skinner@saber.UUCP (Robert Skinner) (01/01/70)
> > > The use of exclusive-OR to draw and erase cursors, rubber band lines, > > > etc. is covered by patent number 4,197,590... > > > It would also be useful to know the name/contact information of the > company holding the patent. (Strangest thing I've heard since Roy Acuff's > lawyer tried to claim his client "owned" _The Wabash Cannonball_. WAAAUGH!) > -- Or how about this one: Lexidata holds a patent on the "concept" of doing solid modeling and rendering on a peripheral device, i.e. not the host. Raster Tech is fighting this one because of its Model 1/25. This patent could be interpreted to include any terminal (or workstation even) connected to a host computer (or simply more powerful computer, in the case of a workstation). ------------------------------------------------------------------------------ The difference between America and England is, the English think 100 miles is long distance and Americans think 100 years is a long time. Name: Robert Skinner Snail: Saber Technology, 2381 Bering Drive, San Jose, California 95131 AT&T: (408) 945-0518, or 945-9600 (mesg. only) UUCP: ...{decvax,ucbvax}!decwrl!saber!skinner ...{amd,ihnp4,ittvax}!saber!skinner
b-davis@utah-cs.UUCP (Brad Davis) (01/01/70)
In article <544@ecsvax.UUCP> hes@ecsvax.UUCP (Henry Schaffer) writes: >> > The use of exclusive-OR to draw and erase cursors, rubber band lines, >> > etc. is covered by patent number 4,197,590... >> >> (Forgot this in my previous response) It would also be nice to know the >> date of the patent. >> >Date: Apr. 8, 1980 In my Principles of Interactive Computer Graphics, Second Edition, William M. Newman and Robert F. Sproull, McGraw-Hill Book Company, Copyright 1979, 1973, ISBN 0-07-046338-7, pages 256-259 the following algorithm is described. F-1 is the inverse of f. "... we simply apply to each pixel value 'x' the function 'f(x) = a - x' where 'a' is the maximum value that the frame buffer can record for a pixel. Then the function 'f-1 (x) = a - x' will invert the transformation." If a = 1 then the functions are f(x) = 1 - x and f-1 (x) = 1 - x. Anyone can show that for 1 bit the function 1 - x is the same as 1 xor x. 1 - x = f(x) 1 xor x = f(x) 0 | 1 0 | 1 1 | 0 1 | 0 Since the Newman and Sproull algorithm also works for color I will continue to use it and not be in violation of any patent, even though the general case seems to have been described in literature 1 year before the patent was awarded for the simple case. I wonder what NuGraphics does about software that was written before their patent was awarded that uses xor to do cursors and rubber banding. I might have even wrote such software in 1978. If someone has some old SIGGRAPH's it might be interesting to look for some references. Also, the bibliography in Newman and Sproull has a number of references that talk about raster graphics. -- Brad Davis {ihnp4, decvax, seismo}!utah-cs!b-davis b-davis@utah-cs.ARPA
doug@terak.UUCP (Doug Pardee) (10/01/85)
> FLIPping is used mostly for cursors and rubber-band lines; its only > useful definition is that there is no net change after doing a FLIP > with both source and destinations the same. This is usually > implemented by performing a generalized exclusive-OR of the source with > the destination. Just so you'll know... The use of exclusive-OR to draw and erase cursors, rubber band lines, etc. is covered by patent number 4,197,590. The holders of that patent are, um, "aggressive" about defending their patent. Nearly as I can tell, the company that holds that patent produces no products; its only income is from license fees on that patent. [Don't flame *me* if you think it's not patentable, I agree. But that doesn't help any when your company is given the choice: license or court, where it'd cost more to win the suit than to buy a license.] -- Doug Pardee -- CalComp -- {calcom1,savax,seismo,decvax,ihnp4}!terak!doug
henry@utzoo.UUCP (Henry Spencer) (10/04/85)
> The use of exclusive-OR to draw and erase cursors, rubber band lines, > etc. is covered by patent number 4,197,590... > Nearly as I can tell, the company that holds that patent produces no > products; its only income is from license fees on that patent. It would be nice to know name, address, and telephone number. -- Henry Spencer @ U of Toronto Zoology {allegra,ihnp4,linus,decvax}!utzoo!henry
