[ODE] Quickstep and patents

[Giel van Schijndel]

erwin at erwincoumans.com schreef:
> Beau Albiston writes:
>> metanet software wrote:
>>     
>>> hi,
>>> I'm working on a physics solver, and have recently discovered that 
>>> several well-known techniques are in fact patented. I'm eager to 
>>> opensource my project, however doing so will make it obvious what 
>>> methods i'm using, and one of them is definitely patented. 
>>>
>>> From discussion on the Bullet forums, it appears that Ageia holds a 
>>> patent related to ODE's quickstep solver: 
>>> http://www.patentstorm.us/patents/7079145-claims.html 
>>>
>>> Has there been any trouble with these patents and ODE? I'm thinking 
>>> specifically about the various big-budget commercial games released 
>>> with ODE.. did Ageia extract any sort of licensing/blackmail fees from 
>>> the developers? 
>>>
>>> I'm just trying to get a sense of whether Ageia actually enforces 
>>> these ridiculous patents or they're just there to make them look good 
>>> in stockholders' eyes.
>>>       
>> I believe they are referring to LCP solvers in the context of being 
>> implemented on dedicated hardware.
>>     
> This patent covers ODE quickstep. Quickstep implements a variant of 
> Gauss-Seidel (SOR). 
>
> This is not about hardware, it is about software.
> Just cross your fingers and hope on some prior art, or that Ageia doesn't 
> pursue this.
>   
Didn't ODE exist before this date? Application: No. 10793899 filed on
2004-03-08
If so then ODE _is_ prior art to this patent. As usual the US patent
office does hardly any checking for prior art (sigh).

-- 
Giel

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 254 bytes
Desc: OpenPGP digital signature
Url : http://ode.org/pipermail/ode/attachments/20070607/3afa4a17/signature.bin