Tech Preview: Hive Wavetables

[cytospur]

more

You can thank Viktor for this

[cytospur]

Big thanks to Mark Holt. Amazing stuff!

Very kind of you to say. Many thanks to you too!

[Howard]

(BTW aux2 isn't used in Bend03, so that line can be removed)

[cytospur]

(BTW aux2 isn't used in Bend03, so that line can be removed)

Correct, it isn't, but it has been left in as an option to swap with any of the aux1's for variation. Sorry, I forgot to mention that! Well spotted though

[cytospur]

A couple more scripts. The first is a naive hard sync:

Code: Select all

Info "MH Sync\nBy Mark Holt\n"

NumFrames=256

Wave target=aux1 "2*pi*((phase%(1-0.941176*table^0.368))/(1-0.941176*table^0.368))"
Wave "sin(sin(1*aux1)/(1-0.9*cos(3*aux1)))"

Spectrum lowest=0 highest=0 "0"
Normalize Metric=peak Base=each

Export Source=main "MH Sync.wav"

The second is the same script, but passed through a waveshaper:

Code: Select all

Info "MH Sync with waveshaping\nBy Mark Holt\n"

NumFrames=256

//Waveshaper transfer function
Wave "0.25*sin(12*pi*phase)+(phase-0.5)"

Spectrum lowest=0 highest=0 "0"
Normalize Metric=peak Base=each

//Input (Sync wavetable)
Wave target=aux2 "2*pi*((phase%(1-0.941176*table^0.368))/(1-0.941176*table^0.368))"
Wave target=aux1 "sin(sin(1*aux2)/(1-0.9*cos(3*aux2)))"

//Output
Wave "main_fi(frame,1024*(aux1+1))"

Spectrum lowest=0 highest=0 "0"
Normalize Metric=peak Base=each

Export Source=main "MH SyncWS.wav"

I expect that most of my future scripts will come with a waveshaping section, given that it's so darned easy to add on!

[cytospur]

Phase distortion of a simple 3 cycle sine wave passed through a simple sine based waveshaper:

Code: Select all

Info "MH PD waveshaping\nBy Mark Holt\n"

NumFrames=256

//Waveshaper transfer function
Wave "0.25*sin(8*pi*phase)+(phase-0.5)"

Spectrum lowest=0 highest=0 "0"
Normalize Metric=peak Base=each

//Distorted phase table
Wave target=aux2 "2*pi*phase^(0.5+0.5*(1-table))"

//Input to waveshaper
Wave target=aux1 "0.99*sin(3*aux2)"

//Output from waveshaper
Wave "main_fi(frame,1024*(aux1+1))"

Spectrum lowest=0 highest=0 "0"
Normalize Metric=peak Base=each

Export Source=main "MH PDWS.wav"

[cytospur]

Some new stuff that I'm working on that is based on the recently invented (ca2009) modified FM that I don't think has yet made it in to the VST world! You can read about it here:

https://pdfs.semanticscholar.org/6039/2 ... 1a4889.pdf

I'm using equation 19, although I've swapped the cos terms for sin and vice versa to force zero crossing points at the start and end points of the waveforms.

Code: Select all

Info "Modified FM\nBy Mark Holt\n"

NumFrames=256

Wave target=aux1 "2*pi*phase"

Wave "exp(table*cos(3*aux1))*sin(1*aux1+16*table*sin(3*aux1))"

Spectrum lowest=0 highest=0 "0"
Normalize Metric=peak Base=each

Export Source=main "MH ModFM.wav"

Waveshaped version to follow tomorrow

[Urs]

Hehehe, I'll give it a shot - but when I implemented modFM in the upcomming Rev3 FMO, it sounded like the most boring version of FM. I think I went as far as to completely delete the code. Either I had a bug in it, or it simply isn't very interesting, and that's maybe why we don't see it often.

[AnX]

he called your work boring!

[Russell Grand]

It's awesome that u-he is keeping the same price for Hive 2 for new customers instead of upping the price like other devs might.

[cytospur]

he called your work boring!

Not my work!

[cytospur]

Hehehe, I'll give it a shot - but when I implemented modFM in the upcomming Rev3 FMO, it sounded like the most boring version of FM. I think I went as far as to completely delete the code. Either I had a bug in it, or it simply isn't very interesting, and that's maybe why we don't see it often.

I agree, by itself, but there's some interesting stuff done with it such as channel vocoding and automatic patch generation for specified sounds using genetic algorithms etc. when using multiple 2op modFM units in parallel. Here modFM appears to out-perform conventional FM/PM. I think its best use will be in combination with other sound synthesis types. Something that I'll explore.

[nichttuntun]

2 OP FM can be very interesting, when the producing chip has enough character. A fantastic example is the unparalled 2 OP FM VST emulation PORTA FM from Plogue. Plogue does re-engineering from the original sound chips to achieve the detailed FM-sound. Please watch to be inspired.

Yes 2 OP can be very boring if there isn't anything special involved considering modulation or a special sound character.

[Ranoka]

Some new stuff that I'm working on that is based on the recently invented (ca2009) modified FM that I don't think has yet made it in to the VST world! You can read about it here:

https://pdfs.semanticscholar.org/6039/2 ... 1a4889.pdf

Wow, this is really interesting - thanks for sharing!

I gave the wavetable a quick spin and thought it sounds quite clean and made me think of the Tone2 synth Nemesis, which is a 2OP FM synth with 'NeoFM' which is designed to be easier to use than regular FM (or whatever marketing description they went with).

I compared them because I though it might be based on Modified FM. I found that NeoFM sounds even cleaner than Modified FM somehow.

The nice thing about NeoFM is it works really well when the modulator signal is a different shape - like a narrow pulse or more complex shapes, it creates really interesting digital fractal type sounds. They sound really nice when set to 4x unison in Nemesis.

I thought I'd share a Hive preset I made using the ModFM wavetable, I'm making some presets for the FB Community Soundset. I hooked the modulation targets to some fun creative sound stuff

[nichttuntun]

Hi. I think NeoFM produces less atonality. It contains more "musical" overtones when modulation starts. It's sounding cleaner, tho not necessarily more interesting. Correct me please. Cheers.