(thread migrated from sourceforge forum - originally posted 3/7/2008)
Can the browser significantly affect load times? If so is there any easy way to get concrete numbers on different browsers hitting a site? I saw that this tool uses IE7 and YSlow can give some information in Firefox, but I want to make sure apples are apples and if possible add IE6 to the equation.
Absolutely. IE6 and IE7 will perform very similarly unless a fair amount of javascript comes into play at which point IE7 usually is faster. Most of those differences come from what show up as “gaps” in the waterfall and are really not all that common in normal web pages but become more important for rich Internet apps like webmail.
Firefox on the other hand is significantly different from IE in the number of parallel connections, CSS/JS dependencies and enough other things that the results can not translate directly between firefox and IE (and if something like fasterfox is installed, it diverges even more). That’s actually the main reason we built pagetest in the first place. Nearly all of our internal developers use firefox because it’s a much friendlier browser to do dev work in (thanks to firebug primarily) and they were not doing any testing in IE.
The optimization recommendations from both Pagetest and YSlow will still apply to both browsers though, none of the rules either is checking for only apply to one of the browsers. What will not translate directly is the waterfall of the load (net panel from firebug) and the absolute load times.
Hi
The browser only makes the connection after that it is purely academic. After that it’s the computer and the connection delivery.
The variations would be RAM, CPU speed, Ethernet or USB connection, The rating of the Ethernet card, Protocols, Modem transference speeds, wired or wireless for the computer.
Regards the Internet side, Packet sizes, Connection speed, where your downloading from, what client you are using [legal or illegal] Proxies, how many on the site you are downloading from, The number of people downloading the same thing, lags, interruptions etc.
No two downloads are the same. You yourself know you can go to download something and it’s so bad you give up. Go back two hours later and zip it’s done in a couple of minutes. Yet your using the same computer and the same browser and the same connection
As wasimasif mentioned different browsers allow different numbers of connections per host, which can affect load time. In addition browsers have different JavaScript execution engines, which can vary widely speed wise (as the author alludes to). So the browser can have a HUGE effect on overall page load time, especially for sites with a lot of content and a lot of JavaScript. Browserscope has replaced gquintana’s link, and that’s a good place to start if you want to see the differences between browsers.