Opera 9.5 Beats Firefox and IE7 As Fastest Browser

Abhinav Peddada writes "Ars Technica takes Opera 9.5, the latest from Opera's stable, for a test run and finds some interesting results, including it being a 'solid improvement to an already very strong browser.' On the performance front, Ars Technica reports 'Opera 9.5 scored slightly higher (281ms) than the previous released version, 9.23 (546ms). And Opera 9.x, let it be known, smacks silly the likes of Firefox and Internet Explorer, which tend to have results in the 900-1500ms range on this test machine (a 1.8 GHz Core 2 Duo with 2GB RAM). Opera was 50 percent faster on average than Firefox, and 100 percent faster than IE7 on Windows Vista, for instance.'"

ok, really?

[markybob]

c'mon...like the average person can even notice a 400ms difference. is this a joke? gimme a break

Re:Wasn't that always the case?

[vipw]

The Opera web browser is no longer ad supported. Just thought you should know.

Re:Article is very misleading - JS benchmark only

[ciroknight]

"Still, I'm interested how comes Opera's Javascript is so fast compared to the other browsers."

Well, they didn't test it against WebKit/Safari/Konq, which blazes through Javascript tests. Firefox's Javascript engine (SpiderMonkey) leaves a lot to be desired, and well, Internet Exploder is just plain terrible at everything. Things will get better for Firefox once Mozilla figures out a way to integrate Tamarin, but this is still a while off.

Re:Wasn't that always the case?

[Macthorpe]

You're only 2 years out - the ads were dumped in Opera 8.5, and that was released on the 20th of September 2005.

If you're going to complain about something, please try and make it relevant.

Re:Those numbers mean nothing...

[hernyo]

It's the ratio what matters. While rendering a random page, opening a huge html, processing arbitrary js code, or whatever: Opera is 2x faster than the others.

Forget the units, use the ratio.

Re:Article is very misleading - JS benchmark only

[Jugalator]

I think the reason they're focusing on Javascript here is because that's a major optimization that took place in Opera 9.5. Actually, the changelog tells that they rewrote the ECMAScript engine. But Opera also had optimizations done to its table renderer, and due to the still all too frequent table layouts on the web, even used by modern web designers, it would be interesting to see more general tests of loading times etc. Opera would probably still come out very close on top though, as it has before in the pre-9.5 versions too.

Re:Grade article: incomplete

[oatworm]

They did a review of Safari 3 back in June [arstechnica.com]. As for comparing against Opera, they probably elected not to due to their opinion of Safari, as noted in the first paragraph:

At the World Wide Developer Conference this week, Apple announced the availability of Safari 3 for the Windows operating system. Today, we put the Safari 3 beta to the test to see how it compares to Internet Explorer 7 and Firefox 2 on Windows. What we found didn't impress us very much. Although Safari offers slightly faster page loading, the beta is extremely unstable and suffers from interface deficiencies that make its value on the Windows platform questionable at best.

In other words, they may not think it's worth reviewing, at least on a Windows platform, especially since it's not a Windows-native browser. Think of it as being similar to comparing browsers on Ubuntu and including IE 6 under WINE.

Re:Resource-conservation, not speed

[RuBLed]

From what I have seen from running Opera over the years (was a total convert since 8.5) is that one of its' consistent nature is a lower memory footprint and yes it even runs on our old 128megs XP Box.

Re:Different market

[SolitaryMan]

From what I've seen the speed rankings in all tests always have Opera and Safari leading with IE and FF being behind.

Opera aims at different market -- small gadgets. This is where the speed is really critical. For IE and FF good enough is enough, since performance on modern desktops is not that critical.

Re:Article is very misleading - JS benchmark only

[arivanov]

Well, they didn't test it against WebKit/Safari/Konq, which blazes through Javascript tests

It may blaze through tests, but in real life Konq is considerably slower than Firefox. I have to deal with a number of javascript ladden juggernauts like the ex-PeopleSoft eBusieness suite on a daily basis and konq is visibly much slower than Firefox.

Re:Article is very misleading - JS benchmark only

[Bert64]

Well it depends at what, i had an MD5 routine and benchmark in javascript that was laughably slow on konqueror/safari...
The benchmark is at:
http://pentestmonkey.net/jsbm/index.html [pentestmonkey.net]

And i get the following results on a macbook pro 2.16ghz core2 duo running osx 10.4.10:

Safari (2.0.4):
MD5 Benchmark took 15.136 seconds for 3000 hashes (198 hashes/second)
MD4 Benchmark took 10.876 seconds for 2700 hashes (248 hashes/second)
SHA1 Benchmark took 19.052 seconds for 1900 hashes (100 hashes/second)

Camino (1.5.1):
MD5 Benchmark took 1.78 seconds for 3000 hashes (1685 hashes/second)
MD4 Benchmark took 1.271 seconds for 2700 hashes (2124 hashes/second)
SHA1 Benchmark took 1.931 seconds for 1900 hashes (984 hashes/second)

