Hartmann Neuron: Does a good demo exist?

[pricklyrobot]

Caliber and stuff - really nice, but Hartmann should've sold its IP to SonicCore or something.

Yeah, the people who can't be bothered (three years and counting since the last firmware updates) to fix any of the numerous bugs in the ASB's that they took over—rebadged and are still selling new—from Creamware; they're the perfect people to refine a glitchy product.

Sorry, I'm still bitter about having to sell my Minimax due to serious MIDI glitches; and the fact that Sonic-core never bothered to respond to any of my numerous e-mails on the subject (you can go check out the the Planet Z forums for further confirmation that Sonic-core's development and support—as least as pertains to certain products—is fairly nonexistent).

[23]

I thought the V-Synth was the king of sample mangling.

The Vs are powerfull no doubt, though I'll confess that it was the resynthesis abilities that really got me with the Neuron. When it came to the Neural network concept, in all honesty, I have no idea what that was all about, so whether or not that it led to some mindblowing new things I have no idea. Actually, all in all I have no idea on anything about the mythical (from where I stand) beast called the Neuron.

Shame to hear about how poorly the software and it's controller may have been rolled out prior to bankruptcy filing.

from what I understood though, I'm not sure whether or not the Neuron was even really a sampler.
I mean once a sample is resynthesized, it leaves the sample domain from what I understand, and instead what your left with is this model that's been built up through a fashion somewhat akin to additive synthesis.
I really can't say I completely understand the concept.

I've worked with resynthesis lightly, and all I can say is that I found the results "weird". I hope to dig into it more at some point, but as stands, I haven't messed with it enough to say anything other than that I think it may be a pretty interesting area to explore.

[urgetoplay]

The Neuron does Resynthesis on a macro level, rather than a micro : Here's a better explanation culled from the helpful people at www.surroundsfx.com
The NAS, short for Neuron Audio System, is the synthesis engine in the Neuron. It was developed by Stephan M. Bernsee at Prosoniq.
File under "resynthesis with a twist". While most other resynthesis techniques are based upon breaking up the input sound into sinusoid ( = "sinewave-like") components and then, after processing the resulting spectral data, adding these back together, the NAS uses neural networks that analyse the input sound to create a physical model that has the original sound as output. It's like listening to a recording and reverse-engineering the gear that was used to make it. You can then manipulate the properties of this model.Obviously this technique can only approximate the real source (or may fail to even get close when using wrong settings, be it intentionally or not), but it gives the user access to soundmangling possibilities that simply wouldn't be feasible to program with sinusoid-based resynthesis methods. While you don't get surgical access to spectral content with the NAS as opposed to sinusoid techniques, you get macro-like access which i find a lot more intuitive.
So from a production situation point of view, you'll probably want to have both in your arsenal.
You can read what the creator has to say about it here:
http://neuron.prosoniq.com/

[ThinkTanx]

The Neuron does Resynthesis on a macro level, rather than a micro : Here's a better explanation culled from the helpful people at http://www.surroundsfx.com
The NAS, short for Neuron Audio System, is the synthesis engine in the Neuron. It was developed by Stephan M. Bernsee at Prosoniq.
File under "resynthesis with a twist". While most other resynthesis techniques are based upon breaking up the input sound into sinusoid ( = "sinewave-like") components and then, after processing the resulting spectral data, adding these back together, the NAS uses neural networks that analyse the input sound to create a physical model that has the original sound as output. It's like listening to a recording and reverse-engineering the gear that was used to make it. You can then manipulate the properties of this model.Obviously this technique can only approximate the real source (or may fail to even get close when using wrong settings, be it intentionally or not), but it gives the user access to soundmangling possibilities that simply wouldn't be feasible to program with sinusoid-based resynthesis methods. While you don't get surgical access to spectral content with the NAS as opposed to sinusoid techniques, you get macro-like access which i find a lot more intuitive.
So from a production situation point of view, you'll probably want to have both in your arsenal.
You can read what the creator has to say about it here:
http://neuron.prosoniq.com/

=P~