• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

Samsung did not infringe Nvidia graphics chip patents: U.S. ITC

Status
Not open for further replies.
C

csbin

Senior member
http://finance.yahoo.com/news/samsung-does-not-infringe-nvidia-184357393.html


NEW YORK (Reuters) - Samsung Electronics Co Ltd has been cleared on its use of graphics chip technology owned by Nvidia Corp without permission, in a U.S. International Trade Commission ruling on Friday.

Judge Thomas Pender said Samsung did not infringe two Nvidia patents, and while it did infringe a third, he ruled that patent is invalid because it was not a new invention compared with previously known patents.

Nvidia spokesman Robert Sherbin said the ruling will be reviewed by the full commission, which will make a final decision on the dispute in February. "We remain confident in our case," he said.

A spokeswoman for Samsung declined to comment.
 
Thank goodness! It's always good to see a patent suit actually have the correct solution. Shame on Nvidia for trying claim that they invented GPUs.
 
Keysplayr said:
What was the first GPU?
Click to expand...

Depends on what you mean by first "GPU" ...

Nvidia considered the GeForce 256 as the first GPU but that would not fit with their modern definition of a GPU. Today a GPU consists of clusters of programmable processors known as shaders with specialized fixed function units ...

The GeForce 256 technically didn't have "shaders", they had hardware accelerated T&L units to do vertex transformations which were precursors to vertex shaders so technically it's not a GPU in the modern sense ...

The "first" GPU could arguably be the SGI's RCP which was first used to power the graphics of the N64 since it had all of the main features of a GPU. It featured a rasterizer, clipping unit, texture samplers, and a blending unit much like a GPU but it also featured a programmable processor too to boot so it can emulate T&L units or execute a vertex program and it released before the GeForce 256 ...

TLDR: SGI beat Nvidia to the idea of a "GPU" ...
 
ThatBuzzkiller said:
Depends on what you mean by first "GPU" ...

Nvidia considered the GeForce 256 as the first GPU but that would not fit with their modern definition of a GPU. Today a GPU consists of clusters of programmable processors known as shaders with specialized fixed function units ...

The GeForce 256 technically didn't have "shaders", they had hardware accelerated T&L units to do vertex transformations which were precursors to vertex shaders so technically it's not a GPU in the modern sense ...

The "first" GPU could arguably be the SGI's RCP which was first used to power the graphics of the N64 since it had all of the main features of a GPU. It featured a rasterizer, clipping unit, texture samplers, and a blending unit much like a GPU but it also featured a programmable processor too to boot so it can emulate T&L units or execute a vertex program and it released before the GeForce 256 ...

TLDR: SGI beat Nvidia to the idea of a "GPU" ...
Click to expand...

This was many many years ago. Why would it have to be in compliance using a modern sense, or a modern definition?
 
Keysplayr said:
This was many many years ago. Why would it have to be in compliance using a modern sense, or a modern definition?
Click to expand...

Many of the foundations of a GPU in the past still apply today ...

The only thing that has changed over the years is the processing power to fixed function unit throughput ratio ...

The idea of a GPU changes a bit since mobile GPUs as well as Intel Gen 9 graphics are capable of render target reads so in the future blending units could be eliminated altogether. ARM's Midgard Gen 3 architecture is able to read and write to depth/stencil buffers which further challenges the definition of a GPU too ...
 
First definition of a GPU was the ability of the graphics core to take some of the load off of the CPU. Geometry processing unit, I believe, was the first definition. I'm still not sure what the current definition of a GPU has to do with a time when the first GPU was launched. Whichever one that was. Everything morphs over time from generation to generation. Who was the first to coin the "phrase" or "acronym" GPU? I am not certain.
 
Keysplayr said:
First definition of a GPU was the ability of the graphics core to take some of the load off of the CPU. Geometry processing unit, I believe, was the first definition. I'm still not sure what the current definition of a GPU has to do with a time when the first GPU was launched. Whichever one that was. Everything morphs over time from generation to generation. Who was the first to coin the "phrase" or "acronym" GPU? I am not certain.
Click to expand...

