Just wanted to know the difference between the start render and first paint…
From the metrics definition it seems to be very similar…
https://sites.google.com/a/webpagetest.org/docs/using-webpagetest/metrics
http://www.webpagetest.org/forums/showthread.php?tid=15
Both says time when the user start seeing the rendered content…
Can somebody pls clarify?
Start render is measured by capturing video of the page load and looking at each frame for the first time the browser displays something other than a blank page. It can only be measured in a lab and is generally the most accurate measurement for it.
msFirstPaint is a measurement reported by the browser itself about when it thinks it painted the first content. It is available from javascript and can be reported from the field. It is usually pretty accurate but sometimes it reports the time when the browser painted nothing but a blank screen.
So can we say like this?
Start render is measuring for the first time the browser displays something other than a blank page. Similar to MSpaint but we are doing the measurement by capturing video…
msFirstPaint is a time when the browser itself thinks it painted the first content.You also said sometimes it reports the time when the browser painted nothing but a blank screen also.
How we can measure the accuracy in that case when the MSfirstpaint is measured with blank screen.
Yes, that is better. As far as improving the accuracy for msFirstPaint, you’d need to open bugs with IE and Chrome. It is a browser-reported value.
Patrick thanks…
Is there is any way we can see the capured videos for any test but i am able to see the Images only captured for Start render… Can you pls clarify…
Screenshot below…
http://www.webpagetest.org/screen_shot.php?test=141003_7Y_5FS&run=2&cached=0
Just click the “Filmstrip view” link from the main test results (far right column) - http://www.webpagetest.org/video/compare.php?tests=141003_7Y_5FS-r:2-c:0
can startRender time can be greater than first paint time ??
I read your earlier comments from 2009 where you were calculating startRender time by calculating whether dom dimensions are changing or not… now you are calculating by observing the video captured… correct ??
Yes - mostly because the browser under-estimates the first paint (doesn’t include some of the graphics pipeline) or if the paint is of legitimate, but white, content. There are a few other cases as well but those are the most common.