Welcome to WebmasterWorld Guest from

Forum Moderators: incrediBILL

Message Too Old, No Replies

Firefox - the most annoying browser in use?



9:47 pm on Dec 8, 2004 (gmt 0)

WebmasterWorld Administrator ergophobe is a WebmasterWorld Top Contributor of All Time 10+ Year Member Top Contributors Of The Month

Heh heh, that should attract some readers.

Let's just agree that Firefox has it's plusses. Visiting my folks over the holidays I was forced to use IE and it was unpleasant. Firefox has been my default browser since back when it still had other names.

More and more, though, I find it the most cumbersome and annoying browser I've ever used for the end user. Granted, NS4 and IE* are far more annoying for the developer. Even as an end user, I find that Firefox has really two things that recommend it to me

1. tabbed browsing (but I can get that with Opera or with IE via something like Maxthon or Crazy Browser).

2. The numerous developer extensions - developer toolbar, Venkman debugger, things like that which most end users don't care about. Judging from the broken pages all over the place, neither do a healthy number of developers.

My biggest gripe is the terrible tendency to lock up when reading PDFs. I have long noticed this, but it has become a real sore point since losing my high-speed connnection. Now I wait forever for PDFs and Firefox trips all over itself trying to handle them.

So my gripes not in the least in brief....

- freezes up when loading a pdf (much much longer hesitation than IE - as in an order of magnitude or two). Then, after showing the first page, if you scroll way down to an unloaded section of the PDF, the whole thing freezes up until the PDF gets loaded to that point. If you've already hit the "stop" button once, it thinks it *is* stopped so you can't even use that to stop the loading. Very frustrating on a slow connection. My browser is stopped up like that as I write this (on Wordpad since, of course, Firefox is stuck) and I can't even look at the content in the other tabs because the whole thing is frozen until it downloads the PDF to the point that I scrolled to. I can't even close the browser. On some occasions I've resorted to shutting down the internet connection and reconnecting. I suspect that PDFs are not becoming more common, but since losing the high-speed connection, it seems like every third page is a PDF - or at least 1/3 of the time I spend waiting for things to load is spent waiting for a PDF.

- many problems with copy and paste. It often doesn't work with the CTRL-C/CTRL-V sequences and I have to right click and select from the menu. It's the only program I have that does this and FF does it so often I don't even try to use keyboard sequences for copy and paste. I often find the same thing is true of CTR-T to open another tab. I often have to click it multiple times before the tab opens.

- sometimes a folder of my bookmarks disappear even though it still shows up in the bookmark manager (and no, it's not a scrolling issue). They reappear upon closing and restarting Firefox.

- often, if it's still loading pages in other tabs, it fails when you try to bookmark the current tab.

-if you're in a window opened by a javascript script that doesn't have a tab bar, it will let you CTRL-click to open links in new tabs, but they will be in the background and you can't get to them or you can't get back to the tab you started with. When you close the window, it will tell you you have several tabs open, but there's no way to see those pages.

- password manager doesn't work if site demands password then username instead of username then password (I reported this to Bugzilla already and has been confirmed as a bug).

- password manager doesn't fill in fields until the page has fully loaded, which can be forever on a dial-up. There may be legitimate security reasons, but PM just doesn't seem as good as in IE.

- the search function. Aside from its annoying placement at the bottom of the screen (not obvious when it pops up; terrible ergonomics since it is as far as possible from top menus), there again I find it often won't open if I'm in one tab and content is still loading in other tabs.

- print preview gives you the same number of total pages, even as you change the sizing. So you have four pages and you're looking at page 1 of 4. Then you shrink the text tiny so you only have one page. Now you're still looking at page 1 of 4. But there are no other pages.

So there it is... the most annoying browser ever?



3:39 pm on Dec 28, 2004 (gmt 0)

WebmasterWorld Senior Member 10+ Year Member

Maybe you should its been out of Beta for awhile now.

Dag nabbit, didn't even notice. I'll update this afternoon and my "slack cutting" will be greatly reduced.


