NVIDIA Prepares 195.xx Linux Driver, Carries Fermi Support [Archive]

View Full Version : NVIDIA Prepares 195.xx Linux Driver, Carries Fermi Support

phoronix

11-06-2009, 09:20 AM

Phoronix: NVIDIA Prepares 195.xx Linux Driver, Carries Fermi Support

It was just last week that NVIDIA had finally released a stable 190.xx Linux driver after this driver series had been in beta for months. The 190.xx driver series brought new hardware support, OpenGL 3.2 support, VDPAU improvements, and a fair amount of other changes...

http://www.phoronix.com/vr.php?view=NzY4MA

droidhacker

11-06-2009, 09:44 AM

Open source?
No?
Not interested.

Heiko

11-06-2009, 09:54 AM

Fermi shipping in December?
All rumors I hear are pointing to February/March 2010. Perhaps a paper launch in December with very limited numbers only going to developers.

rohcQaH

11-06-2009, 12:47 PM

Now they just need to create a fermi that fits the driver. Easy going.

deanjo

11-06-2009, 12:54 PM

Fermi shipping in December?
All rumors I hear are pointing to February/March 2010. Perhaps a paper launch in December with very limited numbers only going to developers.

Actually that sounds more like the ATI 58xx series.
http://www.digitimes.com/news/a20091105PD213.html

Everybody except Charlie says their sources say Late Nov/Dec launch for Fermi.

Back to the 195.17 drivers that the article is about, they are solid as usual and looking forward to posting up the openCL bench results with PTS when I get around to it.

Ranguvar

11-06-2009, 03:33 PM

SLI, anyone?

Game_boy

11-06-2009, 03:55 PM

Now Nvidia is suggesting Q1 2010. "Ramp" usually comes well before retail release for all CPUs and GPUs I've heard of.

“Next year it is going to be an interesting first quarter because, in fact, we will need more wafers than ever in Q1. The reason for that is because – and I mean more 40nm wafers than ever in Q1 –we are […] fully ramping Fermi for three different product lines: GeForce, Quadro and Tesla,” said Jen-Hsun Huang, chief executive officer of Nvidia, in the conference call with financial analysts.

Nvidia’s first quarter of fiscal year 2011 begins on the 26th of January and ends on the 26th of April, 2010.

[from Xbitlabs]

Charlie, not to be outdone, is suggesting it won't be at retail until May.

Heiko

11-07-2009, 03:25 PM

Actually that sounds more like the ATI 58xx series.
http://www.digitimes.com/news/a20091105PD213.html

Everybody except Charlie says their sources say Late Nov/Dec launch for Fermi.

Back to the 195.17 drivers that the article is about, they are solid as usual and looking forward to posting up the openCL bench results with PTS when I get around to it.

I know plenty of people who own a HD5870, so I think you must be mistaken. On the other side: Charlie definately isn't the only one who suggests Fermi won't be available in huge quantities before Q1 2010 (as pointed out above). By the way, looking at some of your other reactions in these forums: we all know you are in the green camp.

Ex-Cyber

11-07-2009, 04:24 PM

Actually that sounds more like the ATI 58xx series.
http://www.digitimes.com/news/a20091105PD213.htmlI'd expect that 40nm yield problems at TSMC (the reason cited by the article for the 58xx shortage) would affect Fermi worse than HD58xx, since Fermi is a considerably bigger chip.

Game_boy

11-07-2009, 05:03 PM

@Ex-Cyber

I agree - look how much yields suffered with the very small chips GT216 and GT218. They should have been launched around the same time as AMD's first 40nm in March (HD4770, which was much bigger than that anyway). But they were OEM-only for months (i.e. poor yields) and are only just at retail.

HardOCP is also suggesting Q1, independent of Charlie. However Charlie's been (semi-) accurate about a lot of things recently. The fake Fermi card at GTC (which Nvidia then admitted to). Evergreen shader count. The GT218 delays. The EOLing of the GTX200 series (stocks are extremely low). The codename of AMD's next chip (Northern Islands).

Ex-Cyber

11-07-2009, 07:07 PM

Melcar

11-07-2009, 07:12 PM

They are the first that started calling it that. All we had before were VPUs :).

BlackStar

11-08-2009, 04:28 AM

They are the first that started calling it that. All we had before were VPUs :).

Where V = Voodoo? Virtual? Video?

Voodoos, Mystiques, Verites, Rivas, Rages were all called "graphics accelerators" back in the day. It's possible that the "GPU" name didn't come into existence until 1999, but Nvidia certainly didn't invent graphics processors - this claim is completely laughable.

I have a feeling that Fermi won't end well for Nvidia... (I hope to be proven wrong.)

Ex-Cyber

11-08-2009, 05:18 AM

Given the year, I assume what they're referring to is having transformation+lighting+rasterization all happening in a single chip. I'm pretty sure that 3DLabs had already been doing this for several years using multiple chips.

deanjo

11-08-2009, 01:35 PM

My guess would be that any Fermi chips shipping in 2009 will be on compute cards for a handful of deep-pocketed HPC customers. NVidia's marketing for Fermi has been extremely GPGPU-heavy, to the point that the only directly graphics-related thing on their Fermi page (http://www.nvidia.com/object/fermi_architecture.html) is a mention of raytracing as a possible application.

Also, the white paper is hilarious:

The fact is that nvidia did invent the GPU in 1999. The term was never used before and was defined by nvidia as "a single-chip processor with integrated transform, lighting, triangle setup/clipping, and rendering engines that is capable of processing a minimum of 10 million polygons per second."

Ex-Cyber

11-08-2009, 02:02 PM

The term had been used before, but it wasn't mainstream (check Google Scholar, for example). Either way, the common use of the term now is not ruled by the self-serving "definition" crafted by Nvidia's marketing department (philosophical question: if you downclock a GPU so that it can't do 10 million polygons per second, does it stop being a GPU?).

And either way, this doesn't say a lot about Fermi, which doesn't look like it'll be shipping in an actual graphics card for a while unless I've missed some big announcement/leak. Given Jen-Hsun Huang's statement that they're "ramping Fermi for [...] GeForce" I assume we will see them at some point, but so far I've seen almost no talk about its launch or features as a GPU except that it will support DX11. All indications seem to be that the first generation Fermi is going to be a top GPGPU performer but also brutally expensive. I wouldn't be surprised if they end up going exclusively for the super high-end HPC and gaming markets and ceding the mainstream gaming market to ATI for a little while.

Game_boy

11-09-2009, 04:21 PM

Fermi is going to be a top GPGPU performer but also brutally expensive.

AMD's Hemlock cards will have perform over 5 TFlops. Fermi is projected to do 1.5 given its shader count (512) and architecture, with a similar die size. FLOPS isn't everything but as a measure of raw compute power, Fermi will be behind.

With its dedicated caches and simpler shader hierarchy (AMD has 5 shaders to a cluster) it will perform better in the real world, but anything you could do with Fermi should run faster on AMD if you optimise it. And the price you pay for the two things I mentioned is much greater die size, since neither feature helps in games but they takes up area.

pavlinux

11-09-2009, 06:45 PM