What’s the Fastest Web Browser in the “Real World?” Chrome.

http://techcrunch.com/2011/08/08/whats-the-fastest-web-browser-in-the-real-world-chrome/

 

not surprised its my default browser V.13

anyway,thanks for the link

 

Fastest when it works properly On topic, I'm glad someone actually stepped away from the irrelevant lab tests and into the actual world, where things can and often are very different. Though I must admit I'm not shocked by the results, considering development of Chrome is concentrated almost solely on speed.

 

The lab tests aren't too irrelevant, this test seems to agree with them.

 

Irrelevant in that a controlled environment will never be the same as the real world. This test was actually worth looking at, as it took into consideration the every day issues a user faces.

 

I suppose. Synthetic benchmarks still count for something.

 

They count if you're looking for what a browser is capable of under perfect conditions...and when was the last time you had perfect conditions?

 

What are perfect conditions? You can run synthetic benchmarks on your own computer and it'll show you which browser is at the top, and it pretty much coincides with what we see in the real world.

 

Synthetic tests leave out the web. They leave out server response times, they leave out ads loading, unique layouts, your own connection, and so on. They don't paint a complete picture, period. You and I both know a website chock full of Flash material and/or images, is not going to load faster than a more simplistic website (we're leaving adblockers and tools like noscript out of it, because then it's an entirely different ballgame).

 

Not saying they do.

 

I'm just glad for once that it's not another "Sunspider says this" or some such nonsense. At least this test bothered to use real people with different setups and different usage habits. It's a test I can look at and get an idea of how the browser will generally perform.

 

Mhm. I definitely like these sorts of tests. I also like that they measures real page load times and perceived page load times.

 

I don't understand why in vitro tests are frowned upon. Provided they're honestly set-up, they should give an indication of how a browser will perform in the real world.

 

I think that's the general consensus. They don't give the whole picture but they are indicative.

 

Tests should only be "real world" tests, if community people or support guys come up with things like "You should run the test with a blank profile" or "with a clean install" they're both cheating on you and themselves. Nobody runs around naked all days, and nobody starts each day with a fresh blank install of his internet browser.

 

My browser loads a webpage .0001 milliseconds slower than the other guy's.

I feel so used . . . .

 

Hmm. I have always thought that you need both types of tests.

A manufacturer needs a consistent environment that doesn't change, to see what happens with a new product. How does it compare to the old product. If there is not an exact same testbed, then how do they accurately determine changes?

A person reviewing products needs both a stable testbed, to see what the manufacturer sees, but also a dynamic environment that mimics the real world.

I don't think you can do without either one. It just so happens that we consumers don't really get the product until the static testsbeds are finished, unless we beta test or something. I put more credence in a dynamic tests that take into account normal variables. But that doesn't mean that the static tests are not important.

In the end though, as has been eluded to, until you put it on your machine with your settings and utilities, it is only a representation of the testers methodology.

Sul.

 

The thing is, it might be just 15ms slower on a light website but when you have heavy javascript it can be quite a bit more.

And I'd rather be on the program that's 15ms faster

 

Time is money as they as say.

 

And you'd be surprised at how ms add up so quickly with different elements of your page. Google did a whole study showing how to optimize RTT on servers and lighten the load and they showed a bunch of different elements and their times in ms to load and they were all like "30ms" here or there but when you optimize it you end up saving a full second or two, which is frankly a big deal.

 

If a second or two of time saved in browser speed, is the difference between a thermonuclear weapon being launched or not launched, you have yourself a point. Otherwise, just no

 

Really?

Find a link you want to click, count to two, and then click it. It's a long time.

 

So if Opera came out with an engine that was 2ms faster than V8 you'd swap?

It's come to the point now that the difference between loading of the various top browsers is in milliseconds, so much so that speed simply isn't a defining factor as much anymore as say, security. At least that's the way I've perceived it, I guess your opinion differs.

 

Of course not. But if two browsers were entirely equal except that one loaded pages 15ms faster I'd naturally use that browser. That's all I meant by it.

Security's definitely where Chrome exceed =p though IE9 is great as well.

I would say that there is quite a difference between page loading time speeds depending on the page. Heavy javascript can turn 15ms into a second, and that's only a piece of the picture. There's rendering with the GPU, ogl, html5, all sorts of different ways to effect the speed of a page.

 

I'd rather be on the one that doesn't load javascript to begin with unless I say so. And I am.
Though I shudder to think of those mass quantities of milliseconds lost as I click on that NoScript icon, allow a site, then reload the page.
Better I don't ponder that one at all, I think. No doubt I'll live longer.

Use whatever browser you want and for whatever reason you want.
If you think a couple of milliseconds here and there ultimately adds up to a major time savings, you have every right to believe that and to choose your browser or any other program accordingly.

To me, this whole 'neener-neener, my browser's faster than yours!' business is every bit as foolish and nonsensical as the recent 'major version release' naming idiocy conjured up by Mozilla.
I don't care what version number gets chosen for a browser, nor do I care which browser loads a webpage by a nanosecond faster than some other.
I care that my browser is secure and functional.