4 ms·

> Chrome generally outperforms Firefox. I don't agree. Chrome is pretty fast at executing JavaScript, but it is hundreds of times slower at executing DOM inst

> Chrome generally outperforms Firefox. I don't agree. Chrome is pretty fast at executing JavaScript, but it is hundreds of times slower at executing DOM instructions compared to Firefox. It is also the slowest modern browser, by very far, at executing CSS animation.

ahartmetz 8y ago
It is ridiculous that JS benchmarks are accepted as "browser benchmarks". They are probably so popular with tech journalists because they are easy to run. How do you properly measure page load time (of real pages over the internet) anyway?
- a_imho 8y ago
  I found benchmarks to be a red herring. For me, the most significant factor seems to be an adblocker. Without one, most sites are barely usable.
  - darkhorn 8y ago
    You can block ads in Firefox without using addon. There is an setting for that. https://support.mozilla.org/en-US/kb/content-blocking https://support.mozilla.org/en-US/kb/content-blocking
    - a_imho 8y ago
      I only skimmed this page but I don't see it mention ads only trackers? Honestly, FF should just integrate something like ublock but this won't happen for reasons. Anyway, the performance impacts alone are so huge, imo adblocking should be enabled by default for every user focused browser not supported by ad money. Simply a superior browsing experience.
      - darkhorn 8y ago
        Ads are considered tracker. I have enabled this setting and I don't see any ads. See "content blocking" https://support.mozilla.org/en-US/kb/content-blocking#w_how-to-block-more-or-fewer-trackers https://support.mozilla.org/en-US/kb/content-blocking#w_how-...
- jefftk 8y ago
  > How do you properly measure page load time (of real pages over the internet) anyway? I usually use webpagetest.org, which lets you run page loads in instrumented browsers.
  - ahartmetz 8y ago
    Nice, that one should be better known.
etaioinshrdlu 8y ago
And Safari has seemed to be best at DOM manip for something like 5 years...
jasonkester 8y ago
Part of Firefox's slowness is intentional on their part. They (as of last time I opened it) still refuse to silently auto-update the browser, and instead insist on telling you about it and making you watch. So my experience opening Firefox always involves it saying "wait. Before we get to what you wanted to do, I'm going to spend a minute downloading a new version. OK, now I'm going to install it. Almost there. Now here are a bunch of browser tabs full of information about the things we changed. If you still want to do what you wanted to do after reading those, you can open a new tab. Because we filled the starting one with a message saying we updated Firefox, in case you hadn't noticed." That means I never use Firefox unless I absolutely need to, to test that a new responsive layout works on them or after implementing a semi-cutting edge bit of the html spec. Which, of course means that they're guaranteed to have updated the thing at least once since I used it last, which means I get the whole two-minute sit and spin to load again. Try the same usage pattern in chrome, and what happens is that Chrome loads up in zero point seven seconds and lets you get on with your life. It may then start updating in the background but you as the end user never need to hear about it. It's nutty that Firefox still does this, 15 years after everybody else got it right.
- ludston 8y ago
  Nice story, but Firefox about:preferences has an option like, "Use a background service to install updates". I never checked it, so I guess it is on by default.
  - acqq 8y ago
    > about:preferences has an option like, "Use a background service to install updates". But even if it's checked, it seems the Firefox managers/decision makers still think that it improves the users awareness on the Firefox brand that they spend your time on these (quoted from jasonkester's post): "wait. Before we get to what you wanted to do, I'm going to spend a minute downloading a new version. OK, now I'm going to install it. Almost there. Now here are a bunch of browser tabs full of information about the things we changed. If you still want to do what you wanted to do after reading those, you can open a new tab. Because we filled the starting one with a message saying we updated Firefox, in case you hadn't noticed." vs.: "what happens is that Chrome loads up in zero point seven seconds and lets you get on with your life. It may then start updating in the background but you as the end user never need to hear about it." Note: Nevertheless I personally avoid Chrome as much as I can because I really believe Google has too much power. Monocultures, political power and all that. For me is Chrome in the position IE was before, even if currently a lot of web developers prefer it.

