KVR Audio

OMU · Post by **OMU** » Sun May 01, 2005 7:00 pm

Is there such a thing? I remember I saw a thread where this was the subject. I'm very interested about this. Please give some details if you happen to know.
Thanks.

popsych · Post by **popsych** » Sun May 01, 2005 7:05 pm

Of course there is. Sintefex FX2000 is an eq based on dynamic convolution and sintefex owns the patent to the process. Focusrite license it from them to use it in the Liquid Channel

tee boy · Post by **tee boy** » Sun May 01, 2005 7:12 pm

Really? I heard about that but I had no idea they actually owned the technology. Has anybody (unofficially) used the technique in VST'land? If not, surely this thing cant stay under raps for ever...

cYrus · Post by **cYrus** » Sun May 01, 2005 7:17 pm

waves q-clone

http://www.waves.com/content.asp?id=1583

OMU · Post by **OMU** » Sun May 01, 2005 7:36 pm

Oh, I meant a vst plugin or something.

Sidhu · Post by **Sidhu** » Sun May 01, 2005 7:51 pm

There is Hydratone , excellent, more flexible comparedto Q-clone from what ihave read.. and the company is also launching a compressor based on the same technology... now.. what was the company called ?

Sidhu

bmanic · Post by **bmanic** » Sun May 01, 2005 10:22 pm

cYrus wrote:waves q-clone

http://www.waves.com/content.asp?id=1583

This is incorrect. Q-clone is NOT a DYNAMIC convolution plugin.. just convolution.

- bManic

bmanic · Post by **bmanic** » Sun May 01, 2005 10:24 pm

sidhu wrote:There is Hydratone , excellent, more flexible comparedto Q-clone from what ihave read.. and the company is also launching a compressor based on the same technology... now.. what was the company called ?

Sidhu

Also, this is NOT dynamic convolution AFAIK. It's just convolution. That is, the volume changes of the input signal does not in any way affect the choise of impulse response. Hydratone is therefore static. It just combines hundreds of static samples that for the EQ in a flexible way.

Cheers!
bManic

bmanic · Post by **bmanic** » Sun May 01, 2005 10:28 pm

Think of dynamic convolution compared to simple convolution like this:

Dynamic convolution, a multi-velocity layer sampler patch that reacts to the keyboards midi velocity and changes samples accordingly.

Convolution, a single velocity layer sampler patch that plays a single sample no matter what velocity a user plays the midi keyboard.

Can this effect be heard? Answer: Yes but it's rather subtle (IMHO the difference is huge but hey.. my middle name is apparently 'subtle').

- b 'subtle' Manic

tee boy · Post by **tee boy** » Sun May 01, 2005 10:32 pm

Anyone tried that Waves plug? Sounds like it could be one to two thing - amazing... or complete gash!

bmanic,
Sounds interesting, I bet a decent programmer could impliment that into a plug nfp. Any idea how many layers they tend to use?

greendoor · Post by **greendoor** » Sun May 01, 2005 10:35 pm

I sincerely hope that convolution technology is simply a temporary step along the road to superior (and much more efficient) modelling technology ...

tee boy · Post by **tee boy** » Sun May 01, 2005 10:42 pm

I dont know mate, I reckon the concept of Q Clone is f**king top notch. I mean, you could buy a signal hi end EQ and use that plug to get a rack full (sound good to me)! Its just a question of how good the sound is. I reckon if you spend the time to make good ir's, then convo verbs are ALMOST passable as the original units. Dont see why this should be any different, although Im sure someone will have a reason.

away from home · Post by **away from home** » Sun May 01, 2005 10:52 pm

listen to some reverb tests (uad, vst and dynamic convolution) at this thread...

http://www.kvraudio.com/forum/viewtopic.php?p=1124251

bmanic · Post by **bmanic** » Sun May 01, 2005 11:09 pm

greendoor wrote:I sincerely hope that convolution technology is simply a temporary step along the road to superior (and much more efficient) modelling technology ...

I agree.. but then again, I've been comparing the our university's Tubetec Compressor (the valve, VCA one, without any fancy VU meters) to the "copy" I made with my sintefex FX 8000 and the difference is minimal. It is very very very very very close.

EDIT: I forgot to add here that the SOUND itself of the copy is very close but the actual compressor behaviour, that is, attack and release, is harder to replicate. It has to be tweaked by ear and can be done very close but not perfect. This is the weakest link of a unit like the sintefex (even though the process of copying the compressor behaviour has nothing to do with convolution, it seems to be an index type of process, analysing the knee and then writing the 'position' into a text file/index). This compressor beharviour process should be actually quite possible to code as a VST plugin! Could be that this is exactly what the TriToneDigital guys are doing with their compressor.. we'll see. Also, does anybody know if this is patented?

.. but, modelling is of course the future, and far more interesting, especially true component modelling as it gives talented coders the freedom to re-create and make hybrids of old cool gear. Also, it should be even better sounding in the end as dynamic convolution is only as good as the amount of samples (2048 in 44,1/48kHz and 1024 at 88,2/96khz) and layers used (in case of the sintefex it's 1 sample taken per dB, starting from whatever output volume you want to set and then counting down 40dB).

Cheers!
bManic

bmanic · Post by **bmanic** » Sun May 01, 2005 11:15 pm

away from home wrote:listen to some reverb tests (uad, vst and dynamic convolution) at this thread...

http://www.kvraudio.com/forum/viewtopic.php?p=1124251

I really don't see the point in using dynamic convolution for reverb.. could you explain why this would be desired? It could work as an interesting 'special effect' but doesn't really help in making it more realistic, or?

- bManic

Dynamic convolution plugin