[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Public WebGL] OES_texture_half_float doesn't allow Tex(Sub)Image2D to take ArrayBuffers as input

On Thu, Feb 13, 2014 at 7:47 AM, Florian Bösch <pyalot@gmail.com> wrote:
> On Thu, Feb 13, 2014 at 2:57 PM, Florian Bösch <pyalot@gmail.com> wrote:
>> what if you input numbers larger than expressable by half-float?
> Now also checks for this case, and adds checks for correctness of
> largest/smallest values to convert correctly. I think it might round a bit
> stupid from double (not picking the nearest expressable, but the floored
> one).

Cool, thanks for putting this together. It's good to know that the
required math operations can be done in JS.

I don't anticipate a Float16Array ever being added to the spec. It
would require a lot of specialized code to be added to JS engines
which would be useful only for a small subset of WebGL applications.
Further, it can be implemented in pure JS using proxies and your code,
and as long as JS engines keep getting faster, that implementation
should be as fast as a handwritten assembly version baked in to the


You are currently subscribed to public_webgl@khronos.org.
To unsubscribe, send an email to majordomo@khronos.org with
the following command in the body of your email:
unsubscribe public_webgl