5 ms·

As a recent Firefox user, I'm super excited about the future evolution of the browser. I switched to Firefox Quantum after using Chrome as mainstay for over 9 y

As a recent Firefox user, I'm super excited about the future evolution of the browser. I switched to Firefox Quantum after using Chrome as mainstay for over 9 years and I was absolutely BLOWN AWAY by the under-the-hood refinements and UX enhancements in v57.0 -- could clearly observe the impact of Gecko's upgraded Servo/Rust components at work. Noticeable visible differences included page-load performance as well as a smaller memory footprint relative to Chrome. Faster WebAssembly[0] starting from v58.0 made the speed optimization more visible -- it was at this point when I truly fell in love with the browser -- Firefox dethroned Chrome as my default. Add to this the hardened tracking protection, Greasemonkey's super massive stash of user scripts and values of the non-profit organization that supports the browser -- Chrome felt so old-school. [0] https://hacks.mozilla.org/2018/01/making-webassembly-even-faster-firefoxs-new-streaming-and-tiering-compiler/ https://hacks.mozilla.org/2018/01/making-webassembly-even-fa...

executesorder66 9y ago
> Faster WebAssembly[0] starting from v58.0 made the speed optimization more visible How many websites do you visit (and which ones are they) that use WebAssembly?
- a9t9 9y ago
  I am not the op, but (faster) WebAssembly could turn out to be a big thing for browser extensions. For example, we are just in the process of adding WebAssembly powered image search to our Kantu Selenium IDE to allow for visual testing. We tried before with normal Javascript and even WebGL, but both results were not usable for various reasons. The websight page gave us new hope ;-) https://websightjs.com/ https://websightjs.com/
  - stromgo 9y ago
    Faster WebAssembly isn't listed on the roadmap, so you are disappointed, right? Wasm is currently considered "around half as fast as native"[0]. The Wasm design, Emscripten compiler, and browser compilers all have a part of responsibility in this, but I suspect that browser compilers have the largest share. Mozilla created The WebAssembly Explorer[1] which shows good-looking ".wat", but bloated "Firefox x86 Assembly" compared to "LLVM x86 Assembly". I hope they intend to use this tool to improve Firefox x86 Assembly. [0] https://github.com/WebAssembly/binaryen/issues/1070 https://github.com/WebAssembly/binaryen/issues/1070 [1] https://hacks.mozilla.org/2017/03/previewing-the-webassembly-explorer/ https://hacks.mozilla.org/2017/03/previewing-the-webassembly...
  - joshstrange 9y ago
    Hmm, on that website I got: Chrome: - WebAssembly: didn't load - Asm.js: 50-60FPS - JS: 2FPS Firefox: - WebAssembly: 30-40FPS - Asm.js: 20-30FPS - JS: 2FPS I have the newest versions of both browsers.
cptskippy 9y ago
As someone who has used Firefox as my default browser since it was called Phoenix, I can tell you that Quantum was a big step backward in terms of memory consumption and CPU utilization for Firefox. On a Lenovo T450s sitting idle with 14 tabs across 4 windows, FF is now using 25-100% of the CPU and over 4GB of RAM. When I resume from Hibernation (on Windows) it can take up to 10 minutes for FF to settle down and for the system mouse to respond. If that's an improvement from Chrome, then OMG you poor bastards who use Chrome as your daily. FF seems to be fine on more powerful systems but on my lower spec'd laptops it's just a battery sucking lap warmer.
- Certhas 9y ago
  If you experience a bug like that, report it! People's mental model (Chrome fast, lean; Firefox slow, bloated) seem to really not update quickly. When Chrome hangs it's "That's weird, something must be broken, the website sucks" when it's Firefox "Well further proof the Firefox is terrible". For the record, my 2012 MBA, which is probably weaker than your Lenovo, has no issues with Firefox at all.
  - cptskippy 9y ago
    I generally do report it and the bugs are generally addressed in followup releases of Firefox. My remarks were not strikes against Firefox's performance, it's stellar, just that since Quantum it has become a resource hog to achieve that high performance.
- blacksmith_tb 9y ago
  Ouch, when I have experienced problems like that in FF it tends to have been something I've done to myself, fiddling with prefs in about:config. Do you see the same behavior if you spin up a fresh profile?
  - cptskippy 9y ago
    That's entirely possible though I would say the profile on my work machine is only a couple months old. Is it possible that Firefox sync brought in settings from another machine that's possibly borking it?
dorgo 9y ago
I don't know. Firefox 59 needs about a second to load a small local html file with 3 small pictures (100x100 pixel) from SSD. How is this possible? Most windows games load instantly. UI in most games works instantly. Why is it so hard to make the same possible in a browser?
- infogulch 9y ago
  Open Dev tools and do a trace? I'd like to see out of curiosity.
  - dorgo 9y ago
    Something like this? https://imgur.com/Y9CkxEh https://imgur.com/Y9CkxEh I don't know how to do it. I can only start a recording and reload the page. But it is not the same as to open it in a new tab (which much slower).

