KVR Audio

discretemachine · Post by **discretemachine** » Sun Jan 30, 2005 4:44 pm

Hi,

i just saw the new Christian Budde dynamic convolution plug. looks very sweet !

But when I went to get it, I got this message:

Dynamic Convolution Plugin
I have read, that there is a patent on this technology, so i stopped publishing this, until i have talked to a lawyer!

Did anyone have a chance to download it ?

Since it's free, it should be no problem for someone to redistribute it, yes ? That way Christian cannot be held responisble....if the patent thing is an issue ??

I would love to test this plug....

d

CinningBao · Post by **CinningBao** » Sun Jan 30, 2005 5:27 pm

I didn't have any problem downloading them from this page

http://www.savioursofsoul.de/Christian/

Innominandum · Post by **Innominandum** » Sun Jan 30, 2005 6:03 pm

I don't think it's possible to patent dynamic convolution. Regardless of who has the patent, I'm willing to bet there's prior art. I personally think that any company who would try to enforce a patent of this nature would be shooting their foot in a sea full of sharks.

Of course this is all speculation! I don't blame him for being on the side of cautious.

CinningBao · Post by **CinningBao** » Sun Jan 30, 2005 6:13 pm

Sorry dude,

I should read - I thought you meant Dual Channel FFT. It looks like he's just playing safe by not distributing this plugin at the moment; lets hope we see it soon.

skOre · Post by **skOre** » Sun Jan 30, 2005 7:21 pm

Possibly waves made that patent stuff... but I'm not sure.

But it was great fun to see that its possible that easy (no big comp needed). Trying to get 1k$ for that thing really seems over the top.

popsych · Post by **popsych** » Sun Jan 30, 2005 8:21 pm

Actually sintefex has the patent. Waves new EQ thingie is not dynamic convolution based from what i understand.

I don't think it's possible to patent dynamic convolution. Regardless of who has the patent, I'm willing to bet there's prior art. I personally think that any company who would try to enforce a patent of this nature would be shooting their foot in a sea full of sharks.

and even if there isn't prior art i'm sure the patent specific bits of dyn convo could be avoided and use substitute methods to achieve the same effect

dburgan · Post by **dburgan** » Sun Jan 30, 2005 8:30 pm

The idea of dynamic convo isn't patented, but a specific algorithm for implementing it is patented. I agree he should be able to work around the patent, but legal advice is not a bad idea at all to cover the ole hiney.

Torben · Post by **Torben** » Sun Jan 30, 2005 10:19 pm

"Dynamic Convolution" is nothing but a trademark owned by Sintefex. It is not a patent.

Torben

unidyn · Post by **unidyn** » Mon Jan 31, 2005 6:28 am

Given this is true, and Torben is certainly one who would understand these things quite well, is there any reason some kind soul couldn't post a link to a copy they have? This allows Christian to sleep well at night and for others that are interested to experience what this interesting plug-in is like.

Also, why not just rename this plug-in...something like Capricious Convolator? I came up with that myself, but use a thesaurus or something to describe the process if you prefer something else. Non-Linear Convolution Project, or just simply Space Emulator. It's a shame we can't enjoy the efforts of a great design because of fear that some corporate entity will rail you. There's your free enterprise system working for you. Well, here's my email in the event someone would be kind enough to send it along, so just in case, thank you in advance.

unidyn@hotmail.com

Aleksey Vaneev · Post by **Aleksey Vaneev** » Mon Jan 31, 2005 7:03 am

Torben, it's not about trademark wording. There is a patent on 'brute force' convolution algorithm which uses linear interpolation between the impulses to achieve non-linearity. It's nothing funky and pretty obvious, but it was patented nevertheless.

egbert · Post by **egbert** » Mon Jan 31, 2005 7:08 am

Aleksey Vaneev wrote:Torben, it's not about trademark wording. There is a patent on 'brute force' convolution algorithm which uses linear interpolation between the impulses to achieve non-linearity. It's nothing funky and pretty obvious, but it was patented nevertheless.

Do you think there is potential for a smarter way to do this - another kind of interpolation function perhaps which would get around this?

Space Boy · Post by **Space Boy** » Mon Jan 31, 2005 7:47 am

Aleksey Vaneev wrote:There is a patent on 'brute force' convolution algorithm which uses linear interpolation between the impulses to achieve non-linearity. It's nothing funky and pretty obvious, but it was patented nevertheless.

I mentioned a similar approach in a thread back in November. It's kind of obvious.

However, as far as I understand, Christian has provided a standalone host like tool which enables you to load a plug-in, excite it with a sine sweep, analyse the (frequency) output and generate from this an impulse response. He has also a plug-in which will load into your host, load the generated IR and utilize it in the usual manner.

Other than the tool for ripping IRs from other effects, this is just a standard Convolution processor (or did I miss something). Where is the dynamic bit exactly?

The ripping tool, though crude, doesn't sound too different in concept to a product recently released by Waves. However, the process of extracting an IR from a room, or piece of equipment, is well known (and that may be patented).

Needless to say, both aspects of this plug have been
around for quite some time (unless I missed the special feature).

By the way, there is a 'plug-in' for Minihost, that does more or less the same thing.

bmanic · Post by **bmanic** » Mon Jan 31, 2005 8:22 am

Umm.. you've definately missed the DYNAMIC part of the plugin. It RESPONDS to velocity aka volume. Just run a bunch of kick drums trough the provided default .dyn patch and you'll see. Pretty friggin amazing IMO and I really hope he can get around the patent. Works very well realtime too.

Because of this plugin and some upcoming stuff from tritone http://www.tritonedigital.com/ I've postponed buying the sintefex fx2000. I'll save up and buy a Distressor EL8X-S version instead but as it'll cost a bit more I'll be broke beyond repair!!@$!%#@

Cheers!
bManic

bmanic · Post by **bmanic** » Mon Jan 31, 2005 8:24 am

Aleksey, put that amazing grey matter that you have in your head to work and figure out another way to do this dynamic convolution, in real time and I'll give you the money that was inteded for the distressor!

Cheers!
bManic

Space Boy · Post by **Space Boy** » Mon Jan 31, 2005 11:33 am

bmanic wrote:Umm.. you've definately missed the DYNAMIC part of the plugin.

Back in November I suggested this:

Wide Boy wrote:You could use a different convolution kernel depending on spectral intensity (signal amplitude). This could realize a non-linear behavior.

Original thread:

http://www.kvraudio.com/forum/viewtopic.php?t=63883

I guess I missed the dynamic part of Christian's implementation all together (I didn't read the instructions). It didn't jump out at me - as you suggest.

But, at least, I didn't miss the concept.

Dynamic convolution plug