IE9, FF4 Beta In Real-World Use Face-Off

An anonymous reader writes "Most browser benchmarks are isolated, artificial tests that can be gamed by browser vendors optimizing those specific cases. With only those benchmarks to go on, the folks at LucidChart were skeptical that the IE9 beta would actually outperform other modern browsers in real-world applications. To separate hype from reality, they built their first browser benchmarking tool, based in LucidChart itself. This benchmark is to SunSpider what a Left4Dead 2 benchmark is to 3Dmark Vantage. Product specs don't matter, only real-world performance on a real-world application. The results were surprising. IE9 held its own pretty well (with a few caveats), and the latest Firefox 4 beta came in dead last."

Comments

Too late for a film at 11 joke...

[LostCluster]

Browser makers design for tests, real world data shows exact opposite results to what you expect. We've got a crew working on the story overnight and will have a full update for you on the weekend edition of Wicked Early News, we start before normal people wake up.

Re:Too late for a film at 11 joke...

[Anonymous Coward]

Your comment doesn't really apply. Microsoft is the only company with enough balls to ignore certain aspects of Acid 3 that aren't official specifications of HTML/CSS/JS.

Re:Too late for a film at 11 joke...

[beelsebob]

Actually, these results look very similar to the ones the WebKit guys (both chrome and safari teams) publish. They're almost always saying chrome and safari are similarly fast in the lead, firefox lags slightly, and IE8 is way slower... These are the same results as here. It just appears that IE9 is now added to the pile, and added at the top.

So what have we learned
1) Mozilla are good at lying about benchmarks (actually, we already knew that, they've been claiming the next big firefox release would be faster than everything for a while now)
2) IE9 is quick

The question is... is IE9 correct. I'll take works correctly but takes time over doin it rong quickly any day.

Re:Too late for a film at 11 joke...

[dotwhynot]

Your comment doesn't really apply. Microsoft is the only company with enough balls to ignore certain aspects of Acid 3 that aren't official specifications of HTML/CSS/JS.

Mozilla also ignores SVG fonts, as they should (enter WOFF), but you are right in that they are not a company. They are both right, and both deserve kudos for standing up to the hysteria of artificial 100 score on Acid3.

Re:Too late for a film at 11 joke...

[GMC-jimmy]

I'm not so quick to forget why I dropped IE in the first place. To get away from ActiveX and general apathetic browser security. Where's that represented in benchmark?

Re:Too late for a film at 11 joke...

[Anonymous Coward]

I'm not so quick to forget why I dropped IE in the first place. To get away from ActiveX and general apathetic browser security. Where's that represented in benchmark?

Which browser do you use? Most security companies have ranked Firefox, and Safari, as having far more security vulnerabilities than IE for quite some time.

One of many (JFGI): "Firefox most vulnerable browser, Safari close second": http://www.net-security.org/secworld.php?id=8489

Re:Too late for a film at 11 joke...

[h4rm0ny]

I'm not so quick to forget why I dropped IE in the first place. To get away from ActiveX and general apathetic browser security. Where's that represented in benchmark?

Well, that's more specifically why you dropped IE 5.5 and IE6. By the time you get to IE8, these aren't really issues and Firefox has traded its early security-consciousness for usability. And it looks from those charts that IE9 runs pretty fast. That lines up with my own experiences of the beta. Though it's worth keeping in mind that the IE9 is a beta so it's not 100% fair to draw conclusions until it goes to release.

The most interesting things in the article aren't in TFS, though. One is that javascript processing is now apparently so fast that it's dwarfed by the time it actually takes the browser to render an updated page. Another is that in order to get the results from Firefox that they did, they actually had to drop a number of anomalous results where it ran vastly more slowly for unknown reasons. I'm not surprised as Firefox has been getting fatter and fatter ever since 3. The third is that Chrome blows everything else out of the water. They used an old version of Opera which is a shame as I have the newer one and my anecdotal impression is that it's snappier than the previous one so it would have been worth using the latest Opera. But still, Chrome apparently renders far faster than both IE9 and FF. I'd love to know why. Is it just that it's a new project designed from scratch without the cruft that other browsers come with? Does it lack support for significant functionality? Has there been a lot of low-level optimization?

I think in a years time, the real battle is going to be between IE and Chrome. Even today I mainly only use FF for web-development due there being some excellent development add-ons for it that IE can't compete with. Opera is a nice general browser and my default, but its cookie management is shit. Much of FF's funding comes from Google, who I presume would want to push Chrome and who I guess fund Firefox as a means of keeping Microsoft from re-establishing browser-dominance. But that may not be sufficient reason to really push it to be the best it can be, just to keep it "good enough".

I don't wish to be disrespectful to the FF developers. I know how complicated a code-base that size is and they're to be commended on producing a browser that serves well. But it's become a big, unwieldy beast in comparison to the leanness of Chrome and the new IE.

Re:Too late for a film at 11 joke...

[FunPika]

And on this very site the exact article you linked to was dismissed as being pro-Microsoft FUD. And I ran a couple Google searches and I still saw little more than FUD from Microsoft sponsored research or security companies grouping vulnerabilities from all Gecko based browsers as "Mozilla" (which as we all know the average person will read as meaning "Firefox").

Re:Too late for a film at 11 joke...

[Bert64]

The prevalence of firefox has helped the security of IE considerably, for a number of reasons...
Before they started losing market share, MS had no intention of improving IE at all.. Also now that no browser has over 90% market share they become far less attractive targets for malware authors, who instead now either target specific areas (eg ie6 is still huge in corporate settings) or other software which has a huge market share (adobe flash/pdf, msoffice).
So long as there are a handful of browsers out there competing with each other.
Were it not for firefox, ie7 would be a slightly warmed over ie6, and would still be getting attacked on a daily basis and users would have nowhere to go.

Incidentally, those benchmarks show ie9 coming in 3rd from last, only firefox 4 beta (by a small margin) and ie8 (by a huge and laughable margin) are slower. Actually released browsers such as firefox 3.6, safari 5, chrome 6 and opera 10 are all ahead of both betas of ie9 and firefox 4.

Now as for why the firefox 4 beta is slower than 3.6, it could be compiled in a debug mode (as betas often are), its new javascript engine has only just been integrated and needs tweaking etc...

The fact that current beta versions of both ie9 and firefox 4 are way behind current non beta versions of other browsers is rather poor, and you would hope that both will be fixed by the time their final releases come around.

As for Google, they make their money from the web and generally don't care which browser you use to access it, so long as that browser is fast enough to deliver a reasonable experience and standard enough so it doesn't force them to spend a lot of time kludging around. Older versions of IE were a threat to google as they made their web based apps appear slow, new versions appear better however IE is still controlled by google's biggest competitor who, given the chance, would use it against google in any way they can. When it comes to firefox/opera/chrome/safari i doubt google really cares which you use.

Re:Too late for a film at 11 joke...

[bunratty]

You can't simply count the number of security vulnerabilities reported in each browser. You need to consider the severity of the vulnerability, whether it was exploited by black hats, how long the vulnerability was known by black hats before it was patched, and whether the browser vendor is admitting to vulnerabilities that were no publicly known when they were fixed. The problem with IE is that they take the longest to fix vulnerabilities, and they are the browser most often targeted by black hats.

Besides, these days plugins are attacked more often than the browsers themselves, so keeping plugins up to date and turning on features such as DEP may be more important than which browser you use.

Re:Too late for a film at 11 joke...

[ePhil_One]

I'm not so quick to forget why I dropped IE in the first place. To get away from ActiveX and general apathetic browser security. Where's that represented in benchmark?

I dropped IE for tabbed browsing. After a company takeover made me revert to IE 6, I was excited to get tabs back and got approved for an early IE 7 roll-out. No open link in new tab option, anything that navigated me away from a form I was completing lost the work I'd done (agonizing when crafting wiki articles), etc. I finally got IE 8, but I still prefer to use Firefox for basic functionality reasons.

I did try Chrome, but for whatever reason Yahoo Mail crashes constantly, so "good bye"

Re:Too late for a film at 11 joke...

[Anonymous Coward]

And on this very site the exact article you linked to was dismissed as being pro-Microsoft FUD.

You are kidding, right? On this site everything remotely positive to Microsoft is routinely dismissed as pro-Microsoft FUD and paid astroturfing.

Not that it will matter, but here is another report, this time from Secunia: http://blogs.computerworld.com/report_firefox_is_the_worlds_most_vulnerable_browser

and a newer one from Secunia: http://www.infoworld.com/t/browsers/internet-explorer-deemed-least-vulnerable-browser-593

Re:Too late for a film at 11 joke...

[Anonymous Coward]

I find that these performance benchmarks are just like the the audio guys who blither about 0.005% distortion vs 0.010% distortion. In other words totally worthless for anyone who actually uses a browser. Unlike most fanatics here I don't automatically assume Microsoft software is defective, on the other hand I use Firefox primarily because of the availability of extensions. The truly important browser metrics of stability, compatibility and security are either too hard to measure or are highly subjective.

Re:Too late for a film at 11 joke...

[Anonymous Coward]

Mozilla are good at lying about benchmarks (actually, we already knew that, they've been claiming the next big firefox release would be faster than everything for a while now)

What are you talking about? The Mozilla team caught the IE team cheating at benchmarks, whereby minor code changes which should result in zero internally generated code results in code which runs 22x slower. And even if MS were to fail horribly at the obvious optimization of producing a nop, at worst which can possibly be imagined would be something like a 1ms; whereby a fraction thereof is extremely more likely. In other words, it appears IE specifically detects some benchmarks and returns a predetermined value rather than actually compiling and executing the code. There is no legitimate reason for a nop instruction to slow a benchmark 22x times. None. Zero. And if its a bug, it would be practical impossible for IE to perform nearly as well at any other benchmark. They'd likely return results at least an order of magnitude slower. The only reasonable, plausible explanation is they are cheating; blatantly!

In other words, how can placing code which results in a nop instruction result in running 22x slower in any language whereby the implementation is supposed to be so good it runs many times faster than is theoretically possible. You can't without smoke a mirrors and blowing lots and lots of smoke up your ass - and yet that's exactly what Microsoft is doing here.

With cheat, benchmark = 1ms.
Without cheat, benchmark = 22.5ms.

Which in fact suggests MS's optimizer fails to identify and optimize some of the more common and easiest forms of optimization which in turn hints they are cheating across the board to be close to the results indicated by common benchmarks.

IE = bullshit, smoke, and mirrors.

Re:Too late for a film at 11 joke...

[Anonymous Coward]

Well, from the first article:

...

That doesn't mean, though, that Internet Explorer is off the hook for security concerns. Far from it. ActiveX remains the browser plug-in or add-on with the most number of vulnerabilites. It had a whopping 366 vulnerabilities, compared to 54 for Java, 30 for QuickTime, 19 for Flash, and one for a Firefox extension. No Opera widgets had any vulnerabilities.

Internet Explorer also has a much longer lag time between when a vulnerability is found, and when a patch is issued for it, again by a wide margin. The lag for Internet Explorer was between 78 days and more than 294 days (some vulnerabilities weren't patched by year's end). For Firefox, the lag ranged between 15 and 86 days. Secunia didn't compare how long Safari and Opera took to patch.

The second article doesn't account for this at all.

Re:Too late for a film at 11 joke...

[Shining Celebi]

1) Mozilla are good at lying about benchmarks (actually, we already knew that, they've been claiming the next big firefox release would be faster than everything for a while now)

Well, the next Firefox beta does include a new Javascript JIT. But I'm sure that couldn't make any kind of difference. Mozilla is clearly just lying.

The current Firefox betas also include hardware acceleration - supposed to be enabled by default now, but I had to do it manually. That makes a huge difference on some jobs that completely blows non-HW accelerated browsers out of the water. Check out this, for example.

Re:Too late for a film at 11 joke...

[beelsebob]

Well, the next Firefox beta does include a new Javascript JIT.

1) Not new, it's Nitro – the one from apple's version of WebKit (as opposed to V8, the one from google's version).
2) Again, the benchmarks for even this new version come out slower than Safari and Chrome, so yes, again they're lying about it, just like they did with 3.5, 3.6, etc.

Re:Too late for a film at 11 joke...

[BZ]

> I'll take works correctly but takes time over doin it rong quickly any day.

You would? Then why are you happy with Webkit, which does some things in CSS purposefully wrong to do them faster?

Re:Too late for a film at 11 joke...

[beelsebob]

Who said anything about being happy about webkit?

Having said that, given the logic outlined above, I should be happier with WebKit than with IE, given that it gets fewer pieces of CSS, HTML and javascript wrong.

Re:Too late for a film at 11 joke...

[BZ]

> Is it just that it's a new project designed from scratch without the cruft that other
> browsers come with?

To some extent, yes. It's easier to write a fast JS engine if you don't have to worry about compatibility with existing applications that rely on your JS engine.

> Does it lack support for significant functionality?

To some extent, yes. Webkit purposefully handles some dynamic changes to a DOM incorrectly in terms of not actually restyling everything that needs to be restyled so that it can do it faster.

In this case there's also the matter of the benchmark just measuring things incorrectly so that the "intra-frame time" (as they call it) is dependent on a browser-specific setTimeout clamping that Chrome happens to set differently from other browsers. So those yellowish bars are all pretty much bogus (except the IE8 one, which is way larger than the 10ms clamping).

Also of interest is that Chrome 7 dev on Mac gets about 4fps for me on that benchmark on a fairly new machine. So I wonder whether the Chrome numbers are very different for 6 vs 7 or Mac vs Windows or what....

Re:Too late for a film at 11 joke...

[tuxgeek]

One must also consider that IE only runs natively on Windows. To switch from IE to FF, on Windows, does not really circumvent the inherent vulnerabilities of the OS. Not really sure how that works now as I dumped windows 10 years ago when it was in it's nightmare heyday.

FF on Linux should be much more secure than the IE running on windows. This is a no brainer.
But FF is becoming slower with each new improved version. Although a quick google search can get you about:config modifications to dramatically speed the thing up. Additionally konqueror can now be switched from khtml to Webkit which also makes the browser amazingly fast. Even quicker than FF

Re:Too late for a film at 11 joke...

[Anonymous Coward]

I must say, in FF4 beta's favor, it runs MUCH better on lean hardware than IE9 beta. IE9 is designed to use the high end hardware it expects from a Win7 computer. FF4 has none of that (that I can tell). FF4 simply smokes rendering any page in comparison to IE9 on my netbook with 1G ram, integrated graphics, and dual atom CPU's.

It still pales to Chrome though.

Re:Too late for a film at 11 joke...

[LO0G]

That quote doesn't make sense. ActiveX isn't a plugin, it's a plugin model. It's like saying that all vulnerabilities in Flash are the fault of XPCOM.

ActiveX as a technology is no more or less secure than any other plugin model.

The actual PDF says that the 366 vulnerabilities are in ActiveX plugins, NOT in ActiveX.

Re:Too late for a film at 11 joke...

[Anonymous Coward]

Right, because the web is all about accuracy, and correctness over swiftness.

It's certainly not about doing crap as fast as possible.

Lolz.

Re:Too late for a film at 11 joke...

[Anonymous Coward]

Bullshit and more bullshit.

From my experience as a technician, almost everyone who gets infected by PC ANTIVIRUS PRO 2011 STANDARD and anything of that ilk, use IE, the malware installs through IE via an advert and uses exploits that microsoft refuses to fix (and the activeX based flash plugin as well) meanwhile. those who use firefox, well, never run into any issues with them. There was an entire office that got hit by fake malware, because an advertiser on yahoo was installing it via temp files, the only machines that didnt get it were either servers that only IT (us) used or anyone who uses firefox as their primary browser.

Sorry, but IE7 and IE8 are still fairly insecure.

I'm hoping that they drop activeX in IE9 like was originally said, would be nice to be rid of that horrid garbage.

Re:Too late for a film at 11 joke...

[Bungie]

Damn right, anyone could author an ActiveX control using "VB5 Control Creation Edition" and package it on their web site, just because you choose to install it doesn't mean Internet Explorer or ActiveX is insecure. An ActiveX control is the same as installing any other software component, and you have to be prepared to accept that it might be vulnerable to exploits. This means you have to keep it up to date, or not use it at all if it's not required. Newer versions of IE with protected mode and running in Vista or higher Windows also have significant security improvements which can prevent most ActiveX exploits from affecting the system.

Java was a huge hole for many malwares and no one stopped using FireFox because the Java plugin could be exploited...

Re:Too late for a film at 11 joke...

[BZ]

See, that's the interesting thing. Compared to the IE9 betas, Webkit implements _more_ pieces of CSS, etc (e.g. CSS3 selectors), but the parts that they both implement Webkit gets more wrong in my testing (e.g. various parts of CSS2.1 Webkit "implements" in ways that are totally broken). The IE team has actually done a really good job of implementing the specs they said they would implement. Never thought I'd say that, just a few years ago!

Re:Too late for a film at 11 joke...

[BZ]

> 1) Mozilla are good at lying about benchmarks

Er.. how so? Note that this test wasn't even using the JS engine that will be in the next big Firefox release... nor has Mozilla been claiming it would be "faster than everything" on every single test you can throw at it. That would be a daft claim, obviously, just like the equivalent claim for any other browser.

Re:Too late for a film at 11 joke...

[CarpetShark]

Microsoft is the only company with enough balls

They certainly do have a load of balls.

dev IE9 and dev FF vs release Chrome?

[Tumbleweed]

I'd like to see Chrome 7 results in there...

Re:dev IE9 and dev FF vs release Chrome?

[icebike]

Me too.

Its odd how both the summary above and the linked article sort of over look the fact that Chrome just blew the doors off of every other browser and the compared the production version to the latest and greatest of the others.

Chrome really deserves top billing, but the story is about the who is going to come in dead last.

Yawn.

Re:dev IE9 and dev FF vs release Chrome?

[Anonymous Coward]

You mean like how they didn't test with Opera 10.70 beta? Or how they didn't even use the latest release version of 10.62?

Re:dev IE9 and dev FF vs release Chrome?

[TheSeaCucumber]

Beat me to it. Personally, I think FF is almost as bloated as IE. I think the only good, usable part of FF is the extension support. In the end, Chrome beats them all in speed _and_ speed, yet we only ever hear about how FF just managed to beat IE or the other way around.

REAL MEN USE WEBKIT DAMMIT!! =P

Re:dev IE9 and dev FF vs release Chrome?

[Tumbleweed]

You mean like how they didn't test with Opera 10.70 beta? Or how they didn't even use the latest release version of 10.62?

That's not surprising - Opera hasn't caught on, and probably never will. You know the difference between Opera and technologies like Amiga and OS/2? Me, neither. "You hear that, Mr. Anderson? That is the sound of inevitability." And lemme tell ya, being an Opera user doesn't make you Neo. *shrug*

Re:dev IE9 and dev FF vs release Chrome?

[moderatorrater]

Modern browsers are so fast that the difference is miniscule. If you're looking at using IE6 or using Chrome, then obviously Chrome needs to be praised. If you're comparing several browsers that are all fast enough that there's no strong difference between them in real world use, then mocking the loser is just more fun :D

Re:dev IE9 and dev FF vs release Chrome?

