Yes, it's intended to work with both depth textures and multiple render targets. All color attachments that are attached with framebufferTextureMultiviewWebGL will be rendered to.
I do agree that WebGL 1.0 support is useful, though WebGL 2.0 support will surely ramp up given time. The main use case of the extension is VR, and especially on devices which are VR capable I'd expect widespread support for
WebGL 2.0 already. We did consider some other alternatives for WebGL 1.0 compatibility for the proposal, but they all seemed fairly hacky. I don't think we want to expose rendering
two viewports side by side in a 2D texture for example.
From: Florian Bösch <firstname.lastname@example.org>
Sent: Monday, August 21, 2017 2:16 PM
To: Olli Etuaho
Cc: Public WebGL (email@example.com)
Subject: Re: [Public WebGL] Public review of updated WEBGL_multiview extension proposal
On Mon, Aug 21, 2017 at 10:02 AM, Olli Etuaho <firstname.lastname@example.org> wrote:
Bit of a bummer since WebGL 2.0 has flatlined at ~30% and isn't growing and WebGL 1.0 for now is the only feasible platform. The prevalence of the OVR_multiview extensions on mobiles also seems particularly poor.
Some additional questions: