5 Way Browser Wars

[Super Fast]

6/9/2012 -> Downloaded newest Chrome/Internet Explorer/Firefox/Opera/Safari (& a few older ones) to test Hardware + HTML5 acceleration.

 

Tested all browsers on a clean 32 Bit 7 profile / P4 3.2 dualcore / 7,200 RPM 160 GB Sata HDD (114 GB free) / 3GB RAM / 1 GB ATI Radeon 3D

 

Used -> http://ie.microsoft....rmance/FishBowl

_____

 

250 Fish (Frames Per Second)

 

Chrome 19 = 30

Chrome 21 = 40

Internet Explorer 9 = 36

Firefox 13 = 60

Firefox 15 = 60

Opera 12 = 30

_____

 

500 Fish (Frames Per Second)

 

Chrome 19 = 15

Chrome 21 = 21

Internet Explorer 9 = 16

Firefox 13 = 33

Firefox 15 = 37

Opera 12 = 15

_____

 

1,000 Fish (Frames Per Second)

 

Chrome 19 = 8

Chrome 21 = 11

Internet Explorer 9 = 8

Firefox 13 = 15

Firefox 15 = 16

Opera 12 = 8

_____

 

1,500 Fish (Frames Per Second)

 

Chrome 19 = 5

Chrome 21 = 7

Internet Explorer 9 = 6

Firefox 13 = 10

Firefox 15 = 10

Opera 12 = 6

_____

 

2,000 Fish (Frames Per Second)

 

Chrome 19 = 5

Chrome 21 = 7

Internet Explorer 9 = 4

Firefox 13 = 7

Firefox 15 = 7

Opera 12 = 5

_____

 

250 + 500 + 1,000 + 1,500 + 2,000 Fish (Total Frames Per Second)

 

Chrome 19 = 63

Chrome 21 = 86

Internet Explorer 9 = 70

Firefox 13 = 125

Firefox 15 = 130

Opera 12 = 64

_____

 

Browsers listed below in order of speed:

 

# 1 Firefox 15 -> 130 points

# 2 Firefox 13 -> 125 points

# 3 Chrome 21 -> 86 points

# 4 Internet Explorer 9 -> 70 points

# 5 Opera 12 -> 64 points

# 6 Chrome 19 -> 63 points

_____

 

The slowest firefox is nearly 2X faster than Chrome/Internet Explorer/Opera.

The fastest non-firefox doesn't even break the double digits.

 

Firefox is the fastest according to my testing, & wastes the least time on the web. Why spend 2X longer on the web?

 

* There are some tests for only hardware acceleration, which will give all browsers higher scores. But that is pointless.The web is going towards HTML5, & the browsers that have the highest scores for HTML5 + Hardware acceleration will be that much more so higher than the rest on simple hardware acceleration tests.

_____

 

The following browsers failed & were eliminated:

 

Internet Explorer 8 / Firefox 1 / Firefox 2.0.20 -> Failed to load fish.

Firefox 3.6.28 250 fish / 2 FPS / Opera 11.64 - 250 fish / 2 FPS / Safari 5.1 - 250 fish / 1 FPS-> Failed acceleration.

Firefox 16a1 = 15a2 / Firefox 14b6 = 13 final -> Identical results for all numbers of fish.

_____

 

The things that really surprised me, are:

 

* Apple Safari 5.1 -> 1 FPS in 2012?

* Opera 12 -> Dusted by all modern browsers, but is (finally!) getting back into the game.

* Chrome 19 -> Opera 12 just released, & it already (marginally) beats Chrome 19?

* Internet Explorer -> It's not last place!

* Firefox -> How does it go 2x faster than the others?

 

EDIT: Thank you for catching the date, Nergal! I meant to post as 6/9/2012, but I must have typed 6/19/2012 instead. Date is now corrected!

 

Firefox 13 = 14b6 in terms of speed, but I am choosing it as my fav. FF14 improvements under 32 bit 7 include slightly smaller icons, (I love this! Finally, the forward & back buttons are the same size! And smaller means more area for the web page to display), corrects a flash container has crashed error that occurs now & then, & seems very smooth. 13 was smooth, but 14 is even more so.

 

For now, FF14 is my fav.

[Winapp2.ini]

Neato testing, but probably varies greatly depending on your hardware.

[hazelnut]

Neato testing, but probably varies greatly depending on your hardware.

 

Agree, nice testing, but too many variables to make it a 'set in stone' set of results.