4:29 pm on Dec 28, 2004 (gmt 0)

10+ Year Member

I switched to Firefox largely so I could resize font sizes specified in pixels. I found many web pages unreadable with the tiny fonts used and was very happy to switch to a browser that helps me out.

I would be happy if browsers simply showed "Invalid HTML Encountered" whenever bad coding was found. Then it would get fixed. Patching it up just causes folks to not realize they messed up.

I have had problems with Java crashing FF. Hopefully, whatever it is will get fixed.

The big winners are the extensions with Web Developer and AdBlock leading the pack, but many other useful ones as well. And many more to come, I'm sure.

Overall, I have been very happy. There is very little that is annoying about FF.


5:26 pm on Dec 28, 2004 (gmt 0)

10+ Year Member

Same here. It's because of Web Developer plug-in, I chose FF.


9:59 pm on Dec 29, 2004 (gmt 0)

WebmasterWorld Administrator ergophobe is a WebmasterWorld Top Contributor of All Time 10+ Year Member Top Contributors Of The Month

heh heh.... one more peeve.

If I open a bunch of tabs, it eventually runs out of room and the little "loading" icon on the tab covers over the "x" that lets you close the current tab. Click on the "x" and it closes the tab that is loading, not the one you're looking at.


10:28 pm on Dec 29, 2004 (gmt 0)

WebmasterWorld Senior Member 10+ Year Member

Try the Scrolling Tabs or Flowing Tabs extensions. (Scrolling Tabs is my choice.) Also, Tab Clicking Options allows you to set tabs to close on middle or right click, which eliminates the need for the close button (which can be gotten rid of by adding a few lines to userChrome.css).

For extensions, my primary source is [extensionsmirror.nl....] It seems to have more extensions and gets updated faster than update.mozilla.org.


2:35 am on Dec 30, 2004 (gmt 0)

WebmasterWorld Administrator ergophobe is a WebmasterWorld Top Contributor of All Time 10+ Year Member Top Contributors Of The Month

You got an answer for everything don't you ;-)

I'll check those out. Realistically though, once the browser left the beta stage, I don't think core functionality like that shoudl require extensions to make it work reasonably. For me, FF should still carry something like a 0.8 version number.


3:42 am on Dec 30, 2004 (gmt 0)

10+ Year Member

Little note - if you do a search for "acrobat speedup" then there's a little app that lets you disable all those annoying and never used extensions in acrobat and/or acrobat reader - makes it quicker to load (in some cases by quite a bit)



6:00 am on Dec 30, 2004 (gmt 0)

10+ Year Member

Walt Mossberg, the Personal Technology writer for The Wall Street Journal says that Firefox has become his Web browser of choice. He also says it's more secure and modern than Internet Explorer and comes packed with user-friendly features IE can't touch.

The significant of this is that Corporate America looks to such publications as the Journal to validate a new technology. Once validated, they begins to adopt it.

[online.wsj.com...] (you will need to be a subscriber)


7:52 pm on Dec 30, 2004 (gmt 0)

WebmasterWorld Senior Member 10+ Year Member

Honestly, how can you describe a browser being more secure than IE? That comment cannot be validated until you turn the heads of IE hackers/virus writers towards another browser. I am a die-hard Opera user, but I still believe that many many problems would occur if the scope of people trying to break it was increased.

[edited by: mattglet at 8:13 pm (utc) on Dec. 30, 2004]


7:59 pm on Dec 30, 2004 (gmt 0)

WebmasterWorld Administrator ergophobe is a WebmasterWorld Top Contributor of All Time 10+ Year Member Top Contributors Of The Month

RobinC - thanks for the tip. Huge difference and much easier than uninstalling and installing Acrobat Reader 3 instead.


10:42 pm on Dec 30, 2004 (gmt 0)

10+ Year Member

Honestly, how can you describe a browser being more secure than IE?