The general definition of a GPU is a processor specialized in graphics, nothing more and nothing less ...

Almost all GPUs do some sort of "geometry processing" with just an edge setup unit and a rasterizer ...

As to how the current definition of a GPU relates to when the first GPU launched, if you change the idea of an invention then it is assumed that the new invention redefines that idea hence the old invention not been valid of the description. Enough focusing on that ...

The first one to coin the term "GPU" was Nvidia ...
 
Keysplayr said:
First definition of a GPU was the ability of the graphics core to take some of the load off of the CPU. Geometry processing unit, I believe, was the first definition. I'm still not sure what the current definition of a GPU has to do with a time when the first GPU was launched. Whichever one that was. Everything morphs over time from generation to generation. Who was the first to coin the "phrase" or "acronym" GPU? I am not certain.
Click to expand...

OK then...
http://forums.anandtech.com/showpost.php?p=35578847&postcount=157
 
Any word on Samsung's counter suit? Hopefully that will fall apart too, and everyone can walk away from this mess unscathed.
 
Can someone tell me how the definition of the term "GPU" has anything to do with this case? The exact wording of the patents is what matters, not whatever marketing term or fancy TLA was given to something. I see no insightful analysis of the patents in this thread. Even if Nvidia "invented GPUs," using whatever definition you want, if the patents relevant to this aren't being infringed and/or are invalid due to prior art, then I guess they still have the warm, fuzzy feeling of popularizing the term "GPU" and nobody can take that away from them.
 
Nvidia invented the name "GPU" but graphics processors have been around for a very long time, whether they fit the modern definition with shaders, TMUs, or not.

You should be able to patent a method or process, not a function. Also you run into issues of how a certain task is to be performed, and it's to be expected that independent parties will come to the same conclusion.
 
Last edited:
ThatBuzzkiller said:
Depends on what you mean by first "GPU" ...

Nvidia considered the GeForce 256 as the first GPU but that would not fit with their modern definition of a GPU. Today a GPU consists of clusters of programmable processors known as shaders with specialized fixed function units ...

The GeForce 256 technically didn't have "shaders", they had hardware accelerated T&L units to do vertex transformations which were precursors to vertex shaders so technically it's not a GPU in the modern sense ...

The "first" GPU could arguably be the SGI's RCP which was first used to power the graphics of the N64 since it had all of the main features of a GPU. It featured a rasterizer, clipping unit, texture samplers, and a blending unit much like a GPU but it also featured a programmable processor too to boot so it can emulate T&L units or execute a vertex program and it released before the GeForce 256 ...

TLDR: SGI beat Nvidia to the idea of a "GPU" ...
Click to expand...

Also, ATI was first to the market with what would be considered a modern GPU. The GPU in the Xbox 360 was the first available product with unified shaders. Nvidia was still making GPU's with separate pixel and vertex shaders.
 
NVidia's response:

We said we would keep you informed of new developments in our cases with Samsung.

In the latest step of our efforts to protect NVIDIA’s intellectual property, an administrative law judge at the U.S. International Trade Commission ruled today that Samsung and Qualcomm did not violate U.S. law with respect to importing certain Samsung products into the U.S.

Judge Thomas Pender issued an initial determination that Samsung and Qualcomm didn’t infringe two NVIDIA patents, and that both did infringe a third patent but that this patent wasn’t valid.

This initial determination is one more step in a long legal process.

We now intend to ask the full commission (which is made up of six commissioners) to review this initial determination and to confirm the previous judgment of the U.S. Patent Office — that the third patent is valid. If they agree, the ITC would issue an order that would preclude Samsung from importing into the U.S. infringing Samsung mobile devices and smart TVs.

We are continuing this case by proceeding to the next step in the process because we believe our patents are valid and have been infringed.

- See more at: http://blogs.nvidia.com/blog/2015/10/09/patent-infringement/#sthash.YylSoF4w.dpuf
Click to expand...

http://blogs.nvidia.com/blog/2015/10/09/patent-infringement/

So it's not all over yet.
 
Status
Not open for further replies.

TRENDING THREADS

Back
Top