[Webkit-unassigned] [Bug 90375] Parallel image decoders

bugzilla-daemon at webkit.org bugzilla-daemon at webkit.org
Wed Jul 11 01:58:42 PDT 2012


https://bugs.webkit.org/show_bug.cgi?id=90375





--- Comment #15 from Kwang Yul Seo <skyul at company100.net>  2012-07-11 01:58:35 PST ---
(In reply to comment #14)
> (In reply to comment #13)
> > You can check out the updated test results in the following URL:
> > 
> > https://docs.google.com/spreadsheet/pub?key=0Ar2smwimcenMdGpnTEw2clZjSTNkbXNFNFM5dkYyRGc&output=html
> 
> The results look promising!
> 
> > A few notes:
> > - Adjust the threshold from 512x512 to 300x300
> 
> This sounds cool, because I made earlier an investigation about used image sizes based on the first alexa top150 sites and I can say that the most images have lower than 512*512 dimension: https://bugs.webkit.org/show_bug.cgi?id=71555#c7

This is really good information. Thanks. We will experiment more with various thresholds.

> 
> > - Measure first painting time to make sure all the images are decoded.
> 
> How do you measure first painting time?

We measure the last call to Image::draw. This ensures that all the images inside the initial viewport are decoded and painted.

> 
> > However, there is still limitation. Even image heavy sites do not show all the images at once. To see all the images, we need to scroll down many pages. Because WebKit decodes images only when they need to be painted, not all the images are decoded in page loading.
> 
> I think it's normal behavior, I see this as an advantage. If we are decoding all images including the never shown ones then we would make extra useless cpu cost on the worker thread.

I do like the current lazy behavior of WebKit image decoding. I just wanted to mention that we also need to consider images outside the initial viewport because they are decoded by parallel image decoders when we scroll. These images are also decoded faster thanks to parallel image decoders!

-- 
Configure bugmail: https://bugs.webkit.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.



More information about the webkit-unassigned mailing list