Firefox (latest nightly build):
MD5 Benchmark took 1.867 seconds for 3000 hashes (1607 hashes/second)
MD4 Benchmark took 1.299 seconds for 2700 hashes (2079 hashes/second)
SHA1 Benchmark took 2.077 seconds for 1900 hashes (915 hashes/second)

Firefox (2.0.5):
MD5 Benchmark took 2.628 seconds for 3000 hashes (1142 hashes/second)
MD4 Benchmark took 1.919 seconds for 2700 hashes (1407 hashes/second)
SHA1 Benchmark took 2.872 seconds for 1900 hashes (662 hashes/second)

Opera 9.23 (current stable):
MD5 Benchmark took 4.561 seconds for 3000 hashes (658 hashes/second)
MD4 Benchmark took 3.163 seconds for 2700 hashes (854 hashes/second)
SHA1 Benchmark took 4.812 seconds for 1900 hashes (395 hashes/second)

Opera 9.50 alpha (build 4404):
MD5 Benchmark took 1.446 seconds for 3000 hashes (2075 hashes/second)
MD4 Benchmark took 1.021 seconds for 2700 hashes (2644 hashes/second)
SHA1 Benchmark took 1.607 seconds for 1900 hashes (1182 hashes/second)

Quite impressive the improvements that have been made in the latest opera... Also, camino wasn't faster than the firefox nightlies last time i tried it (camino 1.0.4)...
I don't have access to msie or konqueror, i would assume konqueror performance would be similar to safari tho.

Re:ok, really?

[meh106]

Last time I looked, 400ms is the better part of half a second - quite noticeable in my experience!

and yet still no fundamental authentication..

[XaXXon]

wake me up when it supports spnego/kerberos auth. Then I can tell my users they use opera at work.

Re:Article is very misleading - JS benchmark only

[othermaciej]

Here's some results on Mac OSX (MacBook Pro Core Duo 2GHz):

Prerelease builds:

Safari 3 Nightly 177ms
Opera 9.5 Alpha 278ms
Firefox 3 Nightly 823ms

Production builds:

Safari 2 423ms
Opera 9.2 684ms
Firefox 2 880ms

Looks like Safari wins this one.

Re:Those numbers mean nothing...

[Opportunist]

The funny thing is that Opera currently has no JS exploits (at least none that I'm aware of, couldn't test the 9.5 build yet), while both IE and FF suffer from a number of bugs that can be abused for privilege escalation (and are exploited with packages like MPack).

Re:Grade article: incomplete

[Hal_Porter]

Opera used to have problems with Digg back in the 8.x days, but since 9.x it works just fine.

Re:its all about the addons

[sqrt(2)]

Automatically updating block lists, Opera doesn't have that. Flashblock displays an inline play button over all flash content so you can choose to play something instantly. Noscript gives you an icon right at the bottom showing what domains are allowed and what are blocked from running scripts and you can white and black list things through the same menu. Opera doesn't even come close to matching these features natively, and if there's plugins that do I'm not aware of them. And I'll kick in Down Them All plugin that I can't live without now. So that's four reasons I can't use Opera, even though I like it better than FF in a lot of ways, the UI is solid and it's very snappy with a low memory footprint.

Re:I'm fed up with the anti-Opera crap here...

[cp.tar]

I tried Opera.

Good browser it may be, but I don't like it. It's better than IE, but then, so is Lynx.

I like Firefox not so much for its speed (I'll admit Opera is faster), but for the extensions.

And yes, some of the more often used extensions do come off as copies of stuff first introduced in Opera, which makes Opera a bit of the Apple of the browser world.

And JFTR: Opera fanboys (the few that I've encountered) are worse than Linux, Mac and Amiga fanboys combined.

Benchmarks be damned

[just_forget_it]

The only reason I use Firefox over Opera and IE 7 is because of Firefox's find feature. Having a separate window pop up for finding a word or phrase is incredibly disruptive, especially when you're looking for multiple instances. As soon as you click outside the find window, it looses your place, and you have to start all the way at the beginning again. I know it sounds silly to most people, but Find is one of the feature I use most often, and if it isn't like Firefox, I'm not switching.

Re:Benchmarks be damned

[VGPowerlord]

Have you tried hitting / in Opera to open the inline find command?

Re:its all about the addons

[AmiMoJo]

Another handy but little know plug-in for FF, which is not available in any other browser, is CookieButton. It allows you to quickly set per-site cookie permissions from the toolbar.

This is fantastic for privacy. I have FF to accept all cookies, but delete all except the ones I specify to keep when the browser is closed. This way all web sites work (some don't if you disable cookies) but all tracking cookies and other crap gets deleted at the end of every session.

Re:Different market

[Anonymous Coward]

It is possible on Windows, by setting up an event with CreateMemoryResourceNotification - see http://msdn2.microsoft.com/en-us/library/Aa366541. aspx [microsoft.com]

Re:Who cares?

[Anonymous Coward]

"I'd like a multi threaded browser" - by Bert64 (520050) on Friday September 07, @05:19AM (#20505053)

Opera runs 8 threads here (per taskmgr.exe &/or process explorer) in the Windows model... check it yourself!