[TheSeaCucumber]

Look, apart from the fact that it was a little too heavy on my ram (FAR less than FF btw) Opera was ok, BUT i thought it was unfortunate that they filled it with superfluous crap. Like the massive toolbar on top, the animated tab switch thingo (which you could disable mind), and that annoying dialer thing. Seemed like they could have squeezed more horsepower out of getting rid of that.

Have to agree with Tumbleweed though, Opera is a decent browser, but compared to IE's default shipping, Firefox's pseudointellectual fanboy support and Chrome lurking in the shadows with its high speed appeal, it just doesn't have anything really distinct you know?

Re:dev IE9 and dev FF vs release Chrome?

[icebike]

Fast enough unless you need to work a major portion of your day using one. Then you find out there are significant and meaningful differences in speed that affect your ability to get things done.

Fast enough for someone who checks in once a day to look at email is easy.

Re:dev IE9 and dev FF vs release Chrome?

[Tumbleweed]

They forever screwed themselves because they used to charge people to use it. When there's a ton of competition, and it's free? Yeah.

So did Netscape, which eventually turned into Firefox. I don't think that has anything to do with it. Opera has great technology, and no understanding of what most users want in a (default) user interface. They also seem to have no clue that most users never change default settings, so it doesn't matter how configurable it is.

Re:dev IE9 and dev FF vs release Chrome?

[CAIMLAS]

Absolutely. I find it humorous how I will occasionally hear coworkers cursing:

1) The speed of their browsers. "Render, god damn it!" echos down the halls.
2) The ability to quickly switch tasks/tabs within the browser (ie responsiveness vs. speed). "Fucking flash!"
3) The stability of the browser. I don't really care so much if a single tab crashes; I'll just reload it. Someone with 40+ tabs in firefox, however, is stuck waiting a minute or so while whatever they were doing crawls back from the dead. (Users who don't have session management in their browser are even less fortunate.)

Meanwhile, I sit there contentedly working away, not distracted by such things, due to using Chrome and a lightweight window manager on Linux. I only start noticing a slow down when I'm being inefficient, anyway - IE, doing too much at once, getting distracted, and not getting anything done.

Of course, the slow users don't complain all that much, either. Seems they can't quite keep up with much of anything. :P

A little speed in the right places makes a huge difference.

Re:dev IE9 and dev FF vs release Chrome?

[Bing Tsher E]

So did IE. It used to sell as a retail-box add-on to Windows 95. But that was a long, long time ago.

Re:dev IE9 and dev FF vs release Chrome?

[NoZart]

Well, some people are not "most of the people", and i like that there is one browser who caters to that group.

Re:dev IE9 and dev FF vs release Chrome?

[derek_m]

No reason to see anything sinister in that; in terms of JS performance which is all this article tests the latest dev build of chrome wouldnt give significantly different results.

Everyone that is going to read a browser JS performance test article already knows who the winner is, the IE9 results are what we click on it for - arent they?

Re:dev IE9 and dev FF vs release Chrome?

[Anonymous Coward]

yea against opera 10.53 and not 10.62

Re:dev IE9 and dev FF vs release Chrome?

[montibbalt]

Look, apart from the fact that it was a little too heavy on my ram (FAR less than FF btw) Opera was ok, BUT i thought it was unfortunate that they filled it with superfluous crap. Like the massive toolbar on top which is now a tiny toolbar that FF4 is copying quite directly, the animated tab switch thingo (which you could disable mind), and that annoying dialer thing which every browser has now

FTFY

Re:dev IE9 and dev FF vs release Chrome?

[CarpetShark]

Chrome really deserves top billing

In the limited category of speed, perhaps. It's only starting to catch up with firefox in terms of adblock support. That may be a small issue for you, but for many of us, it's a modern web necessity, just like anti-spam features are modern email necessities.

Re:dev IE9 and dev FF vs release Chrome?

[Anonymous Coward]

There are only 25 AdBlock extensions available for Chrome. How many does it need to catch up with FireFox?

I like Chrome; it does what it's supposed to do, does it well and does nothing else. If I want something more, I can install an extension for it.

Re:dev IE9 and dev FF vs release Chrome?

[Anonymous Coward]

"That's not surprising - Opera hasn't caught on, and probably never will"

Opera is always the second or third, but in ALMOST EVERY benchmark. That means something.

Re:dev IE9 and dev FF vs release Chrome?

[sdnoob]

Then you find out there are significant and meaningful differences in speed that affect your ability to get things done.

There are 'significant and meaningful differences' in usability, feature set and extendability that affect my 'ability to get things done'. That's why I've used Firefox since before it was even known as Firefox.

Re:dev IE9 and dev FF vs release Chrome?

[Anonymous Coward]

Same here. Is this an conditioning campaign to make all of us think that the game is between IE and Firefox?

Naaaanana.

Re:dev IE9 and dev FF vs release Chrome?

[Anonymous Coward]

I don't really care so much if a single tab crashes; I'll just reload it. Someone with 40+ tabs in firefox, however, is stuck waiting a minute or so while whatever they were doing crawls back from the dead.

Browsers with a process-per-tab often still crash as a whole. Process-per-tab is a good thing but it doesn't work very well yet to prevent browser crashes.

Re:dev IE9 and dev FF vs release Chrome?

[h4rm0ny]

Actually one interesting bit from the article is that in their tests, javascript in most browsers is now so fast that it is rarely significant in comparison to rendering performance. Chrome is fast at javascript yes, but I hadn't realised that its super-performance was because it also leaves all the other browsers' rendering engine's in its dust. I don't know what Google feeds its developers on the Chrome project, but it's working. ;)

Re:dev IE9 and dev FF vs release Chrome?

[skyride]

Clearly you have not used Opera recently. I've personally been using it as my main browser for about 2 years, and the sheer degree of polish in the windows version is just totally unsurpassed by any browser, aside from it being bloody quick. Use it for a week, you won't go back.

(note: That is the windows version I am talking about. The linux version's UI is a bit off and it is a little slow to load)

Re:dev IE9 and dev FF vs release Chrome?

[cp.tar]

Browsers with a process-per-tab often still crash as a whole. Process-per-tab is a good thing but it doesn't work very well yet to prevent browser crashes.

I still use Firefox as my primary browser, but I find that Chrome handles Flash-based websites a bit better. However, dozens of open tabs slow down Chrome to a grind, the same as Firefox. And still nothing really helps save for a browser restart.

Process separation/isolation is a nice thing, but it’s not the end-all and be-all of stability.

Re:dev IE9 and dev FF vs release Chrome?

[bunratty]

I've noticed that when using Chrome and I switch between tabs, it sometimes takes a long time for Chrome to draw the tab I'm switching to. I think it has to do with each tab being in a process, so that perhaps the process for the tab got swapped out to disk. I find Firefox faster at everyday browsing.

Re:dev IE9 and dev FF vs release Chrome?

[Anonymous Coward]

Yeah, everyone always dishing Google the underdog award. Google doesn't get nearly enough attention.

Re:dev IE9 and dev FF vs release Chrome?

[whoop]

Well, a quick search at mozzila.org shows, "Showing 1 - 20 of 64 results for adblock". So I'd say Chrome needs another 39 extensions to do the same thing.

All this talk about Adblock extensions here has me curious why people don't use the hosts file to block all the ad hostnames? I've used 'em for a few years, and it's reasonable. It doesn't matter what browser I start up, and it works. Then again, even if one gets through, I just don't click on it. That's just the kinda guy I am. Maybe I should type up a long rant about it on the Internets.

Re:dev IE9 and dev FF vs release Chrome?

[hedwards]

True, but if you're going to cater to some people, then you damn well better not whine about failing to pick up marketshare.

Re:dev IE9 and dev FF vs release Chrome?

[Anonymous Coward]

Like the massive toolbar on top which is now a tiny toolbar that FF4 is copying quite directly

Like Opera was the first one to come up with that... sheesh.

Re:dev IE9 and dev FF vs release Chrome?

[Tumbleweed]

"That's not surprising - Opera hasn't caught on, and probably never will"

Opera is always the second or third, but in ALMOST EVERY benchmark. That means something.

It's not always about the technology, otherwise we'd all be using Amigas with Dvorak keyboard layouts.

Or OS/2. Or BeOS. It's a long list, even in the computer world.

Re:dev IE9 and dev FF vs release Chrome?

[CarpetShark]

even if one gets through

You virtually never have that problem with adblock plus. Moreover, you don't have to maintain the same hostfile that many others would maintain too. By definition, host files are more limited, more manual, more prone to being out of date, and more error-prone. If I was going to use a system-wide solution, it'd be something like junkbuster, but that's way too slow, and too prone to breaking sites. Finally, the adblock plus lists are automatically updated, and maintained/monitored by a global community.

I just don't click on it.

A lot of ads now are becoming extremely deceptive, designed to look like legitimate page elements. To avoid them, you would have to slow down your browsing process and start checking urls on hover etc. very carefully. That's not productive, when you can simply use a proper ad blocker.

Re:dev IE9 and dev FF vs release Chrome?

[spinkham]

Bartab + FF 4 beta = almost instant restart with 100+ tabs.

Oh, actually, looks like bartab like functionality is now the default in the latest builds, as per the anonymous coward below. http://blog.zpao.com/post/1140456188/cascaded-session-restore-a-hidden-bonus

Re:dev IE9 and dev FF vs release Chrome?

[shutdown -p now]

You know the difference between Opera and technologies like Amiga and OS/2?

Yes. Opera has significant market share in some real-world markets (e.g. 30-40% in Eastern Europe, >50% in CIS countries); just not, apparently, in yours. I'm not aware of anything similar for Amiga or OS/2.

Re:dev IE9 and dev FF vs release Chrome?

[Bungie]

In the limited category of speed, perhaps. It's only starting to catch up with firefox in terms of adblock support.

Give me a break. FF users praise AdBlock like it invented ad blocking. Long before FireFox (in the days of Netscape Navigator) there was a free addon I used for IE which did exactly what AdBlock does now (but with a less slick GUI). I can't remember the name of it now but it used to be a standard thing for many IE power users...the newer versions required paying for it (and the free one broke under Vista) so it died off. There are also many other IE AdBlock plugins like the one included in the IE7Pro addin.

I also remember Opera having add blocking support ages ago when I used it...but there was no GUI at the time for adding entries into the block list file. I wrote one in VB and it worked just fine, but I'm sure Opera has a GUI for it now.

AdBlock for FireFox simply provides a better interface than the IE and Opera versions, and was completely free when others were not. No one is benchmarking how well browsers can run the AdBlock plugin (which was written for FireFox in the first place). Please stop using it to compare them and Google search for how to block ads in other browser if you need to.

Re:dev IE9 and dev FF vs release Chrome?

[Bungie]

IIRC the hosts file fell out of favor in Windows because the system would manually process the entire hosts file before performing any DNS lookup (even ones cached by the DNS Service). If you had a large hosts file this would cause adverse performance issues as it would read through each entry every time. The hosts file also does not support wildcard blocking so you have to manually add each individual server for a domain, which could possibly lead to the large hosts file issue above.

Re:dev IE9 and dev FF vs release Chrome?

[richlv]

ahh, as an opera user on linux i can mostly agree on it being a great browser. too bad latest versions seem to have broken kde/qt3 support... wondering whether that would be any better with 10.70 test builds. nope, kde4 still has a bit to wait for my primary desktop ;)

Re:dev IE9 and dev FF vs release Chrome?

[definate]

I have used it many times, and currently use it as a backup browser.

Everytime I use it I'm reminded how much I rely on extensions like AdBlock and TabMixPlus.

While it might be polished, without a good AdBlock extension, it's shit.

And yes, I know about the methods to block ads it supports. That method is buggy and shit.

Re:dev IE9 and dev FF vs release Chrome?

[CarpetShark]

Please stop using it to compare them and Google search for how to block ads in other browser if you need to.

Please study what you're talking about before ranting. Probably everyone who talks about adblocking in firefox has used junkbuster as a general network ad filter, or some windows equivalent.

The point is that adblock is BETTER. Check the chrome bug reports regarding the need for proper adblocking, even though it already had SOME adblocking, much like other browsers do.

Re:dev IE9 and dev FF vs release Chrome?

[Anonymous Coward]

Irrelevant. They still tested with an outdated version of Opera. If they thought it was as unimportant as you seem to, they should have just excluded it entirely.

Personally I love Opera. It does everything I want a browser to do, without requiring tons of addons. It's already replaced a number of other applications for me (ie. emial client, IRC client, bittorrent client, ftp client, personal web server, notepad). I don't really care about how large the PC userbase is, because it's still large enough. You could add in all of the Wii users and a large percentage of mobile phone users (a lot of phones use Opera or a rebrand of Opera) if that is something that you consider a priority for which browser you use.

Re:dev IE9 and dev FF vs release Chrome?

[Anonymous Coward]

Or use PeerBlock. It auto updates its lists, works for all internet connected applications and works on more than just ad servers.

Re:dev IE9 and dev FF vs release Chrome?

[laddiebuck]

BarTab with SessionManager and TabKit (for vertical, hierarchical tabs) has completely transformed my online experience. I now save sessions of research, trip planning, work, in fact almost all online browsing, into hierarchical sessions in my SessionManager categories. Then when I need them later I pull them up, lightning-fast thanks to BarTab. I probably still have 10 windows open but very few active tabs, so my RAM usage isn't high either (not that I care, but it's snappy, and that's what I care about).

Seriously, if you think of switching to Chrome for speed but would miss your extensions, give BarTab (and perhaps SessionManager and TabKit) a try. Treat yourself, you'll be amazed.

Re:dev IE9 and dev FF vs release Chrome?

[awyeah]

My current browser of choice is Netscape Gold 3 on OS/2 Warp. It renders animated "under construction" GIFs with utmost speed!

Real test?

[haystor]

Unless the test contains porn and all the accompanying popups, it's not a real world test.

Re:Real test?

[Anonymous Coward]

You have the idea.

Re:Real test?

[gerardrj]

Isn't a popup the point of porn?

oh, come on.

[Anonymous Coward]

to be honest, firefox, the darling of the FOSS movement is starting to get a little tiring.

it is lagging wayyyy behind...and it's a pain in the ass to manage in fleets. honestly, there is no point to installing it in the enterprise anymore.

the old argument...unpatched days, well who gives a fuck. if users run unprivileged, firefox can't update anyway (and don't tell me to make the directory writable, what about registry edits?)

it's just worth it anymore.

Re:oh, come on.

[Potor]

I must say that I have pretty much totally switched to Chrome. FF 3 and 4 are really dragging my system down, and often fail to load sites I depend on (gmail passwords are now randomly rejected in FF, but work in any other browser).

Re:oh, come on.

[icebraining]

what about registry edits?

Portable Firefox?

Re:oh, come on.

[gman003]

Same here. I can't completely ditch FF, though, since it has all my webcomic RSS feeds set up just so, and Chrome doesn't do Live Bookmarks. And trying to recreate that list in another RSS reader would take days.

Re:oh, come on.

[feepness]

(gmail passwords are now randomly rejected in FF, but work in any other browser).

Yeah, that would be the phishing malware screwing with you.

Re:oh, come on.

[Jah-Wren Ryel]

FF 3 and 4 are really dragging my system down, and often fail to load sites I depend on (gmail passwords are now randomly rejected in FF, but work in any other browser).

You've fucked up your installation somehow. That ain't the default behavior, if it were even close to the default behavior it would be all over the news.

Re:oh, come on.

[TheSeaCucumber]

Let me assure you, the bloat is inbred, default, ingrained and well-integrated into FF. That being said, the extension support is good. Oh, and the speed does actually have a bearing in that FF doesn't flush its cache terribly often. For windows this is fine, but on linux where a system could be up for weeks, it can get annoying closing and re-opening so many tabs. Browsing the internet almost exclusively from within a virtual machine, really eccentuates the effects of a bloated browser, particularly with the limited ram issue.
Chrome has beautiful WebKit integration, and suffers none of the shortfalls of Midori :D

Re:oh, come on.

[mischi_amnesiac]

I would switch if only those stupid developers didn't support the standard "thou must click three times to select everything in the urlbar" debacle with no option to change it. That and the fact that you can't switch the tabbar to the side. Yeah, that one is really very funny with 20+ tabs open.

Re:oh, come on.

[Shikaku]

Let me assure you, the bloat is inbred, default, ingrained and well-integrated into FF.

What IS the default bloat?!

Nobody ever says what it is! EVER! From the default install of Firefox what can you remove that you can consider bloat?

Re:oh, come on.

[brusk]

Config (bookmark, settings, stored password, etc.) syncing, what Mozilla used to call Mesh, is now in the 4.0 beta. That's a feature I don't need or want.

Re:oh, come on.

[Anonymous Coward]

click three times to select everything in the urlbar

+1 agree with you there; I now pause for a second and switch to Ctrl-L keyboard every time I need to fidget with the URL. But on the FF4 beta its double click. I hope they leave it like that; I can live with that.

Re:oh, come on.

[Shining Celebi]

Config (bookmark, settings, stored password, etc.) syncing, what Mozilla used to call Mesh, is now in the 4.0 beta. That's a feature I don't need or want.

Chrome and Opera have had the same feature for a while. Are they also bloated?

"Has features I don't want" is a funny definition of bloated anyway. Tons of people never use tabbed browsing - does that make all modern browsers bloated because that's a feature they don't need or want? Or are your needs the gold standard of bloatedness?

The syncing feature isn't even on by default, so I'm not even sure how it bothers you. I didn't even notice it was there until you posted that.

Re:oh, come on.

[Anonymous Coward]

http://portableapps.com/apps/internet/firefox_portable

4th post

[Anonymous Coward]

from ff beta 6 of course

Re:4th post

[IB4Student]

I've had the prebeta for 7 for awhile, and it scored over twice as well as 6 did on Kraken. I would have liked to see them test that, especially since 6 did a little worse than 5 did for me.

Re:4th post

[IB4Student]

Yeah, I just ran the test myself, and I went from: FF4b6: Average FPS: 14.388 JS Time: 10.667ms Frame T: 57.500ms FF4b7pre: Average FPS: 24.841 JS Time: 1.683ms Frame T: 38.568ms This is on a 5 year old HP prebuilt (Core2Duo @ 1.86 Ghz, 2 GB DDR2-533, etc.)

Maybe time to move to Chrome?

[Shyfer]

I have been thinking about using Chrome for some time, it seems faster and already has a respectable community around it. But I also would like to avoid google stuff and I'm already used to Firefox, not sure if it is worth the trouble.... Any opinions?

Re:Maybe time to move to Chrome?

[FutureDomain]

Try one of the Chrome forks, such as ChromePlus, SRWare Iron, Comodo Dragon, or a pure Chromium build. They're just like Chrome, but without the questionable Client-ID and RLZ modules that Google put in Chrome. I typically use ChromePlus since it has several features that I like builtin, but I've been trying the IE 9 beta and I like that as well. It's faster than Firefox in my opinion and I absolutely love the UI layout.

Re:Maybe time to move to Chrome?

[Anonymous Coward]

Chrome is open source, and you're not alone in not wanting it to talk to google. So as you might expect, there are several derivatives that strip out the google-centric stuff. The first one that comes to mind is Iron.

Re:Maybe time to move to Chrome?

[ChipMonk]

If you're concerned about Google eavesdropping on your browsing habits, you might try the SRWare Iron browser instead. It's Chrome minus the snooping built-ins.

Re:Maybe time to move to Chrome?

[Anonymous Coward]

Try it, seriously you won't go back.

