Firefox 4's JägerMonkey JIT status | Hacker News Reader

Firefox 4's JägerMonkey JIT status

bailopan.net

60 pointsnatmaster18 comments

Grrr. Graphs without labeled and meaningful axes and graphs without legends bother me a little too much I think :(

Anyone know what the other lines are?

They're given better context here: http://www.arewefastyet.com/

Thanks!

Once this is merged into the main dev branch of Firefox, I'll build that and switch off of Chrome. The main advantage of WebKit at this point is that their JavaScript implementation is an order of magnitude faster.

Considering Chrome's GTK integration is god awful (on my box anyways), I can't wait.

Firefox 4 still has a little ways to go (of course). I've been running the beta since it came out.

I accidentally discovered a neat DoS in Firefox 4 while working on some JS: the alert loop from hell is persistent even after you close the tab. IIRC, in 3 it was possible to close the tab in between alert()s and they would go away; in 4, you'll have to kill and restart Firefox to get rid of them. Chrome still has an advantage there.

They're working on fixing this the Right Way and right now it's on the list for Firefox 4: https://bugzilla.mozilla.org/show_bug.cgi?id=562258

Wow, what do you do with your browser that you absolutely need to have a JS performance which is an order of magnitude better than FF 3.6? Just curious...

What's up with downvoting? It's a great question, I'm intrigued by adbge performance requirements as well.

  1. Facebook
  2. Google Wave
  3. Google Docs
  4. And a bunch of other webapps I'm forgetting ...

Try out Wave on an older version of Firefox (say, 3.0 or even something from the 2.x branch if you're adventurous). It totally destroys the browser.

High performance isn't just about getting stuff done faster, it can also be thought of as getting the same amount done in less CPU time. When I'm running on batteries, my laptop's CPUs are running at 800MHz, and I get about 8-10 hours of battery life. If something's making my computer work, my CPUs go up to 2.4GHz, and my battery life plummets. More efficient javascript engines should be able to do the same work with less battery use, which makes me happy.

http://www.folklore.org/StoryView.py?project=Macintosh&s...

"Well, let's say you can shave 10 seconds off of the boot time. Multiply that by five million users and thats 50 million seconds, every single day. Over a year, that's probably dozens of lifetimes. So if you make it boot ten seconds faster, you've saved a dozen lives. That's really worth it, don't you think?"

Those few seconds add up.

I like the metal umlaut. http://en.wikipedia.org/wiki/Metal_umlaut

Actually "Jäger" is the german word for "hunter" and it's always written with umlaut. No metal umlaut for you. ;-)

Looks promising. Has anyone done benchmarks on Canvas performance in Firefox 4?

Not sure if there's any graphs on the web, but my own benchmarks have shown it (with hardware acceleration enabled) slightly better than ie9 preview 3, which is many orders of magnitude ahead of everything else (opera's actually not bad though). Keep in mind I have a pretty good graphics card, so that helps with the acceleration.

Now they need to address the rendering performance.

Have you tried Firefox 4 betas? (Legitimate question.) It feels significantly faster than 3 on my machines. Much further on the track of a Chrome-like browsing experience, though there's still many areas it can improve before release.

I'm personally waiting on Electrolysis to speed up the browsing experience. Having a monolithic browser, slower or not than separate processes, just makes it feel more sluggish than it is.

They've landed "lazy frame construction" which improves rendering performance by an order of magnitude in some cases: http://tinikkel.wordpress.com/2010/05/24/lazy-frame-construc...

Reply on news.ycombinator.com