peterh@tekgvs.UUCP (Peter Hildebrandt) (10/04/85)
> >The use of exclusive-OR to draw and erase cursors, rubber band lines, >etc. is covered by patent number 4,197,590. The holders of that patent >are, um, "aggressive" about defending their patent. > >Nearly as I can tell, the company that holds that patent produces no >products; its only income is from license fees on that patent. > >Doug Pardee -- CalComp -- {calcom1,savax,seismo,decvax,ihnp4}!terak!doug Yes, there is in fact a patent on the exclusive-or for cursors. The person who holds the patent actively seeks out companies using graphics controller chips with the feature (like the NEC 7220 or the Hitachi 63484). While I was at Vectrix Corportaion, we were approached by the patent holder who demanded royalties. We, too, could not believe he had actually gotten a patent on a mathematical process. But after he told us he only wanted a flat fee (as I recall) of around $100 or $200, we couldn't refuse. Who can afford to fight a legal battle over $100? What a neat way to make money. If he makes $100 on 100 companies, he has just earned $10,000 for nothing. Peter Hildebrandt Tektronix Laboratories
henry@utzoo.UUCP (Henry Spencer) (10/05/85)
> The use of exclusive-OR to draw and erase cursors, rubber band lines, > etc. is covered by patent number 4,197,590... (Forgot this in my previous response) It would also be nice to know the date of the patent. -- Henry Spencer @ U of Toronto Zoology {allegra,ihnp4,linus,decvax}!utzoo!henry
jeff@qubix.UUCP (Jeff Bulf) (10/07/85)
> > The use of exclusive-OR to draw and erase cursors, rubber band lines, > > etc. is covered by patent number 4,197,590... > > (Forgot this in my previous response) It would also be nice to know the > date of the patent. > -- It would also be useful to know the name/contact information of the company holding the patent. (Strangest thing I've heard since Roy Acuff's lawyer tried to claim his client "owned" _The Wabash Cannonball_. WAAAUGH!) -- Dr Memory ...{amd,ihnp4}!qubix!jeff
hes@ecsvax.UUCP (Henry Schaffer) (10/08/85)
> > The use of exclusive-OR to draw and erase cursors, rubber band lines, > > etc. is covered by patent number 4,197,590... > > (Forgot this in my previous response) It would also be nice to know the > date of the patent. > -- > Henry Spencer @ U of Toronto Zoology > {allegra,ihnp4,linus,decvax}!utzoo!henry Date: Apr. 8, 1980 Inventors: Josef S. Sukonick (Cupertino, CA), Greg J. Tilden (San Jose, CA) Assignee: NuGraphics, Inc. (Cupertino, CA) Title: Method for Dynamically Viewing Image Elements Stored in a Random Access Memory Array This patent is 26 pages, with 21 Drawing Figures, and has 16 claims. It is for a computer-graphics display system, and includes various operations. The NCSU library is a US patent repository. They told me that there are 58 in the USA, with most states having one. --henry schaffer
ksbszabo@watvlsi.UUCP (Kevin Szabo) (10/08/85)
.. re the patent holder charging money for use of his idea .. >What a neat way to make money. If he makes >$100 on 100 companies, he has just earned $10,000 for nothing. I think '$10,000 for nothing' is an unfair assessment of the situation. As with most great ideas, once someone shows you how to do something it becomes obvious (also spoken: 'Once someone opens the door everyone can see through it'). Just because a process is obvious and in widespread use this does not lessen an inventor's credit for it. -- Kevin Szabo' watmath!watvlsi!ksbszabo (U of W VLSI Group, Waterloo, Ont, Canada)
julian@osu-eddie.UUCP (Julian Gomez) (10/10/85)
