Why Firefox 3.5 doesn’t rock, and about the childish obsession with browser performance
By Michael Pietroforte | 18 Comments | Permalink | Trackback | Previous | Next
I suppose, you caught the news that Firefox 3.5 is now available. I skimmed over a few news sites to get an idea about its new features. Once again, I was shocked by the lack of objectivity in our Western media. All sites I have read were praising Firefox 3.5 as if the Mozilla guys had just reinvented the Web (like Opera).
Take the article at Channel Web as an example. The author names five features that are reason enough to say Firefox 3.5 rocks. One is an extension of the “awesome bar” that allows you to search in your bookmarks. Now, seriously, is this really a feature that rocks?? I don’t think you need a PhD in psychology to analyze this behavior. The author didn’t just want to write a review about Firefox 3.5; he wanted to belong to the big, Firefox-praising community. Thus, she was desperately trying to find reasons that could support her faulty claim.
Yes, some of the new Firefox features are nice, such as the new privacy functions and session control. However, these are nothing spectacular, either, because other browsers already had these features. So Firefox is just catching up, and there is nothing that “rocks” here.
Of course, the number one feature that all sites mention is improved performance. Channel Web writes that, according to Mozilla, Firefox 3.5 is twice as fast as Firefox 3.0, and 10 times as fast as Firefox 2.0. I had to rub my eyes 10 times to really believe what I was reading there. I have used Firefox since version 1.0 and updated to 3.5 immediately after it was released. I need the browser all day, and not only to view HTML pages. I work with quite a few Web apps that use Javascript heavily.
Now, what really worries me is that it appears that I am the only one who is not experiencing those unbelievable performance improvements. I already opened my computer to see if someone had hidden a Firefox speed inhibitor inside – but nothing.
Just to be sure, I tried a few of my Web apps in IE8 to see if I could at least experience Internet Explorer’s alleged sluggishness. But again – nothing. I clicked and clicked and clicked, and everything performed just like in Firefox.
So what the heck are all these benchmarks actually measuring? I tried Peacekeeper (what a name for such a controversial topic!). It is really nice. Some cool animated graphics floated over my screen and a few other funny things happened in my browser. I really enjoyed watching the moving balls; it is just a pity that the Web apps I use don’t have such cute balls. I probably would appreciate it a lot if they rolled twice as fast now in Firefox 3.5.
I suppose I am one of the few bloggers who actually has run a benchmark. Nevertheless, everyone seems quite convinced that Firefox 3.5 is lightning-fast. Even Microsoft-friendly blogger Paul Thurrott just copies the product overview from Mozilla’s homepage, making us believe that
Web applications like email, photo sites and your favorite social networks will feel snappier and more responsive.
Obviously, we have another meme pandemic here (like the Windows 7 praising meme) that is spreading fast now (even faster than Firefox can load a Web page). Everyone actually believes that Firefox 3.5 is, indeed, much faster than its predecessor. Psychologists call this a placebo effect. I am sure that 9 out of 10 Firefox 3.5 users swear that they really can “feel” the performance improvements.
This reminds me of my teenage years, when I was trying to tune up my moped at least once a week. Of course, it always “felt” faster after these tune-ups (if it started up at all). Even more important were the comparisons with other moped-tuning “experts.” Naturally, everyone wanted to have the fastest moped; and we were running “benchmarks” regularly to determine who the moped king was. Unfortunately, all those speed improvements didn’t really help me to arrive at school on time.




Subscribe via e-mail: 




I don’t notice a huge difference. It’s footprint is still massive. Some thing’s up with bookmarks -They are VERY easy to mistakenly move for some reason.
Doesn’t matter. IE is still awful by comparison. I have no desire to go back.
Well said, I couldn’t agree more.
A few years back, when i migrated my desktop from Linux back to Windows (new job, new tools) i stuck with Phoenix as my browser.
Since then, i’ve been running nightly builds, upgraded every other week. I’ve never seen or felt any speed improvements.
However, i’ve always felt that IE was decidedly more sluggish than Firefox, though any benchmark i mixed up felt within similar margins of error.
But there is some interesting behaviour that’s different between IE and Firefox:
Firefox CTRL-T, long running instance. Tab opens immediately.
IE CTRL-T, long running instance. Tab opens immediately, but it takes a few milliseconds until the new tab is actually marked active.
The latter does actually not change anything, as i can start typing the URL right after pressing CTRL-T and works just as well.
But it’s about the impression it leaves to me, not the facts.
Here, here! Far too many topics have spread so out of hand that it is really sickening to read – Vista sucks, Apple rocks, Win 7 rocks, *browser* rocks, IE sucks… it’s all one big echo chamber from people who want to jump on the bandwagon.
Granted, some of those topics may be true, but the reviews, praise, or disgust is often not objective or unbiased. It’s helped me weed out which feeds I pay attention to at least. Sad to say, there’s not too many feeds that are objective and unbiased.
While I cannot say that I notice too many speed improvements in 3.5, I do see some small increases. I guess I only see it in Gmail and Google Reader, because those are the two apps that I live in while I am on a computer. The browser seems more responsive switching between tabs and especially loading the two sites. It isn’t amazing, but it is a slight increase.
There are a LOT of ditto-heads in tech reporting, some quoting what shows up in their RSS feeds with the minimal attribution, some using copy and paste from the original marketing blurbs.
Firefox 3.5 did NOT change my life, but I did notice that it was better.
