Creative Pressures id Software With Patents 339
Cryect writes "Earlier today it was announced by Creative that they would be adding in EAX 3D sound support to Doom 3, and that they had come to an 'agreement relating to Creative's patented shadowing technique [also known as Carmack's Reverse in some coding circles] and id's cutting-edge 3D graphics DOOM 3 engine.' This seemed somewhat suspicious, almost as if id was being pressured, and a quick email to John Carmack from Reverend @ Beyond3d got this reply: 'The patent situation well and truly sucks... It was tempting to take a stand and say that our products were never going to use any advanced Creative/3DLabs products because of their position on patenting gaming software algorithms, but that would only have hurt the users...' There's also some possible prior art [PPT link] to Creative Labs' patent, from a 1999 talk by Nvidia's Sim Dietrich."
but PCG said....... (Score:2, Informative)
(pg.79) Sept. 2004
"(8) Is it true that Doom 3's audio engine is entirely CPU-dependent, thus negating the benefits of high-end sound cards? If so, what are the benefits? What are the drawbacks?
[bla, bla, bla]
PC Gamer's take: Much to Creative Labs' chagrin, Doom 3 should sound exactly the same (and perform equally well) on your motherboard's built-in audio processor as it will on a high-end Audigy 2 ZS sound card."
so much for that!
Too bad... (Score:4, Informative)
-truth
Re:Future source code release. (Score:5, Informative)
This was done w/ Doom.
Re:Prior art (Score:5, Informative)
a person shall be entitled to a patent unless -
(a) the invention was known or used by others in this country, or patented or described in a printed publication in this or a foreign country, before the invention thereof by the applicant for a patent.
It is difficult to prove "known", however the talk could be considered a printed publication if it was readily accessable to the public. A paper which is orally presented in a forum open to all interested persons constitutes a "printed publication" if written copies are freely disseminated. Massachusetts Institute of Technology v. AB Fortia, 774 F.2d 1104, 1109. Also see MPEP 2128,2132
Re:If I was Carmack.... (Score:3, Informative)
Re:I stab at thee (Score:3, Informative)
A few years back Intel starting selling toys and peripherals. Those bombed.
Now Intel is starting to sell integrated audio on its boards. Maybe this time it will work, maybe Intel will dominate, but I doubt it.
Intel makes its money on CPUs. Anything that takes work away from the CPU takes away from Intel's profits. That's why USB is CPU dependant but FireWire isn't. So I seriously doubt that Intel will ever come up with a gaming audio system that doesn't drag down the CPU, requiring a CPU upgrade to get better performance.
Re:Not just software patents (Score:5, Informative)
OpenAL? http://www.openal.org/
Re:well then... (Score:3, Informative)
http://www.techreport.com/reviews/2004q3/p
For consumer music and movies, CPU usage is much less of an issue. For these uses I would recommend a M-Audio Revolution. It uses the VIA EnvyHT chip and supports for 192/24 resolution throughout the signal path. The sound quality is excellent and the bass management is vastly superior to that of the Audigy 2.
For strictly 2-channel listening, I'd recommend the Chaintech AV-710.
Here's a review on a headphone forum.
http://www4.head-fi.org/forums/showthread
Here's a setup guide on that same forum.
http://www4.head-fi.org/forums/showthread
Re:Prior art (Score:4, Informative)
Re:If I was Carmack.... (Score:2, Informative)
and belive me, I have many of them*
(*just because I'm reading slashdot does not mean I have no friends - thought the statistical corelation of slashdot reading and friend starvation are fascinating)
Re:Will Carmack change his mind about patents? (Score:5, Informative)
If they filed for the patent within one year of the release of the technique, then they could patent it. Yes, with our system, someone can invent something, decide not to patent it in order to let others use it, someone else can see the invention, patent it, and force the original inventor to pay to use it.
Re:Ugh, I hate software patents. (Score:4, Informative)
From the creative web page:
Phil O'Shaughnessy
Director of Corporate Communications
poshaughnessy@creativelabs.com
Lara B. Vacante
Public Relations Manager
Lara_Vacante@creativelabs.com
Jennifer Ellard
Senior Public Relations Specialist
Jennifer_Ellard@creativelabs.com
Katie Meyer
Public Relations Coordinator
Katie_Meyer@creativelabs.com
Re:I why I hate, why I use Creative's cards... (Score:5, Informative)
Also, digging around the site, if you do find one of those other programs to download.. when you try to run them, they say they can't find an older version to update and force you to quit.
