Skip to main content
GameDev.net gamedev.net
🔒 Locked

Stencil Shadows Patented!? WTF!

Started by drewslater Sep 21, 2003 at 1:49 PM 55 replies 56.8k views
Original Post
drewslater
drewslater
It has been brought to my attention that two guys named William Bilodeau and Michael Songy have filed a patent for a stencil shadow technique (Carmack''s reverse I believe). View their patent claim here Anyone care to comment on any implications of this patent or similar ones. I think it is rediculous for people to patent algorithms like this. What are the odds of this patent actually being enforced? Will these guys have the ability to sue all game developers so they have to pay royalties to use stencil shadows? I find this very hard to believe. I am under the impression that basically anything you pay to be patented will be awarded whether there is proven merit for it or not and the place where things are sorted out is when they go to court. This still is a pain in the ass for the people who get sued by people trying to enforce these rediculous patents. Anyone know of any precedents for software patents? ATS
Nemesis2k2
Nemesis2k2
Dude, the people who issue these patents are morons. It really doesn''t matter how long it''s been known, or how simple it is. Half the time you''ll get it regardless. One day I swear to god someone''ll patent the wheel.

BdR
BdR
It says "filed: October 13, 1999" (4 years ago) so..

Sounds to me like a couple of guys who found out creating software is harder then they thought, so they thought they''d make some quick cash by sueing anyone that used ''thier'' algorithm by patenting it ("hey, it''s the American way" )

I don''t have any 3D programming experience, but it sounds like some algorithm to create shadows, which can be done in many ways. And proving, in court, that a certain program draws shadows using the "patented algorithm" is not practicle, and also hardly worth the effort.
Yann L
Yann L
quote:
Original post by cowsarenotevil
Why does no one actually read my posts? They are NOT making money off of it, because it''s illegal. They have as much of a chance of winning a legal battle as a dead monkey.

Please do not spread incorrect information. Software patents are perfectly legal in the US and Japan. They have always been, and they are aggressively enforced over there (for example the marching cubes algorithm, GIF and MP3 compression, CSS encryption, the infamous SCO vs. Linux trial, etc). Until now, software patents are illegal in Europe. They will decide in 3 days, if they will remain illegal or if a US-style model is adopted (I hope not).
cowsarenotevil
cowsarenotevil
quote:
Original post by Yann L
Please do not spread incorrect information. Software patents are perfectly legal in the US and Japan. They have always been, and they are aggressively enforced over there (for example the marching cubes algorithm, GIF and MP3 compression, CSS encryption, the infamous SCO vs. Linux trial, etc). Until now, software patents are illegal in Europe. They will decide in 3 days, if they will remain illegal or if a US-style model is adopted (I hope not).


Oh really? I did not know that (I didn't now that those were US patents, I just thought that many programs refraind from using them because they were illegal in some countries). My mistake.

EDIT: Does that mean I can go and patent a bunch of open source programs now?

[edited by - cowsarenotevil on September 21, 2003 3:39:54 PM]

[edited by - cowsarenotevil on September 21, 2003 3:40:44 PM]
-~-The Cow of Darkness-~-
Yann L
Yann L
quote:
Original post by cowsarenotevil
EDIT: Does that mean I can go and patent a bunch of open source programs now?

If you can prove you invented the algorithms (or you have enough money to shut up the original inventor, which is unfortunately often the case), then yes. But you'll have to go through a long and expensive process, until you get your patent granted (or not). And you will not be able to enforce those patents in countries that do not recognize them.

BTW, to prevent any misunderstandings: you cannot patent programs, only copyright them. You can patent algorithms, concepts or processes.


[edited by - Yann L on September 21, 2003 4:27:06 PM]
Ysaneya
Ysaneya
He first talks about it in 2000, so no, it''s not earlier than these guys.

They both came up with the technique independantly. They just tried to protect their work, yet some of you call them "morons". How nice.

Y.
cowsarenotevil
cowsarenotevil
quote:
Original post by Yann L
If you can prove you invented the algorithms (or you have enough money to shut up the original inventor, which is unfortunately often the case), then yes. But you'll have to go through a long and expensive process, until you get your patent granted (or not). And you will not be able to enforce those patents in countries that do not recognize them.

BTW, to prevent any misunderstandings: you cannot patent programs, only copyright them. You can patent algorithms, concepts or processes.


Wait, so games that use shadow volumes are breaking the law?

EDIT: But I did misunderstand. Algorithms being patented makes sense, but actual programs is still not legal, am I right?

[edited by - cowsarenotevil on September 21, 2003 5:03:43 PM]
-~-The Cow of Darkness-~-
deiterate
deiterate
This seems absurd.

