Tech Preview: Hive Wavetables

[nichttuntun]

I do understand your points and needs. It's a save feel-good zone, like a well known home territory. Always sounding right. Working fast. No risk of entering new territory and maybe get lost or discover stranger things and become consumed by them. Yes please...do go on

Hive 2 already offers lots of new territory and strange things to be discovered. What's holding you back?

Nothing. Except I am in Greece right now

[cytospur]

Here is a 2 operator FM script that wraps the normal FM formula with the arcsin trigonometric function.

It's also possible to replace the asin function with other trig functions such as atan, tan and tanh for even more variations

I'll knock up a script that shows that.

[cytospur]

Here's a script that shows usage of the trig functions sin, asin, tan and atan in all combinations (16 in total) as wrappers to the normal sin terms in the usual phase modulation formula. Set number of tables to 16!

Code: Select all

Info "Set Tables to 16\n2op FM with a twist\nBy Mark Holt\n"

NumFrames=256

Wave target=aux1 "1"	//Carrier frequency
Wave target=aux2 "1"	//Modulator frequency

Wave start=0 end=15 "2/pi*asin(sin(aux1*2*pi*phase+10*table*2/pi*asin(sin(aux2*2*pi*phase))))"
Wave start=16 end=31 "2/pi*asin(sin(aux1*2*pi*phase+10*table*sin(sin(aux2*2*pi*phase))))"
Wave start=32 end=47 "2/pi*asin(sin(aux1*2*pi*phase+10*table*2/pi*tan(sin(aux2*2*pi*phase))))"
Wave start=48 end=63 "2/pi*asin(sin(aux1*2*pi*phase+10*table*2/pi*atan(5*sin(aux2*2*pi*phase))))"

Wave start=64 end=79 "sin(sin(aux1*2*pi*phase+10*table*2/pi*asin(sin(aux2*2*pi*phase))))"
Wave start=80 end=95 "sin(sin(aux1*2*pi*phase+10*table*sin(sin(aux2*2*pi*phase))))"
Wave start=96 end=111 "sin(sin(aux1*2*pi*phase+10*table*2/pi*tan(sin(aux2*2*pi*phase))))"
Wave start=112 end=127 "sin(sin(aux1*2*pi*phase+10*table*2/pi*atan(5*sin(aux2*2*pi*phase))))"

Wave start=128 end=143 "2/pi*tan(sin(aux1*2*pi*phase+10*table*2/pi*asin(sin(aux2*2*pi*phase))))"
Wave start=144 end=159 "2/pi*tan(sin(aux1*2*pi*phase+10*table*sin(sin(aux2*2*pi*phase))))"
Wave start=160 end=175 "2/pi*tan(sin(aux1*2*pi*phase+10*table*2/pi*tan(sin(aux2*2*pi*phase))))"
Wave start=176 end=191 "2/pi*tan(sin(aux1*2*pi*phase+10*table*2/pi*atan(5*sin(aux2*2*pi*phase))))"

Wave start=192 end=207 "2/pi*atan(5*sin(aux1*2*pi*phase+10*table*2/pi*asin(sin(aux2*2*pi*phase))))"
Wave start=208 end=223 "2/pi*atan(5*sin(aux1*2*pi*phase+10*table*sin(sin(aux2*2*pi*phase))))"
Wave start=224 end=239 "2/pi*atan(5*sin(aux1*2*pi*phase+10*table*2/pi*tan(sin(aux2*2*pi*phase))))"
Wave start=240 end=255 "2/pi*atan(5*sin(aux1*2*pi*phase+10*table*2/pi*atan(5*sin(aux2*2*pi*phase))))"

Normalize dB=0 base=each
 
Export Source=main "MH TrigFM.wav"

[Funkybot's Evil Twin]

Here is a 2 operator FM script that wraps the normal FM formula with the arcsin trigonometric function.

Here's a preset I just made with that script (uhm also included in the zip - hope you don't mind). Just wanted to say thanks ( ) and share.

[cytospur]

Here is a 2 operator FM script that wraps the normal FM formula with the arcsin trigonometric function.

