Thread

What does {elapsed_time} really measure? The manual says the time it takes EE to render the page. And I assume that means the time from which EE received the request from the browser until the time it sends the page back to the browser. I assume this doesn’t include the time it takes to load the js, css and images and html. Just the time it takes EE to assemble the HTML page sent to the browser.

So why would different browsers have a different average of elapsed time? I loaded a page in FireFox and refreshed the page 10 times. Then did the same in Safari and Chrome. Then went back thru each browser and refreshed 10 times before moving onto the next browser. I cycled four times thru the browsers. While I didn’t save the times in a spread sheet for stastical analysis, it is clear that Firefox is a slower browser than both Safari and Chrome based on {elapsed_time} delivers. Safari and Chrome are very close but Safari seems to be faster.

Can it be that EE performance is affected by browser type?

What does {elapsed_time} really measure? The manual says the time it takes EE to render the page.

Yes, exactly. The amount of time between receiving the first request and handing off the final result to the browser.

Just the time it takes EE to assemble the HTML page sent to the browser.

Yes.

So why would different browsers have a different average of elapsed time?

How much of a time difference are we talking here?

Can it be that EE performance is affected by browser type?

I don’t see how. The only difference that I can imagine is in the case of multiple resources / requests which are handed back to the server at varying speeds.

Firefox typically displays an elasped times that are 50% to 100% longer. So for http://www.sode.org the elasped time:
-for FireFox ranged 0.9 to 2.2 secs
-for Safari ranged 0.6 to 1.6 secs

The only difference that I can imagine is in the case of multiple resources / requests which are handed back to the server at varying speeds.

Are you saying the server/EE is making requests back the browser while it renders the page?

Is there anything I could to definitely test this?

Is there anything I could to definitely test this?

I’ll ask… Thanks in advance for your patience.