Chrome now has many extensions including some great adblocking extensions.

If you are paranoid about Google, SRWare Iron is Chromium without all the privacy implications.

Re:Maybe time to move to Chrome?

[Shyfer]

Will google some Chrome forks (I should stop using google...) and see how good they are. IE 9 is not an option, closed source, no linux support, no portable version, etc.

From your experience, what is the best Chorme fork? They work with Chrome extensions without problems?

Re:Maybe time to move to Chrome?

[Shyfer]

Oh, also does Chrome and its forks work nicely with linux? I plan moving to linux soon and want to avoid future problems.
And how customizable is it without messing with the source? Does it have decent extensions? And I heard somewhere that now Chrome has native flash support... Is this right? Can it be completly disabled?

Re:Maybe time to move to Chrome?

[Anonymous Coward]

Switch. It is faster and more stable. When flash goes crazy, Chrome kills just flash without the whole browser dying. It is a better experience in every way.

Re:Maybe time to move to Chrome?

[Shyfer]

Wow, so many people saying that Chrome is so much better that I might use it this week to see how good it is.

Re:Maybe time to move to Chrome?

[rmcd]

I use Ubuntu 10.04. I have mostly switched to chrome (not completely; there are still sites that don't work properly with it). My problem with firefox was memory usage. I tend to have *lots* of tabs open and I often don't reboot for weeks. Firefox memory usage creeps up over time and my laptop slows. I keep reading that this is no longer supposed to happen, but it happens to me. Chrome with a comparable number of open tabs does not slow everything else down.

If Firefox were better behaved I would stick with Firefox.

Re:Maybe time to move to Chrome?

[Anonymous Coward]

If you're concerned about Google eavesdropping on your browsing habits, you might try the SRWare Iron browser instead. It's Chrome minus the snooping built-ins.

Why would you trust the SRWare people more than Google? Or at all for that matter?

If you get Chromium from say a trustworthy distro, I would rely on that. Or if I compiled it myself. But who exactly are SRWare?

Re:Maybe time to move to Chrome?

[oddfox]

SRWare Iron was created for the sole purpose of earning the "creator" some money on ad revenue. To borrow from my previous post on the subject:

Everyone mentioning SRWare Iron should know about this little tidbit: The story of Iron. The article and the linked IRC log tell a very interesting story about a guy less concerned with having a good reason to fork and more concerned with making money off of adsense and publicity for creating a "privacy-respecting" Chrome which is basically a perpetually outdated Chromium with a few checkboxes in "Under the Hood" defaulting to off.

The guy who runs that blog does not try to hide the fact that he's a Chrome developer, and he admits that there is the highly unlikely possibility that the person who was asking these questions was not the person who went on to release Iron. I was skeptical as well until I checked out the log file itself and quite honestly it would have to be an incredible coincidence for this guy to be asking such questions and providing the information that he does in his attempts to glean information on the right way to advertise his product as well as how to go about renaming the executable. There's more that makes it very reasonable to believe this is the guy who went on to release Iron, so please don't dismiss it until you've checked out the log file in detail. If this was a supremely unnecessary and elaborate hoax it sure is pulled off convincingly.

Using Iron after reading this information made me feel like I was supporting the wrong guy here and I couldn't do it anymore, it was just too uncomfortable seeing that this guy was looking for adsense revenue and to make a name for himself. The attitude of this developer is not one I would encourage at all.

Re:Maybe time to move to Chrome?

[FutureDomain]

I like ChromePlus the best. They added stuff like builtin Adblock, IE Tab, SuperDrag, Mouse Gestures, and a few other tidbits to the standard Chromium build. Iron was fine, but several of the "privacy infringing" features I actually wanted, such as some sane error pages and automatic updates. I also use the pure Chromium build on my Kubuntu system, and it works almost just like Chrome and it's open source. Chromium is missing the RLZ module (which sends an ID to Google whenever you do a search from Chrome), but I'm not sure whether it retains the Client-ID (a unique identifier generated at installation with potential use for tracking) or not.

Re:Maybe time to move to Chrome?

[satuon]

Alternatively just go to the Options page in google's Chrome version, and uncheck "Use a suggestion service to help complete searches and URLs typed in the address bar". Crome stops the 'snooping' then.

Re:Maybe time to move to Chrome?

[CAIMLAS]

I've been using Chrome for about two years now, I think (mostly Chromium, actually, but the performance improvement does not match the stability/memory use right now, so thinking of going back to Chrome). I gave it a try in version 4, and was underwhelmed. I switched outright when it was first released as available for Linux - in the early alpha stages of 5 - and haven't looked back once.

I was never a 'heavy extension user' in Firefox (at least not since 3.x), but I was almost always "cutting edge" (alpha, beta builds; built it myself a couple times to see if I could get something more satisfactory) up through 3.6 to get the better performance. There were performance improvements. They did not outweigh the problems:

* Javascript would frequently hang the whole browser or cause it to crash. This was true for stable as well as dev builds.
* Flash. Flash was basically useless (Hulu and youtube mainly) unless I ran Firefox with the 'flash page/program' as the only tab.
* Waiting. I was tired of the whole browser freezing up for a period of time (especially on a slower system) while trying to load a heavy page or one with crap javascript (like Slashdot).
* Crashing would frequently lose all open tabs, and occasionally lose tabs even with session manager. No session data was usually saved in such an event, either. Then I'd have to wait as it restarted.

Currently I'm running Chromium 7. Every time I touch my wife's laptop to look up something (she's got Firefox + a bunch of extensions she must have) I tend to get agitated by the slowness and generally inferior feel.

I was initially resistant to Chrome due to the UI (in the earlier versions - not sure which, but they were Windows only), specifically the keyboard shortcuts. Having used Mozilla, then Phoenix in 2003 (which changed names several times until it became Firefox), I was really ingrained in doing several things certain ways, and it was infuriating when they were so markedly different as to not work at all/be possible.

With 5, I can go back and forth between Firefox and Chromium and at least the keyboard input (for what I use, at least) is identical.

The Session Manager functionality was the other biggie for me in switching. I'm the type who has 40+ tabs open normally, so having this feature work as well as it did was a big sell.

Something i'd been trying to figure out how to do in Firefox for some time (on account to having a little fire and losing all but my system with 512MB RAM) was figure out what was consuming so much damn memory, making me hit swap. I've since gotten new hardware, but Chrome also has a built in process/tab manager which tells me nice things like that. I've only had to use it once or twice, on ancient hardware.

Multiprocess browsing, leading to individual pages being more responsive (especially on a multicore system) when one is locked up is also nice (though that happens very infrequently with Chromium due to the apparently better js engine, I suspect).

I'm trying to think of something which displeases me about Chrome/Chromium, and I really can't. It runs acceptably well on all my systems, including those which are really not much use in general (500MHz celeron, W2K, 386MB RAM). It's fast even there, provided I don't open too many tabs.

The biggest "down side" I can think of is that there is not a "stop"/halt button, but in hind sight this isn't a problem like it was with Firefox (where I'd have to hit stop on occasion just to get my browser back from hungry javascript). It just isn't an issue with Chrome.

While I would not say that I prostitute myself for Chrome, I have converted a couple die-hards over to Chrome/Chromium from their previous browsers of choice: a die-hard MS fan who liked IE8 (though he's probably going to stick with IE9 now that it's available in beta); an Opera user; a die-hard Firefox user who used a million hideous extensions, and several others of lesser dedication. They've all been quite pleased, for one reason or another, with Chrome/Chromium.

Re:Maybe time to move to Chrome?

[ADRA]

As a guy who doesn't specifically care what to use besides IE, I can say that I always end up going back to FF from chrome. I try it for a few days, but I can't explain it but there's always something that I need in FF. The browser just seems slightly less likely to work with sites (not necessarily the browser's fault mind you), and I get crashes in it. Despite what I hear from others, I rarely get FF crashing on me. Now that there's plugin isolation, I imagine its less crash friendly. The browser is still a little herky jerky on opening a larger number of flash enabled tabs( like 3-4 youtube pages at the same time), but nowhere near the almost complete freeze of the browser from background loaded tabs at it used to do. I'm using the beta FF right now which seems to work alright.

Re:Maybe time to move to Chrome?

[ADRA]

Browser open several days fully loaded with things? It takes all of 5 seconds to reload a browser. If its really that large of a burden what's the problem? I'm amazed at the behavior changes that I've adapted too now that browsers save open tabs on close/crash. Bookmarks have seen substantial decline, and I don't leave a browser open unless I'm in core development time or looking something up online. With a 5 second start time, there's really no bother for me. Maybe if you have dozens of media rich tabs constantly doing things the load may get larger, but I really can't see a use case for something like that on a regular or even semi-rare basis, though everyone's different.

Re:Maybe time to move to Chrome?

[Anonymous Coward]

Who the fuck modded that up? Doing that doesn't remove half the stuff that SRWare Iron does. You're still under constant Google surveillance even with those options unchecked.

You're still left with the RLZ token (cannot be disabled at run time), the constant ping back to Google, and the unique client ID. None of that can be disabled without using a custom build of Chromium. If you can't do that yourself, then the kind people at SRWare provide it for you.

I mean, it's not like the link was RIGHT THERE for you to click on to see what SRWare Iron actually does. No, wait.

Re:Maybe time to move to Chrome?

[satuon]

Oh, also does Chrome and its forks work nicely with linux?

Yes, the linux version has been available for quite some time now.

Re:Maybe time to move to Chrome?

[Zelgadiss]

I have been thinking about using Chrome for some time, it seems faster and already has a respectable community around it. But I also would like to avoid google stuff and I'm already used to Firefox, not sure if it is worth the trouble.... Any opinions?

Trouble?

Good lord, you make switching browsers sound like you packing up your life and moving to another city or something.

Just install it and import your bookmarks, the UI is pretty much self-explanatory if you have used a browser before.

Re:Maybe time to move to Chrome?

[Anonymous Coward]

SRware Iron is well known to be both unnecessary and shoddily modified. Just use chromium. Why the fuck do people keep suggesting this piece of shit? Ever actually used it?

Re:Maybe time to move to Chrome?

[TheRaven64]

I should stop using google.

I switched to DuckDuckGo a while ago for search. It links to Google at the end of the list of search results, so you can fall back to Google if you don't find what you need. I've only done this a few times, and Google hasn't given helpful results for those queries either. It uses SSL by default, doesn't log anything identifiable, and has a much nicer UI than recent Google.

Re:Maybe time to move to Chrome?

[h4rm0ny]

You just made my day. Chrome runs wonderfully but I don't trust Google. I'm going to try some of these forks which I hadn't even known were there.

Thanks,
H.

Re:Maybe time to move to Chrome?

[codeshot]

Do the chrome forks fix the chrome 6 beta rendering bugs. It fails the acid2 test (acid2.acidtests.org). Like firefox 4 beta 6 it draws the nose half a pixel too high, but then goes on to draw it half a pixel to the right. In addition chrome 6 beta has the eyes and their part of the face flick into the page only substantially later than the rest of the face. This test is supposed to be a static test with no dynamic content.

Re:Maybe time to move to Chrome?

[Elbows]

I'm using Chrome on Ubuntu 9.10. It's definitely fast -- I have an older system and Firefox is just painfully slow, Chrome is pretty useable. But when I get around to upgrading my hardware I'm probably switching back, for a few reasons:

• Zoom/font size settings. Text on most web pages is so small I have to squint (I have this problem all the time: I crank up the minimum font size in Firefox, set my Windows fonts to 125%, etc). Chrome will remember your zoom setting per domain, which is nice, but every time I visit a new site I have to set the zoom again.
• Stability. Yeah, Chrome is supposed to be super-stable with the process-per-tab architecture. But I've crashed the whole app, and it took chunks of my profile with it (had to reimport from Firefox). The flash plugin crashes frequently on videos that play fine in FF. I can't remember the last time FF crashed, and it always recovers cleanly.
• Extensions. There are a few extensions that I miss, and I find myself firing up Firefox now and then just for those.
• Updates. Firefox is in the Ubuntu repositories, so it gets updated automatically. Chrome is a .deb that I have to download and install manually.

Chrome did have flash support right after install -- not sure if it was built in or just found the flash plugin from Firefox that was already on my system. I don't see a way to disable it. There is a flashblock extension, but I haven't tried it yet.

Re:Maybe time to move to Chrome?

[mikechant]

Updates. Firefox is in the Ubuntu repositories, so it gets updated automatically. Chrome is a .deb that I have to download and install manually.

Is there some reason you don't use Chromium instead, which *is* in the Ubuntu repositories (at least at 10.04)?

Re:Maybe time to move to Chrome?

[Anonymous Coward]

I just tried it. 2 pop ups less than 30 seconds after installing. And yes, options are set to block pop-ups. Verdict: I'm uninstalling immediately.

Re:Maybe time to move to Chrome?

[oddfox]

Works fine over here on Chrome 6.0.472.62 beta, the Acid2 test looks exactly like the reference.

Re:Maybe time to move to Chrome?

[Terrasque]

From ChromePlus page:

ChromePlus is considered the virus by some antivirus(such as kaspersky) recently. To resolve this problem, please update the antivirus to its latest version. If the problem still exist, please post the antivirus' product name to our forumcn: (http://forumcn.chromeplus.org/index.php), we will communicate with the antivirus officical as soon as possible.

A project with that bad English on the front page... Tends to make a few alarm bells to start ringing.

Re:Maybe time to move to Chrome?

[codeshot]

I've got the same chrome with both acid2 and its reference loaded in separate tabs so I can flick between them. The nose moves up and to the right by a tiny bit when flicking from the reference to the live render. Also there is a slightly darker yellow vertical line to the left of the nose that shouldn't be there indicating that the section containing the nose is shifted to the right by about half a pixel - which would explain the nose shifting right by about the same amount.

Re:Maybe time to move to Chrome?

[codeshot]

The offending div has auto margins, I guess the browsers are picking half pixel values. Although chromes element inspector shows the correct computed values, its acting like right and top are half a pixel less.

Re:Maybe time to move to Chrome?

[Elbows]

Hey, so it is. :)

I haven't really been paying attention to the countless forks of Chrome, but I'll give that a try.

Re:Maybe time to move to Chrome?

[Goaway]

Do not use SRWare Iron, it is made by a less-than-honest developer who likes to spread FUD about Chrome so he can make money off Iron, by his own admission.

If you want to not use Chrome, use Chromium.

Re:Maybe time to move to Chrome?

[Goaway]

Be aware that the author of Iron lies a whole lot about what Chrome does versus what Iron does.

Re:Maybe time to move to Chrome?

[oddfox]

Right you are, I didn't notice at first that minute difference.

Re:Maybe time to move to Chrome?

[Anonymous Coward]

Try smoking crystal meth and you will find that tabs and windows start opening all over the place. You can end up with massive amounts of tabs and windows that will actually stall FireFox completely for minutes at a time when you reload it.

Re:Maybe time to move to Chrome?

[Kalriath]

No, SRWare Iron is Chromium with a couple of preprocessor directives.

Re:Maybe time to move to Chrome?

[laddiebuck]

One word will fix your problems; BarTab. I was in the same boat as you, I should know.

Re:Maybe time to move to Chrome?

[rmcd]

Never heard of it, it seems to work great. Thanks!

no sir, i didn't like it.

[gandhi_2]

No native Gopher support?

From my cold, dead hands!

Re:no sir, i didn't like it.

[Tumbleweed]

No native Gopher support?

I don't care if it's native Gopher or imported Gopher, as long as it's here legally.

Re:no sir, i didn't like it.

[Anonymous Coward]

No native Gopher support?

From my cold, dead hands!

Yes sir. I am sitting right here. Of coarse, because I am 72 years of age, I have been sitting a long time.

Frames per Second?

[Anonymous Coward]

Frames per second seems like pretty much the opposite of "real-world" for how 99% of users use their browsers.

Re:Frames per Second?

[citizenr]

and it looks like Chrome was the only browser not syncing to screen refresh

Re:Frames per Second?

[cyfer2000]

Especially after the frame rate lock applied to firefox.

Re:Frames per Second?

[BZ]

That rate lock only applies if you use the new "let me know when you plan to paint and I'll update my DOM/CSS so the user sees the right thing" API. This benchmark isn't doing that it's just using a "1ms" timer (which is still rate-locked, to 200Hz in Chrome and 100Hz in other browsers).

Speaking as someone that switched to OS X

[aussersterne]

from Linux this month after using Linux since 1993, I think this applies to all of FOSS.

Somehwere around 2000-2006 FOSS was basically head-and-head with commercial software in practical usability and maintainability, with its own distinct advantages and a relatively small learning curve.

Then there was this veer into "if you ever want all the Windows users to switch..." thinking, and in an effort to eliminate the learning curve FOSS threw away pretty much all of its advantages as well. If FOSS is just Windows/Mac OS/IE by another name, why choose FOSS?

Particularly when Windows/Mac OS/IE win on the polish, compatibility, and accessibility fronts by virtue of their being cathedral-built software?

With Firefox slow and cumbersome, Thunderbird choking on Gmail IMAP continuously while Apple's Mail.app sails along happily, and KDE4/GNOME3 being emblematic of the many ways in which FOSS has lost its way, I just decided I'd had enough of the nonsense. I'm ready to be able to walk into Best Buy, purchase any device, and expect that it will work seamlessly with the current generation of computing devices, without options, without Bugzilla (and condescendingly dismissive developer retorts), and without lots of consulting Google to find out how the gconf infrastructure has changed in the last two years or how HAL has been replaced by DeviceKit or policies moved from /etc tree A to uneditable dynamic filesystem B (but just use this easy command line management tool to set options...)

It just plain saves me a boatload of time and headache to use something else, like OS X plus Google apps plus Chrome. The pending desktopization of FOSS has fizzled thanks to the politics of the bazaar.

Re:Speaking as someone that switched to OS X

[Moridineas]

Hear, hear. I'm right there with you.

As an example along similar lines, a user at my office reported a bug in Thunderbird to me. I tested and found it was definitely a Thunderbird bug. I made a test case file and submitted it to Bugzilla. A few days later my reported bug is deleted, to be merged with the same bug report from *2005*

Nobody who works on Thunderbird felt like working on the bug. It's not a sexy bug, probably doesn't hit too many people, and has work arounds...so it's stayed in the software for ~6 years.

And yeah I know, I should go in and fix it myself. Maybe one day I will. In the meanwhile I'll keep using Mail.app and I'll move more users over to new versions of Outlook that actually seem somewhat decent, and we'll go from there.

Is there any quality email app for Windows??

Re:Speaking as someone that switched to OS X

[Jah-Wren Ryel]

Let me see if I understand your complaint - a rare bug with workarounds is given extremely low priority and that's indicative of a general problem with FOSS software? What do you propose as a change to the FOSS model of development to improve the engineering?

Re:Speaking as someone that switched to OS X

[Anonymous Coward]

You are just getting old. There are distros that do the same as 5 years ago, you just don't care about "freedom" anymore, not now that you can afford a Mac.

Linux on the desktop is farther than ever after selling its soul to proprietary software.
A regular user doesn't use Linux without nVidia, Adobe and a DVD player. They are encouraged to do so as well. They grow in an environment where using proprietary software is acceptable and the only reason to use a free alternative is that it costs nothing.