APK

P.S.=> Some added "FYI" for those of you comparing FireFox/IE/Opera:

Opera security advisories @ SECUNIA (0% unpatched):

http://secunia.com/product/10615/?task=advisories [secunia.com]

FireFox security advisories @ SECUNIA (43% unpatched):

http://secunia.com/product/12434/ [secunia.com]

IE 7 security advisories @ SECUNIA (56% unpatched):

http://secunia.com/product/12366/ [secunia.com]

(As far as security related vulnerabilities remaining unpatched, Opera leads here (super-important in today's online world where security IS a concern))

---

Also, as far as speed comparisons? This is one that also extolls Opera's benefits over FF &/or IE here, & ON MULTIPLE OS PLATFORMS:

BROWSER SPEED COMPARISONS ON MANY TASKS & MULTIPLE OPERATING SYSTEM PLATFORMS:

http://www.howtocreate.co.uk/browserSpeed.html [howtocreate.co.uk]

And, especially on Win32 OS', the most used PC platform/OS there is...

---

(& the best part is, Opera has ALL of the features a body can need, WITHOUT using addons (though it has that via Opera widgets), & YET, Opera is LIGHTER ON MEMORY than FireFox &/or IE typically!)

You can check memory residency yourselves by loading FF, & Opera (& IE for Windows users) & test memory size occupancy via taskmgr.exe (or similar tools like Process Explorer) yourselves & see what I mean... I did so with FF 2.0.0.6, IE 7.x, & Opera 9.23.

---

Opera also passed the "ACID2" test, for standards compliance (it is not alone here, but is over IE & FF, & it was the 6th browser to do so):

http://it.slashdot.org/it/06/03/12/1416222.shtml [slashdot.org]

A descending chronological order in which browsers (and authoring tools) passed Acid2, per a tip I got from by rh0 (member 1110203) here on /.:

http://en.wikipedia.org/wiki/Acid2 [wikipedia.org]

Safari, Prince, Shiira, Konqueror, Opera, & iCab

(Firefox's Acid2 compliant branch has been merged into the trunk, thus, Firefox 3 will likely be Acid2 compliant, but currently FF & IE are not passers of this test.)

---

And, Opera had features (like tabbed browsing) that other browsers (major 2 others in IE/FF) copied from it:

FIREFOX MYTHS:

http://mywebpages.comcast.net/SupportCD/FirefoxMyt hs.html [comcast.net]

(Yes, Opera had tabbed browsing before IE &/or FF, & other features as well. Opera comes FULLY LOADED features-wise, with a built in email client, IRC client, RSS client, & more + yet eats less RAM than others, & addons only bloat IE &/or FF even more memory-occupancy-wise. (AND YES, Opera has addons as well in "opera widgets" (like .xpi addons for FF))... apk

everytime I hear about Opera's amazing speed...

[scorilo]

... I think that the news is not how fast it is, but that it manages to be the fastest while being so far ahead of the pack in terms of everthing else. Consider this:

• Security: According to Secunia, Opera has 0 unpatched holes, compared to IE which has the most and Safari, second worst; unfortunately, Firefox has quite a few left as well.
  
• Features: Integrated email, feed reader, widgets, notes, IRC & bittorrent client; back in the day, Netscape tried to do that with email, then gave up when the code became too heavy and impossible to manage, opting instead for "modularization"; IE followed by introducing menu items for OE and FP Express. Opera is the only one left standing and still the fastest with the smalled footprint.
  
• The only browser that can read pages back to you (Windows->select text, right click, V)
  
• Portability - opera-usb.com, comes with flashblocker button in case you don't know how to set it yourself
  
• Mobile devices, where it's at - Opera rules that market
  
• Someone complained about the find function - the window doesn't actually disappear if you click behind it, it stays on top but loses focus

Wishlist: better integration with Google modules and especially Google Reader, but part of that seems to be addressed with Synchronization I used Opera since version 5. I would not use an OS unless there's an Opera made for it.

Re:Does it have the equivalents of these extension

[porneL]

Opera does have equivalents of many must-have extensions. Some are missing (IETab), some are better integrated (gestures), some are almost-but-not-quite (web dev tools unfortunately).

That sums it up: http://my.opera.com/Rijk/blog/2006/07/04/top-150-p opular-firefox-extensions-and-opera [opera.com]

Out of 113 most popular Fx extensions: 38 are built-in, 38 are not possible, rest can be added by tweaking/hacking/configuring something.

How Opera is Supported

[Kelson]

Then how is it supported?

They have deals with search engines, like Google and Yahoo, to get placement as the default engines in the toolbar, in Speed Dial, and in Opera Mini. (I think these days it's Yahoo in all 3.) Same kind of deal that Firefox has with Google, really.

Plus there are the versions for devices [opera.com] (Nintendo DS, etc.), which they still charge for, either directly or through licensing deals with device manufacturers and mobile carriers. So they pull in revenue from that.

This article is a year out of date, but still informative: Opera making big profits from free software [itwire.com.au].