DailyTech - Silicon Graphics Sues ATI

quick

By Schadenfroh on 10/25/2006 12:01:40 PM , Rating: 3

I see that SGI put their bankruptcy lawyers on the offensive now that they have emerged. I am actually surprised that no one has bought SGI out.

RE: quick

By peternelson on 10/25/2006 1:02:23 PM , Rating: 2

If AMD bought SGI that would rock!

Maybe that's what SGI are trying to achieve here by suing AMD/ATI. It could be cheaper just to buy them and SGI was and is a great company. A bargain at current stock valuations!

Parent

RE: quick

By Pandamonium on 10/25/2006 1:39:58 PM , Rating: 2

It's patents like these that make a mockery of our legal system. You shouldn't be able to patent a theoretical method. You can patent a physical device that executes said theoretical method, but you shouldn't be able to patent an algorithm.

Parent

RE: quick

By Murst on 10/25/2006 2:04:52 PM , Rating: 2

That is absolutely one of the stupidest things I've heard.

So if I came up with a method to cure AIDS or cancer, I couldn't patent it? The only thing I could patent is the machine that mixes the fluids to make it?

The IDEA is the important part. Implementing it is, in most cases, trivial.

Parent

RE: quick

By FITCamaro on 10/25/2006 2:25:35 PM , Rating: 3

You should be able to patent your method. But you shouldn't be able to patent the idea of curing AIDS. They didn't patent their method. They patented the idea of using floating point values to process graphics data and store it in a frame buffer. If they patented their method, fine. I doubt ATI is using the same exact algorithm that SGI did to process data in their driver. Sure, some code is similar or perhaps even the same, but that doesn't mean they copied SGI's implementation.

quote:
A floating point rasterization and frame buffer in a computer system graphics program. The rasterization, fog, lighting, texturing, blending, and antialiasing processes operate on floating point values....The final floating point values corresponding to pixel attributes are stored in a frame buffer and eventually read and drawn for display. The graphics program can operate directly on the data in the frame buffer without losing any of the desired range and precision of the data.

The .... is them just explaining an example case. That patent is similar to Microsoft patenting the idea of an operating system. Or Intel patenting the idea of a CPU. With that patent, they can effectively stop anyone else from creating and selling a graphics card that processes 3D graphics regardless of how it works. Unless you can find another way to represent graphics data as something other than a floating point number and store the data in something other than a buffer.

Parent

RE: quick

By Murst on 10/25/2006 3:01:41 PM , Rating: 2

This patent is the the process of getting data from one "section" of the graphics processor to another. This is in no way similar to SGI patenting graphics cards, MS patentint OS, or Intel patenting CPUs.

It solves a problem which, prior to SGI's solution, did not have a solution.

It may seem obvious to you now, but that's generally the way solutions are. Just because something is obvious AFTER someone comes up with a solution should not, in any way, strip the inventor of their due credit for solving the problem.

Parent

RE: quick

By Spivonious on 10/25/2006 3:06:06 PM , Rating: 2

I have a hard time believing that using floating point numbers for graphics data is really "patentable". There has to be something else.

Parent

RE: quick

By Murst on 10/25/2006 3:09:44 PM , Rating: 2

Its not. Half of the people on these forums are just too stupid to realize that.

Parent

RE: quick

By FITCamaro on 10/25/2006 5:55:21 PM , Rating: 2

Where do you see anything in that patent description about moving data from one place to another?

If it had some kind of a procedure for how the final data eventually ended up in the frame buffer, I'd agree with you. All it says though is that the final data is stored in the frame buffer to be eventually drawn out to the screen.

Maybe in the full patent document theres more but from that description, theres nothing about how data is moved through the GPU.

Parent

RE: quick

By FITCamaro on 10/25/2006 6:00:28 PM , Rating: 2

And based on that patent description, to me the only thing they could patent is storing the data in the frame buffer before its written to the screen. And I really don't think SGI thought that up considering there were displays and computer graphics before SGI existed.

Parent

RE: quick

By patentman on 10/25/2006 8:23:38 PM , Rating: 2

"That patent is similar to Microsoft patenting the idea of an operating system. Or Intel patenting the idea of a CPU."