Old corrupted people like you will assert that using the shortcuts allows for more free software to be used, but the fact is like you people will leave the boat after justifying that they can also use free software on MacOS X or Windows 7. An Ubuntu memory space is mostly filled with proprietary crap anyways.

Switching to Mac only highlights your elitism as the reason to use Linux in the first place. Anyone with any technical knowledge would know that Windows 7 is the best OS hands down, both in technology and in features. If I was to sell my soul to the devil I would certainly buy a Windows 7 machine. The only thing worth MacOS has is Objective-C.

Re:Speaking as someone that switched to OS X

[Anonymous Coward]

Pegasus Mail

Re:Speaking as someone that switched to OS X

[Qubit]

submitted it to Bugzilla

Test case + bug #, or it didn't happen.

Okay, so it's entirely possible that you ran across a bug, but absent us actually looking at the particular bug and seeing how relevant it is to how many users, it's hard for us to judge how "fair" it was for the bug to have existed but not been fixed since 2005.

Or are you going to tell us that every bug found in Mail.app since 2005 has been fixed? (if Apple even has a public bug tracker and never redacts it...)

Re:Speaking as someone that switched to OS X

[guanxi]

I made a test case file and submitted it to Bugzilla. A few days later my reported bug is deleted, to be merged with the same bug report from *2005*

Nobody who works on Thunderbird felt like working on the bug. It's not a sexy bug, probably doesn't hit too many people, and has work arounds...so it's stayed in the software for ~6 years.

And yeah I know, I should go in and fix it myself. Maybe one day I will. In the meanwhile I'll keep using Mail.app and I'll move more users over to new versions of Outlook that actually seem somewhat decent, and we'll go from there.

How have Apple and Microsoft handled your bug reports for Mail.app and Outlook? Did the handle them like Mozilla, where you enter the bug directly in their internal bug databases, monitor the progress, participate in discussions with the developers, and even contribute development yourself? Or do you have no idea what the status is, no influence on the outcome, and no ability to contribute at all? Were the bugs even submitted to development? Were you able to find a way to submit them to Apple and Microsoft at all -- could you communicate with anyone beyond level 1 end user support technicians?

Every application has bugs as old as its first release-- have you seen the age of some Windows security vulnerabilities, going back over a decade? -- and your particular concern won't necessarily get fixed. But if you compare the experience of handling end user bugs at Mozilla with the same thing at Apple or Microsoft, well, there really is no comparison.

Re:Speaking as someone that switched to OS X

[dbcad7]

Outlook.. pretty much all versions including the Expresses, all have issues.. some that have not been fixed.. Do a Google search for "Outlook will not save my password" .. Don't know about "quality" but as far as least number of support calls, actually that would be Windows Mail, or Windows Live mail.. I absolutely hate Outlook calls.., because although it should be a simple matter of filling in the boxes, and checking the right check boxes, it rarely turns out that simple.

Re:Speaking as someone that switched to OS X

[msormune]

You are my hero. That was easily among the top 5 posts I have ever on Slashdot, if not the best.

Re:Speaking as someone that switched to OS X

[Anonymous Coward]

new version of Windows Live Mail is surprisingly good. Yes yes, rebranded outlook express, but just test it, or at least google screenshots for it :-)

Re:Speaking as someone that switched to OS X

[Anonymous Coward]

obvious troll is obvious.

Re:Speaking as someone that switched to OS X

[dbcad7]

I think we started with Linux close to the same time frame.. My first install was Slackware with the root and boot floppies, if that tells you anything.. I went through the phases.. learning to compile from source, trying different distros with different package managers,. trying all the different window managers.. Etc.. But gradually, I developed a preference for all these things, and then found the distro that most matched it.. Where I have stuck.. I don't have to reinvent my system anymore.. been there done that.. Now I just keep my system upgraded as needed. Minor learning curve if any. I'm not bleeding edge or experimental or testing anymore.. I don't feel the need to be, and I just don't have the problems you got fed up with. I work support for the other 2 guys (Windows/Mac), and although I probably wouldn't have most of the problems that most of my customers have with their systems, it still makes me satisfied with what I am doing with my own system.. But to each their own.. I don't regret the things I have gone through and what I learned in "the old days" of Linux, anymore than what I learned with Windows when it had DOS. I don't feel that Linux has lost it's way at all.. If I get left behind with all the new fangled whiz bang "improvements" Than MS and Apple bring,, so be it.. I'll just continue to be the oddball nerd in the family running Linux.

Re:Speaking as someone that switched to OS X

[andersenep]

Do you realize how much FOSS is part of OS X? http://www.opensource.apple.com

Re:Speaking as someone that switched to OS X

[Kjella]

Let me see if I understand your complaint - a rare bug with workarounds is given extremely low priority and that's indicative of a general problem with FOSS software? What do you propose as a change to the FOSS model of development to improve the engineering?

Well, first here we have a user. He sure as hell wouldn't have found the workaround. Then it gets reported to an administrator, who can't find any known bug (assuming he searched, but I would first) who then spends time building a test case. Then it goes to a bug triager who knows about five year old bugs and can identify this as a dupe and provide a workaround.

1. 99% of users would never find that workaround
2. After that effort by three people, nothing will really be done to fix it

Even better are the projects that drive down bug count by attrition, WINE is rather notorious for this I've noticed. Every few WINE versions - who are on a biweekly schedule - they'll ask you to retest even though there's been no patches towards the bug, so when you grow tired of that shit the bug is "solved".

As for rarity, any bug that doesn't happen on a developer's machine for a developer's use case is by assumption rare unless a real shit storm of complaints prove otherwise. Don't get me wrong, it's pretty much central to OSS that people fix the stuff they wanted fixed, no bug is too obscure if you're willing to fix it yourself. But it also very much creates an A-list and a B-list of bugs to get fixed, and you go in B. It's actually slightly easier with commercially paid support who don't really have an agenda of their own, what the customers most want fixed is their priority.

Of course no, a single anecdote from single project doesn't prove anything. Sometimes I've had good experiences, but also many bad. Most annoying ar those where a bug report only leads to more and more work until it far outweighs my interest in fixing the bug. I'm guessing that's often the case for the OSS developer on the other side of the table too, which is why he is pushing 90% of the work right back at me. Often the choices then end up: a) keep using buggy software, b) spend way too much time getting it fixed or c) buy something that works.

I wish there was more of an organized bounty system, you'd pledge towards some bug fix or feature request, then developers could take it on. The pledges would be kept by a trust until the developer claims to be done (or withdraw), then the pledgers vote to confirm / claim incomplete / reject that it has been done. You'd probably need to have some sort of arbitration system to deal with formal disputes, who could take a processing fee off the pledge. Combine that with an eBay-like reputation system and I think it should work out well without too much hassle. I know it sounds a little like rent-a-coder and I don't mean it like that, more of an organized way of doing small custom work inside existing projects.

Re:Speaking as someone that switched to OS X

[NuShrike]

No kidding! I get the same experience with Qt by Nokia where I report a memory bloat bug with Qt Creator or portability bugs with their blog examples. They either delete the bug or ignore it. Wtf?

Seems to push their buttons if you submit bug fixes too.

Re:Speaking as someone that switched to OS X

[PipsqueakOnAP133]

While I haven't tried Microsoft.... since I can't figure out if some things are bugs or features, Apple has this: http://developer.apple.com/bugreporter/
It enters it into Apple's internal bug database.

And when I entered a bug against Safari 5.0.1, it was fixed in 5.0.2. Works for me.

Re:Speaking as someone that switched to OS X

[Anonymous Coward]

"All developers who have registered [for free] as an Apple Developer may submit bug reports and enhancement requests on Apple products and track these submissions using the Apple Bug Reporter."

https://bugreport.apple.com/

Re:Speaking as someone that switched to OS X

[cbhacking]

Windows Mail (in Vista, moved out-of-box to Windows Live Essentials and called Windows Live Mail for 7) is actually a surprisingly good mail client. POP3, IMAP, and SMTP all seem very stable and have handled every server I've tried thim with, ranging from my school's Linux boxes to Gmail. It supports spam filtering, phishing filtering, and filter rules (not as good as Outlook's, but comparable to Gmail's). Decent-quality S/MIME support. Hmm... I'm not sure about LDAP, I'd need to check on that. It starts up fast and closes quickly (unlike Outlook), is very stable, and has a pretty sane memory footprint. It also handles being offline for a while quite well, better than Outlook (haven't tried Thunderbird for this).

It also is a nice newsreader, although the need for such seems to be very rare these days. It doesn't have groupware integration and can't communicate on the Exchange protocol, but does integrate with Windows Calendar and can hand off .ics files to it nicely.

Re:Speaking as someone that switched to OS X

[mwvdlee]

no bug is too obscure if you're willing to fix it yourself.

I once found a bug in the de-facto standard framework for Delphi and Borland C++ Builder internet communication.

It was a simple bug; the dates in SMTP headers used a locale-specific time seperator, ':' in most locales. In Italian locales (and undoubtedly many others) it's '.'. It should always be ':' since that's what the relevant RFC states.

I found the bug, wrote the patch, tested the patch, submitted the patch, explained the problem and one-and-a-half years and a couple of tries later... it still wasn't in the official release. At that point I gave up. It was an open source project, but you couldn't get access unless you wrote major contributions; a one-line bugfix isn't major.

So being willing to fix bugs isn't enough if the project team is a bunch of arrogant douchebags.

Re:Speaking as someone that switched to OS X

[selven]

I'm curious, what did Linux throw away in 2006-2010? And can you get most of it back by moving from Gnome/KDE to a simple window manager?

Re:Speaking as someone that switched to OS X

[Anonymous Coward]

I think Bugzilla still blocks links from Slashdot, but let's try this search. (You may have to copy and paste, Mozilla got annoyed at people pointing out bugs from Slashdot threads.)

That query returned 453 bugs in the "New" status (or "Confirmed" as everyone else calls it) that have been open for at least five years, and that are blocking or critical.

Keep in mind that since they're in the "New" state that means no one has worked on them at all. If someone was working on them, it'd be "Accepted" - these are bugs that have been verified to exist, but no one has done anything with them. Including 10 marked as blocking.

Re:Speaking as someone that switched to OS X

[MrHanky]

You'll find there are known bugs in Windows and OS X that aren't getting fixed as well. It's not a FOSS-only thing. I was pissed off at Apple for not fixing a bug in chsh that would render Terminal.app unusable (and only fixable from a terminal, of course) -- I'm pretty sure it would be labeled Release Critical in Debian and never found in Stable, but I don't think Apple ever fixed it. It probably went away when they abandoned the horrid NetInfo garbage in 2007.

Then again, FOSS software can be more susceptible to developer weirdness, and FOSS nerds tend to be weird. KDE's usenet reader has been broken since forever, and apparently the developer likes it that way.

Re:Speaking as someone that switched to OS X

[wrook]

The funny thing is, I'm kind of in agreement with you, but have a different conclusion. People ask me what desktop environment I use. I tell them I don't use a desktop environment. I hate these bloody integrated monoliths. For example, with Empathy I have to change my sound theme from the Gnome configuration tools and turn on sound notifications for my desktop. If I don't do that I don't get sound notifications in Empathy. But what do I do if I want to run Empathy without Gnome?

Free software used to be about choice -- choice to use whatever I want to use. Distributions like Ubuntu try to integrate things into one monolithic package where they choose what you will run. This is egged on by the desktop environments like Gnome and KDE who have always had that as a goal -- let's reduce choice so as not to confuse the user.

But my conclusion is NOT to switch to OS X, where I'll be presented with even less choice. My conclusion is simply to choose a distribution and tools that meet with my philosophy. OK, so Gnome and KDE aren't my thing. Ubuntu is not my thing. There is still a lot of free software options that don't try to integrate everything into one already-chosen-for-you-so-dumb-users-shouldn't-need-to-touch-this block. Other people are free to choose their integrated tools (and I know a lot of people, probably most people, love them).

Free software is still about choice. Some organizations are purposely trying to create a kind of walled garden where people won't get overwhelmed. But those of us who like the wild world can still thrive. What confuses me is what your beef is. You say that Gnome and KDE have lost their way, and yet you run to OS X which you admit is more of the same. You're trading your freedom for polish. Well, each to their own I guess. Personally, I'm not going to go there.

Re:Speaking as someone that switched to OS X

[koiransuklaa]

You know what they about statistics... just showing a number like that isn't the whole truth. Omitting context here is not lying but it is dishonest.

As an example, you could have mentioned that your list includes bugs from _all_ products in the Mozilla bugzilla, things like Venkmann, a js debugger that hasn't existed for five years or so. You could also have mentioned that 453, while an impressive number of forgotten bugs is actually 0.075% of all reported bugs. Or you could have noted that the vast majority or those bugs only have 0-5 votes which in b.m.o means a very obscure bug, and that most of the remaining bugs don't really belong in Critical...

I'm not claiming that the those bugs don't include real, important bugs, I'm saying your figure alone tells us _nothing_ useful. B.m.o is a massive and ancient issue tracker, things do get lost in there, but that doesn't prove that the process is totally broken. If you can't understand this, then you haven't worked with massive issue trackers. In the corporate world this problem is often "solved" by a QA cleanup: sweep under the rug any issues that are too obscure and where no progress is being made. That looks a lot better in the reports but doesn't really help the users.

Re:Speaking as someone that switched to OS X

[TheRaven64]

I have open bugs in Apple's system that have been there since 2004. It took them two years to fix a kernel bug that I found, and there are some bugs in the Cocoa frameworks that are still unfixed. They quietly fixed a security bug I found in the screensaver about six months after I reported it.

Re:Speaking as someone that switched to OS X

[strikethree]

Yes yes yes yes! However, OS X is not so wonderful either. It is a pretty face on a Unix that is not terribly reliable. I have had Wine/Cider games take out video and keyboard access. I have had apps refuse to launch until after a reboot (reboot a Unix? WTF?!) Checking the logs (since there were no error messages) indicates some sort of graphics overlay error.

Somehow, Linux and the associated software got sidetracked from doing kickass things to doing stupid shit. Linux in a command line only environment is still relatively decent though.

It would be interesting to find out how all of this happened though. How did KDE decide to move their entire platform to something radically unstable and unfinished (and distros ship it!). How did Gnome decide to take away choices from users? How did Gentoo get derailed and a "default by the book" install end up with a non-functioning desktop environment? Seriously, there are so many things that have gone terribly wrong that is seems impossible to think the entire community just lost its sense. WTF happened?

Re:Speaking as someone that switched to OS X

[h4rm0ny]

A bug that has been there for approximately six years, mind you. That's almost always a bad sign. For projects I manage, I operate a priority escalation system. A reported issue might be trivial and get pushed to the bottom for the first couple of weeks. But after that, it's a "minor". Give it another month or so (depends on the project) and I'll bump it up to a "major". If you don't do this, you end up with a lot of cruft and an untidy mindset for the project. Yes, you have to prioritize, no that can't mean that some things are effectively never going to be dealt with. Priority escalation creates a good feel on a project as well, like getting the dishes done that you've meant to do for ages. If management is supportive of you taking the time to fix these older problems (and I am. ;) ), then it makes for a healthier project.

Re:Speaking as someone that switched to OS X

[Anonymous Coward]

I submitted several bugs to Microsoft over glitches in DirectShow (part of DirectX) filters.
I did get a reply that due to the bugs being in an abstraction layer between DirectShow and WMF (Windows Media Foundation), I should use WMF (not an option).

I submitted a SERIOUS bug to NVIDIA ~5 years ago that prevented proper clipping of video windows if the window was partially offscreen. The bug was side-stepped by microsoft with the introduction of AERO which used a different window-composition layer, but finally, after more people raised this issue over and over on NVIDIA forums/support mail, it did get fixed, taking about 4 years from report to fix. The bug was introduced in v8x.xx of their drivers and fixed in v196.xx

Re:Speaking as someone that switched to OS X

[Moridineas]

"Rare" was your word, not mine. The fact that it's been submitted to Bugzilla numerous times over the past 5-6 years should tell you something. (Especially considering what a small portion of bugs users hit actually get reported)

The response their--and your attitude here as well--is perfectly indicative of some of the problems that users have with OSS software. "It's good enough!" or "fix it yourself."

That's exactly what the GP was talking about. Polished vs unpolished. A lot of FOSS software just ends up being unpolished. It may have absolutely tremendous features, but especially on the desktop side so many apps tend towards the unpolished. I can sympathize, but it does make it harder to want to use software like Thunderbird, when you can find plenty examples of known bugs lingering for YEARS.

Re:Speaking as someone that switched to OS X

[Moridineas]

Test case + bug #, or it didn't happen.

Sure. And apologies for getting the year wrong--it's actually been since 2003.

https://bugzilla.mozilla.org/show_bug.cgi?id=222747

I can't tell you how many Mail.app bugs haven't been fixed. I can tell you that I get a lot more complaints from Thunderbird users than from Mail.app users.

Another bug which I don't even know how to submit to Bugzilla and so haven't. User is composing an HTML email. While typing, midsentence, and for no apparent reason, the font changes. I thought the user was doing something dumb like clicking elsewhere in the email, but no, the font / style just randomly changes mid-sentence.

It's these kind of bugs that are SO irritating to users.

Re:Speaking as someone that switched to OS X

[Moridineas]

Here's another example (somebody on Slashdot posted this awhile back and I bookmarked it for hilarity)

https://bugzilla.mozilla.org/show_bug.cgi?id=92165

Re:Speaking as someone that switched to OS X

[Moridineas]

I already replied with the one I submitted, but here's another one that I first saw referenced on Slashdot.

https://bugzilla.mozilla.org/show_bug.cgi?id=92165

It's just _PAINFUL_ to read through this bug! It's almost 10 years old now!

And on the topic of crazy bug reports...here's another (non-Mozilla):

http://sourceware.org/bugzilla/show_bug.cgi?id=4980

Re:Speaking as someone that switched to OS X

[m50d]

Linux has lost its way. FreeBSD just sits here, quietly working the way it always has (still on OSS for sound, would you believe, and it works beautifully).

/recent freebsd convert

Re:Speaking as someone that switched to OS X

[aussersterne]

I haven't replied to many posts in this thread because I am beyond tired with this discussion. That's why I switched to OS X.

But let me tell you: "your figure alone tells us nothing useful" is precisely why some of us switch. Maybe it tells you nothing useful, but to a bug reporter the bug is TREMENDOUSLY IMPORTANT and they very much hope that you FIND IT USEFUL.

I used to contribute patches.
They used to be included.
Then they were included less, and less, with discussions about project aims.
Then I stopped sending patches and would just send bug reports.
And they would at least stay in the system.
Then I started getting detailed instructions on how maintainers wanted bug reports submitted.
Then they started asking me to do things like build the latest revisions of KDE or GNOME subsystems and re-test, or buy another [laptop|display card|monitor|hard drive] and see if the problem persists.
Then bugs started getting closed out more and more quickly with "WONTFIX" and snide comments about how the behavior I desire is ideologically incorrect.

The last straws were a series of "WONTFIX" and "NOT AN INTERESTING BUG" responses, despite me suffering things like massive data corruption.

"Behaves badly," "not that interesting," "won't fix," and "_nothing_ useful" are precisely the sorts of things that will quickly drive a person to another operating system, especially when three or four years ago they had a perfectly working Linux system and workflow but somehow a couple years later they're fighting system and maintainers All The Freaking Time to hold it together.

