Carmack Thinks Nvidia’s Patent Lawsuit Could Spell Trouble, Calls it “Disturbing”

Khalid Moammer
Posted 2 years ago

The 44 year old CTO of Oculus described Nvidia’s on-going patent infringement lawsuit against Samsung and Qualcomm as “disturbing” in a recent tweet.
For those of you who don’t recognize this name. John Carmack is the acting Chief Technology Officer at Oculus. He’s the co-founder of id Software. A studio that propelled PC gaming and defined the first person shooter genre with hallmark titles such as Doom, Quake and Wolfenstein 3D the first ever first person shooter game.

Carmack is a huge Virtual Reality advocate and an outspoken proponent of Nvidia technologies. He starred Nvidia’s G-Sync launch event and took the stage with Nvidia’s CEO Jen-Hsun Huang to promote the technology when it debuted. Carmack publicly acknowledges his close relationship with Nvidia and has rarely worked to hide his fondness for the company. So it came as a surprise to some when he took his criticisms of the company’s on-going patent litigation to Twitter. Going as far as to say that he’s disturbed by Nvidia’s attempts to rally public opinion behind its legal action against Samsung and Qualcomm over patent Infringement claims.

Carmack’s Not A Big Fan Of Nvidia’s Patent Infringement Lawsuit

Carmack’s “disturbance” isn’t difficult to justify either once the claims which form the basis of this legal battle are carefully examined.

In the company’s official statement regarding the patents in question which the company alleges have been infringed upon by Samsung and Qualcomm,  Nvidia states the following :

Seven infringed patents

Our 7,000 issued and pending patents include inventions vital to modern computing. We have chosen seven of those patents to assert in these cases.


Those patents include our foundational invention, the GPU, which puts onto a single chip all the functions necessary to process graphics and light up screens; our invention of programmable shading, which allows non-experts to program sophisticated graphics; our invention of unified shaders, which allow every processing unit in the GPU to be used for different purposes; and our invention of multithreaded parallel processing in GPUs, which enables processing to occur concurrently on separate threads while accessing the same memory and other resources.

Let’s briefly look at the first patent that Nvidia claims has been infringed upon, the GPU, which is short for graphics processing unit. That is any chip that can process graphics and display images on the screen according to Nvidia’s description. What the astute among you will notice here is that this description fits every possible graphics chip sold today in any device whether be it a phone a tablet a game console or a PC. If Samsung or Qualcomm are actively infringing on this patent then so is every single other graphics chip maker. Including ARM, AMD and Intel.

NVIDIA GeForce GTX 1060 Official Specifications and Benchmarks Leaked - 15% Faster Than RX 480, More Than 40% Efficient Than Polaris 10

And what can be characterized as “disturbing” in this is that GPUs as described by Nvidia above have actually existed before the company was actually founded in the early nineties. You will notice similar issues with all the other mentioned technologies. Programmable shading, unified shaders and multithreaded parallel processing as described by Nvidia are all recognized industry wide definitions for established graphics standards, not proprietary technologies.  Now the method by which Nvidia achieves this functionality is proprietary and that’s what’s actually patentable but not the functionality itself.

Unified shaders for example is a hardware feature which allows a shader unit i.e. a GPU core to process all types of shaders e.g. pixel shaders, vertex shaders etc.  And it was ATi, now AMD’s graphics division, that actually brought this technology to market with the Xenos XBOX 360 GPU. ATi was also the first to be granted a unified shaders patent in 2005, a year prior to Nvidia’s patent referenced in the official complaint filing. But again it’s the method by which this functionality is achieved that’s proprietary not the concept itself. Interestingly enough Nvidia was actually the one to show reluctance to moving forward from fixed functions to unified shaders as their rival did. There was a big debate at the time about which approach made the most sense and unified shaders ended up the winner.

After careful examination of some of the “disturbing” facts one would arrive at a realization and a question that can only be answered by reading between the lines. What exactly is going on here ?
Carmack seems to share the same general view that’s being expressed across industry. Nvidia has been struggling for some time to establish a strong position in the hyper-competitive mobile market with its Tegra line of chips. A market that Samsung and Qualcomm are leading. This legal battle is viewed in the industry as a vehicle for Nvidia to undermine its aforementioned mobile competitors and establish a foothold in the mobile market. Something that Carmack was quick to point out.

Apple had created a precedent in patent legal battles that unfortunately is not likely to go away any time soon. This is evident by this very Nvidia suit as it’s being pursued in the same US district court of the state of Delaware. When all is said and done it’s rather frustrating to see companies pouring money into litigation inside a flawed justice system.

Share on Facebook Share on Twitter Share on Reddit