Good god, IDEAS ARE NOT PATENTABLE.

Seriously people, you all need to know what you are talking about BEFORE you start arguing about it. IN this case, it might be a wee bit helpful if you knew something about patent law. For instance, it might be helpful for youto know that the scope of a patent is defined by the CLAIMS. While the claims are interpreted in light of the specification, limitations from the specification are not (generally) brought into the claims (although since the phillips case courts are doing this more often).

Parent

RE: quick

By FITCamaro on 10/25/2006 9:35:18 PM , Rating: 2

Thats my point. Ideas aren't patentable but thats what that patent is. An idea.

Parent

RE: quick

By patentman on 10/25/2006 8:19:56 PM , Rating: 2

READ 35 U.S.C. 101. Ideas ARE NOT patentable.

35 U.S.C. 101 Inventions patentable. Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.

Parent

RE: quick

By Saist on 10/26/2006 7:33:17 AM , Rating: 3

I'm not so sure that SGI is and was a great company.

What you need to remember is that many of the better hardware designers at SGI had already split off several years ago into a Company called ArtX. Now, you might remember ArtX because they created an Electrically Efficient Graphics processor named Flipper. You might remember flipper as being the graphics processor behind the #2 console in world wide sales, the Nintendo Gamecube.

You might also remember then that before the Nintendo Gamecube launched, ATi bought ArtX up. This was news back in 2000 because it meant that many of the developers behind the OpenGL specification were now on ATi's payroll. ATi's turn around from Market-Dog with the 8500 to Market Leader with R300 and Catalyst is widely attributed to ArtX, aka SGI.

There have been several articles over the years between then and now detailing the Impact that ArtX had on ATi, and it widely reguarded that if ATi had not merged with ArtX, ATi would be in the position of a certain company called 3dfx... (which was bought up by Nvidia).

There have also been several editorials between the creation of ArtX, the sale to ATi, and now, about how the split from SGI resulted in SGI's descent into bankruptcy. It is not exactly an industry secret that the movers and shakers in ArtX, were the movers and shakers in SGI.

I think what we are going to find in this lawsuit is that the person who came up with this for SGI is currently on ATi's payroll, and probably is a high ranking withing ATi right now.

Quite frankly, if I was ATi, I wouldn't want to buy out the dirt when the diamonds were already mined... over 6 years ago.

Parent

Figures

By othercents on 10/25/2006 12:00:22 PM , Rating: 5

Wasn't an issue until after the merger was finished. I guess SGI things they can get more money out of AMD/ATI than if they had sued ATI by itself.

Other

RE: Figures

By Behlal on 10/25/2006 12:10:37 PM , Rating: 3

Interesting thought, but I imagine the wait has more to do with the fact that they only just came out bankruptcy protection and re-floated their stock on a new stock exchange. This move is to show to investors that there is something worth investing in.

Parent

RE: Figures

By othercents on 10/25/2006 12:12:07 PM , Rating: 2

Well their press release was on the 23rd, so they sued before the merger was over. Also the patent is about using floating point for graphics processing application.

Other

Parent

RE: Figures

By deeznuts on 10/25/2006 1:01:13 PM , Rating: 2

If they had sued them months ago they wouldn't have an award anyway for years, so I don't see how waiting to file right before or during a closure of a merger announced months ago makes a difference.

Parent

RE: Figures

By marvdmartian on 10/25/2006 3:07:56 PM , Rating: 1

What AMD should do is look at SGI, and say, "I'm sorry.....ATI no longer exists. If you'd like to talk to our lawyers, we'd be happy to entertain whatever argument you have." LOL

THEN AMD should buy out SGI!!

Parent

RE: Figures

By patentman on 10/25/2006 8:17:08 PM , Rating: 4

what you stated does not comport with the law of corporations. In mergers, liabilities of the absorbed company are generally imputed to the company resulting after the merger.