I believe the logic is that because firefox is not deeply integrated with the os exploits in its security are much less likely to compromise other parts of your system. IE on the other hand is an incredibly integrated and a breach in its security potentially compromises much more of the computer. So this actually has nothing to do with how many hackers are writing viruses. I think the point is that even with equal numbers of security exploits in IE and Firefox, Firefox will still be more secure because a compromise in its security will not be anywhere as likely to compromise the OS as an IE security hole. A similar concept would be ships compartmentalizing the hull so that a single hole doesn't make the whole thing sink.

My biggest gripe with Firefox is the location of the close tab x. It should be on the tab. Not only is it initially confusing to figure out what exactly the x does, but you also have to actually select a tab in order to close it. Objects should be actionable. It's like needing to hit a switch on the other side of the building in order to turn off your desk lamp.


11:51 pm on Dec 30, 2004 (gmt 0)

WebmasterWorld Administrator ergophobe is a WebmasterWorld Top Contributor of All Time 10+ Year Member Top Contributors Of The Month

In the case of your gripe, though, you can right-click on a tab to close it. This works even if the tab is not active. Granted, it's still perhaps more annoying than being able to simply click on the tab itself.

I guess we're destined to disagree on all matters FF though, because in this case I actually like the FF method of doing things and think it's an excellent design.

I typically open a *lot* of tabs, generally in the order that I plan to read the pages. Then I often read simply by hitting the "X" continuously to cycle to the next page in the series. From an ergonomic point of view, the placement is excellent, since it's right near the scroll bar. So I keep my mouse over on the same side of the screen instead of the way standard website navigation goes (swing right to click link, swing left to scroll). If the page is compatible with just using the mouse wheel, I can click the X, scroll down the page, click the X, scroll down, etc without ever moving my mouse. Worst case, I move down to the scroll bar and back to the X, but I stay on the same side of the screen.

This works without me having to think about it because at the end, when there is only one tab, the X disappears, so I don't accidentally close the browser altogether.

When I want to close a given tab, active or not, I right click and close it.

If the X were on the tabs, it would still work to a point as long as the tabs never resized (i.e. one tab would disappear and it's place would be taken by another the same size). Once the size starts adjusting, though, the X would be jumping all over.

Though FF mimics Opera, I think it's actually superior, because Opera puts its X up near the main window buttons to close or minimize. Apart from making me travel farther, it also increases the chance that I will accidentally close the entire browser.


5:17 am on Dec 31, 2004 (gmt 0)

WebmasterWorld Senior Member 10+ Year Member

You can also wheel click on any tab to cloes it. It doesn't have to be an open tab, and and won't "shift focus" away from the open tab.


8:51 am on Dec 31, 2004 (gmt 0)

WebmasterWorld Senior Member 10+ Year Member

People who are on this plane of thought when it comes to the behavior of tabs (which I find very important for quick and efficient organization of info) should probably go get the Tabbrowser Extension at texturizer.net [texturizer.net], homepage [piro.sakura.ne.jp].

This combines a bunch of tab options, and gives you about a zillion options for the behavior of your tabs - for example, you can get the X in the tabs (which I really like), but still keep the X to the right of the tabs. You can also set either 'fluid' tabs, which will give you an extra row(s) of tabs if you have a zillion tabs open, or else scroll tabs, or sort of a combination of both.

Your tabs are grouped according to the page they are linked from, and you can close a group of tabs at once. Or all the tabs left of a tab. Or right of a tab. You can control where newly opened tabs go, and which tab you get when you close a tab. Close a tab accidentally? Click 'reopen closed tab'. Very nice. I had this installed a long time ago and for some reason ended up uninstalling it - don't remember if it was stability, I just didn't like it, or if it was a ff update.

I think it's good that this one's an extension since it would make things a bit more confusing for new users with all the new options and it might cause slower computers to crash more often.

I'd guess that one reason firefox is crashing on people is the tabs encourage opening lotsa tabs which remain open. I've got 10 open now, and that's pretty modest for my usual browsing habits. Think about this, though - of those 10 pages, probably one will be refreshing something every x minutes, a few will have js stuff all over the page that the browser is still trying to keep account of somehow - this will cause more crashes. With ie I hardly ever had more than 6 windows open because of all the clutter. 'Close group' which appears on the startbar button is a bit of a help, but sometimes annoying too when everything gets stuck into that one little button.

