Prophet VS sounds with Zebra?

[PietW.]

Had a look at the U-He library now. Are the "Waves Of Plenty" waveforms in the U-He library from a real VS or a Wavestation?

If memory serves, they were from the evolver, extracted by dumping its memory.

Nice to see you here suthnear! I was searching the KVR but didn't find
any thread about your wavesofplenty set. Thanks again!

As i had PietW osc set in my Zebra made a patch to try the MSEG thingy.
Just for fun, with only 2 osc and 2 MSEG.
-Forgive me with the WETNESS of the Fx
even more WET with Pw up.

"VS Vectorbells".

http://www.mediafire.com/?thw49196kmqnkz4

Cheers,
Joseph

Finished to sample the VS-Waveforms. Now Start to Convert 121 Single Cycle Waves to h2p.

[hakey]

Each stage of the envelope would need three settings: the x and y coordinates, plus the time taken. A fourth option could be the type of motion and acceleration - linear vs curved, etc.

Two envelopes would do the same job, I think?

By a recordable I mean the capture and playback of mouse/cursor movements. Similarly, animatable XYs would allow vectors between points to be drawn onto a graph.

Just novel ways of inputting information - hence, 'gimmicky'.

(One of the Reaktor factory ensembles - 'Big Green Synth'? - has a recordable XY. iirc, Howard did the presets for it.)

[Ingonator]

I want to convert all single cycle waves from Korg Wave Station to Zebra.
Importance of good quality. I have tried to convert the files to my good quality one as I was.
The file contains the first 6 VS Welenformen VS35-VS40.
It would be nice if you could assess the quality..
Korg in Marburg has given permission to convert the waveforms.

http://www.file-upload.net/download-794 ... 0.h2p.html

cheers
Piet

Already done and published here yesterday. Here are the Zebra files:

https://dl.dropboxusercontent.com/u/532 ... 0waves.zip

Those were done from my Wavestation EX with bypassed filter.

[

I checked and i found like if you had a softer keytracking in the lower
register. So at C1 it's clearly much softer and more quiet than the Augur or
the wavesofplenty or Ingo wavestation and V set. From C3 and up it's perfect.

All of the sets is different mostly around C1-C2.

Your set is softer a bit, much softer in the low region but keeping a warm body
of the sound.
Somehow Augur also keeping this body besides it's more harsh, crispy (C1-C2)
Wavesofplenty are brigter everywhere
Ingo WvestationVS are also brigther but thinner around C1-C2

I will be happy for a whole wavestation set.

No idea why those from PietW should sound different except if some processing is involved. I sampled the raw waveforms with bypassing the filter. No further processing involved. Was sampled directly into Wavelab Elements 7 and edited there to get a single cycle waveform.

Maybe Piet used the filtered waveforms (at maximum Cutoff) which are more "smooth" as i found too.

I included the waveforms 032-034 (the "standard waveforms" like e.g. Saw) which are not included as the VS versions but i took the standard ones from the Wavestation.


Ingo

[zerocrossing]

I actually think that the Arturia model gets pretty damn close and if you can discard your need for a perfect emulation it's a pretty good sounding synth.



Smokey.



Funky.

[Ingonator]

I more or less agree. Even if we could do vector synthesis in Zebra now it would not mean it is more authentic than the Arturia one (while the waveforms from Augur or the Wavestation are maybe better).

I guess that we should see the approach in Zebra independent from how authentic it really is.

Currently Arturia Prophet V seems to be the best available emulation as other like Augur (which is very nice) and Vector Sector are discontinued.


Ingo

[PietW.]

Hi Ingo,

I can not tell you that. I sample the waveforms, no loop and cut to exactly 128 samples. Normalize to-2dB. At first I was normalized to o dB and as the result was poor. No Filter or Effekt from the WS or Waveosaur.

cheers
Piet

[Ingonator]

Hi Ingo,

I can not tell you that. I sample the waveforms, no loop and cut to exactly 128 samples. Normalize to-2dB. At first I was normalized to o dB and as the result was poor. No Filter or Effekt from the WS or Waveosaur.

cheers
Piet

I had sampled at C1 and length of the single cycles was around 31 milliseconds (= around 1347 samples). 31 ms is the real size of a single cycle of the samples at C1. If yours is 128 samples that would explain a few things. Even at C3 i would have more than 128 samples. Did you reduce the size yourself?
Amplitude in mine was normalized to -1 dB.