Parent

Good

By Crusader on 10/25/06, Rating: 0

RE: Good

By FITCamaro on 10/25/2006 1:48:38 PM , Rating: 3

The point people are trying to make is that it's an invalid patent. The patent in question is similar to patenting to logic behind the if statement. You shouldn't be able to patent using floating point values in a calculation for processing graphics data. Of course you have to do that. It's possible ATI didn't even know the patent existed.

If you were writing software that involved reading in a file, would you check to see if theres a patent on the idea of reading data into a buffer?

If SGI had this patent, they should have disclosed it or brought it up with ATI long before this. To me this is similar to the Blackberry patent lawsuit.

Parent

RE: Good

By DigitalFreak on 10/25/2006 1:56:40 PM , Rating: 2

Just because a company licensed a patented technology doesn't mean the patent is valid. At the time, it may have just been cheaper for them to pay the fee than go through a long and protracted court battle. Hopefully AMD will stand it's ground an crush SGI on this.

Maybe SGI is turning into another SCO...

Parent

RE: Good

By ToeCutter on 10/25/2006 2:19:06 PM , Rating: 2

quote:
Maybe SGI is turning into another SCO...

Too funny!

SGI put up a HUGE stink when SCO terminated SGI's System V license for IRIX a few years ago!

Parent

RE: Good

By s12033722 on 10/25/2006 3:15:19 PM , Rating: 2

No. Nvidia bought 3DFX. SGI is not 3DFX and continued to exist well after 3DFX went down.

Parent

RE: Good

By ToeCutter on 10/27/2006 10:20:41 PM , Rating: 2

quote:
No. Nvidia bought 3DFX. SGI is not 3DFX and continued to exist well after 3DFX went down.

Allow me to clarify:

-3Dfx was formed by former SGI employees. No implication that SGI became 3Dfx.

-SGI acquired Intergraph's Win32 workstation line and re-badged them as the SGI Visual line.

-The SGI Visual line used graphics subsystems by SGI (Cobalt), Number9, and even Intel(!) and ATI(!!!).

My confusion surrounding Intergraph stemmed from Intergraph offering the Voodoo-based Intense3D right about the same time SGI acquired the Intergraph workstation line.

Parent

SGI?

By ixelion on 10/25/2006 12:40:51 PM , Rating: 2

Isn't SGI that company that tried to make dual GPU cards a year or two back but was not successful?

RE: SGI?

By s12033722 on 10/25/2006 12:45:38 PM , Rating: 3

Silicon Graphics is one of the true pioneers of the graphics industry. They were doing high end graphics way before Nvidia and ATI. Do you remember the film Lawnmower Man? The graphics in that film were done on SGI workstations and were groundbreaking for the time.

Parent

RE: SGI?

By ToeCutter on 10/25/2006 1:44:16 PM , Rating: 2

Are you referring to the 3Dfx-based Quantum3D hardware? I recall (vaguely) that SGI incorporated graphics from Quantum or Intergraph years ago. I'm certain that 3Dfx were formed by ex-SGI folks, though.

SGI is mere shadow of what they once were. I work in automotive engineering and the current crop of RISC-based workstations offered by HP, IBM, even Itanium2 blow away Tezro in just about every application.

The days of monolithic rendering are over. WETA used a 1000-node render farm running LINUX to render the stuff for LOTR. They did it in a fraction of the time it would have taken traditional (Read: proprietary) rendering nodes, at a fraction of the cost.

(But they store the rendered output on NetApp filers ;-)

Parent

RE: SGI?

By Spivonious on 10/25/2006 3:08:43 PM , Rating: 3

But did they transfer floating point values into a frame buffer?????

...one patent to sue them all.

Parent

here are the claims of the patent at issue

By patentman on 10/25/2006 8:14:56 PM , Rating: 2

See the topic. Note that the thing claimed is a computer system and a method, NOT an algorhythm.

What is claimed is:

1. A computer system, comprising: a processor for performing geometric calculations on a plurality of vertices of a primitive; a rasterization circuit coupled to the processor that rasterizes the primitive according to a rasterization process which operates on a floating point format; a frame buffer coupled to the rasterization circuit for storing a plurality of color values; and a display screen coupled to the frame buffer for displaying an image according to the color values stored in the frame buffer; wherein the rasterization circuit performs scan conversion on vertices having floating point color values.