Aside from this, though, ff is more crash-prone on some pages - this has been known since the time of that study somebody did on ie's exception handling when it came to buggy HTML etc. On Linux, at least, nothing hangs - just restart the browser and bang, it's back - I remember dreadful Windows sessions where not only did ie cease functioning, I couldn't close anything either, everything in eternal wait mode.

When you get a chance, you "power users" who need lots of aps open in order to have lots of info available, or to do things simultaneously, should at least consider trying out one of the linuces - the desktop switching is so easy that keeping lots of stuff open becomes more organized, and more sensible. The MS desktop switcher in the 'power tools' suite is also nice, but doesn't quite compare, since Linux was built from the beginning with the notion of multiple workspaces. This is my main irritation now when I have to work in Windows - the 'clunky' desktop switching and limited options. Windows managers like fluxbox and pekwm will also allow you to put multiple applications into a single tabbed window - so you have just one window for all the browsers you test stuff in. Very nicely organized, very little clutter.


4:24 pm on Dec 31, 2004 (gmt 0)

WebmasterWorld Senior Member 10+ Year Member

Aside from this, though, ff is more crash-prone on some pages

As opposed to IE, which you can crash with 11 bytes of data [webmasterworld.com]? ;) (See msg. 6 of that thread.)


8:37 pm on Dec 31, 2004 (gmt 0)

WebmasterWorld Administrator ergophobe is a WebmasterWorld Top Contributor of All Time 10+ Year Member Top Contributors Of The Month


That may be true, but in my day to day surfing experience I rarely remember crashing IE (I haven't really used it much for a few years now) or Opera, whereas FF does crash and lock up on me quite frequently. In my experience, it's the least stable browser of the major ones out there (principally comparing it to IE and Opera).

Fair enough, it's barely out of beta compared to the years and years IE has been stuck at version 6. Also, let's assume that almost every web page is tested on IE, so we would expect very few crashes. But still, I don't recall crashing Opera a single time and I think more pages are tested against the Gecko engine than against Opera.


10:03 pm on Dec 31, 2004 (gmt 0)

10+ Year Member

whereas FF does crash and lock up on me quite frequently.

I'm not sure what you are doing, but I've only had Firefox crash on me once since I've started using it early November. And unlike IE, it didn't play havic with my taskbar along the way.


11:34 pm on Dec 31, 2004 (gmt 0)

WebmasterWorld Senior Member 10+ Year Member

I think it depends on the platform a bit. I've encountered some "weirdness" with FF on winXP-SP2, but nothing on win 98, win 95, or Linux KDE.

And I'm not going to get too uppity about the XP problems. I've customized my XP box so heavily, and done far too much USB hot-swapping to really expect it to run uber-stably.

Overall, FF still runs more stably than IE ever did. And it still rocks for overall features and (cureent) immunity to spyware and BHO crappola. (Although, if FF starts to hit 20% + of the market, I'll go back to Lynx :D )


5:03 pm on Jan 1, 2005 (gmt 0)

WebmasterWorld Administrator ergophobe is a WebmasterWorld Top Contributor of All Time 10+ Year Member Top Contributors Of The Month


I could do a long thread about things I love about FF, but it would be redundant since the web is full of such threads. Minimal thread from spyware, BHO, popup blocker, cookie control and so much more are great.

That said, I run Win2K which has been super solid for a long time. In fact, before I had Win2K, I was only switching to Win95 from Caldera Linux when I needed to exchange something with a colleague. Then after changing to Win2K my Windows computer was as solid as my Linux one and bit by bit I just gravitated back to Windows. Anyway, since that time - 4-5 years now - I've found few if any apps as capable of crashing and locking up under Win2K as FF.


5:28 pm on Jan 1, 2005 (gmt 0)

WebmasterWorld Senior Member 10+ Year Member

Oh? I have to admit, Win2K is just not one of the win series that I have up and running (in fact, I don't think I have ever really used that iteration).