When comparing the sample waveforms you published to mine (i had started with 032 as i included the "standard" waveforms") they look quite different. Will be interesting to see the final collection.


Ingo

[pdxindy]

Of course I love Zebra for what it is setup to perform already (limitless IMO) and I'm not trying to deface/derail it by requesting such consideration. I think that anyone who has spent some time programing these vector synths (software and/or hardware) hopefully understands when I say that going from init to a complete patch (using VS) is much more immediate than to fanangle a likewise patching scenario in Zebra, as it is. So, in conclusion, if there is too little or no interest by the development team then absolutely no sweat.

I did not find the Wavestation (I owned the hardware keyboard version) easy compared to Zebra...

One can get into a particular thinking. What I mean is, trying to recreate the VS structure instead of considering the sound itself.

With Zebra we have already remarkable capabilities. With a single Osc you can morph between waveforms (not just crossfade). Then you have the OscFX... 2 of them... also PW/Sync... we already have morphing vector synthesis with a single Osc!

This is one quick example I just made using a single Osc and recording while I dragged the X/Y control around. There is a filter to shape the sound a bit, but no filter modulation at all. All the movement is just the Osc controlled by the X/Y pad.

http://draigathar.org/sounds/XYmorph.mp3

[PietW.]

Hi Ingo,

I can not tell you that. I sample the waveforms, no loop and cut to exactly 128 samples. Normalize to-2dB. At first I was normalized to o dB and as the result was poor. No Filter or Effekt from the WS or Waveosaur.

cheers
Piet

I had sampled at C1 and length of the single cycles was around 31 milliseconds (= around 1347 samples). 31 ms is the real size of a single cycle of the samples at C1. If yours is 128 samples that would explain a few things. Even at C3 i would have more than 128 samples. Did you reduce the size yourself?
Amplitude in mine was normalized to -1 dB.

When comparing the sample waveforms you published to mine (i had started with 032 as i included the "standard" waveforms") they look quite different. Will be interesting to see the final collection.


Ingo

Hi Ingo,
in the Wavestation start the VS waveforms with 35 and ending with 155
I sample with F2 on 128 samples. Jupiter8 wave2zebra of the programmer writes It gets resampled to a length of 128 samples so the closer to that the better.

[Ingonator]

Hi Ingo,
in the Wavestation start the VS waveforms with 35 and ending with 155
I sample with F2 on 128 samples. Jupiter8 wave2zebra of the programmer writes It gets resampled to a length of 128 samples so the closer to that the better.

Not sure if you loose details starting from a waveform with 128 samples as you already loose details while resampling to 128 samples. Anyway i also wanted to use the WAV files and it's better having those at higher quality.


Ingo

[Howard]

So, in conclusion, if there is too little or no interest by the development team then absolutely no sweat.

Vector synthesis wouldn't really add anything to Zebra IMHO, so I reckon it's unlikely to be included in Z3. These days, there are much more exciting avenues to explore...

[xh3rv]

Played around with 4 lanes on 1 XY similar to Howard's 4 Osc on 1 XY and discovered a pretty fatal flaw in my thinking, it breaks the ability to put per-voice parameters on, like, a filter at the end of the signal chain

Just thinking out loud, but there could be a sort of VMix inversion of the MMix modulation source. It could multiplex 2 inputs (like MSEGs) to 2 or 4 outputs (I can see it being a little easier to wrap one's head around 4 - e.g. OSC1-4 just have 4 independent 0->max modulations this way, rather than 0->max and max->0 for opposite poles. Also might be some benefit when doing multiplicative versus than additive application of modulation). And a little diamond with I, II, III, IV labeled on the GUI that tracked the input->output transformation. Pretty sure there are some dangerous implications I don't foresee but it popped in my head. [e] Plus clearly it's not too exciting to Howard and Urs

[Kriminal]

Double post

[Kriminal]

Is that legal ?

Copying is illegal. Convert is not illegal.
When the waveform is converted, it will also be changed automatically. This waveform so that no longer matches the original.
If you copy the waveforms directly from the ROM of the Wavestion this is illegal.

Says the German representation of Korg in Marburg.

So you are copying the original, then converting it

Sounds illegal to me...

[electro]

I actually think that the Arturia model gets pretty damn close and if you can discard your need for a perfect emulation it's a pretty good sounding synth.



Smokey.



Funky.

How bad can this thing be if people are trying to get Zebra to do a VS emulation over it?