[Nergal]

wait these tests either happened in the future, or in the fabled 19th month

 

yes all browser acid tests and what not depend greatly on, not only the brand/make/model of a device but also on the unique properties of your exact device(s)

 

also try ie10 in win8 if you can

[Corona]

SuperFast, you constantly mention having "hundreds" of tabs up on your browser. Can you please post a pic of that? Just for laughs? Thanks.

[Winapp2.ini]

I'm curious too. I usually have tens of tabs open, but I can't say I've consistently had hundreds open.

[Andavari]

The most tabs I've ever had open at once was about 12, can't even begin to imagine 100 of them.

[Super Fast]

Corona, but how would you see all the tabs?

 

Sure, they scroll off the screen, but how would you know if it was 50 or 2,000?

Firefox does, however, tabulate the # open tabs when you close it, so I suppose I could do that for you one day when I get time.

The most I ever had open in a single session is 1,611.

 

It took a lot of RAM & my machine started slowing...

[Andavari]

The most I ever had open in a single session is 1,611.

 

It took a lot of RAM & my machine started slowing...

1611?

 

That's allot of them wouldn't you say. If I did that on this old PC of mine the mobo might jump out of the case and smoulder on the floor.

[login123]

Googled for "wide screen desktop monitors", found a few ranging in price from a mere $6,000 to a sort of pricey $49,999. They range in size from big to "build an addition on your house".

The good thing is that you could see most of your tabs.

[Winapp2.ini]

It's worth it, invest in it for our amusement.

[Super Fast]

Googled for "wide screen desktop monitors", found a few ranging in price from a mere $6,000 to a sort of pricey $49,999. They range in size from big to "build an addition on your house".

The good thing is that you could see most of your tabs.

 

Best laugh I had all day! I wonder what 1,611 tabs would look like? -> Each tab = 1 inch. 12 inches = 1 ft. 1,611 divided by 12 = 134.25 feet of tabs!

 

I could see all my tabs on a 135 ft wide monitor! Whew! That's almost 2x longer than my trailer!

 

Never took time to think of it this way before, but you got me thinking with your comment...

[Winapp2.ini]

Rent out one of those really big football stadium monitors for a day

[TheWebAtom]

Using a test created by Microsoft will inevitably have a bias towards IE's capabilities. It's interesting, because if I do the exact same test you did, IE9 wins every single time - on every test.

[Super Fast]

Using a test created by Microsoft will inevitably have a bias towards IE's capabilities. It's interesting, because if I do the exact same test you did, IE9 wins every single time - on every test.

 

Cool. Can you provide the following information?

_____

 

1. OS Name / 32 or 64 Bits / RAM amount / CPU (also, single/dual/quad core?) / GPU 3D? / Programs running / AV used / HDD (7,200/5,400 RPM?) or SSD?

2. VM, or running it natively?

3. Browsers used, & version # for each.

_____

 

I am very interested, as each of the factors may have a part in your results.

 

Also, what link did you use for the testing? Please be sure you ONLY use: -> http://ie.microsoft....mance/FishBowl/

_____

 

That link will test HTML5 + Hardware acceleration. There is ANOTHER fishtank test that doesn't test HTML5 acceleration on the link below:

 

http://ie.microsoft....nk/Default.html

_____

 

Please specify which you use. Also, most browsers tend to stabilize after around 10 to 15 seconds, & may display an initial high burst.

You want the stable test results.

_____

 

EDIT: There are a few other important things to know. Running multiple open tabs? Many web browsers will slow CPU time to background tabs, & give more to the active one. Some people test with all browsers open at once. This is never a good way to test, because some web browsers may request/get more CPU or GPU time to run.

 

I ran all my tests with ONLY the HTML5 window open, & ONLY that particular browser open until testing was complete. In addition, I waited 10 to 15 seconds for the browser to stabilize & settle down on the final # of fish. Some browsers will start high, then drop massively after around 10 seconds or so. Chrome seemed especially guilty of this.

 

I am very interested, so I will check back later to see your reply. Great to have others interested in testing!

 

@Nergal -> I thought about testing IE10 under Windows 8, but I am not too sure yet. I heard Win 8 might not even come with a DVD player, unless you buy that separately!

 

I might end up testing it still, when Win 8 finalizes, so I can get a true score for it. As of now, the betas have a lot that could be changed before release.

I believe I will wait for Win 8 Final to test, so the results show the true Win 8 score.

[Nergal]

I believe the test you ought to be running (as was incenuated above with the ofcourse MS passes MS test) is the industry standard html5 test such as http://html5test.com/

[Winapp2.ini]

my firefox waterfox 13 gets 345 + 9 points.

 

My chrome 20 gets 414 + 13 points.

