Comments

• NVIDIOT

  2008-03-16 12:03:02

  Reply for sending companies the Open Source NVIDIA pledge letter :
  
  Leadtech :
  
  Thank you for your inquiry
  
  We will try to remind nvidia to concern about providing more open-source for Linux Users at proper occasion.
  
  BFG :
  
  I will forward your complaint to Nvidia.
  
  Thank you, xxxx BFG Support

• CoolGuy

  2008-03-16 12:03:24

  hehe !!

• Roy Schestowitz

  2008-03-16 12:16:40

  There's quite a saga behind this rumour about NVIDIA. For context (and FWIW), here are some of the more recent references/notes:
  
  1. NVIDIA Plotting Open-Source Strategy
  
  

  According to an AIB partner, NVIDIA is planning an open-source counterattack against ATI/AMD.

  
  
  2. OpenTheBlob.com - An Open Letter To NVIDIA
  
  

  We the GNU/Linux community and the undersigned, kindly request that you, NVIDIA Corporation, increase your efforts in better enabling the open-source community to develop free software drivers for your graphics hardware. Your major competitors in this market, AMD/ATI and Intel, have not only supported the community in open-source driver development efforts but they are now openly releasing hardware programming documentation.

  
  
  3. Doing More For An Open NVIDIA
  
  

  Following the open letter to NVIDIA at OpenTheBlob.com that takes aim at NVIDIA's lack of a reliable open-source driver, now out is a letter geared for NVIDIA's board partners (ASUS, Dell, BFG Tech, etc). This happens to be based off of a strategy I discussed before for frustrated ATI customers prior to the new driver code-base. If you're interested in taking a stand for an open NVIDIA, the page with links and a sample letter can be found here.

  
  
  Also of interest (less relevant):
  
  4. PhysX For CUDA, Linux Support A Given?
  
  

  CUDA has been supported under Linux from the start with their graphics driver now bundling the CUDA driver and version 1.1 of the CUDA Toolkit being supported under Fedora 7, RHEL 3/4/5, SLED 10, OpenSuSE 10, and Ubuntu 7.04 (binary downloads).

  
  
  5. Is NVIDIA Buying AGEIA Good For Linux?
  
  

  Though with this afternoon's announcement of NVIDIA acquiring AGEIA, we wonder how this could impact the level of PhysX support for Linux. As NVIDIA actively supports a binary driver for their GeForce and Quadro products, we'd hope that once these GeForce+PhysX graphics cards are introduced we will find support within the NVIDIA Linux driver. We also hope the future bundling of PhysX IP into NVIDIA's GPUs won't hamper their rumored open-source strategy.

  
  
  The vision (and future) of open source graphics drivers/cards everywhere is no longer far-fetched. Deployments on Linux are encouraged owing to high performance.
  
  6. Interview: The driver behind NASA's Mars Rovers
  
  

  It all runs on a collection of high-end Linux boxes -- nice systems, but commodity PC hardware. Since 3-D visualisation is a big part of the job, the Linux boxes sport bonzer NVIDIA graphics cards.

  
  
  7. NVIDIA Workstation Performance: Windows vs. Linux vs. Solaris
  
  

  Ubuntu 8.04 Alpha 5 with the 169.12 driver had overall produced the fastest results within SPECViewPerf. In only three benchmarks had Solaris Express Developer 1/08 outpaced Ubuntu Linux, but with two of these tests the results were almost identical. These three tests were Maya (maya-02), UGS NX (ugnx-01), and UGS Teamcenter Visualization Mockup (tcvis-01). In the rest of the OpenGL tests, Solaris Express and Windows Vista were fighting for second place. In six out of the nine tests, Windows Vista was in last place. For those interested in NVIDIA's Quadro series for CUDA (Compute Unified Device Architecture; NVIDIA's GPGPU implementation), this technology is supported in NVIDIA's binary Linux driver.

  
  
  8. Hollywood, Linux, and CinePaint at FOSDEM 2008
  
  

  At first, studios had no choice but to develop Linux applications in house. The film industry has millions of lines of Linux code, both desktop applications and server pipelines. Studio Linux programmers write code in C++, Python, Perl, and Java. What does all this secret proprietary Linux code do? And, why do they do parallel programming with grid computing? How does the film industry use renderfarms with thousands of CPUs?