I still wonder if it's something to do with how NT (which is behind both Win2K and WinXP, if I'm not mistaken) deals with external software. I have a sneaking suspiscion it's something in WinSock, but for the life of me, I can't figure out what..

That doesn't make it Windows' fault. FF should be made to be able to deal with the operating systems it's released for use with.


6:21 pm on Jan 1, 2005 (gmt 0)

WebmasterWorld Senior Member 10+ Year Member

Just to add to the anecdotal evidence-

I've been running Firefox on Win 2K since around the .7 days and it, like just about everything else on 2000, is very stable. I doubt I have had half a dozen crashes in that time.



11:47 pm on Jan 1, 2005 (gmt 0)

10+ Year Member

<<< Only then did it become clear that many of the annoyances on FF and TB were because they don't work as well as... yes, I'm going to say it... Microsoft products on slow speed lines. >>

This is true, unfortunately. Firefox/Thunderbird both suffer a little because most open source developers simply are not using dialup connections, so this aspect falls a little to the wayside. I had to use dialup and a slow old computer a few months ago, and the only products that were really useable were Outlook express and IE 6.

Thunderbird, which I use, is definitely not a 1.0 release. It has far too many known, documented, and unresolved functional level bugs to be considered as a 1.0 release. I think of it as a 0.7 roughly, getting better, but far too many potentially fatal bugs to recommend to most users. I'm waiting for around 1.5 to start switching people to it. That's about what happened with Gecko too, by the way, it was only at 1.5 that it was really starting to have a bug free rendering engine, then Firefox took that 1.5 and worked with it, cleaning out the interface bugs until it was ready for a 1.0. And Gecko is at 1.8alpha 6 I think now.

Thunderbird should not have released their 1.0 version in my opinion, since that suggests a relatively safe and bugless product, which isn't the case, I won't list the serious unresolved bugs but they are too serious to be treated as unimportant. And I like this product, I just think they've massively jumped the gun, trying to keep up with Firefox I think, which was ready, but tbird isn't.

As for the claims that IE doesn't crash, I don't agree, I figured out several ways to make IE 6 crash, like clockwork, used to be a problem for me because I'd be testing the code that crashed it, can't remember what it was but it's wasn't that complex, it was a basic thing.

Sometimes the tabbed browsing can cause Firefox problems, but as has been mentioned, the 'tabbed browser extension' resolves many of those issues, many of which come from the tab session caching, which you want turned off, a memory leak seems to be present somewhere that can cause instability if you are caching a tabbed session.

I haven't experienced any of the problems other people are reporting here however with Firefox. All browsers have their problems, personally, I think that Firefox just has fewer annoying ones than other browsers. But Firefox has very poor Javascript support, I have several scripts that completely overload Gecko/Opera, but that IE runs perfectly. But this type of complexity would never be used in a real world commercial application, it's just eye candy.

<<< I've been running Firefox on Win 2K since around the .7 days and it, like just about everything else on 2000, is very stable.>>

Same here, since 0.6 days. Only instability I've run across is misconfiguring the tab session caching, having it on that is. Gecko is very stable, other problems seem to be around external stuff.

I think some of the DNS issues might possibly be related to having the pages refresh themselves automatically with tabbed browsing, that's a guess, but that's something I turn off in tabbed browser extensions, and I never have that problem.

You can't blame Firefox if Adobe can't code their plugins correctly, this was an issue a few years ago with their IE plugin, a massive issue, some of the biggest websites in the world could not get pdf files to download correctly when the pdf file reached a certain level of complexity. The trigger? Slow download speeds. Sound familiar ergophobe? Don't blame firefox for that, that's been an ongoing adobe problem, call their tech support and complain. Not that that will do any good, but at least you're complaining to the right place.


1:17 am on Jan 2, 2005 (gmt 0)

WebmasterWorld Senior Member 10+ Year Member

Re FF and javascript. One of the reasons may be because (contrary to the received wisdom) FF is a slow browser. IE cycles at the speed of the OS, so a js script is executed every OS cycle. But FF runs at a slower rate and as a result js is executed less often.

