nightly, not that useful without the support for the hardware (Like the Mac Touchpad), but at least Firefox now sends the correct events on those platforms, vs. nothing at all.

It's interesting that this type of support took so long, the whole point of Pointer Events was that it unified the existing mouse and touch APIs, so a single event is triggered regardless of the type of input, i.e. a mouse event is a pointer event. But until now Firefox only sent those events if you were running on something like a Surface, it wouldn't send any mouse only events even though the API supported it.

It's not like this support offers any extra functionality (It's literally just the same data as normal mouse events, but on a different object), but it reduces the amount of code pages have to write to support this "new" type of input (And here's hoping that OS X and Linux get full support soon enough, I want to try gestures out)

New things in Nightly:

 

HTML5 Canvas Hit testing support (only basic support landed - gained few points on html5test.com - although its behind preference "canvas.hitregions.enabled"): https://bugzilla.mozilla.org/show_bug.cgi?id=966591

Improving little performance of DOM: https://bugzilla.mozilla.org/show_bug.cgi?id=976272 (it improved basically performance of this bug: (https://bugzilla.mozilla.org/show_bug.cgi?id=684305)

Moving Moz2D in libxul: https://bugzilla.mozilla.org/show_bug.cgi?id=976392 (If people here remember old PGO linker hitting limits on 32 bit) [Not landed but same: https://bugzilla.mozilla.org/show_bug.cgi?id=977701]

Oldie bug fixed: https://bugzilla.mozilla.org/show_bug.cgi?id=401564

Golden Bug removing Main Thread I/O (quite big contender so now its down UI and Firefox will feel snappier for sure): https://bugzilla.mozilla.org/show_bug.cgi?id=975750

 

 

CSS GRID basic support is also landing soon behind preference.. Above mentioned bugs already landed.

Landed in mozilla-inbound:

Interesting No Audio bug on driver with vendor enhancement (more specifically SoundBlaster sound card): https://bugzilla.mozilla.org/show_bug.cgi?id=971693

CSS Blending bug fix with border-radius and overflow:hidden : https://bugzilla.mozilla.org/show_bug.cgi?id=952051

Native themeing for Help Button on Mac OS X: https://bugzilla.mozilla.org/show_bug.cgi?id=970079 (Image: https://bug970079.bugzilla.mozilla.org/attachment.cgi?id=8380796)

JS Debugger 2 related bug: https://bugzilla.mozilla.org/show_bug.cgi?id=969786

Code cleanup (I don't know how much it affect Desktop but still Desktop share code base with Firefox OS so posting here): https://bugzilla.mozilla.org/show_bug.cgi?id=458300

CSS Transform Perspective + Box Shadow bug fix: https://bugzilla.mozilla.org/show_bug.cgi?id=970294

Rewrite of WebConsole API in C++ to take out about 3 different console API: https://bugzilla.mozilla.org/show_bug.cgi?id=965860

Eviction Algorithm for Media Source Extensions: https://bugzilla.mozilla.org/show_bug.cgi?id=976037

Probably a dumb question, but the nightly updates that you mention, they apply to UX Nightly as well, right? If so, you said that Firefox will feel snappier. How so? Lower memory usage, more responsive UI, etc?

Probably a dumb question, but the nightly updates that you mention, they apply to UX Nightly as well, right? If so, you said that Firefox will feel snappier. How so? Lower memory usage, more responsive UI, etc?

It depends on how often they merge the main codebase into the UX branch 9They probably do it every few days, so you'll be behind a bit)

And the "snappier" stuff really depends on what type of computer you have, my PC is pretty good so the snappy work hasn't had any effect (A 2fps increase from 58 to 60 is pretty much noise, while from 2fps to 4fps is doubling performance, etc.)

Probably a dumb question, but the nightly updates that you mention, they apply to UX Nightly as well, right? If so, you said that Firefox will feel snappier. How so? Lower memory usage, more responsive UI, etc?

 

Firefox is single threaded and all process run on main thread and all processes are run in queue, if one bottleneck from main thread is gone, it surely feels smoother and sometime improve performance throughout. (Many stuff and processes has been taken down and run in off the main thread now and it is continuous work even after electrolysis lands since they need to move process still then to other process).

The main thread I/O gone above is related to HTTP Debug code use to measure stuff, its was taken out now. So every connection does not has its overhead now.

Guys, the Metro version of Nightly doesn't work. Even on clean install Windows 8.1

 

Resolution: 1680x1050

CPU: Intel Core 2 Quad Q6600

RAM: 3GB(lame frequencies)

 

Latest Nightly version.

 

It shows the full-screen Nightly Metro logo and then closes by itself. On Task Manager I see that it opens, uses around 35mb of RAM and closes.