2. A computer system, comprising: a processor for performing geometric calculations on a plurality of vertices of a primitive; a rasterization circuit coupled to the processor that rasterizes the primitive according to a rasterization process which operates on a floating point format; a frame buffer coupled to the rasterization circuit for storing a plurality of color values; a display screen coupled to the frame buffer for displaying an image according to the color values stored in the frame buffer; a texture circuit coupled to the rasterization circuit that applies a texture to the primitive, wherein the texture is specified by floating point values; and a texture memory coupled to the texture circuit that stores a plurality of textures in floating point values.

3. A computer system, comprising: a processor for performing geometric calculations on a plurality of vertices of a primitive; a rasterization circuit coupled to the processor that rasterizes the primitive according to a rasterization process which operates on a floating point format; a frame buffer coupled to the rasterization circuit for storing a plurality of color values; and a display screen coupled to the frame buffer for displaying an image according to the color values stored in the frame buffer; wherein the floating point format is comprised of sixteen bits in a s10e5 format.

4. A computer system, comprising: a processor for performing geometric calculations on a plurality of vertices of a primitive; a rasterization circuit coupled to the processor that rasterizes the primitive according to a rasterization process which operates on a floating point format; a frame buffer coupled to the rasterization circuit for storing a plurality of color values; a display screen coupled to the frame buffer for displaying an image according to the color values stored in the frame buffer; and a fog circuit coupled to the rasterization circuit for performing a fog function, wherein the fog function operates on floating point color values.

5. A computer system, comprising: a processor for performing geometric calculations on a plurality of vertices of a primitive; a rasterization circuit coupled to the processor that rasterizes the primitive according to a rasterization process which operates on a floating point format; a frame buffer coupled to the rasterization circuit for storing a plurality of color values; a display screen coupled to the frame buffer for displaying an image according to the color values stored in the frame buffer; and a blender coupled to the rasterization circuit which blends floating point color values.

6. A computer system, comprising: a processor for performing geometric calculations on a plurality of vertices of a primitive; a rasterization circuit coupled to the processor that rasterizes the primitive according to a rasterization process which operates on a floating point format; a frame buffer coupled to the rasterization circuit for storing a plurality of color values; a display screen coupled to the frame buffer for displaying an image according to the color values stored in the frame buffer; and logic coupled to the rasterization circuit which performs per-fragment operations on floating point color values.

7. A computer system, comprising: a processor for performing geometric calculations on a plurality of vertices of a primitive; a rasterization circuit coupled to the processor that rasterizes the primitive according to a rasterization process which operates on a floating point format; a frame buffer coupled to the rasterization circuit for storing a plurality of color values; and a display screen coupled to the frame buffer for displaying an image according to the color values stored in the frame buffer; wherein the processor, the rasterization circuit, and the frame buffer are on a single semiconductor chip.

8. The computer system of claim 7, wherein the processor, the rasterization circuit, and the frame buffer reside on a same substrate of the single semiconductor chip.

9. In a computer system, a method for rendering a three-dimensional image for display, comprising the steps of: performing geometric calculations on a plurality of vertices of a plurality of polygons; scan converting a plurality of pixels according to the vertices, wherein scan conversion is performed on floating point color values; applying a texture to the image by reading floating point texture values stored in a texture memory; simulating fog effects, wherein fog is simulated by modifying floating point color values; drawing the image for display on a display screen coupled to the computer system.

10. The method of claim 9, wherein the floating point values are comprised of sixteen bits.

11. The method of claim 10, wherein the floating point values are specified by a s10e5 format.

12. The method of claim 10 further comprising the step of storing the floating point color values in a frame buffer.

13. The method of claim 10 further comprising the step of blending at least two floating point color values.

14. The method of claim 10 further comprising the step of performing antialiasing on floating point color values.

15. The method of claim 10 further comprising the steps of: reading data from the frame buffer; modifying the data; writing modified data back to the frame buffer.