I'm a user. I'm not interested in being a developer. I want to USE.

Elsewhere in this thread someone said that in Windows/Mac OS you never actually get to personally interact with the maintainers, implying that this makes FOSS better.

Two responses to that:

(1) In other systems you know quite simply what the system does and what it doesn't do and you buy/choose accordingly. In FOSS you live in the fuzzy gray world of questions: How is it supposed to work? Is this a bug? Is it a feature I'm not using correctly? Should I file a report? Will the feature I'm using now disappear next month because my use case isn't interesting? There is an undercurrent of instability here demonstrating that the fact that you feel the need to talk to maintainers all the time is not necessarily a Good Thing[TM].

(2) 90 percent of the time these days, what you get from FOSS maintainers is: WONTFIX, UNINTERESTING, BEHAVES BADLY, or UWANT-UFIX. That's not actually all that much better than simply sending off a bug report to M$ and then crossing your fingers for the next three years that someone helps you out while doubting that it will occur.

Re:Speaking as someone that switched to OS X

[knarf]

It is all well and good that you feel the need to justify your investment in Apple's products but that does not mean that 'FOSS has lost its way'. All it means it that you, for whatever reason, were swayed - by what or whom I don't know - to buy into said companies product line.

Meanwhile that free software which according to you - a new Apple user - suddenly has lost its way is happily making inroads in more and more markets. That Apple phone which you might feel the urge to buy after your switch to an Apple PC? It is in heavy competition with not one but several free software based phones. If you have not yet felt the need to replace all other computer-related products by Apple-branded versions there is a big chance that that router also is based around free software. You might even have experienced one of the benefits of this fact when you installed a more featured version of the operating system on that router - suddenly it could do much more than what was stated on the box.

There will be people who tell the opposite story of what you just told. They will, for their own reasons, have moved from Apple - or Microsoft or who knows which - products to free software. 'Apple has lost its way' they will proclaim, stating that in the beginning of the Mac OS X time they were doing their best to outcompete Microsoft and how they lately just seemed to feel the need to own everything and control every aspect of the use and where possible also the users of their products. That those people tell this tale does not make it true - you are a shining example of how Apple still appeals to some people.

I hope you enjoy using your Apple products as much as I enjoy using - and creating - free software. You are always welcome to use them, hardly any strings attached...

Re:Speaking as someone that switched to OS X

[iris-n]

Oh that's amazing!

I hope they don't ever fix it; and Thunderbird dies a slow and painful death.

Another example is this:

https://bugzilla.mozilla.org/show_bug.cgi?id=299346

Thunderbird would randomly corrupt and then delete your address book. It's five years old.

Re:Speaking as someone that switched to OS X

[Anonymous Coward]

But if you compare the experience of handling end user bugs at Mozilla with the same thing at Apple or Microsoft, well, there really is no comparison.

You shouldn't make general statements on top of your ignorance: http:connect.microsoft.com

Re:Speaking as someone that switched to OS X

[Jah-Wren Ryel]

"Rare" was your word, not mine.

Are you seriously giving me shit over a synonym for "probably doesn't hit too many people?"

The response their--and your attitude here as well--is perfectly indicative of some of the problems that users have with OSS software. "It's good enough!" or "fix it yourself."

Actually it is indicative of what happens when you have limited resources. That applies to every software project known to man and has nothing to with FOSS. Maybe you've just never seen the inside of a case tracking system for non-free software but having worked for a few of the biggest software developers in the market I can assure you that they are packed full of minor bugs that are unlikely to ever get fixed unless a customer with lots of clout decides to make a stink.

The fact that you can cite examples of similar bugs in free projects doesn't mean FOSS projects are less polished it means FOSS projects, by definition, let their dirty laundry air in public.

Re:Speaking as someone that switched to OS X

[Bungie]

Actually the reporting process on Microsoft Connect is relatively active and you do participate with the developers and others in discussion. I have submitted other product feedback for the Windows Live suite which are just submitted via web and you are only contacted if the developers need to get ahold of you...but the issues always get fixed in the next version which is fine for me.

OSS projects needs to understand that the end user doesn't give a shit about participating in developer discussion or following it every step of the way...they just want their problem fixed.

As for the speed of Mozilla bugfixes, FireFox 3.x still can crash and lose your bookmarks and profile data which is an age old bug which has greatly frusterated many FireFox users for as long as I can remember. I don't care about the bug reporting experience when I'm rebuilding my lost shit, and it's never made me less pissed off when it happens.

Re:Speaking as someone that switched to OS X

[Anonymous Coward]

Claws mail is worth a looksee.

Re:Speaking as someone that switched to OS X

[Moridineas]

Are you seriously giving me shit over a synonym for "probably doesn't hit too many people?"

I'm not quite sure why you're getting upset over this (and don't think it's worth it) but it's a valid distinction. If the spectrum is Common, Uncommon, and Rare, I would suspect this bug hits people in the "uncommon" range. Since it's been reported so many times over the years, it's more than rare. I couldn't have expected you to know that right off since I didn't post the link initially, so, my mistake.

Actually it is indicative of what happens when you have limited resources

No, I think you're wrong here bymissing the point of what I and several others in this thread are saying. It's beyond obvious to say it's a question of limited resources--with unlimited resources we could hire a programmer to fix every bug that ever arose. The problem is FOSS allocation of resources. Secondly, we are saying that open source software--and specfically desktop software--tends to be unpolished. This largely means UI. I really think defending FOSS UI is a difficult task, but if you want to take that one on, you're welcome to!

Programmers like to come up with new indexing algorithms/engines, or they like to come up with clever solutions that make xyz 5 times faster! There are certain programming tasks that more people like to work on. Not nearly so many people want to do boring, rote UI work, or the "details" that make some programs nice to work with, and some programs a bear. Look at the other Bug I referenced about renaming files, and look at the discussion. It's PAINFUL.

And please note, I'm not whining. I think FOSS is fantastic for many things, and some foss communities are fantastic about most things. FreeBSD I think does a good job up and down the line (including documentation). The server programs -- Apache, Samba, etc -- tend to be really fantastic. I just don't think the virtues of foss development seem to translate nearly so well into some other venues.

Re:Speaking as someone that switched to OS X

[Jah-Wren Ryel]

I'm not quite sure why you're getting upset over this (and don't think it's worth it)

If that's how you feel, why did you come out swinging about word choice in the first place?

Furthermore when I looked at the bug it sure seemed rare to me - only happens when replying to and quoting certain types of messages with one specific formating method from one mail client and all it does is spell-check the htmls tags in addition to the real text. Rare and non-critical - it belongs at the back of the line.

The problem is FOSS allocation of resources.

Best I can make out is that you want allocation of resources to tasks that are less than critical. Maybe you don't realize it, but that's the only end point available. The whole bit about "programmers like to invent new and faster algorithms" is a canard - there have been a ton of rote-work fixes in the UI and elsewhere in tbird and all of the major FOSS apps. You only see the empty half of the glass.

Look at the other Bug I referenced about renaming files, and look at the discussion. It's PAINFUL.

Nothing I haven't seen on case trackers for proprietary software. The occasional outliers don't indicate anything.

And to take the issue back to assersterne's post which you were "right there with" he was complaining about stuff not "just working" out of the box. Hardly something you can blame on FOSS when the only reasons stuff "just works" on other systems is because the manufacturer has paid someone to make it so and hasn't bothered with anything else. That mis-directed complaint is older than linux.

Anyone who has been around a while can go anecdote to anecdote about proprietary stuff that is "unpolished" too - drivers that fail in corner case scenarios, anti-virus software causing all kinds of applications to behave mysteriously. Hell my current winxp box has had screen/window redraw problems for years and its regularly updated with patches from MS and nvidia - the damn sound card driver blue-screens the system if I feed it 5 channel sound instead of 5.1 channels and its been that way for nearly four years now, despite being current on patches. That vendor doesn't even have a bug-tracking system I can look at.

In summary, my criticism of your point of view is that you have unrealistic expectations that are founded on cherry-picked anecdotes.

Re:Speaking as someone that switched to OS X

[Moridineas]

If that's how you feel, why did you come out swinging about word choice in the first place?

Sorry, I'm really not sure what you're getting so upset about here? All I said is that you used a word different from what I said--and that this changed meanings. Don't know what you mean by "came out swinging" but I'm sorry if I offended you.

Furthermore when I looked at the bug it sure seemed rare to me - only happens when replying to and quoting certain types of messages with one specific formating method from one mail client and all it does is spell-check the htmls tags in addition to the real text. Rare and non-critical - it belongs at the back of the line.

Nope, that's not it. It's that spellchecker spellchecks the contents of the tag. More than just "one particular mail client," and common enough to be reported time and time again over 8 years. 8 years is a long queue!

The whole bit about "programmers like to invent new and faster algorithms" is a canard - there have been a ton of rote-work fixes in the UI and elsewhere in tbird and all of the major FOSS apps. You only see the empty half of the glass.

Sure you're right, and again, what you're saying is very obvious. Neither I nor anybody else here has claimed that NO work goes into the UIs of open source software, just that they almost uniformally lack polish and lag in usability.

I cheerfully grant that in a programmers view of priority, UI often comes last. That's part of the problem.

And to take the issue back to assersterne's post which you were "right there with

Take a look at his later posts about bug reports.

Anyone who has been around a while can go anecdote to anecdote about proprietary stuff that is "unpolished" too - drivers that fail in corner case scenarios, anti-virus software causing all kinds of applications to behave mysteriously. Hell my current winxp box has had screen/window redraw problems for years and its regularly updated with patches from MS and nvidia - the damn sound card driver blue-screens the system if I feed it 5 channel sound instead of 5.1 channels and its been that way for nearly four years now, despite being current on patches. That vendor doesn't even have a bug-tracking system I can look at.

In summary, my criticism of your point of view is that you have unrealistic expectations that are founded on cherry-picked anecdotes

Still missing the point. I've not once mentioned drivers, or blue screens or any such. I am talking about primarily desktop apps. Things like Thunderbird. Firefox. Openoffice. Even KDE/GNOME as entire entities I think fall into this too.

It's not an indictment of FOSS as a whole, I think many bazaar style open source projects cdo utterly fantastic work. What FOSS projects that are deskstop programs do you think have a great interface with great polish?

I think Firefox comes closest, nothing else in the Mozilla suite would I call polished or even particularly usable for non-techies.

I really am somewhat surprised...I've seen it admitted over the years even by hardcore FOSS believers that open source apps frequently lack good user interface design. I do think things are better now than a decade ago, but -- IMHO! -- usability remains a big problem for desktop apps.

Re:Speaking as someone that switched to OS X

[Jah-Wren Ryel]

All I said is that you used a word different from what I said

A declarative statement like "your word, not mine" generally indicates that the writer believes the word to be deliberately misleading.

So far, nope all evidence points to "rare" being pretty accurate. You say that this bug has been reported a handful of times over the course of 8 years - I see that as pretty freaking rare. You see "8 years" - I see hundreds of thousands if not millions of users of which only a handful have been bothered enough to report it - maybe representing 20x as many who haven't bothered to report it. 200 out of say a very conservative 200,000 - that's still extremely rare.

You say "checks the contents of the tag" does that mean something other than checking the tags? It's not a critical failure if the spell checker hits on things like "href" and URLs. Is there something else that is being checked that I'm missing here? Furthermore, according to my reading of that bug it is JUST one client - outlook or exchange configured to do rtf formatted messages that causes the spell-checker to do that.

Take a look at his later posts about bug reports.

huh? What does that matter? You were "right there with" something he had not even written yet?

I cheerfully grant that in a programmers view of priority, UI often comes last. That's part of the problem.

No, you don't cheerfully grant - you declare and base your entire argument on that declaration. I don't agree - see the part where I called it a canard. Doing UI work well is hard, but that's true of all the other areas of programming too, they are all hard to do well - its the doing well part that is difficult. You just don't see all the sloppiness in the non-UI parts of the code because if it works fast enough with few enough errors you have no idea what cruft is under the hood. You may not realize it, but you are cherry picking your anecdotes.

Still missing the point. I've not once mentioned drivers, or blue screens or any such.

But the guy whom you were "right there with" did when he complained about stuff not "just working" out of the box. All those things are examples of stuff just not working in such a way as to be perceived as very "unpolished" to the user.

Re:Speaking as someone that switched to OS X

[Moridineas]

A declarative statement like "your word, not mine" generally indicates that the writer believes the word to be deliberately misleading.

Ok, since that's your interpretation I can do nothing more than again apologize if you felt I was being unfair to you.

You say "checks the contents of the tag" does that mean something other than checking the tags?

Slashcode stripped out what I wrote. It checks the contents of <style> tags. So if a message, pasted content, your signature, a message you're replying to (etc) has a <style> tag in it, you get spelling matches. Are you looking at the same bugzilla report? I'm on a different computer from the one I have it bookmarked on now, but I think I linked to it from a previous message.

I agree it's not a "critical" error, but I do think it's a non-minor one, and a major annoyance for users!

huh? What does that matter? You were "right there with" something he had not even written yet?

No, no, you misunderstand--if you read his other message I think you might find some interesting stuff. (relating to bug reporting)...that was why I referenced it. Sorry if that was unclear.

No, you don't cheerfully grant - you declare and base your entire argument on that declaration. I don't agree - see the part where I called it a canard. Doing UI work well is hard, but that's true of all the other areas of programming too, they are all hard to do well - its the doing well part that is difficult. You just don't see all the sloppiness in the non-UI parts of the code because if it works fast enough with few enough errors you have no idea what cruft is under the hood. You may not realize it, but you are cherry picking your anecdotes.

Sorry, but I think you're still missing the point about UI design and polish. For the average user (heck, for the vast majority of all users, average or not!), "slopinness in the non-UI" parts of the code don't matter if it works. For instance in this example of the spellchecker bug. It's been around for 8 years and you say correct, it's rare, non-critical, and goes to the end of the queue. But if you're a clueless non-programmer user it's a BIG bug. If you don't know what these "rgb" or "background-color" or whatever else things are or where they are coming from, that's the kind of inexplicable impossible to solve error that will drive you away from FOSS! That's not something I think anybody wants...

In this example, by your standards Thunderbird works just fine -- email is sent correctly. It's just a bumpy road for the users!

I guess it ultimately comes down to this though. I point to the corpus of FOSS desktop applications and say by and by, the FOSS model has not produced many polished apps. You say well not all cathedral developed apps are good either. And of course we're both right. As an example I've been aggravated by Quark XPress bugs for about two weeks. Good software on one level, but unpolished for the last decade.

I do believe that in general FOSS programs are not as polished nor do they have as well developed user interface design. I do believe that in general programmers working on FOSS software tend to be less interested in user interface design and less interested in what I'm calling "polish." But I think at this point we just agree to disagree.

But the guy whom you were "right there with" did when he complained about stuff not "just working" out of the box. All those things are examples of stuff just not working in such a way as to be perceived as very "unpolished" to the user.

Oh I see. Let me put it this way -- when he complains about "Bugzilla (and condescendingly dismissive developer retorts)," I agree. When he complains about "polish, compatibility, and accessibility," I agree. When he complains about "Firefox slow and cumbersome, Thunderbird choking," I agree. And that was the main thrust about which I've written since then.

The other comments about OS issues are more tangential, but ones with which I also happen to agree (though I haven't said a thing about them here).

Re:Speaking as someone that switched to OS X

[guanxi]

"All developers who have registered [for free] as an Apple Developer may submit bug reports and enhancement requests on Apple products and track these submissions using the Apple Bug Reporter."

https://bugreport.apple.com/

Interesting, and I wasn't aware of it. In fairness though, it's for developers only, and does it provide the transparency and other benefits that Mozilla does?

Re:Speaking as someone that switched to OS X

[guanxi]

Actually the reporting process on Microsoft Connect is relatively active and you do participate with the developers and others in discussion

Well, I admit I didn't know about that. Is it for Technet or MSDN members only, or can anyone participate? Is there transparency about the status of your bug?

OSS projects needs to understand that the end user doesn't give a shit about participating in developer discussion or following it every step of the way...

I run a business that provides end user support and my experience differs. Of course some feel that way, but many do want transparency and participation. Per your comment above, apparently Microsoft seems to see demand for it too.

As for the speed of Mozilla bugfixes, FireFox 3.x still can crash and lose your bookmarks and profile data which is an age old bug which has greatly frusterated many FireFox users for as long as I can remember. I don't care about the bug reporting experience when I'm rebuilding my lost shit, and it's never made me less pissed off when it happens.

I don't think this is true. I submitted a bug about eight years ago on a similar problem, I think with Mozilla Suite (Firefox and Thunderbird's predecessor, now called Seamonkey), but haven't seen it in probably 7 years. We support many Firefox users and they don't have the problem. I used to volunteer in Mozilla's support forums, and while there were crashes, very few caused dataloss.

However, if you make your arguments more angrily, disparage others, and swear a little, I might be convinced!

Re:Speaking as someone that switched to OS X

[guanxi]

As I responded to someone else, I was unaware of that and it is interesting. What are the differences? It seems to be open only to developers. Also, is there transparency -- can you see the status? Can you participate, either in the discussions or by contributing code?

Regardless, thanks for pointing it out.

Re:Speaking as someone that switched to OS X

[dbcad7]

I tried to reproduce your spell check error in replying to an HTML email.. everything worked just fine.

Re:Speaking as someone that switched to OS X

[Moridineas]

Was there a <style> tag?

It's a bug that's been present since 2002 I think, but in the case I submitted, the user had a html sig file that hadn't been modified in years, and only in a fairly recent update to Thunderbird did the spellchecker start tagging it. Not sure why...

I could replicate on multiple comps.

Re:Speaking as someone that switched to OS X

[PipsqueakOnAP133]

Yes, you require an account, which is free. Actually, I haven't tried but I wouldn't be surprised if a standard user's iTunes account works for it too. Otherwise, you simply register at developer.apple.com

I have minimal experience using this particular interface.
Given that the bug I filed was pretty direct and comprehensive, there was no discussion nor response. It simply got done.
From what I understand, if the developer who gets the bug assigned wishes to, he can contact you back through it or email. But there isn't much in transparency; I wouldn't be surprised if there were log entries I can't see. Friends who work on campus have mentioned that this is typical. The reasoning being that if you don't directly work on the project, given the secrecy, there's little difference between the rest of the company (outside the project) and the rest of the world (anybody who happens to file a bug report).

I've never tried submitting code through it. I'd think that if somebody submitted code, and a developer saw it and really liked it, it could be picked up. But like any larger organization, people usually have stuff they're busy working on, so they probably don't have the time to integrate outside code.

It's not going to be as interactive as an open source project, no doubt about that.
But, given what I know, it does allow a person to feed directly into the project developer's workflow. I think that is pretty cool. Especially for a company where employees arn't even supposed to know about the existence of other projects.

Re:Speaking as someone that switched to OS X

[koiransuklaa]

Maybe it tells you nothing useful, but to a bug reporter the bug is TREMENDOUSLY IMPORTANT and they very much hope that you FIND IT USEFUL.

Easy cowboy, no need to shout. I only explained why the number is not a useful statistic for evaluating the quality of development. Your experiences, as interesting and useful for improving the Mozilla development process as they may be, are not relevant to that argument.