As for normalization, I wouldn't leave that up to your soundcard. Use a decent audio player like FooBar2000, and set replaygain levels on your tracks. Enjoy the stability.
Or, if you want, buy the OctiMax plugin for Winamp2/5.. it's a multiband compressor/limiter, and does a pretty good job at keeping things steady. I use it for radio broadcasts, or LAN parties.
my letter to creative public relations (Score:2, Informative)
I'm just writing to inform you that you will not receive anymore of my business regarding your position on gaming software algorithms patents. I have canceled my order for the Maximum-power 6.1 sound system and will take my business elsewhere. I have supported Creative since I first got my computer, but I do not approve of this disregard for gamers and I'm quite saddened by your position.
MP3Gain (Score:4, Informative)
Cheers,
Toby Haynes
Derek Smart, PhD: "software patents are necessary" (Score:2, Informative)
Yes, it's difficult for me to justify patenting the intangible. I find it equally laborious equating copyright infringement with theft.
And for the sake of adding fuel to the fire, I present you this:
I'll think I'll allow some of you Linux guys handle that one for me as I'm, admittedly, more of a Windows dweeb.
Countdown to Derek Smart, Ph.D. responding to this post with expletives in 5...4...3...
Intel has the highest volume shipments (Score:3, Informative)
http://news.com.com/Intel%2C+AMD+market+shares+re
Intel sell more graphics cards than anyone else (Score:3, Informative)
Of course, this is not the same as saying they are the widest used graphics card or that they will sell a large amount of standalone cards but still.
Most people don't buy external sound cards any more but once upon a time everyone did. So those cheapo AC97 based things are ALL over the place - OK Intel don't make them all but they did come up with the AC97 codec [intel.com].
Here's a Register story which mentions that Intel have 31.7% of the graphics card market [theregister.co.uk].
I've talked to people of various importance who feel that in a certain number of years the graphics card market will go a similar way to the sound card market. The impression I was given was that only people wanting high end quality/speed will go for an extra card but most others will be satisfied by onboard.
Email Addresses to Voice Your Opinion (Score:5, Informative)
Director of Corporate Communications
poshaughnessy@creativelabs.com
Lara B. Vacante
Public Relations Manager
Lara_Vacante@creativelabs.com
Jennifer Ellard
Senior Public Relations Specialist
Jennifer_Ellard@creativelabs.com
Katie Meyer
Public Relations Coordinator
Katie_Meyer@creativelabs.com
Re:Not just software patents (Score:3, Informative)
As for OpenAL titles that are actually used, let me see:
From www.openal.org:
AlienFlux (Windows, Linux, Macintosh)
America's Army: Operations (Windows, Linux, Macintosh)
Bridge Construction Set (Windows, Linux)
Escape From Monkey Island (Macintosh)
FlightGear (Windows, Unix, Macintosh)
Harry Potter and the Chamber of Secrets (Windows, Macintosh)
Jedi Knight: Jedi Academy (Windows, Macintosh)
Jedi Knight 2 (Windows, Macintosh)
Marble Blast (Windows, Linux, Macintosh)
MegaCorps Online (Windows, Linux, Macintosh)
Orbz (Windows, Linux, Macintosh)
Postal 2 (Windows)
Soldier of Fortune 2 (Windows)
Unreal 2 (Windows)
Unreal Tournament 2003 (Windows, Linux, Macintosh)
Unreal Tournament 2004 (Windows, Linux, Macintosh)
Re:John, just do it! (Score:4, Informative)
Nvidia Soundstorm. The Audio solution that comes with the Nforce2 MCP-T. And it's a Nvidia Product, So ID Getting full support From Nvidia would be easy as Pie since both companies like each other so much.
There's also VIA's Vinyl Audio Solution, as well as Analog Devices Soundmax.
Re:The sad part is EAX sucks compared to Aureal (Score:3, Informative)
Re:Too bad... (Score:2, Informative)
Yes and no. The idea is that the claim you make that you invented something someone else filed a patent on has to be based on fact. You must be able to provide documents like dated lab manuals or whatnot that "prove" you really did invent the invention first.
What is to prevent people from patenting other peoples ideas?
If they can prove that the filing you have is, and I forget the term actually used, in bad faith, then your patent is unenforceable (which is worse than being invalid). But the kicker is proving it and bascially is a lawsuit to get at that information.