16. The method of claim 10 further comprising the step of modifying color values for lighting, wherein lighting calculations operate on floating point color values.

17. In a computer system, a method for operating on data stored in a frame buffer, comprised of: storing the data in the frame buffer in a floating point format; reading the data from the frame buffer in the floating point format; operating directly on the data in the floating point format; and writing the data to the frame buffer in the floating point format; wherein the steps of writing, storing, and reading the data in the frame buffer in the floating point format are further comprised of a specification of the floating point format, wherein the specification corresponds to a level of range and precision.

18. The method of claim 17 wherein the specification is comprised of 16 bits of data and the data are comprised of one sign bit, ten mantissa bits, and five exponent bits.

19. The method of claim 17 wherein the specification is comprised of 17 bits of data and the data are comprised of one sign bit, 11 mantissa bits, and five exponent bits.

20. The method of claim 17 wherein the specification is comprised of 16 bits of data and the data are comprised of ten mantissa bits, and six exponent bits.

21. The method of claim 17 wherein the specification is comprised of 32 bits of data and the data are comprised of one sign bit, 23 mantissa bits, and eight exponent bits.

22. A computer system having a floating point frame buffer for storing a plurality of floating point color values; wherein the floating point color values are written to, read from, and stored in the frame buffer using a specification of the floating point color values that corresponds to a level of range and precision.

23. The computer system of claim 22, wherein the floating point color values are comprised of 16 bits of data and the data are comprised of one sign bit, ten mantissa bits, and five exponent bits.

24. The computer system of claim 22, wherein the floating point color values are comprised of 17 bits of data and the data are comprised of one sign bit, 11 mantissa bits, and five exponent bits.

25. A computer system, comprising: a processor for performing geometric calculations on a plurality of vertices of a primitive; a rasterization circuit coupled to the processor that rasterizes the primitive according to a rasterization process which operates on an s10e5 floating point format; a frame buffer coupled to the rasterization circuit for storing a plurality of s10e5 floating point color values; a display screen coupled to the frame buffer for displaying an image according to the s10e5 color values stored in the frame buffer.

26. The computer system of claim 25 further comprising: a texture circuit coupled to the rasterization circuit that applies a texture to the primitive, wherein the texture is specified by s10e5 floating point values.

27. The computer system of claim 25 further comprising a lighting circuit coupled to the rasterization circuit for performing a lighting function, wherein the lighting function executes on s10e5 floating point color values.

28. The computer system of claim 25 further comprising a fog circuit coupled to the rasterization circuit for performing a fog function, wherein the fog function operates on s10e5 floating point color values.

29. The computer system of claim 25 further comprising an antialiasing circuit coupled to the rasterization circuit which performs an antialiasing algorithm on s10e5 floating point color values.

30. The computer system of claim 25 further comprising a blender coupled to the rasterization circuit which blends s10e5 floating point color values.

31. The computer system of claim 25 further comprising logic coupled to the rasterization circuit which performs per-fragment operations on s10e5 floating point color values.

RE: here are the claims of the patent at issue

By raven3x7 on 10/25/2006 11:20:45 PM , Rating: 2

So they patented a Computer System with a 3d Graphics Cards. How nice of them... I'm wondering if they still get licence fees from nvidia for their linux driver...

Parent

Floating-point framebuffers and object data

By PrinceGaz on 10/25/2006 12:16:10 PM , Rating: 3

http://www.freepatentsonline.com/6650327.html

Essentially the patent is to do with using floating-point values throughout the whole of the rendering process; from geometry data through rasterization and the values stored in the frame-buffer.

here are the claims of the patent at issue

By patentman on 10/25/2006 8:14:57 PM , Rating: 2

See the topic. Note that the thing claimed is a computer system and a method, NOT an algorhythm.

What is claimed is:

1. A computer system, comprising: a processor for performing geometric calculations on a plurality of vertices of a primitive; a rasterization circuit coupled to the processor that rasterizes the primitive according to a rasterization process which operates on a floating point format; a frame buffer coupled to the rasterization circuit for storing a plurality of color values; and a display screen coupled to the frame buffer for displaying an image according to the color values stored in the frame buffer; wherein the rasterization circuit performs scan conversion on vertices having floating point color values.

