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

Re: [Public WebGL] ATC texture size computation




On Sep 11, 2012, at 8:31 AM, Daniel Koch <daniel@transgaming.com> wrote:


On 2012-09-11, at 11:02 AM, Chris Marrin wrote:


On Sep 11, 2012, at 7:55 AM, Vladimir Vukicevic <vladimir@mozilla.com> wrote:

On 9/11/2012 3:59 AM, Florian Bösch wrote:
On Tue, Sep 11, 2012 at 2:30 AM, Chris Marrin <cmarrin@apple.com
<mailto:cmarrin@apple.com>> wrote:

   But we really do need to make sure the one(s) chosen can be included
   in the spec free of any undeclared encumbrances. IANAL, but I know
   that not being careful about this will almost certainly bite us in
   the ass later...

That's true. Although, either you put it in the spec, or you'll have to
live with that JS libs do it.

The compression algorithm should definitely not go into the spec.  Just what's needed to implement, which is the buffer size formula.  The vast majority of users will compress their assets offline and deliver the right one based on the device they're running on.

Once again, IANAL, but I'm not even sure it possible to include the availability of a proprietary format in the spec. I suppose it depends on how the licensing for that format is done and for that we need to get lawyers (or at least Khronos) involved.

I don't suppose ATC could be any more ip-encumbered than S3TC, and yet that extension is available in a good many GL and GLES implementations, and seems to heading for webGL support as well.

And we will need to take the same precautions there as well. Of course, since they are owned by two separate companies, issues surrounding them may be different.

-----
~Chris Marrin
cmarrin@apple.com