Korg Forums

Anybody interested in a sound design challenge:

I propose designing just one new sound, to be demo'd here and to be given away free to all, with the following specification:

____________________________________________________________
1. The designed program _must_ use:
Mod-7

2. May include:
PCM Oscillator input

3. May include:
Vector Synthesis

4. May include:
Karma - but with one sting in the tail - Karma _cannot_ be used to create a groove, instead it must form an integral part of the actual sound design.

5. There _must_ be an emphasis on real-time dynamic variation on the sound - so if you are not going to use PCM oscillator, Vector Synthesis or Karma there must be lots of novel reatime control of timbre and modulation!!

____________________________________________________________


Deadline for demo is April 30th. You must provide -

1. Audio demo of your program (even a single note)
2. The program for all to download and use
3. Provide some sort of description of the key aspects of how the sound was made


If you wish we can then have a poll vote on the favourite of the designed sounds.

Here's my thinking - I am keen to hear programmed sounds which:

- Go deep into Mod-7 programming
- Combine PCM oscillators with Mod-7 algorithms
- Explore PCM oscillators in Mod-7 in innovative ways - such as using Vector Synthesis, LFO's and MIDI to 'engage' both the PCM oscillator and the Mod-7 program in general, and through keyboard velocity, realtime controllers and the vector joystick.

Some thought that come to mind include:

- PCM oscillators emulation Realtime Convolution and Modulation (RCM) on Yamaha's SY77 and SY99
- PCM oscillators being modulated extraordinarily though Vector Synthesis, and AMS, and hence radically altering the timbre of the Mod-7 algorithm they entre
- Kamra being used in strage ways - manual advance for example, or running MIDI at maximum speed while being radically enhanced by specific Kamra GE's to affect the timbre of the sound. Here I could imagine, for example, Karma affecting Vector Synthesis, or LFO's synced to Karma/MIDI and causing 'comb' type affects as inputs to individual Modulating oscillators - the sky is the limit.

I thought it's be interesting if some of us developed even one program each, gave it away and provided an explanation - it might open up the current palette of sounds available and spark future ideas and variations for all concerned.

I'll do one anyway - hopefully some of you are on for a challenge - come on - show off!!

Cheers,
Kevin.

Now this sounds like fun ! Thanks Kevin.

Hi Kevin,

+1

good Idea

KARO are on board...

Best wishes

KARO Sound Development
medusaland & EWBR

good idea!! will give it a try but i'm a lot on the arturia origin (programming a bank)

Kevin Nolan wrote:PCM oscillators being modulated extraordinarily though Vector Synthesis

That's kinda tricky since there is only one PCM oscillator (or two if you use two EXi), and vector synthesis doesn't really mean anything other than varying the mix levels of a number of oscillators concurrently using the VJS or the vector envelope. It's damn useful as an extra controller, but "vector synthesis" makes it sound like something so much more exciting

Daz.

Don't forget to get designing a MOD-7 sound!!!

You have just over a week left.

Don't worry how sophosticated it is - just use this as an opportunity to learn about MOD-7, create even one sound and then give it away to the forum (unless you've uncovered 'the lost sound' and want to patent it!)

I'm working on about a half-dozen sounds and doing these have helped hugely in undestanding MOD-7 the engine - it's been a very valuable exercise. My sounds will be far from perfect but I'll aim to make them usable and will place them for download on the 30th of April.

If others make sounds I'll be happy to put them all in one bank and perhaps we can put on the OASYS download page?

So get working - and don't forget to read the MOD-7 tutorial which Dan wrote. It's in the parameter guide MOD-7 synthesizer section and it really is top notch.

Kevin.


PS - forget my 'rules' - just make a MOD-7 sound!!

Fantastic idea Kevin.
Lets go for it-

Dear Colleagues,

Please find 20 new MOD7 programs at:

http://www.knect.ie/OASYS.html

These programs were developed from the ground up - from initialised MOD-7 programs. They are inspired by the Yamaha SY77 and SY99, which were Yamaha's most capable incarnation of FM synthesis. In particular, apart from an extended DX range FM capability (called AFM), Yamaha also introduced 'Realtime Convolution and Modulation' or RCM, whereby PCM samples can be used as modulators in FM algorithms. This is a difficult synthesis technique to get to grips with but ultimately it is extremely powerful, offering a highly original, acoustic-like and dynamic synthesis capability. In my view, due to the 57,000 sounds developed for the SY range of synthesizers, they currently outshine MOD-7 significantly in terms of what's available at the upper echelons of FM synthesis.