> I think '$10,000 for nothing' is an unfair assessment of the > situation. As with most great ideas, once someone shows you how to > do something it becomes obvious (also spoken: 'Once someone opens the door > everyone can see through it'). ... Well, not quite. Einstein showed us relativity, but how many people think it's obvious? > ... Just because a process is obvious and > in widespread use this does not lessen an inventor's credit for it. Anyway, as I understand it, in order for something to be patented, the applicant must show that it is not obvious to a practitioner (i.e. expert) in the field. Thus no one can patent the Pythagorean Theorem because it's obvious to anyone past junior high (no flames, they didn't have junior high thousands of years ago :-). I would say that XOR cursors are obvious, because people have been using them since no later than 1975, when the E&S frame buffer was built. It is very easy to question whether those guys invented the process, since it was around for at least five years before the patent was granted. It is clear that a number of people thought it up independently. However, it would probably take a court battle to clear things up, and like Peter Hildebrandt <1238@tekgvs.UUCP> points out, what company is willing to fight a court battle over $100? -- "If Chaos himself sat umpire, what better could he do?" Julian "a tribble took it" Gomez Computer Graphics Research Group, The Ohio State University {ucbvax,decvax}!cbosg!osu-eddie!julian
dave@onfcanim.UUCP (Dave Martindale) (10/11/85)
Does anyone know whether this guy actually first thought of the use of XOR, patented it (which may have taken a long time), and is only now licensing it, or whether he patented something which was already known and nobody else thought of patenting because it seemed so obvious?
rcd@opus.UUCP (Dick Dunn) (10/19/85)
> >What a neat way to make money. If he makes > >$100 on 100 companies, he has just earned $10,000 for nothing. > > I think '$10,000 for nothing' is an unfair assessment of the > situation. As with most great ideas, once someone shows you how to > do something it becomes obvious (also spoken: 'Once someone opens the door > everyone can see through it'). Just because a process is obvious and > in widespread use this does not lessen an inventor's credit for it. Only true if the inventor actually "invented" the process--i.e., that the process had been unknown. (The key phrase above is "once someone shows you". Sure, if you can't figure it out yourself, there's some substance there.) Otherwise it's just a legal game. I think that any decent programmer could figure out the xor cursor. It's nothing more than a pixel version of an old trick in machines with fast registers and slow memory--how do you exchange two registers without using memory or other registers? Simple--takes three xor's; if the registers you want to exchange are R1 and R2, the two-address form (op src,dst) is xor r1,r2 xor r2,r1 xor r1,r2 I saw this in about '68 or so, and it was an old trick then. I've posed it as a riddle to students taking their first assembly-language course. BFD. For the graphic equivalent, it seems to me that as soon as you start thinking about manipulating pixels with bitwise ops, you come up with something that will run along the lines of RasterOps. Once you get to that point (whether you've actually seen all the material on RasterOps or not), the next natural thing to think about is what all of the 16 possible transformations of (source*dest) bits mean. You find the obvious functions like clear-to-white, clear-to-black, copy, paint white, paint black, invert, etc.--and lo and behold, xor is in there; any sophomore will recognize it. In other words, the technique and the usefulness is obvious as soon as you look at the problem. Why in hell this could be patentable (if, indeed, it is an explicit part of the patent!) is beyond me. Can I patent the use of a pencil for taking up the slack in a cassette tape? -- Dick Dunn {hao,ucbvax,allegra}!nbires!rcd (303)444-5710 x3086 ...Simpler is better.
england@unc.UUCP (Nick England) (11/04/85)
the patent originally granted to NUGRAPHICS covers a lot of things most of us take for granted as being obvious.the patent includes pan,scroll, and zoom in hardware on a frame buffer as well as EXOR. CADTRAK now holds rights to the patent and are asking a minimum of $25,000 for a license.they claim IBM and HP have signed licenses. In the same vein - LEXIDATA has patented the z-buffer and E&S has got clipping matrix multiplier transformation and other stuff patented.there are legal proceedings by LEXIDATA against RASTER TECH currently and E&S has had SILICON GRAPHICS and MEGATEK in court. it's not the friendly old graphics world it used to be. --nick england (ex-IKONAS)