This may not seem very important but a lot of Asian commercial pages use js. Usually pretty crudely, but who knows if more sophisticated browser-side scripts could become more popular in future? The Windows OS cycle time is an ipso facto standard that any modern browser should be able to meet.


1:30 am on Jan 2, 2005 (gmt 0)

10+ Year Member

<<< IE cycles at the speed of the OS, so a js script is executed every OS cycle >>>

This sounds like pure conjecture, completely unsupported by any fact. If this isn't the case, feel free to post some links that document this statement as a fact.

I think the reason IE has better javascript support is that Microsoft, being a programmer oriented company, simply implemented very good Javascript support, it's easy for them to do that. Firefox, on the other hand, being a web developer oriented browser in its beginnings, has outstanding CSS support. According to your logic, IE's bad CSS support would then also be a direct result of being directly tied into Window's frequencies. Sounds spooky, but no fear, this is just a wild guess that overshot the mark.

I'd let things like being able to resize fonts go, that really bothered me too at first when I realized it, I was furious, but now I'm starting to realize that very often the mozilla group is making better decisions than I've given them credit for. In this case, the decision was to let someone with bad eyesight expand your 10 or 11px nav fonts if they needed to do that to see what they said. If your way was implemented, they wouldn't be able to see them. I try to support 1 or 2 sizes up before serious degradation occurs, that's usually adequate. But not always, sometimes stuff will explode all over. But that's life.


2:23 am on Jan 2, 2005 (gmt 0)

WebmasterWorld Senior Member 10+ Year Member

lizardx, HarryM: I've had just the opposite experience. The minute I changed to FF from IE, my load speed was WAY faster; I would probably not go back to IE for any conceivable reason. (This is on dialup of course - that's all I have available. My connections on a 56k modem are ABYSMALLY slow - average is 24k on a GOOD day. This is more to do with 30 year old phone lines than anything else, no doubt....)


2:35 am on Jan 2, 2005 (gmt 0)

WebmasterWorld Senior Member 10+ Year Member

This sounds like pure conjecture

Not so. It's simple to measure. Write a js script that record the time when the page was first loaded, cycles say 50 times, records the time then, and give an alert with the time difference/cycles. Win2000 cycles at (if memory serves me right) at about 10ms/cycle whereas Win98 is 35 ms/cycle. It's usually referred to as the OS granularity. We're not talking about chip speed - the OS is set to cycle at the same speed irrespective of the CPU.

If a browser isn't cycling at the same speed as the OS, it shows in the results. I haven't run the check in FF but I did with earlier browsers. NN6 for instance only cycles at Win98 speed irespective of the OS it is running on.

To see the effect go to the site in my profile and then link to the satire site at the foot of the page. The animation runs at the correct speed on IE5-6, but much slower on Mozilla and FF. (Use your mouse to irritate the spider, then watch him climb the page and you will see what I mean.)

OK, it's only eye-candy but I think it proves the point.


2:54 am on Jan 2, 2005 (gmt 0)

10+ Year Member

It was only in an extremely slow computer running Firefox over dialup that this problem was visible, you'd never see it I think on anything over about 300megahertz or so.

HarryM, I did find some information on OS timing, but any application running on that OS would have the same timing I'm assuming. I think you might be mistaking simply less efficient javascript execution for this OS timing thing. Most of the stuff I've found problematic isn't executing much more than I think every 50-100 milliseconds, well above say Windows 2000 10 millisecond timing thing. It's just not as well programmed is my guess, a weaker dom component.

For example, one complex script starts out ok, but slow, but then quickly bogs down completely, until it barely works at all. This isn't a timing thing, it's a programming thing, Gecko just doesn't handle complex javascript as well as IE, although Firefox doesn't let javascript open active x security settings and so on, so it seems like a good trade off.


