First we have
Presumably this refers to Nvidia's binary driver. Then we haveQuote:
A rumor appeared from the heart of Beijing that due to the performance of its GPU architecture and its Linux drivers, NVIDIA was approached by one of the leading Chinese CPU teams to use an NV GPU in a pilot school PC project.
Again, presumably talking about recompiling Nvidia's proprietary code. So we have the situation where the customer wants code compiled for their new architecture, but is quite happy if that code is proprietary.Quote:
To cut the story short, the NV team appeared there, and in very arrogant manner told the Chinese side that they are a large US corporation, and that recompiling the Linux drivers would cost the Chinese a lot of money.
Then we have
My interpretation of this is that AMD is offering source code access to fglrx. The reason I think this is that a) the Chinese didn't seem to care if the code they got was proprietary, b) "mediocre Linux drivers" almost certainly refers to fglrx, and c) "fuss over source code access" would be redundant if they were talking about the open Radeon driver.Quote:
With NVIDIA back in Santa Clara, California and Southern China, there was no doubt as to who the Chinese would call next. The other GPU vendor, while having mediocre Linux drivers, at least did not make any fuss over source code access.
So, yes, perhaps they chose AMD because of open drivers, rather than AMD's willingness to give them access to their closed code. The article isn't clear; maybe the author doesn't actually know either. My reading of it, however, is that the Chinese have gained access to fglrx.
I'm happy to hear arguments to the contrary.
Lots of hate and fanboyism here to contend with.
Basically we have:
1) A source article that is mostly garbage and hearsay and says nothing of confidence.
2) Michael coming along and reinterpreting this article to draw his own wild-ass conclusions. (Though, in fairness, he claims another "source". But who's to know how reliable that source is?)
3) All the fanboys lining up in this thread to congratulate themselves for yet another proof that 100% FOSS is full of win. (Missing the point that, if the story is to be believed, NVIDIA was actually the first to be contacted, despite the fact that it's well known that they don't work with open drivers.)
Is it wrong to wait until there are some actual facts before drawing any conclusions? And, even if true, this "loss" to NVIDIA might not be such a loss... if they place super high value on keeping their tech close to their vests.
2. They alredy use FOSS radeon driver on boards with Radeon IGP.
3. I think port fglrx to MIPS is much harder than use existence FOSS driver and also pointless for weak IGP GPU.
"fuss over source code access" is not redundant if they were talking about the lack of it for the nVidia driver.
Hopefully this will now light the fire under nVidia to at least help the nouveau team out in the same fashion as AMD did for the radeon driver. SO here's hope AMD will have the incentive to improve their blob as well as getting the FOSS radeon driver into tiptop shape
Good for the Chinese to be the catalyst (no pun intended)
Now if nvidia can port their blob to ARM then that'll be another market for them...so they should wake up now
No it's not. Fglrx is easily one of the worst possible drivers you could use on Linux.
It may look half-way decent when you are running full screen benchmarks, but in terms of actual real world use it's compatibility with applications and desktops is miserable and it's crash prone. Not to also mention that it's miserable to install.