Please realize that my argument had _nothing_ to do with the source license or development model, just the difficulty of evaluating software QA process efficiency. Bugs get lost, forgotten and buried in proprietary development and open source, working with any massive bug database will teach you this. Saying "453 BUGS!!!" does not tell us whether Mozilla is doing well or not.

Re:Speaking as someone that switched to OS X

[guanxi]

I support many Thunderbird users and I've definitely seen the spellcheck bug. I've never seen the font change bug. Since you've seen it more than once, as a guess, maybe it's some customization implemented across the office (e.g., and addon?)

I don't think the spellcheck is very consequential -- just ignore it and move on -- and I support many Thunderbird users and they have not complained about it. I still wish it were fixed and Thunderbird could use some spit and polish, but overall our users love it. It's fast and efficient.

FOSS is a tool like any other: Good for some things, less good for other things. If you want a highly polished app, FOSS is often a poor choice; it doesn't seem to be in its DNA, at least for community-based projects. If you want professional support, that can also be a challenge with FOSS. If you want low purchase price, good functionality, security, confidentiality, access to highly technical resources (bug databases, developers, etc.), adherence to standards, etc., FOSS is pretty good.

Re:Speaking as someone that switched to OS X

[Moridineas]

I support many Thunderbird users and I've definitely seen the spellcheck bug. I've never seen the font change bug. Since you've seen it more than once, as a guess, maybe it's some customization implemented across the office (e.g., and addon?)

We don't generally have any addons installed, and the user in with the font change bug definitely doesn't have any. I assume it's some quirk in the HTML composer, and that probably most of the devs don't use HTML email (I know I despise it). Never have been able to replicate it with any real repeatability...

I don't think the spellcheck is very consequential -- just ignore it and move on -- and I support many Thunderbird users and they have not complained about it. I still wish it were fixed and Thunderbird could use some spit and polish, but overall our users love it. It's fast and efficient.

The user (now users) who complained about it to me were ones who have the spellcheck set to spellcheck every message before sending. Fortunately there are workarounds for some of the cases.

FOSS is a tool like any other: Good for some things, less good for other things. If you want a highly polished app, FOSS is often a poor choice; it doesn't seem to be in its DNA, at least for community-based projects. If you want professional support, that can also be a challenge with FOSS. If you want low purchase price, good functionality, security, confidentiality, access to highly technical resources (bug databases, developers, etc.), adherence to standards, etc., FOSS is pretty good.

I absolutely 100% agree with you here. One of the other people in this thread took great umbrage when I asserted the same thing, but I definitely feel as you do. I love many foss packages but in many cases do wish for some more polish. Having said that, it's not like all commercial software is perfect or polished!

Mozilla needs to get Jamie Zawinski back...

[slyborg]

...to ragequit again.

"Real World"?

[farnsworth]

I'm not sure what is "real world" about spinning a UML box around another UML box in a giant (presumably) canvas-based javascript app.

For me, "real-world" means: is gmail fast enough? is opening a new tab fast? is image rendering fast enough? is html video fast enough? is the occasional embellished html5 animation fast enough? is typing into the address bar fast enough?

I'm sure their diagramming app is cool and everything, but I don't think I've ever seen anyone use anything like it, so I'm not sure what is "real world" about using it for a benchmark.

They even said that they altered the test in the middle to fix IE's performance problem. Come on.

Re:"Real World"?

[Danieljury3]

For me, "real-world" means: is gmail fast enough? is opening a new tab fast? is image rendering fast enough? is html video fast enough? is the occasional embellished html5 animation fast enough? is typing into the address bar fast enough?.

I use gmail in basic HTML mode. It takes away some of the things I never use and, in my opinion looks much cooler (none of that modern looking nonsense). I also barely ever type into the address bar.

Re:"Real World"?

[Anonymous Coward]

Gmail is an example of a "nearly everything is javascript" type of an app. More and more, people are adhering to such a philosophy, similar kind of application development is gaining lot more traction. This has had some interesting effects for e.g. people have started building things like Aves Game Engine. Going forward, being performant/efficient in areas such as rendering might become crucial.

I guess what I am trying to say is "This may not be the real world of today, but it could be the real world of tomorrow". Does it justify the tests as being critical at this point? Meh, no! All it says right now, is chrome can do some shit better than firefox, but in my opinion firefox still has time and dedicated userbase to play catch up (or move beyond). Hopefully, it wont end up being our next IE.

Re:"Real World"?

[wrook]

I've got old and underpowered hardware. For me "real-world" means "how much bloody memory does it use". It doesn't really matter how fast they make it if my machine starts thrashing.

Re:"Real World"?

[Anonymous Coward]

I'm not sure what is "real world" about spinning a UML box around another UML box in a giant (presumably) canvas-based javascript app.

For me, "real-world" means: is gmail fast enough? is opening a new tab fast? is image rendering fast enough? is html video fast enough? is the occasional embellished html5 animation fast enough? is typing into the address bar fast enough?

I'm sure their diagramming app is cool and everything, but I don't think I've ever seen anyone use anything like it, so I'm not sure what is "real world" about using it for a benchmark.

They even said that they altered the test in the middle to fix IE's performance problem. Come on.

I'm sure that, for most people, real world performance is "fast enough" in whatever browser they're using. GMail is probably "fast enough" for most people in IE 6, otherwise GMail would have a lot fewer users. Ditto pretty much everything else.

Performance testing is always about pushing the envelope.

Translated from Redmondese this means...

[schmidt349]

... it took Microsoft to finally produce a web browser that performs as well as other browsers were doing in their previous generations, and meanwhile the one they actually have in production (IE8) sucks eggs. So what else is new? Have they got a modern filesystem yet?

Re:Translated from Redmondese this means...

[schmidt349]

Er, correction: "it took Microsoft this long to..."

Re:Translated from Redmondese this means...

[Anonymous Coward]

Too busy spreading anti-MS FUD to proof read? You suck an ass. LOLZ!!!!!

Re:Translated from Redmondese this means...

[DAldredge]

What features do you think needed to be added to NTFS to make it a modern file system?

Re:Translated from Redmondese this means...

[RightSaidFred99]

Obvious troll is obvious...and very, very boring.

Re:Translated from Redmondese this means...

[RightSaidFred99]

Nothing. What's funny is if you want to talk about some primitive ass shit, 90% of the real UNIX world still uses NFSv3. 16 group limits? Literally allowing the client to tell the server "Oh, I'm Joe Bob - trust me!". Seriously?

Re:Translated from Redmondese this means...

[JonJ]

How about making the fucking filesystem _not_ fragment and piss itself regularly? That would make it atleast semi-modern.

Re:Translated from Redmondese this means...

[Super_Z]

NFSv4 was released in December 2000, almost 10 years ago. The fact that people still use NFSv3 is that it simply works and it suits most peoples needs. And NFSv3 is not the only protocol that is vulnerable to spoofing.

Re:Translated from Redmondese this means...

[Anonymous Coward]

People use NFSv3 because migrating to a real fix like Kerberized NFSv4 is a huge change in the way they work.

NFSv3 sucks, and it's inferior in most ways to a modern CIFS environment.

UNIX is hardened to external attackers pretty well, but if you've got a normal, non-root login in a large UNIX environment the odds are with just a little work you can get root and access anything you want locally or over NFS.

Windows is the opposite. The attack surface for external attacks is far larger than UNIX, but it's fundamentally superior in terms of intra-organizational security.

Re:Translated from Redmondese this means...

[Super_Z]

Read up on the features of ZFS and Btrfs. Copy-on-write would be a start.

Re:Translated from Redmondese this means...

[thaig]

performance

Re:Translated from Redmondese this means...

[TheRaven64]

Amusing. The post above you requested a filesystem that didn't fragment. You request copy-on-write, which massively increases fragmentation. With ZFS, every single write adds a new fragment. If you write a file, then modify a few bytes in the middle of it, CoW means that the file will now be in 3 fragments. This cripples performance on mechanical disks, which is why ZFS needs a lot of RAM for the ARC and recommends a big blob of flash for the L2ARC.

The separation of policy and mechanism in NTFS is actually quite similar to the design of ZFS, so implementing CoW semantics would be relatively easy to do at the policy layer. Like ZFS, NTFS differentiates between the low-level on-disk storage mechanism and the high-level user-visible layout.

Re:Translated from Redmondese this means...

[Shikaku]

Newsflash: all filesystems fragment. The idea that extX doesn't is a fantasy and if you believe it it's no better than people believing Linux/Mac doesn't get viruses and trojans.

Ext3 and 4 however are resistant to it though, but currently there's no way to defragment them.

Re:Translated from Redmondese this means...

[Anonymous Coward]

Any modern file system should have case-sensitive file and directory names.

Re:Translated from Redmondese this means...

[TheNetAvenger]

Have they got a modern filesystem yet

Yes they did, it is magical and called NTFS and still is the FS that outperforms most *nix FS technology and offers 10x the features, and is considered the holy grail of non-MS OS FS technology that all *nix users get excited when a FS gets close to NTFS, like when ZFS was to be the next generation for *nix, and it even fell short of NTFS on features.

Thanks for playing, "Not only a Troll, but a Dumb Troll."

Re:Translated from Redmondese this means...

[Anonymous Coward]

How about making the fucking filesystem _not_ fragment and piss itself regularly?

Define "piss itself regularly".

Re:Translated from Redmondese this means...

[DAldredge]

How do I make NTFS "piss itself"?

Re:Translated from Redmondese this means...

[shutdown -p now]

NTFS supports that - it's controlled by a switch. It defaults to case-insensitive because the entire Win32 API layer depends on this assumption (and, naturally, most third-party DOS/Windows software does, too). But e.g. when you install SUA (Unix compat layer) on Vista/2008/7, one thing that it'll ask you is whether you want to switch the target NTFS partition to be case-sensitive.

Re:Translated from Redmondese this means...

[Anonymous Coward]

[citation needed]

Re:Translated from Redmondese this means...

[Bungie]

NTFS does avoid fragmentation in the same ways that ext3, HFS+ any any other modern filesystem try to do. It will naturally write files to sequential blocks when they are available. To completely avoid fragmentation the system would have to find or organize a large enough number of sequential free blocks every time it tried to do a write. If you changed a file so that it requires another free block the whole thing would have to be moved to a sequential area of free space. Imagine how things like the swap file would be impacted by that kind of thing.

As for pissing the file system I don't know what you could be doing because the design of NTFS is pretty robust. The MFT is mirrorred on completely different areas of the disk and structured in a binary tree (instead of chains like FAT). The filesystem itself is journalled and uses bitmaps, as well as supporting things like file snapshots. I've never seen NTFS "piss itself" unless there was a hardware failure somewhere.

If you stop unplugging your drives before they can flush the write cache (hint: Safely Remove Hardware) your NTFS pissing problems will probably go away.

Re:Translated from Redmondese this means...

[marcosdumay]

"NTFS does avoid fragmentation in the same ways that ext3, HFS+ any any other modern filesystem try to do. It will naturally write files to sequential blocks when they are available."

You should take a look on how ext3 works. Yes, it does write files to sequential blocks when they are available, but that is not enough to avoid fragmentation.

Re:Translated from Redmondese this means...

[Super_Z]

In fact, when you modify a file, you write a whole new block, leaving the old block which contained the old data in place. This will not introduce more used blocks (as the old block is freed) and will in fact give a better write performance as the disk head does not jump around seeking blocks to edit.

Whether copy on write can be introduced in NTFS or not is completely beside the point. NTFS does not currently have this feature.

Re:Translated from Redmondese this means...

[TheRaven64]

You are assuming that the workload is composed entirely of writes. Back in the real world, workloads are generally read-heavy, by a fairly large margin. Fragmentation caused by CoW degrades read performance. You hard drive can do around 100-200 seeks per second, so if you are reading heavily fragmented files the speed can drop to around 100KB/s. Since each read will displace the head away from the current insert point (with ZFS, by the way, there isn't just one place where writes can happen - it's not LFS), you still need to seek back to it to write, just as you'd have to seek to the middle of a file to write. If you are using rotational media, you can't do CoW without a very large cache or a performance hit. ZFS recommends 1GB of RAM for the kernel (around 600MB for the ARC) and a large flash drive for the L2ARC to mitigate this.

When the majority of desktops and laptops have SSDs, then CoW will make a lot of sense in a desktop filesystem. At that point, the fact that it's trivial to add it to NTFS will be important. At the moment, the fact that it is not present is not.

I really don't understand this irrational Microsoft bashing. I haven't used Windows for about 7 years, but even I can accept that they did some things right.

Re:Translated from Redmondese this means...

[Super_Z]

Hyperbole much?

ZFS concatenates many small random writes into large sequential write operations. This clearly gives better write performance. As for file defragmentation, WAFL features automatic defragmentation. Similar defragmentation is also in the works for ZFS.

Re:When do you declare a browser dead?

[Shikaku]

Netcraft has to confirm it first.

Ugh, poor benchmarking

[Anonymous Coward]

Firefox 4 beta 6 doesn't have the new JavaScript engine in it. Beta 7 will have it. But there's no particular need to wait for beta 7 as they could benchmark a nightly now. They also don't mention what kind of video card they've got in that laptop. IE9 and Firefox 4 can take better advantage of a good video card on Windows 7 than the other browsers tested and that may significantly influence a charting benchmark like this one.

Re:Ugh, poor benchmarking

[Lennie]

They should also check when doing a benchmark if it was actually engaged and test again with the hardware acceleration disabled.

Figure of Speech

[Anonymous Coward]

I love how a difference of a few milliseconds (looks to be 5ms) means a browser "tanks" and its position, when compared to other browsers, can be described as "dead last." Oh no, we're not painting a bias picture here.

Re:Figure of Speech

[Gamer_2k4]

That's a good point. Once you get to the point that the difference between the slowest browser (not counting IE8) and the fastest browser is ONE ONE-HUNDREDTH OF A SECOND, how much does it really matter any more? And Chrome gets nearly 120 FPS? Do monitors even refresh that quickly?

I understand that some people really do go nuts about benchmarks; my old roommate was one of them. But at the end of the day, those differences are truly negligible. What matters most is how functional the browser is.

Re:Figure of Speech

[BZ]

Modern OSes refresh LCD monitors at 60Hz.

Not Surprising

[Local ID10T]

The results were surprising. IE9 held its own pretty well (with a few caveats), and the latest Firefox 4 beta came in dead last.

No. Not surprising at all... but I still prefer Firefox for everyday use.

Re:Not Surprising

[Teknikal69]

I agree and their wasting their time messing with the GUI instead of fixing speed issues and memory bugs I've switched from Firefox myself now and I've been using it since the first betas. To be honest the upcoming GUI change was the straw the broke the camels back I'm still using Gecko based browsers but not firefox. It's a shame really what everyone liked about firefox was the leanness and now it's just bloat and getting worse with every decision they make.

Re:Not Surprising

[Lennie]

I can tell you why this wasn't surprising, because the currently latest Firefox Betas do not yet include the new Javascript engine which will be in the release version.

How comprehensive ...

[Anonymous Coward]

It's a beta. Don't waste your breath.
Also, 2 charts?

jaegermonkey

[yakumo.unr]

I'd be much more interested to see it being done with the builds of FF 4 that have jaegermonkey enabled. Though that should be merged into the main branch fairly soon with any luck.

http://www.conceivablytech.com/2673/products/first-look-firefox-4-jaegermonkey/

Re:jaegermonkey

[BZ]

It has been, already. Trunk builds of Firefox have it, as will Firefox 4 beta 7.

Re:jaegermonkey

[Bj�rn]

And it looks like optimization work is still very much ongoing. http://arewefastyet.com/

Re:jaegermonkey

[yakumo.unr]

The js-preview builds have it, as shown in the article that I linked.

It's not in Trunk yet, I run the latest nightly build, about:buildconfig is still lacking the --enable-methodjit option

Re:jaegermonkey

[Lennie]

But beta 7 hasn't been released yet.

Re:jaegermonkey

[RebelWebmaster]

JM is very much on trunk already. It's not something that's enabled via .mozconfig. Look for the methodjit options in about:config. Content is on, Chrome is off currently.

Re:jaegermonkey

[BZ]

There is no such build option anymore. It's on by default. There's a --disable-methodjit to disable it, but if you don't see that in your about:buildconfig then it's compiled in. And at runtime whether it's on is controlled via preferences.

Re:jaegermonkey

[BZ]

Yes, that's why I said "will". Future tense and all.

About that link

[dbIII]

Your country functions by having a huge cheap workforce of illegal immigrants so of course nobody in Federal Government is really going to do anything to stop that - not Democrat and not Republican. Also whoever wrote that stuff needs to learn how to use English and what a dictionary is for - "communist" means something completely different instead of just being a generic swear word.

Re:About that link

[Anonymous Coward]

Your country functions by having a huge cheap workforce of illegal immigrants so of course nobody in Federal Government is really going to do anything to stop that - not Democrat and not Republican. Also whoever wrote that stuff needs to learn how to use English and what a dictionary is for - "communist" means something completely different instead of just being a generic swear word.

?

Re:About that link

[Anonymous Coward]

?

I know, right?

Re:About that link

[MichaelSmith]

Well lets talk about my country for a minute. 200 years ago group A (the English) were over here taking property from group B (the Aboriginals). I don't think that was communism. And now that some members of group C (descended from group A) want to give some property back to group B, I don't that that is communist either.

Re:About that link

[wavedeform]

But if you take property from group A and keep it, you're either a capitalist or a thief, depending on your PR person and/or lawyer.

Re:About that link

[Tumbleweed]

However, anyone who think it is okay to take property from group A and give it to group B is a communist. You can't take it from group A unless you believe property is yours to take.... held in common.

Okay, then, so you're giving up: medicare, social security, non-toll roads, all forms of insurance, public schools, libraries, the armed forces, etc? Guess what, all those things are possible ONLY by taking stuff from Group A and giving it to Group B. Throwing around words like socialism and communism is just further proof you're impossibly ignorant. But that's okay, because I'm sure it FEELS true to you.

Re:About that link

[gandhi_2]

the basis of this free nation is the limited government to secure life, liberty, and property. you take such a casual stance with respect to loss of liberty.

Name one appointee

Van Jones.

The red scare was nonsense even when it was happening.

Except that the Soviet Union actually WAS attempting to infiltrate public institutions and influence US actions thru the entire cold war. Man, even NPR talked about that.

Re:About that link

[gandhi_2]

For fucks sake, you argue like a kid. Like everything's a dichotomy.

If you take money from the rich and give it to the poor, that is fundamentally wrong. If you take it from the middle class to give to the corporations, that is fundamentally wrong.

If all groups agree to contribute towards the creation of common property (roads, armed forces), then that is part of the social contract formulated among free people. But why should upper-middle class pay a greater percentage of their earnings toward roads than the lower-middle class? Do rich people wear out roads faster? Do the cops spend more or less of their budget year dealing with rich or poor people?

There are problems with the Bush tax cuts, since there are rich people paying nothing. Everyone should pay in their share to the social contract, but NO ONE should be cashing out.

Obama, et al, really do think your money belongs to them first. To be doled out as they see fit. It's YOUR money, you just agree to let them have a portion to support the social contract.

I listen to people like you, a good chunk of slashdotland....it's like the Enlightenment never happened. Like Rousseau, Voltaire, and Jefferson didn't get thru to y'all the importance of government as a social contract and consent of the governed.

Re:About that link

[CAIMLAS]

Except it isn't Group C giving things back - it's a subset of group C, giving for everyone. They're taking from others in group C in order to be generous.