The awesome bar searching my bookmarks DOES rock. My bookmarks file is pretty big and some sites or pages defy categorization, and recording and searching for tags is extra effort and wasted steps. Just typing in a snippet into the awesome bar is a Good Thing(tm).
The DNS prefetching adds a little when I’m surfing but for the most part I stay on one site for a while so the gain isn’t consistent. It’s most noticeable when ads are not being blocked, or when the network isn’t using fast resolvers (i.e. NOT using OpenDNS).
The faster Javascript processing makes a huge difference on SharePoint sites compared to IE7, but IE8 also gained on IE7 (and came out before Firefox 3.5).
The Javascript boost also shows, I think, in the bookmarks manipulation which used to be sluggardly on my worst machine. Now the bookmarks performance is fine on that worst machine.
Bookmarks also got an update in drag and drop editing; the flyout no longer closes when any change is made. But IE7 had that first.
I keep a dozen to two dozen tabs open all the time… and the Firefox 3.5 memory consumption has dropped. It’s mostly around 200 MB now but used to rise to over 400 MB.
As Lukas noted, the IE8 tabs are nice, but there’s still a lag when trying to open a new one and actually use it. Tabs have always been fast with FF and I use Tab-Mix-Plus to make the tabs a little nicer.
I notice an equal delay with Chrome, Firefox and IE when I go to a noisy site like CNN.com, in my subjective opinion, Firefox is the best, then Chrome, then IE8.
LOL
No you are not the only one to not notice the speed improvement.
Anyone who writes reviews whilst trying to sound like an exited teenager should star in a remake of the song ‘Pretty fly for a white guy’.
It is indeed true, that FF 3.5 feels and behaves just like a minor update. “so many important changes that we have to change the version number from 3.1 to 3.5!!” – yeah right. 3.1 would’ve be been about right for the update FF got.
While performance does have increased on some java script/AJAX heavy sites for me (e.g DeviantART), it still performs like FF 3.0 on most other sites. I also can’t say that the affected sites load faster – saying they load smoother with less hickups or getting stuck seems more correct.
FF is still way behind in terms of speed when compared to the Webkit based browsers. Using Chrome/Iron, the performance increases while rendering sites are actually a lot more noticable than they are with FF 3.5, but those browser have problems of their own (still not rendering a lot of sites correctly for example).
Still FF is a good browser and my browser of choice, but not because of the functionality it provides out of the box. It’s more that the extensions I use can’t be found in other browsers. IMHO Mozilla should deliver with the next major version so they deserve all the hype again.
I just removed the stupid 3.5. There is something completely wrong with the javascript engine. It hangs time after time. For example: http://spots.autogespot.com/stad.php. In the beginning I was very content with the browser but now I have to put myself back to 3.0.11.
Thanks for saying what I’ve been thinking. Browsers may differ when it comes to features, but in terms of real-world performance, it’s virtually impossible for the average user to notice any difference in performance.
In other words, Firefox 3.5 seems just as fast as 3.0, and Internet Explorer 8 seems just as fast as both.
I personaly like firefox 3.5 for high speed but sum times some website do not open in fire fox
I’m trying out FF3.5 because there are some great extensions that make my life easier. But I just opened up a Chrome window to compare memory consumption. One tab in each browser:
Firefox, 144MB
Chrome, 29MB
That’s a big, big difference.
Fred, features need code.
Alan, thanks.
Lukas, tabs in IE and Firefox open exactly the same way on my machine. Do you use IE8?
Jared, “echo chamber”, that was the word I was looking for.
Daniel, I use also Gmail and Google Reader. It is all the same, regardless which browser I use. I also tried some of the Google apps.
Andrew, did you ever try to search for bookmarks in the sidebar? I like it better.
moiecoute, perhaps it is just that behaving like teenagers makes us that the old punk rock times come back.
Sebastian, I also use Firefox only because of the plugins. In my view, that is the only significant difference to IE.
Ruud, the site you linked to works fine in Firefox 3.5 on my computer.
Simplify PC Solutions, telepathy was always my strength.
Rohit, perhaps these sites are “optimized” for IE. Try IE tab.
Jay, Chrome has a smaller footprint but less features.
Michael, Yes, IE8 / FF 3.5 on Windows 7 x64 7100. It’s just a few milliseconds on my desktop at home (i7-920, 6GB), a bit more on my laptop (C2D-T9500, 4GB) but really noticeable on my desktop at work (PIV 3Ghz, 2GB).
Page rendering speed is the same in IE8 and FF3.5.
Chrome, actually, feels much faster than either IE8 or Firefox, but i’ve never bothered to switch.
Lukas, I admire your sense of time. I am unable to notice a a few milliseconds time difference.
The new drag and drop behavior does not rocks at all!
Back in the days when I was saving links into folders on my desktops. It was just drag-dropping a tab on the folder/desktop.
And now… It’s a pain in *** ***!
Jay – There must be something seriously wrong with your FF for it to be taking 144MB with only one tab open.
I currently have 2 tabs open and I’m only seeing 55MB. Firefox was much worse in the past with managing memory when it comes to additional tabs. They’ve gotten a lot better at allocating and dumping memory since the firefox 2 days.
You might look into trashing all of the useless addons that you have, which by the way, slow the browser down substantially. Try running FF in safe-mode to see the difference.
@Jay: your FF uses 144mb of ram?
my FF3.0.13 uses (with 30tabs and always online) like 1GB