Here's a preset I just made with that script (uhm also included in the zip - hope you don't mind). Just wanted to say thanks ( ) and share.

Of course I don't mind! Thanks for sharing

[nichttuntun]

Okay guys. The focus seems to be - as I understood correctly - based on simplicity. So to say not making things overly complicated. It seems to be important for you to stick with this concept plus having a maximum of fantastic sounding and most interesting output.

Okay. Here is an example for this concept. 2 OSCs. Both loaded with a sine wave. Let the OSCs ring modulate each other. Apply pitch drift to the OSCs and play with extreme octaves settings. You get nice results from mega fat bases to very freaky beating sequences which can be a heavenly base for your granular samplers or re-synthesis in other sound design tools.
To go deeper and more interesting you could apply slight distortion, wave folding or some sort of PW modulation directly to one OSC, not globally. As simple as it can be. You would be amazed.

Now for me the next logica and extreme simple step would be to use 2 very simple wavetables in both of hives OSCs. Both having just 2 single cycle waveforms. Keep the ring modulation and all other described stuff and set the OSC scanning in one OSC just at the point where the transitioning occurs and play with different speed and modulate the speed. Now I would modulate the speed modulator so that it would be just about the edge to go into audio rate territory very slow. After that I would modulate this modulation so the time it stays in audio rate would vary each time a bit. Super simple, super awesome... endless options. Wound be sounding killer with hives sound engine.

None of that very basic sound design options are pin board with HIVE.

Why are there 2 wavetable OSCs on board when I just can use them as simple Rompler (, beside their nice but few extra features they offer).

Have joy and be sound.

[cytospur]

Two ring modulated sines (second sine at different pitches for each subtable) with a bit of detuning and additional feedback parameter:

Code: Select all

Info "Set Tables to 16\nRing Modulation with Feedback\nBy Mark Holt\n"

NumFrames=256

Wave target=aux1 "0.85"	//Feedback term: range= -1<FB<1

Wave start=0 end=15 "sin(4*pi*phase)*sin(2*pi*phase+sin(2*pi*table))/(1+aux1*sin(4*pi*phase)*sin(2*pi*phase+sin(4*pi*table)))"
Wave start=16 end=31 "sin(4*pi*phase)*sin(4*pi*phase+sin(2*pi*table))/(1+aux1*sin(4*pi*phase)*sin(4*pi*phase+sin(4*pi*table)))"
Wave start=32 end=47 "sin(4*pi*phase)*sin(6*pi*phase+sin(2*pi*table))/(1+aux1*sin(4*pi*phase)*sin(6*pi*phase+sin(4*pi*table)))"
Wave start=48 end=63 "sin(4*pi*phase)*sin(8*pi*phase+sin(2*pi*table))/(1+aux1*sin(4*pi*phase)*sin(8*pi*phase+sin(4*pi*table)))"
Wave start=64 end=79 "sin(4*pi*phase)*sin(10*pi*phase+sin(2*pi*table))/(1+aux1*sin(4*pi*phase)*sin(10*pi*phase+sin(4*pi*table)))"
Wave start=80 end=95 "sin(4*pi*phase)*sin(12*pi*phase+sin(2*pi*table))/(1+aux1*sin(4*pi*phase)*sin(12*pi*phase+sin(4*pi*table)))"
Wave start=96 end=111 "sin(4*pi*phase)*sin(14*pi*phase+sin(2*pi*table))/(1+aux1*sin(4*pi*phase)*sin(14*pi*phase+sin(4*pi*table)))"
Wave start=112 end=127 "sin(4*pi*phase)*sin(16*pi*phase+sin(2*pi*table))/(1+aux1*sin(4*pi*phase)*sin(16*pi*phase+sin(4*pi*table)))"
Wave start=128 end=143 "sin(4*pi*phase)*sin(18*pi*phase+sin(2*pi*table))/(1+aux1*sin(4*pi*phase)*sin(18*pi*phase+sin(4*pi*table)))"
Wave start=144 end=159 "sin(4*pi*phase)*sin(20*pi*phase+sin(2*pi*table))/(1+aux1*sin(4*pi*phase)*sin(20*pi*phase+sin(4*pi*table)))"
Wave start=160 end=175 "sin(4*pi*phase)*sin(22*pi*phase+sin(2*pi*table))/(1+aux1*sin(4*pi*phase)*sin(22*pi*phase+sin(4*pi*table)))"
Wave start=176 end=191 "sin(4*pi*phase)*sin(24*pi*phase+sin(2*pi*table))/(1+aux1*sin(4*pi*phase)*sin(24*pi*phase+sin(4*pi*table)))"
Wave start=192 end=207 "sin(4*pi*phase)*sin(26*pi*phase+sin(2*pi*table))/(1+aux1*sin(4*pi*phase)*sin(26*pi*phase+sin(4*pi*table)))"
Wave start=208 end=223 "sin(4*pi*phase)*sin(28*pi*phase+sin(2*pi*table))/(1+aux1*sin(4*pi*phase)*sin(28*pi*phase+sin(4*pi*table)))"
Wave start=224 end=239 "sin(4*pi*phase)*sin(30*pi*phase+sin(2*pi*table))/(1+aux1*sin(4*pi*phase)*sin(30*pi*phase+sin(4*pi*table)))"
Wave start=240 end=255 "sin(4*pi*phase)*sin(32*pi*phase+sin(2*pi*table))/(1+aux1*sin(4*pi*phase)*sin(32*pi*phase+sin(4*pi*table)))"