ahartmetz 8y ago

It is ridiculous that JS benchmarks are accepted as "browser benchmarks". They are probably so popular with tech journalists because they are easy to run. How do you properly measure page load time (of real pages over the internet) anyway?

a_imho 8y ago
I found benchmarks to be a red herring. For me, the most significant factor seems to be an adblocker. Without one, most sites are barely usable.
- darkhorn 8y ago
  You can block ads in Firefox without using addon. There is an setting for that. https://support.mozilla.org/en-US/kb/content-blocking https://support.mozilla.org/en-US/kb/content-blocking
  - a_imho 8y ago
    I only skimmed this page but I don't see it mention ads only trackers? Honestly, FF should just integrate something like ublock but this won't happen for reasons. Anyway, the performance impacts alone are so huge, imo adblocking should be enabled by default for every user focused browser not supported by ad money. Simply a superior browsing experience.
    - darkhorn 8y ago
      Ads are considered tracker. I have enabled this setting and I don't see any ads. See "content blocking" https://support.mozilla.org/en-US/kb/content-blocking#w_how-to-block-more-or-fewer-trackers https://support.mozilla.org/en-US/kb/content-blocking#w_how-...
jefftk 8y ago
> How do you properly measure page load time (of real pages over the internet) anyway? I usually use webpagetest.org, which lets you run page loads in instrumented browsers.
- ahartmetz 8y ago
  Nice, that one should be better known.

etaioinshrdlu 8y ago

And Safari has seemed to be best at DOM manip for something like 5 years...

jasonkester 8y ago

Part of Firefox's slowness is intentional on their part. They (as of last time I opened it) still refuse to silently auto-update the browser, and instead insist on telling you about it and making you watch. So my experience opening Firefox always involves it saying "wait. Before we get to what you wanted to do, I'm going to spend a minute downloading a new version. OK, now I'm going to install it. Almost there. Now here are a bunch of browser tabs full of information about the things we changed. If you still want to do what you wanted to do after reading those, you can open a new tab. Because we filled the starting one with a message saying we updated Firefox, in case you hadn't noticed." That means I never use Firefox unless I absolutely need to, to test that a new responsive layout works on them or after implementing a semi-cutting edge bit of the html spec. Which, of course means that they're guaranteed to have updated the thing at least once since I used it last, which means I get the whole two-minute sit and spin to load again. Try the same usage pattern in chrome, and what happens is that Chrome loads up in zero point seven seconds and lets you get on with your life. It may then start updating in the background but you as the end user never need to hear about it. It's nutty that Firefox still does this, 15 years after everybody else got it right.

ludston 8y ago
Nice story, but Firefox about:preferences has an option like, "Use a background service to install updates". I never checked it, so I guess it is on by default.
- acqq 8y ago
  > about:preferences has an option like, "Use a background service to install updates". But even if it's checked, it seems the Firefox managers/decision makers still think that it improves the users awareness on the Firefox brand that they spend your time on these (quoted from jasonkester's post): "wait. Before we get to what you wanted to do, I'm going to spend a minute downloading a new version. OK, now I'm going to install it. Almost there. Now here are a bunch of browser tabs full of information about the things we changed. If you still want to do what you wanted to do after reading those, you can open a new tab. Because we filled the starting one with a message saying we updated Firefox, in case you hadn't noticed." vs.: "what happens is that Chrome loads up in zero point seven seconds and lets you get on with your life. It may then start updating in the background but you as the end user never need to hear about it." Note: Nevertheless I personally avoid Chrome as much as I can because I really believe Google has too much power. Monocultures, political power and all that. For me is Chrome in the position IE was before, even if currently a lot of web developers prefer it.