Korg Forums

[Daz]

[Daz]

This is like the mother-in-law joke of FM programming ... the cliche bell/marimba like tone using just two sine oscillators with one modulating the other :

http://www.korgforums.com/misc/gt_fm.mp3

Just playing a couple of notes in each octave ... the sample rate is 48kHz so if plays with an odd speed/pitch that's why.

Daz.

[ski]

In my previous post I wrote:

[domc]

Thanks to all of you who posted replies while I was away last night.

I think I get it now - I was getting confused thinking that the 2 inputs to each of the VPMs would be assigned to be either carriers or modulators.

Now its clear that each (at its basic level) sine wave generated by an oscillator acts as a carrier (when connected to the mixer) - and is modulated by both the inputs if lets say they come from other VPMs.
And at its more advanced level the sine wave can be changed as ski just mentioned to sine+waveshaper etc.


So.........

Lets say VPM3 is connected to the mixer. And VPM 1 is connected to input 1 of VPM3 thereby acting as a modulator. Does the sound stay the same if I switch the input from input 1 to input 2 of VPM3?
What confuses me now is that on the signal path diagrams shown on each VPM Osc page it looks like there's a different path for input 1 and input 2, implying some kind of different output which is why I was thinking one was a carrier, one a modulator, rather than both acting as modulators.

Kind rgds, Domc
_________________
Oasys 88
Kronos 88
Virus TI Keyboard
Octopus

[ski]

[Sina172]

...........

[billysynth1]

I'm very close to starting a fight with Kurzweil/Sonikmatter/David Weiser.
This person has a bad habit of making derogitory comments about all other manufacturers. Someone stop me please.


http://community.sonikmatter.com/forums/index.php?showtopic=37159


Billy
_________________
Yamaha C1 Grand Piano.
Korg Oasys 88, Jupiter 80
Kronos 88, V Synth GT
I am a student of classical piano...I am not a classical pianist.

[danatkorg]

[Mike Conway]

They seem to be stuck on the 4 way sample split, which is in VPM mode.

Billy, you don't have to argue, but you could point out that a single wave sequence can be used as a 64 layer sample split, via velocity. Of course, a single oscillator can have 2 wave sequences...

[danatkorg]

[Mike Conway]

I was thinking of Layer mode. Up to 8 in a dual osc program - even better!!

[jerrythek]

There's a good path to take in these instances, it's called "the high road".


Kurzweil has made and I'm actually hopeful will continue to make some great products. We have great respect for them and wish them success. I know Dave W. and others there and wish them well. Each company in the industry that innovates and develops good ideas only serves to help the category as a whole, and you the users.

We all have innovations and strengths, and I find it a bit foolish to start picking out one element of a spec and battling them. For every one in the win column there might be one in the loss column. And it goes on and on, in circles.

And then there's the age-old forum habits of saying things like:

"I like Motts Apple Sauce. Anyone who disagrees with me is wrong."

And you know what? They are!! Think about it...



Regards,

Jerry

[domc]

Thanks ski for your helpful explanations. I think I understand enough of the rudiments of FM to be dangerous on Oct 1st.

and so.........

next up for some brave soul is to explain the waveshaping. How exactly does this modify things - and what happens when you process PCM through the waveshaper - won't that mix in the sine wave of the VPM Osc - as the PCM will be acting as a modulator on the that VPM OSC. Or this expected to happen and is what you mean when you talk about processing PCM with a waveshaper.

and then finally ring modulation.

Sorry for all the questions but I thought I might as well pass these days while we're waiting for the new module finding out about it so that I hit October running
_________________
Oasys 88
Kronos 88
Virus TI Keyboard
Octopus

[Daz]

Hey Dom,

With the waveshaping of PCM don't forget that each of the VPM oscillators can run in a variety of modes, one of which has only the waveshaper active (i.e. the oscillator part is turned off). In that mode the VPM oscillator become just like an effect, you feed audio in and a processed version comes out the other side.

Hows does a waveshaper work ... tricky one to explain. Are you familiar with the Karma note mapping tables ? It kinda works like that. With the note mapping table you can specify how incoming notes should be mapped to outgoing notes. So you might specify that a C4 coming in maps to A6 going out, a C5 is mapped to G5, an A4 to a F#5 etc. etc. A waveshaper works similarly but acts on individual sample points of an input waveform, mapping their amplitude to a new value which is then output. As a waveform, in digital audio terms, is just a stream of numbers representing the amplitude at a given point in time, if you start remapping those numbers to different numbers in realtime the result is that you will change the shape of the waveform.

The effect this has on more complex PCM waveforms varies substantially based on the PCM waveform you choose and the Waveshaper shape (i.e. mapping table) you select. The results can vary from just straight garbage, to rather pleasant timbral transformations. Experimentation is of the essence. I have had the best results so far using quick attack/decay sounds like guitars and basses. I have also found that putting a filter between the PCM oscillator and the input of the waveshaper can help as the simpler the signal going into the waveshaper, the less chaotic the output is

The most immediate use of the waveshaper is to transform the simple sine wave output from a VPM oscillator into other more interesting waveforms. When you do the Drive Parameter varies the overall transformation and the results are more coherent and interesting timbral shifts.

As with most of these kinds of things subtlety is key, getting too enthusiastic with the waveshaper will just generate a lot of rather noisy yuck.

Daz.

[Daz]