 

 

Not that I care, but I want to have it as an option. I guess Mozilla coders are kind of lame.(can give some examples but not in the mood, so if possible, do not comment on this, just help me with my problem)

Firefox is single threaded and all process run on main thread and all processes are run in queue, if one bottleneck from main thread is gone, it surely feels smoother and sometime improve performance throughout. (Many stuff and processes has been taken down and run in off the main thread now and it is continuous work even after electrolysis lands since they need to move process still then to other process).

The main thread I/O gone above is related to HTTP Debug code use to measure stuff, its was taken out now. So every connection does not has its overhead now.

There's a rather early implementation of electrolysis in the nightly builds at the moment, only works on platforms where the rendering has been separated onto another thread though (So, not Windows)

File > New e10s Window

There's a rather early implementation of electrolysis in the nightly builds at the moment, only works on platforms where the rendering has been separated onto another thread though (So, not Windows)

File > New e10s Window

 

I already know but I deliberately not mentioned it because its buggier and if he tried he might be little disappointed.

Firefox is single threaded and all process run on main thread and all processes are run in queue, if one bottleneck from main thread is gone, it surely feels smoother and sometime improve performance throughout. (Many stuff and processes has been taken down and run in off the main thread now and it is continuous work even after electrolysis lands since they need to move process still then to other process).

The main thread I/O gone above is related to HTTP Debug code use to measure stuff, its was taken out now. So every connection does not has its overhead now.

Good to know. Thanks!

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

    • "...a low price of just $340..." I don't think it means what you think it means.
    • This Corsair Vengeance DDR5-6000 32GB RAM with RGB is a great deal for limited time by Sayan Sen Memory prices have been through the roof for a while, though it seems like things might finally be getting better. If you are in the market for one, then grab this Corsair Vengeance DDR5 32GB (2x16GB) DDR5 6000 CL36 kit with RGB for a low price of just $340 (purchase link under the specs table down below). The kit is compatible with both AMD and Intel systems as it supports both EXPO and XMP overclocking profiles, respectively. 6000 MT/s is often the sweet spot for many systems as it provides ample data transfer speed while still being on Gear 1 mode. This Vengeance variant has RGB so if you love bright setups with such lighting, this is a win-win for you. The technical specifications of the Corsair Vengeance memory kit are given in the table below: Specification Value Memory Type DDR5 Memory Size (Total) 32GB Kit Configuration 2 × 16GB Form Factor UDIMM (Desktop) Pin Count 288-pin Speed (Data Rate) 6000 MT/s Speed Rating PC5-48000 Tested CAS Latency 38-44-44-96 Voltage (Tested) 1.35V Performance Profile AMD EXPO & Intel XMP Heat Spreader Aluminum heatspreader Cooling Type Passive (Heatsink) Lighting Ten Zone RGB Software Support Corsair iCUE Get it at the link below: CORSAIR Vengeance RGB DDR5 32GB (2 x 16GB) 6000 CL38 – Gray (CMH32GX5M1E6000Z38): $339.99 (Sold and Shipped by Woot US, Fulfilled by Amazon US) This Woot deal is US-specific and not available in other regions unless specified. This is a first-party seller link (at the time of article publishing); ensure that you also purchase from a first-party seller link only. If you don't like it or want to look at more options, check out the previous deals that we have covered, OR you can also visit Amazon US deals page. Get Prime (SNAP), Prime Video, Audible Plus or Kindle / Music Unlimited. Free for 30 days. As an Amazon Associate, we earn from qualifying purchases.
    • The very fact that a TPM (v2.0 specifically which is part of the issue I suspect) is now a baseline for any supported Windows installation will naturally mean other vendors will start to leverage it as they know it'll be there. It's called progress, and it's always been the way. A TPM isn't a windows thing, it's just a module designed to securely store keys. Secure boot isn't a Windows thing (although MS are the TCA as I recall hence the upheaval this year as the 2011 certs expire), it's just a way to verify a bootloader is signed. Windows simply leverages them.
    • It's a local account with the ability to reset a password at a very base level. I really don't get the issue that gets whipped up around it But you do you
  • Recent Achievements

    • Proficient
      Eric Biran went up a rank
      Proficient
    • Dedicated
      Conjor earned a badge
      Dedicated
    • Week One Done
      Windows Guy earned a badge
      Week One Done
    • Dedicated
      Mark Spruce earned a badge
      Dedicated
    • Collaborator
      conkir earned a badge
      Collaborator
  • Popular Contributors

    1. 1
      +primortal
      479
    2. 2
      PsYcHoKiLLa
      244
    3. 3
      Steven P.
      72
    4. 4
      +Edouard
      66
    5. 5
      Skyfrog
      65
  • Tell a friend

    Love Neowin? Tell a friend!