Changes To Look Forward To With Firefox 52

[Michael_S]

Web 3.23 is still beta, it's not even in Elementary yet so you have to build it from scratch (provided that you have GTK 3.22, it won't build on anything lower than that). And for the record, Chromium uses Blink, not Webkit. Browsers like QupZilla, who use qt5-webengine, are more similar to Chrome (as qt5-webengine is very similar to Blink) than Webkit.

(Sorry for the late response.) I know Blink is a fork of WebKit, but they share enough code that I consider them the same.

[Anvil]

blink is just a fork of webkit1

[Vistaus]

(Sorry for the late response, I didn't see this post for a while.) Thanks for clarifying that. That's pretty cool. As I've said elsewhere, I think a browser monoculture is bad even if it's built mostly on open source / free software. Anything that might challenge Chrome/Chromium is good.

I had to admit I'd still favor Firefox because Web uses WebKit and possibly (?) V8, and I'd rather see competition instead of adoption vs. Chrome/Chromium on all levels instead of with shared components. But I'll give Web on Elementary a try again.

Web 3.23 is still beta, it's not even in Elementary yet so you have to build it from scratch (provided that you have GTK 3.22, it won't build on anything lower than that). And for the record, Chromium uses Blink, not Webkit. Browsers like QupZilla, who use qt5-webengine, are more similar to Chrome (as qt5-webengine is very similar to Blink) than Webkit.

[Anvil]

problem with Chrome it still uses Webkit1 = forked to Blink, Webkit2 is getting much better . Chromium still uses old outdated libraries? compared to Chrome

[Michael_S]

Web (Epiphany) is getting a hell of a lot of better though now with 3.23.x It has seen tremendous amount of progress. There's more customization (though admittedly, some of those things can only be customized through dconf-editor right now, but at least it's there!), there's more privacy focus (AdBlock and web tracker blocking by default), it's way faster, more stable and there's even experimental support for libhttpseverywhere and Firefox Sync. They also have plans to support WebExtensions in the future. So don't write of Web just yet; it's becoming way, way, way better now with 3.23 and will get better after that as well.

(Sorry for the late response, I didn't see this post for a while.) Thanks for clarifying that. That's pretty cool. As I've said elsewhere, I think a browser monoculture is bad even if it's built mostly on open source / free software. Anything that might challenge Chrome/Chromium is good.

I had to admit I'd still favor Firefox because Web uses WebKit and possibly (?) V8, and I'd rather see competition instead of adoption vs. Chrome/Chromium on all levels instead of with shared components. But I'll give Web on Elementary a try again.

[liam]

that's the result on the same linux system where chrome runs:Graphics
Features
Compositing Basic
Asynchronous Pan/Zoom wheel input enabled; touch input enabled
WebGL Renderer X.Org -- Gallium 0.4 on AMD RV670 (DRM 2.46.0 / 4.8.0-37-generic, LLVM 3.8.1)
WebGL2 Renderer X.Org -- Gallium 0.4 on AMD RV670 (DRM 2.46.0 / 4.8.0-37-generic, LLVM 3.8.1)
Hardware H264 Decoding No
Audio Backend pulse
GPU #1
Active Yes
Description X.Org -- Gallium 0.4 on AMD RV670 (DRM 2.46.0 / 4.8.0-37-generic, LLVM 3.8.1)
Vendor ID X.Org
Device ID Gallium 0.4 on AMD RV670 (DRM 2.46.0 / 4.8.0-37-generic, LLVM 3.8.1)
Driver Version 3.0 Mesa 12.0.3
Diagnostics
AzureCanvasAccelerated 0
AzureCanvasBackend skia
AzureContentBackend skia
AzureFallbackCanvasBackend none
CairoUseXRender 0
Decision Log
HW_COMPOSITING
blocked by default: Acceleration blocked by platform
OPENGL_COMPOSITING
unavailable by default: Hardware compositing is disabled

ps: chrome take benefit on opengl core profile 3.3.