How would they proved that they had the idea first, beyond saying "yes we did", if they did not publish it?
See above about lab manuals. Publishing will _never_ help you acquire patent rights. The best you can do is publish a technology someone else is trying to patent and hope that they don't have something on file yet (and that they cannot swear behind your publication).
-truth
Because... (Score:2, Informative)
As for its being an accepted standard (for gaming), there is no sense using up processing time compressing and encoding multi-channel audio into a Dolby Digital stream when you can simply output the digital audio as a raw PCM stream which most digital audio decoders are capable of handling directly. It would only be of use for pre-rendered cut scenes/FMV where all the audio would be the same each time.
Re:playing with fire... (Score:3, Informative)
Terratec are the only ones I can think of at the moment. They make quality parts, and they have a game-oriented line.
I have an old Addonics SV550 (YMF724 chip) myself, and it's all I need. It doesn't have any fancy features, but it works, and the output sounds great, and that's all I need.
I'm beginning to think that onboard sound has finally reached a point where sound cards have become as good as obsolete. Hell, the only reason I use my YMF724 card is because the onboard sound in my machine doesn't have hardware mixing support in Linux.
Re:Prior art (Score:3, Informative)
If people understood what this patent was about, they'd realize how stupid this is. Here is a short trip down computer graphics history.
Shadows are an important visual effect for photorealism, which has been the goal of computer graphics for a very long time. But, shadows are computationally expensive. You have to answer the following question, is there something in the way between the part of the object that I know I can see and the light source. Which of course depends on other geometry in the scene. The most obvious was of answering the query is through a ray-trace. Draw a directed line from the point that you know you see, to the light source, is there something in the way? Yes, then this point is in shadow. Call this the shadow determination problem. Unfortunately, raytracing is rather slow.
Franklin Crow wrote a paper for SIGGRAPH in 1977 called "Shadow Algorithms for Computer Graphics" which he describes the idea behind shadow volumes. Basically instead, of doing a raytrace, you create volumes that represent a space of shadow. Shadow determination becomes, is this point in a shadow volume? Yes, then this point is in shadow. In 1977, it wasn't clear why anyone would use the shadow volume method. Crow was just doing an academic survey of all methods you could use to do shadow determination.
In comes the development of dedicated graphics hardware. Graphics hardware is generally made up of a number of buffers. A depth-buffer, color-buffer, and stencil-buffer being some examples. You only see the color-buffer. The stencil-buffer allows you to fill in some integer value for each pixel on the screen. And later render a scene based on some condition based on the value in the stencil buffer and possibly values in others, this is called the stencil-test. The stencil buffer is useful for a number of effects such a mirrors, or portals and the like.
In 1991, Tim Heidmann wrote "Real shadows, real time" in Iris Universe in which he describes how the stencil buffer and stencil-test can be used to to implement the shadow volume test.
Now, here is where it gets hella stupid. One of the problems you run into with shadow-volumes is that they have to be water-tight. Sometimes, the computer graphics hardware will clip the volume. That is, what you see in 3D scene actually exists in a six sided volume, called the view-volume. There is a near, far, left, right, top, and bottom plane to it. And geometry gets choped off if it happens to be on the wrong side of one of these planes planes. In the straight-foward implementation of the idea, the plane that ends up causing the most problems is the near plane. It tends to happen when the view point gets too close to a shadow volume. This causes incorrect shadow determination and suffice it to say, in interactive graphics, people will move their view point too close to shadow volumes. So dealing with shadow volumes getting clipped at the near plane is a problem that you have to deal with. The problem is really messy.
However, Carmack popularized the idea of changing the stencil test a little bit. It moves the problematic plane from the near plane to the far plane. But, if the far plane clips the shadow volume, you still end up with incorrect results. It isn't obvious how this is much of an improvement, but it is. You generally have much more freedom on where you place the far plane than the near plane. The far plane can be basically anywhere after the near plane, the near plane has to be somewhere infront of the view point. In fact, it is possible to place the far plane infinitely far away and thus totally get rid of the clipping problem. Now, Heidmann made the suggestion in the orginial paper that you could flip the test around. Carmack just popularized the idea.
In comes Creative Labs, 20 years hence. They didn't come up with anything. Just filed some paperwork. Not that Carmack's stencil-test flip is that earth-shattering either. I honestly believe that any smart person would come up