AMD vs. nVidia binary driver?

[Lennart Sorensen]

On Fri, Jan 21, 2011 at 01:49:15PM -0500, Thomas Milne wrote:
> That's always been my question, really:  why not? What is the
> advantage, really, to keeping the code closed? I know what NVidia
> would say, but aren't they kinda being disingenuous? For the average
> person, of course, the stuff they don't want to share is protected.
> But for anyone who knows what they're doing, isn't it trivial to take
> apart the 'blob' and see what's inside? Or is it encrypted or
> something? And even if someone _did_ get a peek at the code, they
> would still be legally prevented from using it, no?

Well if someone wants to disassemble their driver, they can try.
It is huge.  Making sense of the assembly would be very hard.

I think they have a lot of 3D optimizing code to make the opengl requests
be handled efficiently.  No point passing stuff to hardware if you know
it isn't needed in the first place.  If their optimizations are better
than what ATI currently has (No idea if they are or not), then open
source would let ATI see how they do that.  Of course it may be that it
isn't even ATI they are worried about, but other smaller companies that
currently don't stand a chance at writing a good 3D driver from scratch.

The Mesa project does seem to have done some decent work on 3D drivers,
but how they compare to what nvidia is doing, I have no idea.  Hopefully
someday openGL and 3D in general will be a generic commodity and hardware
makers can go back to making better hardware and not worrying about
driver optimization secrets.

-- 
Len Sorensen
--
The Toronto Linux Users Group.      Meetings: http://gtalug.org/
TLUG requests: Linux topics, No HTML, wrap text below 80 columns
How to UNSUBSCRIBE: http://gtalug.org/wiki/Mailing_lists