3:26 am on Jan 2, 2005 (gmt 0)

WebmasterWorld Senior Member 10+ Year Member


I agree a complex js script could give some browsers a problem. But the speed test I use is very simple - just a few lines - so that would hardly be the problem with the browsers that return the slow times. Netscape 6 definitely always returned exactly the same time for Win98 and Win2K, so I assume it is slugged to skip cycles on high speed OS's. Opera returned exactly the same speeds as IE.

I haven't measured FF so I can't prove that it is skipping OS cycles. However FF executes my animation script perfectly, but the speed of the visual display is exactly what I would expect if cycles were being skipped.


5:38 am on Jan 2, 2005 (gmt 0)

10+ Year Member

<script type="text/javascript">
var date1 = new Date();
var milliseconds1 = date1.getTime();
var j=0;
for (i=0; i < 1000000; i++)
var date2 = new Date();
var milliseconds2 = date2.getTime();

var difference = milliseconds2 - milliseconds1;

I tried this on several Operating systems, all on the same machine. Your times will vary widely depending on your system, but the relationship between the times would stay the same I think. The differences are wide, between IE versions there's a noticeable difference:

windows 98 IE 4: 770
Windows 2000 IE 5.0: 691
Windows 2000 IE 5.5: 770
Windows XP IE 6.02: 942

As you can see, OS cycles have almost nothing to do with the execution speed. What makes the difference is how the javascript engine is handling the programming. In fact, the results are exactly what I would expect from seeing browsers work. IE 4 was not perfect but good, lightweight. IE 5 was quite improved. IE 5.5 was very good. IE 6 starts bogging down again.

Because there are so many iterations, you can actually see how fast each browser's javascript engine handles the javascript it's been given.

Opera is the slowest by far, at between 2300-2600 on all OS's. Opera 6 was a few hundred milliseconds faster than 7.54.
Firefox 1.0 comes at about 1450 for all OS's, almost exactly the same as Firebird 0.6 on windows 98, which was about 1590, suggesting the javascript engine has been improved slightly.
Mozilla 1.6 is about 1650, suggesting Mozilla is as slow as people think.

These results are primarily a function of processor speed and how well written the browser's internal javascript engine is.

Opera will have to stop saying it's the fastest browser on earth. The only slower one I found was Konqueror. If this isn't off topic I don't know what is. Happy new year you all.


12:09 am on Jan 3, 2005 (gmt 0)

WebmasterWorld Senior Member 10+ Year Member


A belated reply, but it took me some time to find the script which I used.

Yes, you are right some of my conclusions did not take account of poor js performance, but even allowing for that FF and Moz still do not have the basic speed of IE or Opera.

The js in this script is so insignificant it shouldn't affect the results. Adding the timer should I believe ensure the script is only called once per OS cycle, so it measures the cycle rate rather than the overall performance.

<script type="text/javascript" language="javascript">
var osspeed;
var cyclecount = 0;
var date1 = new Date();
var date2;
var msec1 = date1.getTime();
var msec2;

function checkspeed() {
if (cyclecount >= 500) {
date2 = new Date();
msec2 = date2.getTime();
osspeed = (msec2 - msec1)/cyclecount;
else {
// -->
<body onload="checkspeed()">

On Win98 both IE4 and NN4.8 return 55 msec

On Win2000 all the following browsers return just over 10 msec. IE5, IE5.5, IE6, NN4.8, NN6.1, NN6.2.3, Op5.02, Op6.05, Op7.20, Op7.50, and NJStar1.26.

On Win2000 Moz 1.2.1, Moz 1.3, Moz 1.6, and FF 1.0 returned just over 14 msec.

The results from both sets of browsers are independant of the timer interval, which can be set between 1 and 10 msec.

However N7 is different to all the above. It's cycle time varies with the timer interval, and for a 1 msec interval
returns 1.6 msec. Presumably it is not handling the timer correctly.

It seems as if Moz and FF are taking more than one OS cycle to exececute, but this is a supposition. If anybody knows the real relationship between these browsers and Windows cycle rate, please enlighten me!

This 129 message thread spans 5 pages: 129

Featured Threads

Hot Threads This Week

Hot Threads This Month