[music-dsp] low-latency convolution and patent

[koen vos]

Stefan Meyer wrote:
> One DSPler mentioned a patent by Lake for this task which was granted in
> the early 90ies,
...
> Why there is a patent, if the technology was not state of the art (In my
> understanding, patents are only granted if the technology is new.)

The important thing is the filing date of the patent. If the echo
cancellation paper was received by the journal before the patent was aplied
for, then the patent (or part of it) can be made invalid in a law suit.
However, be aware that often several years pass between filing of a patent
and its being granted.


> Is it allowed to build a program with this trick, without geting in
> trouble with Lake, if the patent was not granted legaly?

Lake may sue you, but you would probably win the case. But most likely they
wouldn't take any steps after you've shown them there exists 'prior art'.

koen


dupswapdrop -- the music-dsp mailing list and website: subscription info,
FAQ, source code archive, list archive, book reviews, dsp links
http://shoko.calarts.edu/musicdsp/