View Single Post
Old 08-11-06, 09:26 AM   #14
Registered User
Join Date: Jul 2005
Location: Munich
Posts: 910
Default Re: Winds of freedom for graphic drivers!!!

Originally Posted by MamiyaOtaru
In light of recent announcements from Intel and potential developments at AMD, this is a false dichotomy.
Again: intel's on-chip graphic is not an option for serious 3D applications
or gaming - i.e. i have a tripple head setup (4800x1200 desktop
resolution) and there is no way to reproduce that with intel hardware
at the time being (not taking into account reduced performance).

I certainly would appreciate it if AMD would open source the ATI drivers,
so that there could be a high performance well debugged 3D solution
for Linux in future, and i would dump nvidia hardware the second i think
AMD/ATI is the better solution - but there is no schedule or even a
confirmation from AMD for this to happen.

So at least for me and today, nvidia is the only way to go. As
already told in an earlier post: "average" Linux users that can life
with lower 3D performance on single head setups may likely opt
for intel HW in future - they propably did that in the past, anyway,
because why should somebody buy an expensive 3D card when
he intends to only use the OSS 2D driver?

Originally Posted by MamiyaOtaru
Thanks for the links, but this discussion is not about patents or copyright
issues that prevents nvidia from open sourcing their drivers: it's about
the PC architecture requireing a kernel module for DMA to work - and
non-GPL kernel modules is what this fuzz is mostly about, right?

Originally Posted by MamiyaOtaru
I'd have to go with a vendor that doesn't have those crippling fears (ie Intel, maybe AMD later).
In this context: although intel published driver sources for the 82Q965,
you can't download the data sheet for the GMA3000 (on-chip graphic).
The 965 data sheet includes a section for the GMA3000, but this only
includes a white paper describing the features for the 82Q965.

Means: bugs in the intel driver can be *much* easier fixed than for
object code drivers, but you hardly can implement new features,
simply because you don't know which register to access in which way.
So "in the spirit of Linux" this is actually not very open! (nevertheless
it's a good move from Intel).

Originally Posted by MamiyaOtaru
Part of that fear seems to have to do with PCIe (which is why I brought it up), at least according to the links provided, which is why I lament the current lack of action with hypertransport GPUs and look forward to possible future such action from AMD. I am of course not an industry insider, so I cannot vouch for the correctness of the information provided, but it certainly would explain a lot.
I briefly studied the HT spec some times ago and i'm not sure how
HT could help avoiding a kernel module, because you still need to
handle DMA and interrupts on kernel level. It might be feasible to
write a generic (GPLed) DMA virt/phys address conversion and
"interrupt redirection" kernel module that would allow to implement
a 3D driver completly in user space - but this would then also
apply for PCIe.

Originally Posted by MamiyaOtaru
I'd love to spend hundreds of dollars to get faster 3d performance, but I'm no longer willing to do so if it comes with a binary blob, and flies in the face of the spirit of the OS I choose to use.
I for one realy like OSS, but at the same time i enjoy using
i.e. UT2004 and Googleearth and have to respect that those are
closed source software packages.

Originally Posted by MamiyaOtaru
If someone else is willing to do so, I postulate that Linux is the wrong choice for that person.
There are thousands of closed source server applications that admins
happily use on top the kernel (oracle, DB2 etc.)

Originally Posted by MamiyaOtaru
Linux can hardly be said to be easier than Windows (though it is for me, even if closed drivers make it worse), and a carefully tended Windows box would be far better for gaming anyway. I just don't get why some feel pragmatism should trump idealism in these debates.
So you advice somebody should go for a 100% closed source windows
solution, just because for the Linux alternative, 1% is closed source?
Sounds like a strange logic to me ...

Originally Posted by MamiyaOtaru
If one pragmatically wants his games to just work, openness be damned, one simply has to use Windows.
ACK - that's what probably 99.9% of all gamers do, anyway.

Originally Posted by MamiyaOtaru
In short: if one must have 3d drivers, and binary drivers were the only option, Linux would have no point
Sorry, but in my ears, this sounds like complete non-sense!



Last edited by JaXXoN; 08-11-06 at 09:31 AM. Reason: fixing typos
JaXXoN is offline   Reply With Quote