All said scenarios are theft. It just so happens that the current scenario happens to do so through the wealth redistribution methods outlined by Karl Marx, a man who wrote a manifesto which became the basis for modern "democratic socialism" and communism alike. The problem? Even people like Hitler and Mussolini were nice guys who came off favorably at first. Then they got power and control of the state through Marxist agendas. Oops!

Re:About that link

[Weasel Boy]

But why should upper-middle class pay a greater percentage of their earnings toward roads than the lower-middle class?

Marginal income. A lower-middle-class family might have to spend, say, 85% of their post-tax income on basic necessities, whereas an upper-middle-class family might be able to meet those same needs on 60% of their income. Wealthier people have more cash left over after their basic needs are met. That's why we have progressive marginal tax rates.

Do rich people wear out roads faster?

Yes, they do. They take public transit less, and they drive more and bigger cars.

Do the cops spend more or less of their budget year dealing with rich or poor people?

Rich people have a greater expectation of service from the cops than poor people do.

Obama, et al, really do think your money belongs to them first. To be doled out as they see fit.

The only difference between Obama and Bush is that Obama thinks the rich should pay taxes.

Re:About that link

[amorsen]

If you take money from the rich and give it to the poor, that is fundamentally wrong.

You can't make a statement like that and expect it to be universally agreed with. You'd be in the far minority around here with that opinion.

Re:About that link

[Kjella]

It's YOUR money, you just agree to let them have a portion to support the social contract.

Don't abuse the words "social contract" as if it were an individual contact. If so, show me where I signed it. Nor did I vote for it, and even if I boycotted the election that is hardly consent. It is important in the sense that the right to govern is given by the people, not some divine right, royal heritage, nobility, military force or something like that but the people forces their will on the individuals. That I could escape the jurisdiction of the contract by leaving the country is not the same as being able to abstain from it, nor is it certain that I could, that any other country would accept me or that it'd be better there. Being the smallest evil does not make it voluntary.

Some don't recognize the contract at all (anarchists), some only recognize negative rights, some only at the most basic level of positive rights like the right to a fair trial (which would require neutral courts, which would require taxes). On the other extreme of the scale are those who'd give up all property rights and most any other right if the social contract required it, communists and Marxist socialists, with most people somewhere in between. You're beating down an open door, regimes without consent of the governed are universally despised. The question is rather how much the governed may impose on the individual, as opposed to individual freedom.

I live in one of the more socialist countries in Europe, and there's plenty that I would call undue meddling in my life. I can't go buy a beer in the shop on 7 PM on a Saturday, because the sale closes at 6 PM by law. I can't in any way complain about the democratic process (we're a monarchy, but it's not practically relevant), we have a good selection of political parties, regular and reliable elections every four years and while I in some cases can blame the government for running their own way in this case I'm convinced it's because we have too many busybodies and nanny state proponents in the people.

I just don't think it's right even if a majority thinks so, that kind of interference just shouldn't be part of the social contract. Who will stop the people from grabbing too much power? What stops the people from taking too much taxes, from taking your property simply because they're in the majority? There is a lot of talk of the balance of power within the branches of government, but there's very little balance outside it, the government is an instrument of the people over the individual. In some ways I guess I want it too, I wouldn't want every lone nut be able to do anything. But I definitively don't want too much of it either, and it seems the slope is only slippery in one direction...

Re:About that link

[drinkypoo]

If you take money from the rich and give it to the poor, that is fundamentally wrong

The rich can't get rich without the poor to feed upon, therefore giving back to the poor is simply perpetuating the system. It's for everyone's good, including theirs. If the rich would support those they depend upon without being forced then we wouldn't have to force them.

Obama, et al, really do think your money belongs to them first. To be doled out as they see fit. It's YOUR money, you just agree to let them have a portion to support the social contract.

Yeah, where "et al" includes Bush, Clinton, Reagan, and almost every other president we've had, or political party. It's not your money, though. Money is a bullshit fiction. Indeed, it is the government's money. They issue it and without them it's ass paper.

Re:About that link

[Jedi Alec]

If you take money from the rich and give it to the poor, that is fundamentally wrong.

Says who? We've been taking things from one another since we first fell out of the trees. Granted, for 99% of human history it was the rich and powerful taking it away from those who already had far less, but that's a minor detail.

Taxation is the price for living in a civilized society. So please, if you are of the opinion that all forms of taxation are theft, why not piss off and literally sit by yourself on an island?

I listen to people like you, a good chunk of slashdotland....it's like the Enlightenment never happened. Like Rousseau, Voltaire, and Jefferson didn't get thru to y'all the importance of government as a social contract and consent of the governed.

Quite the contrary. Which is why we ridicule the exceptionally small percentage of selfish dirtbags that seem to think they can reap all the benefits of society without paying for any of it.

Obama, et al, really do think your money belongs to them first. To be doled out as they see fit.

Populist claptrap. Citation needed. And no, Glenn Beck doesn't count.

Re:About that link

[Fallingcow]

They are also taxed the same on the same income, thanks to marginal tax rates. Everyone pays nothing (well, no income tax anyway) on the first $10,000 (or so). Everyone pays the same on the next $20,000, then on the next $40,000 after that (just making up numbers) and so on. Everyone pays the same on the same chunks of income.

The wealthy also pay far less in social security tax, so their rates can actually come out to about the same as the middle class. The ultra wealthy likely don't pay over 20% on average, and possibly less, meaning they have a lower tax rate than most of us.

Broken time measurements of the inter-frame time

[BZ]

The way this benchmark measures "intra-frame time" is broken. In particular, it uses a setInterval with a 1ms delay. No browser actually respects that 1ms. Chrome clamps it to 5ms; others clamp it to 10ms, all to avoid the website thrashing the CPU pointlessly.

The upshot is that Chrome's interframe delay in the graph is about 5ms and Firefox 3.6's interframe delay is aboug 10ms. Which this particular benchmark can't tell apart from "no delay at all", given its methodology.

Firefox 4 beta, IE9 beta, Safari, and Opera seem to have delays greater than 10ms, so they're clearly doing some work they can't finish in 10ms.... or have slightly buggy timer implementations. Or both.

Of course in practice frame rates above 60fps or so are pointless since the screen doesn't redraw that often. ;)

On the other hand, on Mac, on modern hardware, I get 4.5fps in Chrome 7 dev on a random trial document I just tried, with JS render tiems on the order of 7ms (with a 7ms standard deviation) and "intra-frame time" of 224ms with a 900ms standard deviation (yes, those numbers are nuts). Firefox 4 beta comes in at about 11s for the JS (with 3ms stddev) and 125ms for the "intra-frame time" (with a claimed stddev of 0, which looks really suspicious).

It'd be nice if there were non-obfuscated source for this benchmark so its number-crunching could be evaluated; that 0 stddev is ... highly improbable.

Bullshit Slashvertisment

[phantomcircuit]

This benchmark can be run by anyone in LucidChart. First, sign up for a free account here.

Nuff said

Re:Bullshit Slashvertisment

[inflex]

Agree++

Always makes me wonder how these people manage to push the 'articles' through when you see a lot of other vastly more decent articles get voted down on the Firehose.

Re:Bullshit Slashvertisment

[SheeEttin]

username: bugmenot
password: bugmenot
http://www.bugmenot.com/view/lucidchart.com

Useless

[AmberBlackCat]

I think if it were truly a test of "real world" use, they would find that all of the browsers have the same performance level and none of this matters.

Re:Useless

[AmberBlackCat]

Some idiot modded me "overrated" but, you know, you could browse a website on a netbook or a 1000-core supercomputer with a monitor and keyboard attached, and either way you'll spend most of your time waiting for things to download. Everything you've already loaded will render before the next element loads. The only difference you'll notice is how fast it takes your browser to load in the first place, and that has nothing to do with graphics acceleration.

Odd definition of "dead last."

[whoever57]

TFA (yes, I actually read it) says: "Firefox 4.0 Beta 6 came in behind all other browsers except for IE8". That's quite different from "dead last".

Re:Odd definition of "dead last."

[Vandilizer]

Well they were comparing modern browsers and I don't know any one who considers IE8 in that category.... Personaly I think it was there just to show how much the IE team has really improved performance, to be homiest I am impressed. As for beating chrome, just keep in mind that you are going up against a bunch of PHD's who probably work on it for fun...

Re:Odd definition of "dead last."

[ifiwereasculptor]

Not really. Come on... IE8 is paleolithic technology.

Re:Odd definition of "dead last."

[gmuslera]

IE8 counted as "braindead last"... so the next place was that one for FF4

Re:Odd definition of "dead last."

[RedWizzard]

Well they were comparing modern browsers and I don't know any one who considers IE8 in that category....

Just what do you consider modern? IE 8 was released in March of last year. It's actually the newest piece of software from Microsoft on my PC.

Personaly I think it was there just to show how much the IE team has really improved performance, to be homiest I am impressed.

Well I'm not. Why should I be impressed when a company the size of Microsoft takes 9 versions and 15 years to achieve acceptable performance? If they were on a par with Chrome I'd be impressed but it's not, it's slower than every other competitor except the Firefox 4.0 beta.

Re:Odd definition of "dead last."

[shutdown -p now]

In a test of assault rifles, outperforming a stone club is not a notable achievement.

Re:Odd definition of "dead last."

[Bungie]

Why should I be impressed when a company the size of Microsoft takes 9 versions and 15 years to achieve acceptable performance?

During the days of IE4 and IE5, IE would smoke every competitor (except for Opera) in rendering pages and overall performance. Performance was so fast that they they could use web page code for the content of every window in Explorer (and have the IE engine render each one) on the processors of the day (like origional Pentiums or Pentium II's). So no, it wasn't 15 years and 9 versions. They just got lazy after IE6 which allowed their competition to overtake them, and have been playing catchup for three versions. Chrome is the totally new player right now and is ahead just like Opera and FireFox back in the day, when it becomes as ancient as IE we'll see how it performs.

Re:When do you declare a browser dead?

[FrostDust]

Firstly, the only thing including more samples in a test does is give you clearer results. It doesn't cost the tester that much time or money to simply run a web browser.

Furthermore, even if someone were to accept your claims and assertions, the matter is simply that the selection of browsers in the article covers all the actively developed rendering engines currently in use. No one would argue to include Seamonkey, Flock, or Galeon, even if they had a higher usage share than Opera, since Firefox already represents Gecko.

If anything, Safari or Chrome should be dropped, since they are both based on Webkit.

Re:When do you declare a browser dead?

[BZ]

The thing with "real-life" benchmarks is that "based on Webkit" only gets you so far. Safari and Chrome use totally different JavaScript engines, for example. They use totally different drawing libraries. Heck, Chrome on Windows uses totally different drawing library than Chrome on Mac (which makes a difference in "real-life" benchmarks, since drawing is anywhere from 30% to 80% of the total benchmark time).

The less synthetic the benchmark the more silly details (exact browser, not just rendering engine, exact graphics driver version, exact Xorg version on Linux, exact graphics hardware, etc) start to matter...

GPU acceleration

[DigiShaman]

Now, will IE9 scale with a faster GPU? I'd like to see some benchmarks with different video cards.

Re:GPU acceleration

[cbhacking]

It matters, but less than you might think. I tried it with my computer and a friend's box. Both are Core 2 CPUs at 2.8 GHz with the same amount of cache per core and the same model line (though his is a quad-core and mine is dual). However, mine is a 3-generation-old mid-range GPU (GeForce 9600) while his is a 1-generation-old gaming GPU (RadeonHD 4850). GPU benchmarks generally show his performing at least twice as fast as mine. The little "hardware accelerated rendering" tests on the IE9 testdrive site show a non-trivial difference, but it's more like 30-40%, not 100-160% that other benchmarks show.

Just for fun, I also tried it on my tablet. The thing has a tragically underpowered CPU (Core 2 Duo, but 1.20 GHz to the 2.80 GHz of our main boxes) and Intel Integrated graphics (X3100, I think). Most benchmarks put that hardly in the same order of magnitude as even my GPU, especially on such a slow CPU (integrated graphics is partly dependent on CPU speed). However, it still got about 40% what my main box got (call it 30% of my friend's) and was able to run all the demos. All other (non-hardware-accelerated) browsers did much worse. Based on that, I'd say that even on a crappy graphics chip, hardware rendering is a non-trivial advantage.

Who cares?

[Goonie]

I use Firefox and IE regularly, have played with Chrome, and occasionally use Safari on the Macs at work.

I honestly can't notice any difference between any of them in rendering speed.

99.99% of the time, web browsing performance is network-limited anyway.

Surely standards support and browser stability are more important features, at least on platforms with more grunt than an iphone?

Re:Who cares?

[shutdown -p now]

If you're using IE regularly, I presume that would be IE8 (or below). If so, are you seriously saying that you are not noticing and/or not annoyed to by something like ~0.5s delay when creating a new tab, or when navigating from page to page?

Re:Who cares?

[Goonie]

Perhaps it's just lack of experience with anything faster, but, honestly, no.

But then, my ping time to Slashdot is 265 ms. That may explain part of the reason why my browser wait times are dominated by the network.

Why take a 6 month old built of opera?

[Anonymous Coward]

He seem to have used the most recent build of all browser, but he uses a 6 month old built of Opera when 10.62 was release just a couple of weeks ago.
Doesn't seem fair to me.

But then again maybe he just didn't want Opera to best the fastest

Intel Graphics?

[vandan]

The story doesn't mention the GPU in use, but it does mention it's an i7 processor. So I assume it's using an i965-class GPU. These aren't exactly known for speed or stability on linux. I believe FF4 uses Cairo, which in turn uses XRender, and my experience with integrated Intel GPUs and XRender is that pure software ( ie X on FBDev ) is faster. I would have liked to have seen a system used which could actually accelerate the drawing operations.

Chrome: Very Beta

[DogDude]

I read the article. I thought, "Hey, I should try Chrome". I try to install. I get "unknown installer error".

Color me impressed.

Re:Chrome: Very Beta

[dbcad7]

I'm impressed with your ability to get an unknown error installing something that hundreds of thousands of people can install with no problem.

Not Surprising

[dangitman]

The results were surprising. IE9 held its own pretty well (with a few caveats), and the latest Firefox 4 beta came in dead last."

Why is this surprising? Firefox has been the slowest of the current crop of browsers (IE8 excepted) for quite a while now.

FF4 was not dead last

[Anonymous Coward]

According to TFA, FF4 was not "dead last", IE8 was. From thier charts, IE8 was far inferior to all the other browsers. Of course, you might be able to argue that IE9 is the new IE, so that wouldn't be fair, but IE8 is still the one being used. If you want to use the latest browser, what about the FF nightly builds? Once again, even with a bias, MS products don't beat the free ones.

Re:FF4 was not dead last

[Anonymous Coward]

They tested Firefox 4 Beta 6. That came out one day earlier than the IE9 Beta. That's not a bias, that's fairer than you can usually expect. The stars have to align for such fairness. Nightly builds are generally poor comparisons except when under lockdown in the runup toward release, but here, we don't have to when comparing Firefox to IE9.

In fact, it's that Beta of Firefox that lost the comparison, not the "old" version.

Re:FF4 was not dead last

[Lennie]

Well, the suggestion of trying the nightly builds isn't such a bad idea, because it does include the have the improved engine (Jeagermonkey). The currenty latest beta does not have the improved engine. So the any performance test you do with Firefox 4 Beta is going to be nothing like the release version.

Lag and latency

[Twinbee]

And once again I see no mention of lag or latency behind everyday controls, one of the real factors which affect a user's perception of speed and responsiveness. The kind which gets them to say, "It feels faster, but I don't know how". I'm talking about switching between tabs, closing tabs, clicking the browser's back or forward button, and general UI navigation. You want hundredths of a second or less for these kind of actions.

useless benchmark, horrible summary

[macshit]

I'm not sure how they get off calling this a "real world benchmark", as it seems to bear almost no resemblance to what people normally use web browsers for: "The benchmark works by simply dragging a part of the diagram around the page for five seconds." WTF?

It certainly doesn't seem to be any more useful than the other browser benchmark being touted these days, and arguably it's much less useful, because it measures a single very narrow aspect of browser operation, one which has little connection with typical browser usage.

Moreover, the slashdot summary seems to go to great lengths to emphasize how "badly" FF4 did on this (useless, remember) benchmark, and to pump up IE9: "The results were surprising. IE9 held its own pretty well (with a few caveats), and the latest Firefox 4 beta came in dead last" -- but if you actually look at the results that emphasis is misplaced: almost all the browsers were quite close to each other, with a few outliers, but in no cases was FF4 an outlier, and indeed was pretty much identical to IE9 (on this test).

The only clear result I can see is: When doing a certain very specific type of javascript rendering, most modern browsers have pretty much identical performance, though chrome's particularly fast, and IE8 particularly slow.

Of course, that isn't very interesting to anybody except LucidChart users, of course, nor very likely to generate any controversy...

Re:useless benchmark, horrible summary

[Dputiger]

You're confusing 'real world' benchmark with 'entirely inclusive' real world benchmark. A benchmark is a valid real-world test if it models an activity browsers are actually used for. If someone benchmarked a bunch of modern video cards in Half Life 2, that's a real-world test. That doesn't mean HL2 should be the only gaming test, or that readers can look at the HL2 figures and extrapolate comparative performance data for Metro 2033 or World of Warcraft.

Re:useless benchmark, horrible summary

[macshit]

Er, sure, but this benchmark bears almost no resemblance to typical browser usage, and the correlation with performance on any other web app/page is likely to be pretty near zero.

Thus: useless

Test is pointless

[Trerro]

Chrome is faster because it massively favors speed over customization and features. FF is slower because it favors customization, and assumes, correctly, that no one actually actually gives a flying fark if it needs slightly more than a 1/50th of a second to render a page that Chrome can do in 1/100th of a second. This isn't a problem, nor is it news. Now of course, you may do the occasional task where those milliseconds actually matter because your browser is processing something enormous, but then just install both browsers, use Chrome in those rare instances, and use FF for a primary browser.

Frame rate isn't really an issue either. You can point out that the human eye sees at roughly 60 FPS, so going under 60 is undesired, but let's be realistic. Those Flash games are usually built at 20-25 FPS, because running at 60 would make them freaking huge. Video on the web likewise runs at 60 FPS roughly never, because it needs to stream. Downloaded video WILL run at 60, but your browser isn't playing that.

This doesn't mean there isn't a couple of very specific tasks that FF is abnormally slow at and could use a code cleanup on, but for the most part, FF's speed difference vs. Chrome is utterly negligible in actual use.

Fuck this shit

[Massacrifice]

Really, fuck it. I've had it with corporate-sponsored dick-fighting contest about which browser is the fastest. I really, really couldn't care any less. Features, openness, security, standards compliance, yeah. But If I want a fast app, I'll go native, thank you. Maybe I'm too old, but I've always thought HTML sucked as a programming paradigm. As an information distribution mechanism, sure. But for interactivity? Please. It's about time somebody called bullshit on this. Hell, a goddamn Visual Basic app from fifteen years ago kicked the butt of most modern web sites in usability, performance and ease of maintenance. The only thing that makes the web so attractive is the barrier to entry : free, nothing to install, immediate access to the average brains of millions. Just like TV. No thanks for dumbing it down to this. And now you wanna make it faster? Piss off. Go write real code that does something, not just another abstraction layer.