However, revolutionary ideas would occur more slowly if we were not given limiting parameters such as patents. They force us to develop more original methods. It seems to me that if the two gentlemen who hold the patent for "Carmack''s Reverse" ever bother to make it binding, there will be a revolutionary mind out there who will develop an even more elegant solution (and hopefully that person will not patent it as well).

Although, I must admit, the limitations of current hardware are enough of a pain to deal with. We would do just as well without this patent.

Here is a link to a file on the nVidia developers website that seems to indicate that "Carmack''s Reverse" is not as old as the patent.

http://developer.nvidia.com/attach/3413
Nemesis2k2
Nemesis2k2
quote:
Original post by Ysaneya
He first talks about it in 2000, so no, it''s not earlier than these guys.

They both came up with the technique independantly. They just tried to protect their work, yet some of you call them "morons". How nice.

Y.

Correction, I called the people who issue patents morons on a seperate issue. At the time I was not aware that this was four years old either, though as I mentioned, my comments did not relate to the people who filed for this patent.

python_regious
python_regious
Most large companies use patents as a defence more often or not. Believe it or not, Microsoft was actually against the legalisation of software patents in the US. AFIAK, IBM is currently the largest owner of software patents ( in the US - something stupid like 10% ), and is one of only a few that actively enforces them.

I just hope to god they don''t legalise it here ( Europe )...

You have to remember that you''re unique, just like everybody else.
If at first you don't succeed, redefine success.
chronos
chronos
quote:
Original post by Anonymous Poster
It means don't worry about it... the patent is simply not enforced or there would have been trials by now.

Don't count on that. Once they have the patent they can sue whenever they like (within the 17-year period of exclusivity).

[edited by - chronos on September 21, 2003 6:47:03 PM]
SimmerD
SimmerD
Don't worry about it fellas. I described this technique publicly a few months before they filed the patent - hence Prior Art. Ironically, it was at a Creative Labs developer's forum.

During my stencil buffer talk, I described doing shadow volumes the 'reverse' way. At the time, I didn't realize the major reason why the z fail method is better than the z pass method, although I did realize they were logically equivalent, which is why it's now known as 'Carmack's Reverse' and not 'Dietrich's Reverse'! ;)

-- Update : Actually I first presented this at GDC '99 in March, although I may have also presented at at the Creativity '99 conference later in the year as well.

<SPAN CLASS=editedby>[edited by - SimmerD on September 21, 2003 7:00:12 PM]</SPAN>

<SPAN CLASS=editedby>[edited by - SimmerD on September 21, 2003 10:49:10 PM]</SPAN>

[Edited by - SimmerD on July 29, 2004 4:50:28 PM]
Yann L
Yann L
quote:
Original post by cowsarenotevil
Wait, so games that use shadow volumes are breaking the law?

In the EU: no (currently). In the US: legal greyzone. Theoretically, every application using a patented algorithm must get written permission from the patent holder, if not stated otherwise (there are also free to use patents). In practice, however, it's extremely hard to keep track about which algorithm exactly is patented, and by whom. Also, keeping track of all royalities to pay would be a nightmare - considering that things like scrollbars, editboxes, and many aspects of 3D graphics are actually patented...

Most companies do not enforce software patents. Interestingly, most companies are against them. In most cases, it wouldn't make sense financially, since the costs for the license logistics are generally far greater than a potential gain from royalities. There are some exceptions, unfortunately, who actually enforce their patents. In that case, you could get sued if you use the algorithm in question in your project - even if you came up with the idea on your own.

quote:

EDIT: But I did misunderstand. Algorithms being patented makes sense, but actual programs is still not legal, am I right?

You don't need to patent a program, it is automatically protected by intellectual property laws (copyright). A full program could only be patented, if it is an integral part of a machine or process (for example certain types of firmware, especially for embedded devices).

BTW, did anyone here ever got a US software patented granted ? I (or better: my company) was thinking of patenting a some algorithms from our engine (the ABT, and a couple of others). But the costs seem to be very prohibitive, and you don't get anything back, should the patent be rejected.


[edited by - Yann L on September 21, 2003 8:18:04 PM]
cowsarenotevil
cowsarenotevil
quote:
Original post by Yann L
You don''t need to patent a program, it is automatically protected by intellectual property laws (copyright). A full program could only be patented, if it is an integral part of a machine or process (for example certain types of firmware, especially for embedded devices).


What about gameplay? Can the idea of say, an RPG, be patented, copyrighted, or neither?
-~-The Cow of Darkness-~-

Topic Locked

This topic has been locked by a moderator. New replies are not allowed.

Sign in to reply to this topic.