 

The biggest difference seems to be in "forms" (56v81)

 

Edit:

 

Nightly (fx16) 363 + 9

 

IE9 (64bit) 138 + 5

 

IE9 138+5

 

Aurora (fx15) 346 + 9

 

Firefox beta (fx14) 345 + 9

 

I have an installer for Chrome 21 (apparently, mine is out of date in terms of the dev channel) but it just installs Chrome 20. Weird, else I'd test that for you folks as well. I just went with what I have installed.

[Super Fast]

I believe the test you ought to be running (as was incenuated above with the ofcourse MS passes MS test) is the industry standard html5 test such as http://html5test.com/

 

I tried that test, & I passed, however, the reason I used the HTML5 fish bowl test, is because the test you listed only shows whether your browser supports this or that.

The fishbowl test actually is a comparison, of sorts, of the hardware acceleration. And it gives you some idea which browser handles hardware acceleration the best for multimedia web browsing.

 

Hopefully, the future will give us even better browsers, but we must wait till it gets here...

[Super Fast]

my firefox waterfox 13 gets 345 + 9 points. Etc.

 

I got different results than you (Just slightly). What is your OS? And are you on 32 or 64 Bit? What kind of CPU/GPU do you have?

[Winapp2.ini]

Windows 7 Professional x64 Service Pack 1

 

AMD Phenom II X4 955 Black Edition (CPU)

 

ATI Radeon HD 5770 (GPU)

 

12GB DDR3-1600 RAM

[TheWebAtom]

Yes, I used only the fishbowl test.

 

Windows 7 Professional x64 Service Pack 1

Intel® Core i7 CPU 950 @ 3.07GHz

16GB DDR3 RAM

ATI Radeon HD 5770 (GPU)

[Super Fast]

Yes, I used only the fishbowl test.

 

Windows 7 Professional x64 Service Pack 1

Intel® Core™ i7 CPU 950 @ 3.07GHz

16GB DDR3 RAM

ATI Radeon HD 5770 (GPU)

 

That is very interesting.

Things that make a big difference on yours, are your processor is a quad core with 8 threads.

Also, your running 64 bits.

 

They have 64 & 32 bit IE9. I wonder which you are/were running, because 64 bit can make a difference.

Also, if your using Firefox, it would probably be better to test Waterfox & other 64 bit browsers vs each other.

 

Comparing 32 to 64 bit is simply not gonna work, because 64 bit on a high powered system will blow 32 bit away.

[hazelnut]

All this goes to show is that all results are relative to what hardware/software that you test with.

 

Very few people have exactly the same hardware and software installed.

[Super Fast]

@Hazel, I agree.

 

It would be awesome if he could/would post his results in a manner similar to as I have, that I can compare the differences.

I may update this system to a newer 64 Bit AMD machine later to test how it does with the browsers.

 

I still think Opera will beat Chrome, & Firefox will beat both of those, but I will have to test to verify.

I am thinking of an AMD machine, because AMD bought ATI, & ATI is known for superior graphics.

_____

 

NVidia (was) known for faster rendering, while ATI was higher quality rendering, but in latest reports I have seen, not only is ATI better graphics quality, but it also is around 22% faster (on average).

 

In addition, I was looking to see which was better, iCore7 Vs AMD newer multi-core chips, & from what I have read, iCore will win (slightly) on running single applications, or maybe 2 or 3 applications.

 

AMD, however, is supposed to be better with running multiple tasks, & still keeping it smoothly. AMD has built in 3D graphics acceleration on some of their newer chips, which means it gets a performance boost from that alone.

_____

 

I watched a video of comparing intel iCore 7 VS AMD newer chips, & what I saw amazed me. They both started out with processing video, then running another task, then doing real time filtering on the video to make them look better. They were similar at first, but the AMD blew the intel away after stacking more processes for it to run. After say, 5 decent load apps, AMD was still running very smoothly, & intel was stuttering badly.

 

My understanding is that the AMD has way better load balancing, & can run far more things simultaneously with no slowdown. And that is what I am looking for in a multi-tasking machine.

_____

 

Let's say, for example, I have over 50 youTube videos open (in background tabs). If AMD is lots smoother & handles lots more youTube/Video games/3D content than intel with no lagging, then that will greatly accelerate my browsing/gaming/tinkering.

 

Very, very interesting!

[hazelnut]

As I already said Super Fast, everyones system specs are different, therefore different results.

 

'Which is better' usually ends up being a pointless exercise with browsers. Most folk choose a browser because it has features they like... nothing else.

 

Also getting someone to test with 50 youtube videos open (as you say you do along side the other open 1550 tabs) is never going to happen.....ever.