2. A computer system, comprising: a processor for performing geometric calculations on a plurality of vertices of a primitive; a rasterization circuit coupled to the processor that rasterizes the primitive according to a rasterization process which operates on a floating point format; a frame buffer coupled to the rasterization circuit for storing a plurality of color values; a display screen coupled to the frame buffer for displaying an image according to the color values stored in the frame buffer; a texture circuit coupled to the rasterization circuit that applies a texture to the primitive, wherein the texture is specified by floating point values; and a texture memory coupled to the texture circuit that stores a plurality of textures in floating point values.

3. A computer system, comprising: a processor for performing geometric calculations on a plurality of vertices of a primitive; a rasterization circuit coupled to the processor that rasterizes the primitive according to a rasterization process which operates on a floating point format; a frame buffer coupled to the rasterization circuit for storing a plurality of color values; and a display screen coupled to the frame buffer for displaying an image according to the color values stored in the frame buffer; wherein the floating point format is comprised of sixteen bits in a s10e5 format.

4. A computer system, comprising: a processor for performing geometric calculations on a plurality of vertices of a primitive; a rasterization circuit coupled to the processor that rasterizes the primitive according to a rasterization process which operates on a floating point format; a frame buffer coupled to the rasterization circuit for storing a plurality of color values; a display screen coupled to the frame buffer for displaying an image according to the color values stored in the frame buffer; and a fog circuit coupled to the rasterization circuit for performing a fog function, wherein the fog function operates on floating point color values.

5. A computer system, comprising: a processor for performing geometric calculations on a plurality of vertices of a primitive; a rasterization circuit coupled to the processor that rasterizes the primitive according to a rasterization process which operates on a floating point format; a frame buffer coupled to the rasterization circuit for storing a plurality of color values; a display screen coupled to the frame buffer for displaying an image according to the color values stored in the frame buffer; and a blender coupled to the rasterization circuit which blends floating point color values.

6. A computer system, comprising: a processor for performing geometric calculations on a plurality of vertices of a primitive; a rasterization circuit coupled to the processor that rasterizes the primitive according to a rasterization process which operates on a floating point format; a frame buffer coupled to the rasterization circuit for storing a plurality of color values; a display screen coupled to the frame buffer for displaying an image according to the color values stored in the frame buffer; and logic coupled to the rasterization circuit which performs per-fragment operations on floating point color values.

7. A computer system, comprising: a processor for performing geometric calculations on a plurality of vertices of a primitive; a rasterization circuit coupled to the processor that rasterizes the primitive according to a rasterization process which operates on a floating point format; a frame buffer coupled to the rasterization circuit for storing a plurality of color values; and a display screen coupled to the frame buffer for displaying an image according to the color values stored in the frame buffer; wherein the processor, the rasterization circuit, and the frame buffer are on a single semiconductor chip.

8. The computer system of claim 7, wherein the processor, the rasterization circuit, and the frame buffer reside on a same substrate of the single semiconductor chip.

9. In a computer system, a method for rendering a three-dimensional image for display, comprising the steps of: performing geometric calculations on a plurality of vertices of a plurality of polygons; scan converting a plurality of pixels according to the vertices, wherein scan conversion is performed on floating point color values; applying a texture to the image by reading floating point texture values stored in a texture memory; simulating fog effects, wherein fog is simulated by modifying floating point color values; drawing the image for display on a display screen coupled to the computer system.

10. The method of claim 9, wherein the floating point values are comprised of sixteen bits.

11. The method of claim 10, wherein the floating point values are specified by a s10e5 format.

12. The method of claim 10 further comprising the step of storing the floating point color values in a frame buffer.

13. The method of claim 10 further comprising the step of blending at least two floating point color values.

14. The method of claim 10 further comprising the step of performing antialiasing on floating point color values.

15. The method of claim 10 further comprising the steps of: reading data from the frame buffer; modifying the data; writing modified data back to the frame buffer.

16. The method of claim 10 further comprising the step of modifying color values for lighting, wherein lighting calculations operate on floating point color values.

