The More Popular the Browser, the Slower It Is 367
demishade writes "Peacekeeper, the browser benchmark from the makers of 3DMark, comes out of beta and shows an interesting (though perhaps not surprising) tidbit — the more popular a browser, the worse its performance. While it should not be surprising to anyone that IE slugs at the last place, the gap between Firefox and Chrome, is. Once IE's market share goes the way of the Dodo will web developers start cursing Firefox? How long until Google comes out with a JavaScript intensive application that will practically require Chrome to function?"
chrome experiments (Score:5, Interesting)
Not so surprising (Score:5, Interesting)
Do the words "TraceMonkey" mean anything to the authors? It's the core Javascript engine of the upcoming revision of Firefox. And it is fast. Some benchmarks suggest that it is highly competitive with V8 (Chrome) and SquirrelFish (Safari).
(Speaking of which, isn't it a bit disingenous to compare Safari 4 BETA to the current version of Firefox? Why not compare the Firefox beta then? Smells of yeller-bellied journalism to me.)
Javascript is currently a hugely competitive area. Every browser revision is trying to boost performance. (Including Microsoft.) It only makes sense that the older and cruftier engines would have a harder time competing with the newer and more nimble engines created by these upstart competitors. However, with the exception of Microsoft who's stuck updating JScript (haha, bundle FAIL!), all the other competitors can and are swapping out engines for faster and faster performance.
Features Create Popularity... (Score:5, Interesting)
Then again, over time, isn't this what happens with almost all software? They get more and more features as time goes by, and get bigger and consume more resources. Look at the size/requirements of any linux distro with a graphical system over the past 10 years.
No one wants to lose features, and users complain too much, so the only way to get a faster thing with less features is to fork it, or start anew (which is what the lesser popular browsers have often done).
In other words... (Score:4, Interesting)
Either:
1) up and coming browser makers see speed as an easy differentiating factor and target their browser for it; or
2) Newer products tend to be faster since they have the older ones to compare to. And newer products also are "up and coming" and thus have lower uptake than "old and entrenched" ones. or;
3) the public puts very little value on browser speed. Those spending their resources optimizing for it rather than other features get few users as a result.
CPU Usage Observations with FF and Xorg on colors (Score:2, Interesting)
Some of the ones where they had the graphics/colors rolling around:
CPU Usage: 96% Xorg, 2% Firefox.
I've seen that happen on several other sites that have javascript doing funny things with the colors/images. Makes the entire machine/interface hard to use.
Of Course, the Google Web Toolkit (Score:5, Interesting)
Yeah, the Google Web Toolkit [wikipedia.org] (which I believe they are all using for a front end) basically produces code that produces one metric ton of Javascript and HTML that gets dumped on the client's browser. It's not just an application, it's a whole library of Java APIs that produces a ton of Javascript that could become the de facto standard one day. I'm betting it won't but I've asked why more sites aren't using it [wikipedia.org] on Slashdot before.
At least Google eats their own dog food on a large scale.
application that will practically require Chrome (Score:1, Interesting)
Then I, and many others will probably never use said app.
By the way, have any of you guys tried the "basic HTML" version of GMail? I actually find it to be quite nice and I greatly prefer it over the default JavaScript version.
Re:No surprise (Score:5, Interesting)
I installed SRWare Iron the other day. According to the publishers, it's basically a Chrome de-Googlified with a few other downstream tweaks (eg using a slightly newer version of WebKit). It seems to run all right, but I'm still typing this on Firefox because Adblock trumps Chrome/Iron's performance & user interface design advantages.
I *like* Chrome/Iron, and when it gets a decent extensibility model I think it'll tear a huge hole in Firefox's market share - but until then, it's going to be not much more than a cool tech demo.
Re:No surprise (Score:5, Interesting)
Sadly, half of the firefox plug-ins I now consider indispensable are the ones that disable functionality or return the user interface to an older style and functionality. Firefox was destined for bloat once they committed to building-in features that would have been more useful as pre-bundled, official extensions (like the anti-phishing technology).
Re:No surprise (Score:5, Interesting)
Don't use plug-ins unless absolutely necessary. I just uninstalled a bunch of bloatware. One of those was NoScript because I felt it created more hassle than the advertising it blocked. Yes popups are annoying, but not as annoying as NoScript constantly disabling websites and me being forced to select "trusted site". It got to the point I was spending more time futzing-around with Noscript than with actually browsing the net.
Also I suspected NoScript of slowing down my computer as it gobbled more-and-more memory. Now the only plugins I use are a flash downloader for Youtube, and ImageZoom so I can see pics on Ebay more easily. That's it. And my broswer runs much better.
As for the article:
The conclusion is ridiculous because the sample size is too small. It's also ridiculous because those of use who remember the early 90s recall that Mosaic and Netscape Navigator were not only #1, but also extremely fast. QED the conclusion that a number one browser is slow.... is false.
Re:Not so surprising (Score:4, Interesting)
Well here are some simple reasons for this trend.
1. The more popular the browser the more well established the code is trend. Meaning Popular Browsers are older thus have an older code base. And because it is popular you are more careful not to break anything.
