Physical modelling in 2017

[woggle]

I'm planning on checking put Madrona Labs Kaivo, but was wondering what else was out there that's based on physical modelling, but takes a step up from the previous generation?

Kaivo is great

Is it capable of going beyond the sounds I hear on youtube demos (including Madrona's own channel)? I think the various demos all sound similar. I have an Aalto license but have not sprung for Kaivo because of this.

I've not worried about the demos, for me kaivo has a lot of versatility - but I also have AAS chromaphone too and would use that for more conventional sounds. What sounds are you looking to make that Kaivo does not seem to cover?

[cron]

check this out, for instance :

https://physicalaudio.co.uk/PA3.html

Wow, a new comer to the Physical Modelling world. So exciting!

Oh wow, just clocked this is by some of the guys from the NESS research group. The NESS project was pretty much right on the cutting edge as far as physical modelling synthesis is concerned. Great to see some of those concepts trickling down into a more user-friendly format. Gutted there's no Windows version. Hopefully that's coming.

Incidentally, I believe Kaivo is using some of the FTDT (finite difference time domain) concepts that NESS was mostly focused on, although optimised to run real time on a general purpose processor rather than the massively parallel graphics-type hardware NESS was targeting.

[aMUSEd]

check this out, for instance :

https://physicalaudio.co.uk/PA3.html

Wow, a new comer to the Physical Modelling world. So exciting!

Oh wow, just clocked this is by some of the guys from the NESS research group. The NESS project was pretty much right on the cutting edge as far as physical modelling synthesis is concerned. Great to see some of those concepts trickling down into a more user-friendly format. Gutted there's no Windows version. Hopefully that's coming.

And they also have a modelled plate reverb that sounds great

[beely]

check this out, for instance :

https://physicalaudio.co.uk/PA3.html

Wow, a new comer to the Physical Modelling world. So exciting!

Oh wow, just clocked this is by some of the guys from the NESS research group. The NESS project was pretty much right on the cutting edge as far as physical modelling synthesis is concerned. Great to see some of those concepts trickling down into a more user-friendly format. Gutted there's no Windows version.

I had a quick play with this - it sounds in the ballpark to Sculpture, and seems to do similar things, although there's a cymbal percussion preset that while not perfect is scarily impressive. It shares the slightly chaotic and unpredictable pitch behaviour that I've often had with Sculpture, which seems to be a part of the modelling and is very hard to dial out, making something more pitch stable often times a slightly frustrating process. The user interface is pretty baffling but it would need a bit more time and manual reading to get into it.

I prefer Sculpture overall at this point but that's because I'm more into creating playable musical sounds that just experimenting with PM for just sound design purposes, and I'm a lot more familiar with it.

[Gribs]

I'm planning on checking put Madrona Labs Kaivo, but was wondering what else was out there that's based on physical modelling, but takes a step up from the previous generation?

Kaivo is great

Is it capable of going beyond the sounds I hear on youtube demos (including Madrona's own channel)? I think the various demos all sound similar. I have an Aalto license but have not sprung for Kaivo because of this.

I haven't checked out the sound demos, but I've had Kaivo for ages and it really depends on which bits of it you're using and how you're using them.

There are basically 3 ways to get playable pitch out of Kaivo: repeating grains at audio frequencies (e.g. 440 times per second for A), using samples that already contain pitched tones and repitching the grains themselves, or using the resonator. It just depends on where you've routed pitch from the key section (although you can route to any combo of these or even elsewhere).

Thank you for the discussion and interesting information.

Maybe Inwill give the demo a try and see what it can do.

[Gribs]

I'm planning on checking put Madrona Labs Kaivo, but was wondering what else was out there that's based on physical modelling, but takes a step up from the previous generation?

Kaivo is great

Is it capable of going beyond the sounds I hear on youtube demos (including Madrona's own channel)? I think the various demos all sound similar. I have an Aalto license but have not sprung for Kaivo because of this.

I've not worried about the demos, for me kaivo has a lot of versatility - but I also have AAS chromaphone too and would use that for more conventional sounds. What sounds are you looking to make that Kaivo does not seem to cover?

I am not looking for anything in particular. I just like to experiment. I have chromaphone and Pianoteq stage. I use the latter for ordinary piano playing... or at least I was using it until I had a major job change and cross-country move. It is taking me a while to get my home studio set up again.

[cron]

Good stuff. As an Aalto owner, you'll probably already be reasonably familiar with the mode of operation as they're pretty similar in terms of the user experience. For instance, in Aalto you can route pitch from the key section into the delay time, crank the feedback, and 'play' the delay like an oscillator (perfectly tuned if you leave the modulation amount at the default setting). It's much the same in Kaivo, except with a highly tweakable resonator. Routing pitch to the delay time was generally how I played Aalto in the first place, so Kaivo was like mana from heaven for me as it really took that process to the next level.

[Ploki]

damn. Physical modeling is progressing slower than I had hoped it would when I first discovered it.

[masterhiggins]

damn. Physical modeling is progressing slower than I had hoped it would when I first discovered it.

Probably because it's not useful enough for trance or dubstep to increase demand.

[Krakatau]

damn. Physical modeling is progressing slower than I had hoped it would when I first discovered it.

Probably because it's not useful enough for trance or dubstep to increase demand.

likely...

[cron]

damn. Physical modeling is progressing slower than I had hoped it would when I first discovered it.

Probably because it's not useful enough for trance or dubstep to increase demand.

likely...

I'd say that physical modelling has kept up with other forms of synthesis just fine in terms of DSP innovation. I suspect the real problem is that the name "physical modelling" instilled/instils unrealistic expectations of what was/is to come. As though a perfect copy of an instrument, somehow expressed in terms of code and mathematics, would be available to use. Cutting edge techniques like the FTDT methods used in Kaivo are unfathomably complex compared to the Karplus-Strong and digital waveguide methods used 20 years ago, yet they're still enormously simplified abstractions of the things that happen when you pluck a guitar in the real world. And that's just one string we're talking about.

[Tony Ostinato]

agreed, it needs chaos and turbulence modeling.

i have 2 vl70ms for use with windsynth and that would be my gripe with them.

the guy who does this:

http://www.vl70m-editor.com/site/

says he is working on a vl synth that will go beyond what the yamaha vl series does.

[electro]

damn. Physical modeling is progressing slower than I had hoped it would when I first discovered it.

IK's MODO Bass and Moddart's Pianoteq 5 are examples of extremely good PM.

[pdxindy]

damn. Physical modeling is progressing slower than I had hoped it would when I first discovered it.

IK's MODO Bass and Moddart's Pianoteq 5 are examples of extremely good PM.

Sure there are a few examples, but still, the pickings are on the slim side

[electro]

WIVI does Orchestral Brass pretty well and sampled perc is always best. Bowed Strings are the big void.