executesorder66 9y ago

> Faster WebAssembly[0] starting from v58.0 made the speed optimization more visible How many websites do you visit (and which ones are they) that use WebAssembly?

a9t9 9y ago
I am not the op, but (faster) WebAssembly could turn out to be a big thing for browser extensions. For example, we are just in the process of adding WebAssembly powered image search to our Kantu Selenium IDE to allow for visual testing. We tried before with normal Javascript and even WebGL, but both results were not usable for various reasons. The websight page gave us new hope ;-) https://websightjs.com/ https://websightjs.com/
- stromgo 9y ago
  Faster WebAssembly isn't listed on the roadmap, so you are disappointed, right? Wasm is currently considered "around half as fast as native"[0]. The Wasm design, Emscripten compiler, and browser compilers all have a part of responsibility in this, but I suspect that browser compilers have the largest share. Mozilla created The WebAssembly Explorer[1] which shows good-looking ".wat", but bloated "Firefox x86 Assembly" compared to "LLVM x86 Assembly". I hope they intend to use this tool to improve Firefox x86 Assembly. [0] https://github.com/WebAssembly/binaryen/issues/1070 https://github.com/WebAssembly/binaryen/issues/1070 [1] https://hacks.mozilla.org/2017/03/previewing-the-webassembly-explorer/ https://hacks.mozilla.org/2017/03/previewing-the-webassembly...
- joshstrange 9y ago
  Hmm, on that website I got: Chrome: - WebAssembly: didn't load - Asm.js: 50-60FPS - JS: 2FPS Firefox: - WebAssembly: 30-40FPS - Asm.js: 20-30FPS - JS: 2FPS I have the newest versions of both browsers.

cptskippy 9y ago

As someone who has used Firefox as my default browser since it was called Phoenix, I can tell you that Quantum was a big step backward in terms of memory consumption and CPU utilization for Firefox. On a Lenovo T450s sitting idle with 14 tabs across 4 windows, FF is now using 25-100% of the CPU and over 4GB of RAM. When I resume from Hibernation (on Windows) it can take up to 10 minutes for FF to settle down and for the system mouse to respond. If that's an improvement from Chrome, then OMG you poor bastards who use Chrome as your daily. FF seems to be fine on more powerful systems but on my lower spec'd laptops it's just a battery sucking lap warmer.

Certhas 9y ago
If you experience a bug like that, report it! People's mental model (Chrome fast, lean; Firefox slow, bloated) seem to really not update quickly. When Chrome hangs it's "That's weird, something must be broken, the website sucks" when it's Firefox "Well further proof the Firefox is terrible". For the record, my 2012 MBA, which is probably weaker than your Lenovo, has no issues with Firefox at all.
- cptskippy 9y ago
  I generally do report it and the bugs are generally addressed in followup releases of Firefox. My remarks were not strikes against Firefox's performance, it's stellar, just that since Quantum it has become a resource hog to achieve that high performance.
blacksmith_tb 9y ago
Ouch, when I have experienced problems like that in FF it tends to have been something I've done to myself, fiddling with prefs in about:config. Do you see the same behavior if you spin up a fresh profile?
- cptskippy 9y ago
  That's entirely possible though I would say the profile on my work machine is only a couple months old. Is it possible that Firefox sync brought in settings from another machine that's possibly borking it?

dorgo 9y ago

I don't know. Firefox 59 needs about a second to load a small local html file with 3 small pictures (100x100 pixel) from SSD. How is this possible? Most windows games load instantly. UI in most games works instantly. Why is it so hard to make the same possible in a browser?

infogulch 9y ago
Open Dev tools and do a trace? I'd like to see out of curiosity.
- dorgo 9y ago
  Something like this? https://imgur.com/Y9CkxEh https://imgur.com/Y9CkxEh I don't know how to do it. I can only start a recording and reload the page. But it is not the same as to open it in a new tab (which much slower).