That said, MOD-7 is in many ways far more sophisticated than the SY range. It too can engage RCM because you can feed PCM samples into MOD7 FM algorithms. Furthermore, MOD7 algorithms are completely user configurable. However, one of the most important aspects of FM synthesis is the careful shaping of Envelope Generators to individual oscillators (because modulator oscillators affect the timbre), and alas MOD-7 EGs are not as detailed as the SY range, rendering MOD7 a little weaker in this important regard.

A found creating the 20 provided sounds - programming MOD7 is a lengthy process. None of the sounds provided are the complete article - but time is limited and I did say 30th April is the deadline for anyone wanting to join in this effort. Never the less I hope the sounds are complete enough to be useful and educational. Overall they are quite straight forward – there’s not a lot to get too excited about in them – but hopefully they demonstrate the RCM side of MOD-7. Below is a brief description of each program. Note that by default they install into slots 0-19 in User bank


1: KN-MOD7-BahMallet1
This program uses a vocal sample as a modulator, producing an intriguing and vocal-like keyboard-performance sound. It is inspired by a preset on the SY77 called BahMallet, but the SY77's is far more polished and dynamic (due to better programming but also better EGs). I hope to improve this program over time with better velocity sensing and envelope shaping.

2. KN-MOD7-BahMallet2
Variation of 1

3. KN-MOD7-BahMallet3
Variation of 1

4. KN-MOD7-Pipe
Wind instrument used as modulator

5. KN-MOD7-SpaceBell1
Based on an SY99 program SpaceBell. Quite dynamic in velocity response

6. KN-MOD7-SpaceBell-VelLFO
I spent quite a lot of time on this (and Dan Phillips kindly helped information-wise). This uses a stunning feature of MOD-7 (and indeed AL-1 and STR-1) - polyphonic LFO's. That is - for each of the 52 note polyphony of the MOD7, each note has its own unique set of LFO's. So when you adjust LFO1 in a MOD-7 program, you're actually configuring 52 LFO1's. Where this is vital is in modulating the LFOs themselves via note velocity – providing in effect 'polyphonic modulation'. In particular this program adjusts the speed of the LFO according to note velocity - hit faster and the LFO rate is faster. I modify the LFO rate via an AMS Mixer (thanks Dan) where both the note velocity and an EG are multiplied together to affect the LFO rate according to velocity _and_ via an EG (which slows the LFO over the sustain of a held note). Try this by hitting a note fast and then slow to hear the different LFO rates. And of course this is polyphonic because there are 52 separate LFO's being affected - so different notes struck simultaneously with different velocities will have differing LFO rates. This sound is inspired by a similar sound (I believe to be from the Yamaha GS1) used by Vangelis as he solo's at the end of Movement 4 on his album 'Mask'. He uses the effect masterfully and very soulfully. The LFO applied in this program is reasonably constrained; and surely can be used to greater and wilder effect elsewhere?

7. KN-MOD7-SpaceClav-VelLFO
A kind of clav sound based on program 6

8. KN-MOD7-SpaceBell2-VelLFO
Variation on 6 - with aftertouch also affecting LFO rate (though the Velocity LFO rates are still polyphonic, the aftertouch is not polyphonic so it speeds all note up by the same amount)

9. KN-MOD7-SpaceBell3-VelLFO
Variation on 6

10. KN-MOD7-Wurl-VelLFO
Wurlitzer type sound with polyphonic LFO's

11. KN-MOD7-Guitar1
Pure FM plucked guitar sound based in 'Ramirez' - a program for SY99
(still needs refinement)

12. KN-MOD7-Guitar2
Variation on 11

13. KN-MOD7-Guitar3
Variation on 12

14. KN-MOD7-Guitar4
Variation on 13

15. KN-MOD7-Guitar5
Variation on 14

16. KN-MOD7-Nay
An a mono FM sound using a 'Nay' as a modulator

17. KN-MOD7-Bah-Chaos
This sound is barely scratching the surface of gargantuan sounds capable from MOD7 where harmonically rich sounds are fed into FM algorithms full on, and with slowly evolving EGs. I hope to demo more of these in the future - but this sound takes the earlier 'controlled' BahMallet and forces the vocal sound up front in the modulating process. It’s a crude example, but hopefully makes the point. Push the vector joystick to +X to have more presence. As said, this program is barely scratching the surface of what's possible. One SY99 similar sound is on a par with the sound of thunder - as in it seems to offer the experience of the expansive 'might' that thunder can deliver - achieved by low and overpowering harmonics 'breaking into' the FM algoright and shaped by careful EG settings. Forcing PCM harmonics through FM algorithms is capable of huge soundscapes (but takes lots of time to program - and this program, as said, is but a very basic and crude example).

