Slashdot Mirror
Valve & Intel Collaborating On Open-Source Drivers
An anonymous reader writes "It looks like Valve's Linux team that's still growing has found much interest in open-source graphics drivers. Intel Linux graphics driver developers and Valve's Linux team were meeting for the past week to look at each other's code, work out performance goals, and collaborate on new features. Ian Romanick of Intel blogs, 'The funny thing is Valve guys say the same thing about drivers. There were a couple times where we felt like they were trying to convince us that open source drivers are a good idea. We had to remind them that they were preaching to the choir. :) Their problem with closed drivers (on all platforms) is that it's such a blackbox that they have to play guess-and-check games. There's no way for them to know how changing a particular setting will affect the performance. If performance gets worse, they have no way to know why. If they can see where time is going in the driver, they can make much more educated guesses.' Perhaps the companies are paying attention to Linus Torvalds' memo to NVIDIA?"
Above link is gay porn
Parent is spam. Link is to porn.
If Intel gives a shit about open source graphics drivers, where are the open drivers for their Atom IGP?
Licensed from PowerVR so not their IP, but next year it looks like they'll replace it with their in-house Ivy Bridge graphics in the "Valley View" Atoms. But if you got an Atom today and want good open source support, you're shit out of luck.
Live today, because you never know what tomorrow brings
There's more ways in which a driver can be buggy than poor framerate - such as graphical corruption, buggy shader compilers that crash, excessive CPU usage, ...
I don't know exactly what they were thinking; but Intel's licensing of PowerVR GPUs somehow seemed to exclude video drivers that don't suck in any form, much less fully open. One would have thought that chipzilla could have gotten better terms, especially if they were planning a part for the embedded market...
The 'GMA500' and 'GMA600'(SGX 535, at different clock speeds) and 'GMA3600' and 'GMA3650'(SGX545, also differing in clock speed) all have tottering heaps of crap for drivers. Even if you don't care about license, they aren't exactly catching Nvidia in the 'actually works while tainting your kernel' department.
The rest of the GMAs are pretty unexciting; but are in-house designs and don't seem to have the same epic driver woes.
You don't need to download them from anywhere. They are in the mainline kernel.
3DLABS not Intel developed the Permedia graphics cards.
Don't forget the PowerVR GPUs they used in several chipsets which last I checked are black box as well.
Lets face it folks, if you care about FOSS drivers you really only have one choice when it comes to GPUs, and that is AMD. they are opening up all their code as fast as their lawyers can sign off, they have gone coreboot over UEFI, it seems since taking over ATI that AMD has gone above and beyond trying to be FOSS friendly, even hiring devs to help the FOSS devs get up to speed quicker on the drivers.
So if the FOSS community wants FOSS drivers they need to put their money where their mouths are and buy AMD across the board. only by showing that supporting FOSS increases sales will you get other corps to sign off on opening up their drivers. The fact that damned near every forum talking about FOSS and GPUs ends up with a bazillion "LOL buy Nvidia" tells me that frankly you might as well accept binary blobs and a hardware API, because obviously the community doesn't care about FOSS over convenience.
ACs don't waste your time replying, your posts are never seen by me.