[Public WebGL] OT: exception handling

Gregg Tavares (勤) [email protected]
Fri Apr 6 09:39:10 PDT 2012


On Fri, Apr 6, 2012 at 9:27 AM, Florian Bösch <[email protected]> wrote:

> On Fri, Apr 6, 2012 at 6:15 PM, Gregg Tavares (勤) <[email protected]> wrote:
>
>> You can rant all you want. Unfortunately it's today's reality. Lost
>> context is not going away anytime soon.
>>
> I know
>
>
>> That has nothing to do with WebGL. It has to do with the current state of
>> GPUs, drivers and OSes. GPUs are not CPUs. Their memory isn't easily
>> swapped or protected and currently none of them are preempt-able. None of
>> the browser vendors have control over that. Even Apple doesn't write their
>> own drivers or make their own GPUs.
>>
> I know that too. I'll nevertheless note a couple of things:
> 1) No WebGL framework in existence handles context lost, or will handle it
> in anything like the forseeable future, if at all, ever.
> 2) It might just be nigh-impossible to write any kind of framework that
> can handle it.
> 3) Requring the average interested WebGL developer (or even the average
> framework writer) to deal with this, is relegating the writing of WebGL
> code to the elite of the elite, and even so, their code won't be pretty,
> reusable or understandable.
> 4) No, none whatsoever, WebGL demo/application beyond the hello-world
> equivalent handles context-lost so far, I posit, it's impossible anyhow for
> anything slightly non-trivial.
>
> So I know I can rant whatever I like, and the problem's not gonna go away.
> But you can introduce context-lost all you want, the problem's *also* not
> going away.
>

Every DirectX app from DirectX1 through DirectX9 handled this. There's lots
of precedent for dealing with this.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://khronos.org/pipermail/public_webgl_khronos.org/attachments/20120406/50736d96/attachment.html>


More information about the public_webgl mailing list