Normalize dB=0 base=each
 
Export Source=main "MH SineFBRM.wav"

[nichttuntun]

Two ring modulated sines (second sine at different pitches for each subtable) with a bit of detuning and additional feedback parameter

I think you missed my point.

Using wavetables is like using samples.

If you can't make the synthesis engine actually interact with each other OSCs it's nothing more than stacking sounds. The foundation - even though your wavetables are moving - will stay a relatively static one.
My example would allow making organically movements by creating a foundation achieved with actual synthesis. My above example only is a starting base from where you can go wild in ALL possible directions sound design wise; and not only take the way the wavetables direct you in their limited kind of behaviour.
Have joy and be sound.

Ps.
In the ring mod synthesis example experimenting with very extreme octaves settings will result in veeery cool pulsating sequences. In HIVE there is even a limitation how low or high you can go octave wise. Why?

As you may see. I am asking for very simple and most basic things here concerning HIVEs price and concerning that it's a modern synthesizer.

What I find totally absurd is to make a very deep wavetable formulary which you first have to study like a language with which you can bring HIVE to a state it's mimicing all that cool forms of synthesis. But it can't. Spending time programming something out of the synth to make HIVE sound "as if" ... I think this concept is absurd.

[cytospur]

I wasn't trying to make a point. You just inspired me to make a new wavetable

[nichttuntun]

That's a good thing I assume

[pdxindy]

As you may see. I am asking for very simple and most basic things here concerning HIVEs price and concerning that it's a modern synthesizer.

I love doing the kinds of stuff you are talking about so I understand what you are asking for.

It would take a whole lot more cpu to add what you are asking for to Hive and have a sound quality that is satisfying, at least to my ears. For that, I use Bazille. The Osc's are complex and can interact with each other in lots of ways and the sound quality is great.

But that is not what Hive is. If you do not enjoy Hive for what it is, it may just not be a synth for you... nothing wrong with that.

[cytospur]

Here's a version of the above set of tables that creates just the one, higher resolution, wavetable:

Code: Select all

Info "Ring Modulation with Feedback\nBy Mark Holt\n"

NumFrames=256

Wave target=aux1 "0.95"	//Feedback amount: range = -1<FB<1
Wave target=aux2 "1"	//Carrier coarse frequency - should be an integer

Wave "tanh(0.5*sin(aux2*2*pi*phase)*sin(4*pi*phase+sin(2*pi*table))/(1+aux1*sin(4*pi*phase)*sin(5*pi*phase+sin(2*pi*table))))"

Normalize dB=0 base=each
 
Export Source=main "MH FBRM.wav"

Note that the feedback term (after the "/") is the same as the base term (before the "/"). However, that doesn't have to be the case. We can break the rules here and put in any old numbers. They don't even have to be round integers! That is also the case with the modulator in the base term. This does actually allow me to make a point about this type of synthesis. We can break the rules and make up our own set of rules entirely. So, whilst you might not be able to do audio-rate modulation, you can explore areas of synthesis that are just not available anywhere, albeit within the confines of wavetables.

