W3C home > Mailing lists > Public > www-style@w3.org > May 2013

Re: [css-display]? Compositing, expensive things, and laziness

From: Ali Juma <ajuma@chromium.org>
Date: Thu, 23 May 2013 15:26:45 -0400
Message-ID: <CANLC6v0PUd_i5rL7YMLXrwiAambcfSrm-tSeBrprpH-Hg809xQ@mail.gmail.com>
To: Simon Fraser <smfr@me.com>
Cc: "Tab Atkins Jr." <jackalmage@gmail.com>, "Robert O'Callahan" <robert@ocallahan.org>, www-style <www-style@w3.org>, Jon Rimmer <jon.rimmer@gmail.com>, Ian Vollick <vollick@chromium.org>
On Thu, May 23, 2013 at 2:43 PM, Simon Fraser <smfr@me.com> wrote:

>
> But each paint represents a particular state of the render tree. The
> request
> seems to imply that you want to queue up an expensive paint earlier than
> a cheap paint, so they can be done at the same time. That then implies that
> you’re able to generate the render state for the expensive paint
> independently,
> which is the hard part.
>
> Simon
>
>
Another way to look at this is that we want is a way to mark parts of a
page as being allowed to be omitted from what’s displayed on the screen
until a more convenient time (in other words, have a way to say “this div
is optional; when you’ve drawn it, let me know”).

The browser implementation could decide where/how to take advantage of the
optionality of the content. In Blink/chromium, we’d still paint to an
SkPicture in Blink, but we’d defer rasterization in chromium.
Alternatively, an implementation might choose to omit creating the render
tree subtree corresponding to the optional content. Or it might choose to
only omit texture upload.

-Ali
Received on Thursday, 23 May 2013 19:27:12 UTC

This archive was generated by hypermail 2.3.1 : Thursday, 23 May 2013 19:27:13 UTC