Re:Fuck this shit

[Gamer_2k4]

Meanwhile, in the real world, people actually want the source of all their information to be fast. Go figure.

Re:Fuck this shit

[yelirekim]

You mean write real code that does something... like a web browser?

Re:Fuck this shit

[AnalyticaL]

Agreed, real applications for the win. Browsers might become ultrafast, doesn't negate the fact that most website will throw so much stuff at their users that you cant navigate fast anyway. Give me the structure, speed, security and functionality of a well built app...

Re:Fuck this shit

[strikethree]

"The only thing that makes the web so attractive is the barrier to entry : free, nothing to install, immediate access to the average brains of millions."

Ah. You forgot one other very important thing that make the web so attractive as a platform: Control. "They" have all the control over the app/service. They can changes terms of service, price, availability, etc. at any time they want and you have NO RECOURSE.

Control baby. Control. It is good to be King.

Re:Fuck this shit

[AmiMoJo]

HTML turned out to be pretty good for interactivity. Web apps are automatically cross-platform, even on phones and other devices. They are mostly secure if you do it right. Available anywhere, both the app and your data, saves the user having to worry about backups. Plenty of processing power available on the server side or the client side depending on your needs.

I find being able to run apps like Google Docs without installing bloated crapware on my PC and not having to worry where I saved the latest version of the file pretty useful.

Re:Fuck this shit

[Junior J. Junior III]

Hell, a goddamn Visual Basic app from fifteen years ago kicked the butt of most modern web sites in usability, performance and ease of maintenance.

That, and it could even trace IP addresses.

Re:Fuck this shit

[Stick_Fig]

The only thing that makes the web so attractive is the barrier to entry : free, nothing to install, immediate access to the average brains of millions. Just like TV. No thanks for dumbing it down to this. And now you wanna make it faster? Piss off. Go write real code that does something, not just another abstraction layer.

Fun fact: Programmers don't choose the platform. Users do.

Re:Fuck this shit

[CastrTroy]

There are some good things about the web.

Instant updates to all your users at exactly the same time. No more worrying about if your users are using old, insecure, incompatible versions.

For times when users need to be connected to eachother, having everybody go through a standard HTTP server is the easiest way to getting rid of networking problems, not having to worry about firewalls, and not having to expose your users computers to the web.

Re:Fuck this shit

[Anonymous Coward]

Meanwhile, in the real world, people actually want the source of all their information to be fast. Go figure.

Sure, and a lot of people want their cars to go fast, but at some point it becomes meaningless. What's the practical difference between 6.3 and 5.5 seconds to get to 100 km/h? What's the practical difference between 23 and 20 ms of render time as shown in one of the graphs?

I'm all for improvement, but at some point things become "fast enough" for most purposes.

Re:Fuck this shit

[hedwards]

Which means doing well on cooked up bench marking? Most of the idiots bad mouthing Firefox aren't doing it because it sucks as a browser. Yeah, Firefox isn't the fastest thing out there, but that's largely because with its marketshare it would be irresponsible to pull some of the stunts that the other browsers are pulling. As well as I haven't seen anybody make reliable claims that the Firefox devs are cheating in benchmarks beyond the possible making theirs play more to Firefox's strengths.

Fast is one thing, but it's hardly the only issue. I've tried the others and quite frankly found them to be immature and lacking. It's kind of a bullshit complaint to cite benchmarks existing only in cooked up situations when most people don't use the browser like that. Firefox could be doing a lot better in those tests if they were focusing more on that instead of things that actually matter.

Re:Fuck this shit

[strikethree]

There are definitely upsides. I was merely listing a downside which is a killer in some circumstances. Ask COMPTIA about the downsides (about this time last year).

Regards,
strike

Re:Fuck this shit

[Oligonicella]

Maybe at some point, but that point ain't now. I develop apps for local use, and I would want my response times to ideally be as low as a compiled app. Any javascript or rendering speed improvement is still a plus to me.

Re:Fuck this shit

[Oligonicella]

Actually, no they don't. A few houses worth of friends and mine all interconnect without the use of anyone's services. Now, if I want to read The Tokyo Daily, I have to use something that connects to it. Can't afford the 12,000 mile of cabling m'self. They have every right to charge me for the use of all the shit they had to do to get my browser linked up with Tokyo.

Re:Fuck this shit

[jellomizer]

That is why Microsoft Became the Dominate Operating System. The recent successes in the Rise of Macs and Linux and iPhones and Androids over windows during the past decade is the simple fact people no longer need to run OS only applications. Their browsers work and runs the apps they want to run.

Now please explain Structure of a well build app vs a structure of a well build web app? When they are both well built they both run quite well.
Speed, Yes that is a trade off, but web browser speeds are about the same as any other Run Time based languages. Also behind you Web App you could have a huge cloud behind it to do massive amount of processing that could take you little computer a long time to run.
Security, This is an other of those Well Built vs non well built app things. I see about a few dozen Linux Apps updates for apps that could possible cause Local Root code execution that needs to be fixed. If you need to have your app do something outside your own PC then it will have all the same conservers as it did before.
Now for your last bit Functionality, Before the web 90% of the Apps out there were forms that you filled in and hit save. Web Apps are good at that and have been for a decade. Using modern Standard Web Technologies you can fill in about an other 8% to fill in the other areas.
As of right now the only Non-Web Apps we need are...
Web Browsers
CAD
High End Games
Developer tools
Office/Graphics Processing stuff.

Actually I don't like making the distinction between Web Apps and Normal Apps, They are really not that different, the one key difference is that we are using a different Interface engine.

Re:Fuck this shit

[shutdown -p now]

Meanwhile, in the real world, people actually want the source of all their information to be fast.

Not really. People generally want the source of information to be fast enough. Past a certain threshold, nobody really cares.

It's worth noting though that e.g. IE8 is certainly slow enough to be irksome. But then, Opera isn't, and it has neither JIT for JS, nor hardware accelerated graphics.

Re:Fuck this shit

[theArtificial]

Instant updates to all your users at exactly the same time. No more worrying about if your users are using old, insecure, incompatible versions.

I agree in principle but what you're stating exists only in an ideal world. Realistically not everyone upgrades and patches; otherwise we wouldn't still be seeing IE6, bugs exploited from patches that were released 6 months ago etc.

You can lead a horse to water but you can't make it drink.

Re:Fuck this shit

[Anonymous Coward]

Use HTML 2.0 problem solved.

Re:Fuck this shit

[suomynonAyletamitlU]

Meanwhile, in the real world, people actually want the source of all their information to be fast.

Not really. People generally want the source of information to be fast enough.

Once upon a time, a 486 was "fast enough". If an operation was going to take a few minutes, take a walk, grab a notebook and paper, do something else interesting or constructive. It'll be done "soon enough".

I mean, I get your point. However, for me and lots of others, render time is obliterated by slow connections. Until every page fully fetches AND renders faster than you can perceive, "fast enough" doesn't mean much more than "what we're used to" plus or minus for innovation.

Re:Fuck this shit

[CarpetShark]

Hell, a goddamn Visual Basic app from fifteen years ago kicked the butt of most modern web sites in usability, performance and ease of maintenance.

You underestimate the value of HTML as a GUI. After all, what was the point of gradually developing a standard set of well-tested GUI widgets over decades in OS X, Unix, NeXT, MGR, Amiga, GEM, DOS, Windows, if we're not going to let a bunch of un-aware flash "developers" re-attempt all the early mistakes?

Re:Fuck this shit

[Anonymous Coward]

But then, Opera isn't, and it has neither JIT for JS

Uhh, yes it does and has since the introduction of Carakan in 10.50.

Re:Fuck this shit

[CastrTroy]

That's a symptom of the desktop app. The browser itself is a desktop app. The applications that run inside the browsers are the ones I'm referring too.

Re:Fuck this shit

[marcosdumay]

If people really wanted a fast browser, they'd use linx. Most people I know want features and security.

Re:Fuck this shit

[Anonymous Coward]

Slashdot seriously needs an "aging geek" mod item. It's kind of depressing. I'd hope that people working with such objective things would be at least a bit resistant to the "I'm scared of change" mentality. Doesn't seem to be that way, however. At least going by people on slashdot.

Re:Broken time measurements of the inter-frame tim

[Anonymous Coward]

few timers can really measure 2 or 3 ms accuracy (you have to use high performance counters). Your standard dev of 0 probably means the resolution of the timer was less than the actual std dev.

Browser speed is irrelivant

[flyingfsck]

Stability and web standards implementation is what matters most.

... and even older Opera?

[TeXMaster]

The versions used for tests were quite debatable. For IE and FF they put up both the final version and the current beta, while for the other browsers they only show the results for the current release ... mostly: the latest released Opera version is 10.60. A fair comparison should have included Chrome 7 and Opera 10.70 and replaced Opera 10.53 with 10.60.

I guess the actual selection of versions shows how the point of the article was more about bashing FF4 compared to IE9 (in which it also failed, given the very small difference between them) rather than doing a honest comparison of all of the browsers.

Re:... and even older Opera?

[hkmwbz]

What lawsuits did Opera file in the past, exactly?

Re:... and even older Opera?

[shutdown -p now]

Opera initiated anti-monopoly proceedings against MS in EU (which resulted in fines and browser selection screen).

Re:... and even older Opera?

[hkmwbz]

Actually, what happened was that Opera asked the EU to look into Microsoft's anti-competitive practices. After an evaluation, the EC initiated anti-monopoly proceedings against Microsoft. The EC, not Opera. Opera does not have the power to initiate antitrust proceedings against anyone.

The browser selection screen was Microsoft's own idea.

So it seems that the claim about Opera filing lawsuits is bogus. Just more FUD. And Mozilla, Google, and numerous other companies joined the complaint, but these clueless morons only whine about Opera. Pure hypocrisy.

Bloat is in the way

[uofitorn]

I've said this before. I think it's time that Mozilla release a stripped down, lean, mean version of Firefox without all of the bloat.

They can call it something like Firebird, or Phoenix, to distinguish it from the main branch.

Re:Bloat is in the way

[John Hasler]

> They can call it something like Firebird, or Phoenix...

No they can't. Trademarks.

That is *NOT* real world

[aepervius]

Real world people use flashstop, noscript, adblock plus. I use firefox only because of those. If the other browser had the same functionality as easily accessible built in, I would maybe use them. As it stands, only firefox interrest me. As for performance... Talk to me when the majority of the people bandwidth and access time stops being *the* limiting factor.

Untapped opportunity

[rantomaniac]

I'm surprised none of the browser vendors hired the author of this virtual machine, which beats both V8 and TraceMonkey in the shootout benchmarks by a wide margin.
I'm hoping Google or Mozilla does that eventually, it'd be a pity if he got hired by a closed source browser vendor.

Mistitled

[Anonymous Coward]

The abbreviation of Firefox is not "FF", but "Fx".

update..

[Anonymous Coward]

3) is now obsolete in the latest Firefox 4 nightlies - http://blog.zpao.com/post/1140456188/cascaded-session-restore-a-hidden-bonus

Re:When do you declare a browser dead?

[TheRaven64]

It has 1-2% market share in the English-speaking world. It has a much larger share (over 20% in some countries) in Eastern Europe and Russia.

Stupid testers

[Lennie]

The Firefox beta does not even have the new Method JIT. So they are comparing oranges and apples, not nearly finished products. Maybe Mozilla shouldn't have called it Beta, maybe that is the problem. But rerunning the tests in a few weeks will give very different results.

Upgrade? Why upgrade? Slashdot in 1 sec at 32K.

[Anonymous Coward]

(Grumpy old codger ahead)
'Real world torture test'. Yeah, right. Spoiled broadband using whippersnappers have no idea what a real-world browser torture test is. Try this one on for size, you spoiled basement dwellers. Here's torture test for your 'real world' that I run in every day.

Bandwidth: Max of 32K shared on a 30+ computer network with 4 analog phones using the same bandwidth.
Connection: 357ms ping time to the DMZ. >600ms ping to the 'real world'.
Connection Type: Throttled Inmarsatt B.
Browser: IE 6

Yet somehow I can still manage to download Slashdot in 1 second over this connection.

http://www.google.com/gwt/n kicks ass over all your snooty high-faluting bells-and-whistle browsers!

Not quite as fast as good old Lynx though. I miss that. 300 baud.. monochrome... keyboard navigation... those were the days. Could hold all your opcodes on a three by five, and write your machine code by hand.

Re:Broken time measurements of the inter-frame tim

[multipartmixed]

How does the timer in firefox "kick off"? It's obviously not on the branch callback anymore; surely you guys aren't firing an operation callback every 10ms?

Do you know how often in the tracing JIT checks operation callback? How about JM? I'm wondering if this style of benchmark will become increasingly innaccurate as the JS JITs get tighter. Is a super-accurate 10ms timer a goal @moz?

IE Vs FF

[Anonymous Coward]

For me speed is secondary, both deliver adequate speed subjectively. If I don't care that Google, Yahoo and all Ad companies know that I'm ah.... say date online, then IE is OK. When my privacy is important to me, FF+noscript, hands down.

Re:When do you declare a browser dead?

[Rockoon]

In the former Soviet republics, Opera is #1 in many cases.

#1 in Uzbekistan
#1 in Ukraine
#1 in Georgia
#1 in Belarus

..and this is just a short sampling of the former Soviet republics. Its #1 in more of them, and where it isn't #1 its often a close #2.

Re:When do you declare a browser dead?

[TheRaven64]

I hadn't realised it was quite so popular - for those too lazy to click the link, or who don't have Flash, it has 30-50% market share in those countries. The original poster's comment reminds me of the poster in the last iPhone story claiming he'd never seen a Nokia Smartphone in the wild. It's important to remember that the US market is often not reflective of the global market.

Just try submitting patches

[aussersterne]

and seeing how many of these fruits of your labor actually end up in future releases. (Hint: none, even if the bug does things like solve massive NTFS corruption or critical on-screen corruption in the ATI 2D driver, both of which are real attempts of mine.)

They say "U Want? U Fix!"

Then you do and they say it doesn't:

Fit with project goals
Adhere to project style or standards
Offer regression data about other use cases
Solve a big enough problem to justify effort to include

Or they'll just ignore the hell out of you and eventually (as you noted) mark the bug as "solved" simply because bug submitters stop responding to repeated nonsense bug labor/reply requests after 2-3 years... even if there are replies (SOMETIMES DOZENS OF THEM) in the Bugzilla threads linking to WORKING PATCHES.

God I've had it with having to rebuild half of my packages from .src.rpm each release using hacked and rehacked patches, version after version, from Bugzilla discussions that never, ever seem to make it into the code year after year and release after release because arrogant maintainers have their heads up their asses.

THIS is why I'm done with FOSS. Just done. I loved it and the community in the '90s. Now it's mostly arrogant young hotshots with no particular interest in getting actual work done apart from the work of coding for coding's sake, implementing new experimental unstable (if not useful) features at the expense of old, stable, useful ones that most users rely on.

After all, u want... u fix! (But not in my project -- build your own codebase from scratch!)

Apparently writer cannot read or interpret graphs

[Anonymous Coward]

Looks like IE8 came in dead last on both charts provided... How does that make FF4 dead last?

Re:Broken time measurements of the inter-frame tim

[BZ]

Timers run from the event loop. They have all along; they didn't fire via the branch callback at all. So to get your timers to fire you have to stop whatever you're doing and return to the event loop.

I don't know how often the operation callback runs, sorry.

Having a 10ms timer that actually runs after 10ms (accurate to under a ms) is definitely a goal. But again, you'll have to go to the event loop to get the timer, and if something else is running when the timer is supposed to fire (e.g. if you have 1000 timers all scheduled for the same millisecond), chances are your timer will run late.

I can't make enough sense of the benchmark's obfuscated code so far to say anything useful about its accuracy past what I already said.

IE9 is good

[chafey]

I just tested the latest beta of IE 9 with our application and it beats out all the other browsers. Our web application heavily depends on image display and canvas and the difference is likely due to IE9's rendering integration with DirectX. If you are looking at writing web apps that heavily use Canvas, IE9 is going to be more than relevant.

Why should I care about browser speed?

[John Hasler]

I can't recall when my "user experience" has been limited by anything other than download speed (and the abysmal quality of most Web sites, of course). So why should I care about browser speed?

infinitely slower

[Xtifr]

In my tests, on the average, Ie9 was infinitely slower. Of course, I knew that if I just posted raw cross-platform results, people would object that Windows has a disproportionate market share, so I took that into account. In my final numbers, Windows counts for a generous 99% of the result, and other OSes for only 1%. Of course, 1% of infinity is still infinity, so, on the average, any given user will experience an infinite slowdown with IE9. It's all there in the numbers! :)

Not always

[Anonymous Coward]

Try opening a generic website .... lets say CNN.com or even Slashdot.org on FF and IE at the same time. Then tell me if there is no difference in speed.

The Real Benchmark

[Anonymous Coward]

The real benchmark between browsers should be the time taken to perform a search on a site like isoHunt. The Flash adds alone are an IE8 killer even on a Core2duo, and I'm sure they'll only get worse.

Microsoft?

[dogzdik]

Microsoft Operating system/s? No.

.

Microsoft Software? No.

.

Microsoft Web Browsers? No.

.

So who cares? I don't.

IE9?

[zrbyte]

Bah. I'll wait until it's ported to linux.

Oh Chrome

[Anonymous Coward]

Chrome runs in the background when you start your PC, so naturally it loads faster - it has slowed down the rest of your computer to make sure of it.

Also even after you uninstall Chrome, the program continues to connect to Google periodically, slowing down your computer and wasting your time even more as you try to figure out how to fully remove it.

Big difference

[justthinkit]

What's the practical difference between 6.3 and 5.5 seconds to get to 100 km/h?

Big difference. As a commuter I value the acceleration (along with other properties) of my vehicle. If everyone else takes 6.3 and I take 5.5, this difference will help me numerous times per commute to merge more easily or to be able to take advantage of the hole created in front of some grey beard, or in some cases to make the next light due to dumb configuration of light sequence. So it is not 6.3 vs 5.5 with car acceleration, it is relative acceleration that matters. In the dog-eat-dog world of commuting, acceleration matters quite a bit. For me, more acceleration equals less stress. My brother's "zero to 50 in 30" two-cylinder Toyota has no place on today's roads.

And FWIW this is at least partially true with other things -- if browser x is the slowest, whenever we run it we will notice. For me, IE v6, 7, 8 are all so slow vs Opera & Firefox that I simply never run them. Yet if I did an absolute comparison of IE .GE. 6 versus Netscape Navigator I would probably want to trash Navigator all over again. The weakest link gets kicked off the island, even if the difference is measured in fractions.

Actually I find this interesting

[AlgorithMan]

Actually I find this interesting

• Firefox 3.6 (the browser everyone calls bloated and slow nowadays) still beats IE9 beta
• Firefox 4 beta, the bloated and slow browser is not "dead last" (that's IE8 by FAAAAAAAAR!) and it's no big difference to the "oh so fast" IE9 or even the "lightning-fast" opera
• to reach this performance, IE9 has to utilize much more hardware (GPU acceleration, all CPU cores) which will be a bitch in real real-world usage (unless websurfing is everything you ever do)
• did they use Adblock in FF? otherwise I wouldn't call it a "real-world" benchmark...