Fantastic, you found it.
ok, you've a few options to get hwaccel (FF doesn't yet have video hw decode, but there's a bug open and they plan to get it working (I'd suggest creating a new profile while you experiment with this, just in case something goes wrong):
1) toggle the relevant key/values in about:config (so, search for composite, accel, omtc --- I think that will find everything)--- I think you'll have to create the azureacceleratedcanvas one, however
2) download Firefox nightly - this should have most of these things enabled

Btw, you should find a graphical benchmark to test your baseline so can verify if any of changes actually help.

[Azrael5]

Ok. I think i said as much (regarding the feasibility of hwaccel). That's why the decision to not support hwaccel Linux is a political, not technical, one.
Ff, btw, does make use of hwaccel if you enable it and/or you have good drivers. Also, e10s makes things a good deal snappier. Lastly, once they finish the move to skia, on Linux, you should see a pretty rough parity between chrome and ff (as far as drawing is concerned).
Checkout: the graphics section at about:support to see what's enabled. Hopefully you'll have opengl compositing, omtc , apz, and skia enabled for both content and canvas. The one i haven't seen enabled yet, by default on Linux, is accelerated canvas.

that's the result on the same linux system where chrome runs:Graphics
Features
Compositing Basic
Asynchronous Pan/Zoom wheel input enabled; touch input enabled
WebGL Renderer X.Org -- Gallium 0.4 on AMD RV670 (DRM 2.46.0 / 4.8.0-37-generic, LLVM 3.8.1)
WebGL2 Renderer X.Org -- Gallium 0.4 on AMD RV670 (DRM 2.46.0 / 4.8.0-37-generic, LLVM 3.8.1)
Hardware H264 Decoding No
Audio Backend pulse
GPU #1
Active Yes
Description X.Org -- Gallium 0.4 on AMD RV670 (DRM 2.46.0 / 4.8.0-37-generic, LLVM 3.8.1)
Vendor ID X.Org
Device ID Gallium 0.4 on AMD RV670 (DRM 2.46.0 / 4.8.0-37-generic, LLVM 3.8.1)
Driver Version 3.0 Mesa 12.0.3
Diagnostics
AzureCanvasAccelerated 0
AzureCanvasBackend skia
AzureContentBackend skia
AzureFallbackCanvasBackend none
CairoUseXRender 0
Decision Log
HW_COMPOSITING
blocked by default: Acceleration blocked by platform
OPENGL_COMPOSITING
unavailable by default: Hardware compositing is disabled

ps: chrome take benefit on opengl core profile 3.3.

[liam]

I say that chrome has the possibility to implement hardware acceleration without any problems if only they want besides I say that chrome is better than firefox on all the platforms both microsoft or linux because developers of firefox are incompetents. firefox is not able to take benefit of hardware system stressing the hardware because it is not harmonized with it so to be slow and problematic.

Ok. I think i said as much (regarding the feasibility of hwaccel). That's why the decision to not support hwaccel Linux is a political, not technical, one.
Ff, btw, does make use of hwaccel if you enable it and/or you have good drivers. Also, e10s makes things a good deal snappier. Lastly, once they finish the move to skia, on Linux, you should see a pretty rough parity between chrome and ff (as far as drawing is concerned).
Checkout: the graphics section at about:support to see what's enabled. Hopefully you'll have opengl compositing, omtc , apz, and skia enabled for both content and canvas. The one i haven't seen enabled yet, by default on Linux, is accelerated canvas.

[nomadewolf]

You have not measured. Google has improved Chrome since old days.

Neither have you.
FF still uses way less RAM than Chrome.

[nomadewolf]

it requires to use pulseaudio, uses more ram than Chrome and Google Hangouts does not work. It is garbage software now, used firefox for several years.

I don't know what kind of extras you use, but in my setups FF always uses way less RAM than chrome...
You're probably looking at only one chrome process. But if you sum them all, you'll see it's actually the other way around.