18. KN-MOD7-Fairlight1
FM sound with a Xylophone as a modulator - sounding quite Fairlight like on the bass end of the keyboard

19. KN-MOD7-Fairlight2
FM sound with a Vibrophone as a modulator - sounding even more Fairlight like on the bass end of the keyboard

20. KN-MOD7-Fairlight3
FM sound with an Orchestral Bass Drum as a modulator - sounding quite Fairlight like on the bass end of the keyboard and intriguing on the upper end of the keyboard.


Kevin.

Ok, I just uploaded my entry to the downloads section. Not sure when it will be available though. Is there a better place for us to upload our entries?

My sound is User bank B - 0013, and there is a .MP3 demo.

This was a really great idea - thanks Kevin. I learned so much doing this. Mod-7 is an amazingly powerful engine. I would recommend that everyone try to come up with at least one program, even if you miss the deadline.

We should then continue with the other engines. Working together as a community we can get so much more out of the Oasys.

Thanks for the Mod7 programs, Kevin! It's too bad there isn't a way to automatically convert SY99 sounds to Mod7... (or at least to automate some of the conversion process). I have a 99 and a 77, and there's definitely some patches on them that I'm still attached to.

Thanks for creating these sounds! I particularly like "Wurl-VelLFO" and "Fairlight1."

Kevin Nolan wrote: However, one of the most important aspects of FM synthesis is the careful shaping of Envelope Generators to individual oscillators, and alas MOD-7 EGs are not as detailed as the SY range, rendering MOD7 a little weaker in this important regard.

I'm guessing that this refers to the SY's seven-segment EGs, in comparison to the MOD-7's five. Looking through the sounds, I noticed that the envelope segment curvatures were all either set to the default values, or to linear. If you feel limited by the EGs, it might be helpful to experiment with curvature.

Often, two or more segments of multi-point envelopes are used to approximate curvatures; with parameterized curvatures, such shapes can be created more accurately, and within a single segment.

Best regards,

Dan

Hi Dan -

Thanks for the pointer on the EG's. You are of course correct - there are other things I can do to affect the EGs - and the slopes are veryimportant. Another thing I want to look into is assigning AMS to, say a square wave or the 'step function' LFOs and have them adjust EG stages (I think you suggested this sometime ago in one of the forums 'lost' posts?).

So yes - the SY range offer more stages (and the capacity to repeat stages which is very important) - but hopefully with more experimentation on the MOD-7 EGs there will be more sophosticated outcomes to some of the programs. EG's are just so important to FM synthesis because the shape the timbre over time and can respond to performance dynamics.

Overall MOD-7 is a beast - and a sleeping one at that. It seriously needs lots of programmers and programmig hours yet (though Korg's programs are faubulous). I was a little harsh about MOD-7 over the past year or so, but it is just fantastic and enormous in potential and I take my hat off to you. And like with the SY range, it simply will not date - I know that in 30 years time I can return to it and it will still feel fresh and enormous in potential.

Cheers,
Kevin.

Kevin, many thanx for the new mod7 sounds - looking forward to hearing them!!

Just listened to you creations. My favorites are Fairlight3, Nay, SpaceClav, Pipe and BahMallet1. I wish I could take the time to program Mod7-sounds ...

@Kevin
OASYS retriggers the EG's by EG reset parameter, do you mean that SY's can repeat for example the attack or decay stage n-times before continuing?

@Dan, All
It has bothered me from the beginning that the EG curve parameter goes only one way; I have no means to put the curve to the negative direction. Or have I missed it? In the LFO section the shape parameter works in both direction and I have found it useful.

Re: 7-stage EG
The step sequencer can work similarly to an EG, if the attack/decay smoothing in the step-seq is too fast put it through the AMS Mixer smoothing as it is much slower. Then you have crude 32-stage EG.