Eventually something like THIS?: (That would be really, really great and forward-looking!!!)Urs wrote:We do have a pretty cool concept for some sort of modal synthesis, which aligns nicely with the resonator concept. Wait and see
Zebra3 Info
-
- KVRian
- 1353 posts since 31 Mar, 2014
-
- u-he
- Topic Starter
- 30186 posts since 8 Aug, 2002 from Berlin
Hmmm, this is an interesting concept, but I'm not sure if we would like to dive into academia that much. I'm pretty sure what they do might be possible in Zebra3, but not necessarily on the level of material and shape. It'll rather be on the level of excitation and body, with methods to influence both dynamically.
-
- KVRian
- 1353 posts since 31 Mar, 2014
Yeah, that's exactly what I meant. Sure you can't implement 3d bodies and so on. But what I find interesting about that is that you can hit the bodies at different points and the sound varies in a quite natural and 'human' way. So in a software instrument it would be an intuitive way.
Of course it should meet the requirements if it's implemented as a knob to turn, XY-pad or whatever you're planning to do...
The other thing I like about it is that you can chose different excitors and resonators (what you'll have in mind already), but not only based on algorithms - also on sample-analysis. That's also really intuitive...
Another thing I came across is the Step sequencer/arpeggiator of Dune 2 where you can import whole MIDI sequences...
Of course it should meet the requirements if it's implemented as a knob to turn, XY-pad or whatever you're planning to do...
The other thing I like about it is that you can chose different excitors and resonators (what you'll have in mind already), but not only based on algorithms - also on sample-analysis. That's also really intuitive...
Another thing I came across is the Step sequencer/arpeggiator of Dune 2 where you can import whole MIDI sequences...
-
- u-he
- Topic Starter
- 30186 posts since 8 Aug, 2002 from Berlin
Dunno yet. Enough developers to make a stand alone we might have.joeyluck wrote:Will Zebra3 have an option to be standalone? =D
-
- KVRAF
- 3329 posts since 18 May, 2003 from Sweden
Just use Savi Host on Windows or Audio Plugin Player on the Mac. That way, you can also add any additional FX that you want.
/Joachim
/Joachim
If it were easy, anybody could do it!
-
- KVRer
- 1 posts since 12 May, 2014
Having the XY controls as an overlay to the main synthesis window would be extremely useful (could be a lot smaller) as would something similar to the macro controls in massive.
Audio in would also be a nice feature and (imo) multiple outs would be a godsend
.
Standalone has already been mentioned but I'll give it a +1 (with a fullscreen option).
Audio in would also be a nice feature and (imo) multiple outs would be a godsend
.Standalone has already been mentioned but I'll give it a +1 (with a fullscreen option).
-
- KVRian
- 1115 posts since 6 May, 2010 from Munich, Germany
And don't forget a little Thank you to Herman. The Father of SaviHostSpitfire31 wrote:Just use Savi Host on Windows or Audio Plugin Player on the Mac. That way, you can also add any additional FX that you want.
/Joachim
-
- KVRAF
- 3329 posts since 18 May, 2003 from Sweden
Herman is The Man!
/Joachim
/Joachim
If it were easy, anybody could do it!
-
- KVRAF
- 1895 posts since 13 Oct, 2002
I had this thought months ago, but at the time I wasn't sure it was appropriate or useful. However after thinking about it for some time, it seems like the idea has some value for Z3:
For years I wanted to see distortion at the voice level; how better to get heavy per-voice distortion without the usual IM distortion crud? And after seeing how well that worked in Zebra, it occurred to me that other effects could be implemented at the voice level, like delay/verb, chorus/flanging/phasing, granular effects, even simple EQ, etc... The parameters on these effects could be modulated by the usual sources, but the most useful would be key follow which would scale the effect parameters appropriately to each voice. Along with a wet/dry mix...
In essence what I'm suggesting is real polyphonic "effect" modules. I think the rich creative potential for this is huge, but unfortunately I also realize that the CPU consumption will likely be as well.
Anyway: "For your consideration..."