17. In a computer system, a method for operating on data stored in a frame buffer, comprised of: storing the data in the frame buffer in a floating point format; reading the data from the frame buffer in the floating point format; operating directly on the data in the floating point format; and writing the data to the frame buffer in the floating point format; wherein the steps of writing, storing, and reading the data in the frame buffer in the floating point format are further comprised of a specification of the floating point format, wherein the specification corresponds to a level of range and precision.

18. The method of claim 17 wherein the specification is comprised of 16 bits of data and the data are comprised of one sign bit, ten mantissa bits, and five exponent bits.

19. The method of claim 17 wherein the specification is comprised of 17 bits of data and the data are comprised of one sign bit, 11 mantissa bits, and five exponent bits.

20. The method of claim 17 wherein the specification is comprised of 16 bits of data and the data are comprised of ten mantissa bits, and six exponent bits.

21. The method of claim 17 wherein the specification is comprised of 32 bits of data and the data are comprised of one sign bit, 23 mantissa bits, and eight exponent bits.

22. A computer system having a floating point frame buffer for storing a plurality of floating point color values; wherein the floating point color values are written to, read from, and stored in the frame buffer using a specification of the floating point color values that corresponds to a level of range and precision.

23. The computer system of claim 22, wherein the floating point color values are comprised of 16 bits of data and the data are comprised of one sign bit, ten mantissa bits, and five exponent bits.

24. The computer system of claim 22, wherein the floating point color values are comprised of 17 bits of data and the data are comprised of one sign bit, 11 mantissa bits, and five exponent bits.

25. A computer system, comprising: a processor for performing geometric calculations on a plurality of vertices of a primitive; a rasterization circuit coupled to the processor that rasterizes the primitive according to a rasterization process which operates on an s10e5 floating point format; a frame buffer coupled to the rasterization circuit for storing a plurality of s10e5 floating point color values; a display screen coupled to the frame buffer for displaying an image according to the s10e5 color values stored in the frame buffer.

26. The computer system of claim 25 further comprising: a texture circuit coupled to the rasterization circuit that applies a texture to the primitive, wherein the texture is specified by s10e5 floating point values.

27. The computer system of claim 25 further comprising a lighting circuit coupled to the rasterization circuit for performing a lighting function, wherein the lighting function executes on s10e5 floating point color values.

28. The computer system of claim 25 further comprising a fog circuit coupled to the rasterization circuit for performing a fog function, wherein the fog function operates on s10e5 floating point color values.

29. The computer system of claim 25 further comprising an antialiasing circuit coupled to the rasterization circuit which performs an antialiasing algorithm on s10e5 floating point color values.

30. The computer system of claim 25 further comprising a blender coupled to the rasterization circuit which blends s10e5 floating point color values.

31. The computer system of claim 25 further comprising logic coupled to the rasterization circuit which performs per-fragment operations on s10e5 floating point color values.

RE: here are the claims of the patent at issue

By mindless1 on 10/26/2006 9:16:43 AM , Rating: 1

URLs are your friend, no need to post all this, twice.

Parent

SGI is not the same company any more.

By Milliamp on 10/25/2006 3:21:17 PM , Rating: 2

Most of their 3D graphics patents were xfered to MS back in 2002. Their IRIX and MIPS lines are effectively dinosaurs.

All of their best talent has gone away to work at other company’s.

The company literally has no other future outside selling AMD/Linux systems, a segment that is already effectively cornered by so many other players.

They literally emerged from bankruptcy with little more than a name and some old patents.

Given that AMD now owns ATI and this move isn't like to make them any friends in the industry, I think they are probably less interested in a rebirth of the company as a tech superpower and more interested in the business model that is working so well for SCO right now.

Now lets see if we can find some nVidia or Intel money involved in standing this company back up on its feet. I sure wouldn't be surprised.

hmm

By ttnuagadam on 10/25/2006 6:43:35 PM , Rating: 2

I dont get why they are just now suing ATI. havent framebuffers been used since the beginning? Its not like ATI tried to hide it or something.