2. Popular Browsers have more safeguards and checks. IE runs a ton of Broken Code, hence why it loads up so much junk and renders it OK. Firefox does a middle ground at this. The less popular browsers the more it follows the stricter standard creating it to load faster as it has less checks.
3.Popular Browsers have more features which can slow it down
Re:No surprise (Score:3, Interesting)
Use Adblock to block ads, and noscript to block scripts. I used to have both, but as you say it ends up just being a pain navigating to a new site, wondering why something isn't working and then remembering about noscript (some poorly designed pages won't render at all without javascript..). but Adblock is enough for me by itself.
Re:No surprise (Score:2, Interesting)
FireFox fail (Score:3, Interesting)
I'm not even slightly surprised, FireFox runs like crap under Mac OS X. Safari is soo much faster
Re:Mosaic (Score:3, Interesting)
Aw, come on. Most of 'em don't know that you can talk to a lot of things with telnet. I've amazed people by connecting for HTTP, POP3, and IMAP using just telnet. You should see them when I send them an email, to them, from them, through their own mail server. I always get a "you can't do that!", which I always follow by "check your mail, and tell me that again."
{sigh}
It's not rocket science. It's even documented in the RFC's. Then again, most people on the Internet have never heard of an RFC, and glaze over when you start explaining it to them. At least they know how to clickie their way through putting enough bling on their MySpace page to blind almost anyone, or give epileptics seizures.
I actually had an interview at a big hosting company for a SysAdmin position, where they asked "How would you verify a site is working?". I responded "First whois hostname to make sure we're authoritative. Then nslookup hostname to make sure the ip is right. Then telnet hostname 80 GET / HTTP/1.0[enter][enter]. Then test in a browser to make sure the content is right." They looked at me, then each other to figure out if I was right, and then with a surprised look said "ok", and continued with the questioning. I guess they had a different way, like firing up a web browser and checking there. My way verifies the whole way through, but hey, they weren't looking for the best answer, they were looking for their answer, which I failed to give.
My answer to "How do you gain access to a MySQL database, if all of the passwords have been forgotten?" wasn't exactly what they expected either. That's when they started taking notes on my answers, and the questions went from interviewing to asking me for their own knowledge.
They tried to put me in the call center though, instead of only fielding hardcore admin questions. I'm too old and grumpy to be warm and friendly on stupid support calls. I'd only make it a few hours before I went ballistic on the 100th caller who said "My interwebs ain't working."
Re:portable Chrome (Score:4, Interesting)
I'm really surprised at the number of people who use noscript. Adblock is not surprising, since sites often fail to follow the common sense rules of advertising. But what is up with noscript? I've pretty much never had javascript freeze the page or anything of the sort. (On the other hand I don't use dialup). The security implications of letting javascript run are actually pretty minimal.
Now, as for flash adverts, I've never had an issue with the download size, although sound is an issue.
As for advertisements, if they fellow the following rules I don't mind them.
Rule 1: The advertisment must be on the relevant page.
Rule 2: The advertisement must remain entirely within a rectangle on the page that does not overlap the page content.
Rule 3: The advertisements must be reasonably sized.
Rule 4: While animation is permissible, rapid flickering of any sort is not permissible, nor is automatically playing video. Playing video as a result of clicking on the advertisement is acceptable, as is a limited amount of automatic pre-buffering.
Rule 5: Sound is not permissible.
Rule 6: If a user chooses to interact with the advertisement by clicking on it. (Simply moving the mouse over it is not sufficient), the advertisement may do any of the above.
Rule 7: Automatic video playing and audio are permissible despite rules 4 and 5 if the advertisements are part of video playback.
Rule 1 explanation: It must not be a popup or pop-under.
Rule 2 explanation: None of those flash adds that project an appendage over the page, that can only be closed after the animation is finished, and usually by explicitly clicking on some part of the advertisement. Further, no pseudo-popups (utiling CSS to create what looks like a popup, but is actually part of the page), unless they do not cover the page content.
Rule 3 explanation: Non of those double height horizontal advertisements.
Rule 4 explanation: Those flicking adverts are obnoxious. Actual video can be processor intensive. However reasonable vector based animation is fine.
Rule 5 explanation: Obvious.
Rule 6 explanation: Once a user clicks on the advertisement, it may play video, play audio, overlap the page, etc.
Rule 7 explanation: That is to say, that advertisements like Hulu's and ABC.com's Full episode player are permissible.
However, there are all too many advertisements that violate those rules, which is why I do run adblock.
Re:No surprise (Score:4, Interesting)
I keep thinking that we need a forum to discuss slashdot, but I've never seen one.
When I created Seen on Slash [seenonslash.com] a few years ago I thought the forum [seenonslash.com] might become a place for slashdotters to discuss slashdot. We get some decent traffic to the rest of the site, but the forums never caught on.
Re:NoScript is about Security NOT Ads (Score:2, Interesting)
What kind of pages are you browsing if having protection against those attacks is necessary? Assuming you're not constantly going to unknown porn and warez sites, is there really much of a risk?