For years I wanted to see distortion at the voice level; how better to get heavy per-voice distortion without the usual IM distortion crud? And after seeing how well that worked in Zebra, it occurred to me that other effects could be implemented at the voice level, like delay/verb, chorus/flanging/phasing, granular effects, even simple EQ, etc... The parameters on these effects could be modulated by the usual sources, but the most useful would be key follow which would scale the effect parameters appropriately to each voice. Along with a wet/dry mix...
In essence what I'm suggesting is real polyphonic "effect" modules. I think the rich creative potential for this is huge, but unfortunately I also realize that the CPU consumption will likely be as well.
Anyway: "For your consideration..."
-
- u-he
- Topic Starter
- 30186 posts since 8 Aug, 2002 from Berlin
Yep, *some* per-voice fx stuff may be possible.
The main reason for me to not do it is not CPU usage. It's memory consumption. If we have to pre-allocate, say 2 MB of sample RAM for each delay, reverb and comb filter, times maximum number per voice times maximum number of voices, then we end up with hundreds of MB per instance.
We have however introduced many new technologies "under the hood" - in preparation of things to come. One of which is a timer-based solution to allocate memory for delays etc. on the fly. Because that's really tricky. If one asks the system to reserve some memory during a process call, a MIDI event or during a parameter/preset change (latter only in VST3), then the host will crash. Dynamic memory allocations - as well as de-allocations - have to happen "on the side", in a low priority thread.
Stuff like that is being investigated and tested. Once it's "good to go" we'll certainly look into using it for exactly such ideas as per-voice-delays
- Urs
The main reason for me to not do it is not CPU usage. It's memory consumption. If we have to pre-allocate, say 2 MB of sample RAM for each delay, reverb and comb filter, times maximum number per voice times maximum number of voices, then we end up with hundreds of MB per instance.
We have however introduced many new technologies "under the hood" - in preparation of things to come. One of which is a timer-based solution to allocate memory for delays etc. on the fly. Because that's really tricky. If one asks the system to reserve some memory during a process call, a MIDI event or during a parameter/preset change (latter only in VST3), then the host will crash. Dynamic memory allocations - as well as de-allocations - have to happen "on the side", in a low priority thread.
Stuff like that is being investigated and tested. Once it's "good to go" we'll certainly look into using it for exactly such ideas as per-voice-delays
- Urs
-
- KVRAF
- 1769 posts since 30 Jul, 2007
+1Urs wrote:Yep, *some* per-voice fx stuff may be possible.
The main reason for me to not do it is not CPU usage. It's memory consumption. If we have to pre-allocate, say 2 MB of sample RAM for each delay, reverb and comb filter, times maximum number per voice times maximum number of voices, then we end up with hundreds of MB per instance.
We have however introduced many new technologies "under the hood" - in preparation of things to come. One of which is a timer-based solution to allocate memory for delays etc. on the fly. Because that's really tricky. If one asks the system to reserve some memory during a process call, a MIDI event or during a parameter/preset change (latter only in VST3), then the host will crash. Dynamic memory allocations - as well as de-allocations - have to happen "on the side", in a low priority thread.
Stuff like that is being investigated and tested. Once it's "good to go" we'll certainly look into using it for exactly such ideas as per-voice-delays
- Urs
-
- KVRAF
- 26937 posts since 3 Feb, 2005 from in the wilds
I'd so love to have the HZ resonator as a per voice effect... plus 1 eq...Urs wrote:Yep, *some* per-voice fx stuff may be possible.
The main reason for me to not do it is not CPU usage. It's memory consumption. If we have to pre-allocate, say 2 MB of sample RAM for each delay, reverb and comb filter, times maximum number per voice times maximum number of voices, then we end up with hundreds of MB per instance.
We have however introduced many new technologies "under the hood" - in preparation of things to come. One of which is a timer-based solution to allocate memory for delays etc. on the fly. Because that's really tricky. If one asks the system to reserve some memory during a process call, a MIDI event or during a parameter/preset change (latter only in VST3), then the host will crash. Dynamic memory allocations - as well as de-allocations - have to happen "on the side", in a low priority thread.
Stuff like that is being investigated and tested. Once it's "good to go" we'll certainly look into using it for exactly such ideas as per-voice-delays
- Urs