Re: [svgwg] Note that buffered-rendering should be removed.
"Tab Atkins Jr." <[email protected]>
| Newsgroups | gmane.comp.web.svg |
|---|---|
| Message-ID | <CAAWBYDA5+GZRzt+7b29VSvPp65gN6WYxRxatswX4iZ8T+-_NMA@mail.gmail.com> |
On Wed, Sep 2, 2015 at 1:08 PM, Amelia Bellamy-Royds <[email protected]> wrote: > If the decision is to rely on CSS `will-change` for buffered rendering > hints, should we ask CSS to introduce a special keyword value to allow > authors to declare that SVG geometric properties will change? For example, > if you will be animating path data or polygon points, indicate it with > `will-change: svg`. > > There is precedent: the current will-change definition includes keywords for > changes to scroll position and to text contents, which are not otherwise set > in CSS. > > Or would SVG geometry be covered under `will-change: contents`? If that's > the case, perhaps we could add a short note somewhere in the SVG 2 specs > (the scripting section, maybe?) advising authors to use `will-change: > contents` to warn browsers about scripted changes to SVG attributes. If you're gonna change attributes that map to properties, you can just indicate those in will-change. If you're gonna be changing things willy-nilly, yeah, "contents" is the thing. ~TJ