u-he drum synth please!

[Urs]

they're using, afaik, a prefab CMS system, so the programming of the web page has nothing to do with them also—it must be said—you're replying to a post that's 3 years old

No CMS. We have a home grown template system that renders into static web pages. No active content equals ultimate data protection.

The long loading times are due to preloading Youtube videos. Because we've been eagerly producing them, some pages now have like twenty of those. So we might switch to load-on-click instead, if it really is such a big problem

[sleepcircle]

oh, tut tut

(i'm kidding. i've never actually noticed any delay, personally, but also I grew up on a dial-up modem so, like, everything after that has been "fast as lightning" for me.)

[sascha]

Maybe a bit offtopic, but today I wondered again why there still is not a single drum VSTi available which provides a similar quality like the korg wavedrum, using physical modelling + mic + samples.

Not virtual, but there's a physical thing, and it's called eMerge, a collab between Pearl and Korg. Based on the Wavedrum, and not released yet, but you can already pre-order a full-sized kit for a laughably cheap retail of around 4200€.

[Ploki]

i'd just like to remind people that at one point this thing existed, and gently prod u-he to reconsider it

[pierb]

i'd just like to remind people that at one point this thing existed, and gently prod u-he to reconsider it

Would love to see that released, but it seems IK already released their physical modeling drum plugin first.



https://www.ikmultimedia.com/products/mododrum/

[bonch]

i'd just like to remind people that at one point this thing existed, and gently prod u-he to reconsider it

My understanding based on previous posts is that research will be used to expand the physical modeling capabilities of Zebra 3.

[sascha]

As if it were that easy.
They don't drop their pants about the actual technique in MODO (you can use the term 'modeling' for almost anything), but I'm 99% sure it's basically combining small snippets of samples and some delay-based resonators (if at all). And for those familiar with top-of-the-line Roland modules (TD-30, TD-50(x)) this doesn't feel novel at all.
You can also tell by the cymbals (all samples) and the still quite large library footprint that they never considered applying non-LTI techniques. This was the very reason why we, back at the time, put the project on-hold. You just can't replicate a complex nonlinear system like cymbals with strictly LTI approaches. It's applied systems theory, and it's non-trivial. There's still few academic work around on this subject, with poor sonic outcome, plus it's not realtime-capable. At least one has to work with finite elements / finite difference methods (like what other industry fields use for, say, simulating jet streams over a bending aircraft wing). But there's definitely no shortcut using clever delay-based approaches, banded waveguides, modal synthesis and what not. They're all LTI systems. You can make these things sound nice and musically interesting but you'll never make it sound like analogue drums. That's where we considered a hybrid, but coming up with a non-LTI system isn't something you can do one the side while keeping up normal business, and chances are you invest 2 years or more and still fail. It was clearly stated as 'research project', and that's what it was. We've gotten pretty far with simulating membranes, along with the phenomenum like wave impedance change on impact (which is the cause of toms changing pitch with force, and it took me a while to grasp that it's essentially the same as nodal analysis in EE circuits). But it becomes a nightmare when you propagate waves through a change in medium (like membrane to shell), and wavefronts become dependent of a remote medium, and change speed and dispersion. Then the system a) has memory and b ) has additional inputs, so it's no longer linear and numerically predictable.

But I'd love to see some of the stuff in Zebra, too. That's the far better domain, where creativity is the key & limit.

[Juanjo]

...
I'd love to see some of the stuff in Zebra, though, too.

Salivating at the thought of a Zebra3 with some of these algorithms in one form or another (I'm aware I might be salivating for quite a while, but still...)

[Funkybot's Evil Twin]

As if it were that easy.
They don't drop their pants about the actual technique in MODO (you can use the term 'modeling' for almost anything), but I'm 99% sure it's basically combining small snippets of samples and some delay-based resonators (if at all). And for those familiar with top-of-the-line Roland modules (TD-30, TD-50(x)) this doesn't feel novel at all.
You can also tell by the cymbals (all samples) and the still quite large library footprint that they never considered applying non-LTI techniques. This was the very reason why we, back at the time, put the project on-hold. You just can't replicate a complex nonlinear system like cymbals with strictly LTI approaches. It's applied systems theory, and it's non-trivial. There's still few academic work around on this subject, with poor sonic outcome, plus it's not realtime-capable. At least one has to work with finite elements / finite difference methods (like what other industry fields use for, say, simulating jet streams over a bending aircraft wing). But there's definitely no shortcut using clever delay-based approaches, banded waveguides, modal synthesis and what not. They're all LTI systems. You can make these things sound nice and musically interesting but you'll never make it sound like analogue drums. That's where we considered a hybrid, but coming up with a non-LTI system isn't something you can do one the side while keeping up normal business, and chances are you invest 2 years or more and still fail. It was clearly stated as 'research project', and that's what it was. We've gotten pretty far with simulating membranes, along with the phenomenum like wave impedance change on impact (which is the cause of toms changing pitch with force, and it took me a while to grasp that it's essentially the same as nodal analysis in EE circuits). But it becomes a nightmare when you propagate waves through a change in medium (like membrane to shell), and wavefronts become dependent of a remote medium, and change speed and dispersion. Then the system a) has memory and b ) has additional inputs, so it's no longer linear and numerically predictable.

But I'd love to see some of the stuff in Zebra, too. That's the far better domain, where creativity is the key & limit.

Nice seeing you around Sascha. Out of curiosity, have you seen this:

https://www.kvraudio.com/product/exc-te ... n-research

...there's a free version too. Physically modeled snare drum. I thought immediately about the U-he drum project when I saw that. Even checked to see if Sascha's name was on it.

[Cyoon]

Time is such a wonderful thing. A couple of years back I would have placed a u-he drum machine on the top-ten of my wanted list. Today, I don't feel it's that important really. Been messing with making drum sounds in Repro the whole week and it's just plain amazing. Punchy. Deep. Hard. Soft. Wimpy. Plastic. And you can even get quite a cool variety of hihats/cymbals/shakers modulating the noise with an oscillator.

Fantastic stuff.

/C

Do you have some patches to share maybe?