Here is an example of non-integer (actually all irrational numbers) unbalanced feedback ring modulation in wavetable form:

Code: Select all

Info "Ring Modulation with Feedback\nBy Mark Holt\n"

NumFrames=256

Wave target=aux1 "0.99"	//Feedback amount: range = -1<FB<1
Wave target=aux2 "2"	//Carrier coarse frequency - should be an integer

Wave "tanh(0.5*sin(aux2*2*pi*phase)*sin(sqrt(2*pi)*pi*phase+sin(2*pi*table))/(1+aux1*sin(pi^2*pi*phase)*sin(exp(pi)*pi*phase+sin(4*pi*table))))"

Normalize dB=0 base=each
 
Export Source=main "MH FBRMX.wav"

EDITED: Added tanh to the formula. The exported wave from the Uhm script was clipped. Fixed now.

[cytospur]

[nichttuntun]

As you may see. I am asking for very simple and most basic things here concerning HIVEs price and concerning that it's a modern synthesizer.

I love doing the kinds of stuff you are talking about so I understand what you are asking for.

It would take a whole lot more cpu to add what you are asking for to Hive and have a sound quality that is satisfying, at least to my ears. For that, I use Bazille. The Osc's are complex and can interact with each other in lots of ways and the sound quality is great.

But that is not what Hive is. If you do not enjoy Hive for what it is, it may just not be a synth for you... nothing wrong with that.

Thank you for your reply. I can't clearly say it's not my kind off synth. I do certainly like the sonic character and it has some nice features. I could imagine using HIVE more as a very versatile percussion instrument. It has everything on board for that purpose, especially the A - D generators.
A great addition for that of course would be multiple outputs to different DAW tracks.
The wavetable sound- engine itself is nice and I have some use for that.
But HIVE always will show me it's borders relatively fast when I am into something I like and can't push it to the next level. Beside that's it's great for what it is and I still think the new additions are good four HIVE.

I am just not this sequencer/ supersaw/ unison / Rompler guy. That's all.

Ps.
Yes. Bazille is awesomely great sounding and I am experimenting around with it extensively and each time I come out with something strange but interestingly enough to catch my attention.
But Bazille too is a bit limited, especially in providing modulation sources.

I really hope and wait for ZEBRA3 and my biggest hope is that there will be absolutely no limitations of how many modules I want to use and how many modulators I want to use and that I can modulate each and every parameter. You guessed it. ZEBRA2 is a bit limited in that matter.

Have joy and be sound.

[pdxindy]

Thank you for your reply. I can't clearly say it's not my kind off synth. I do certainly like the sonic character and it has some nice features. I could imagine using HIVE more as a very versatile percussion instrument. It has everything on board for that purpose, especially the A - D generators.
A great addition for that of course would be multiple outputs to different DAW tracks.
The wavetable sound- engine itself is nice and I have some use for that.
But HIVE always will show me it's borders relatively fast when I am into something I like and can't push it to the next level. Beside that's it's great for what it is and I still think the new additions are good four HIVE.

I am just not this sequencer/ supersaw/ unison / Rompler guy. That's all.

Ps.
Yes. Bazille is awesomely great sounding and I am experimenting around with it extensively and each time I come out with something strange but interestingly enough to catch my attention.
But Bazille too is a bit limited, especially in providing modulation sources.

I really hope and wait for ZEBRA3 and my biggest hope is that there will be absolutely no limitations of how many modules I want to use and how many modulators I want to use and that I can modulate each and every parameter. You guessed it. ZEBRA2 is a bit limited in that matter.

Have joy and be sound.

Of course Urs only can speak for sure about this, but based on past statements, I think it highly unlikely Zebra 3 will be unlimited. I think it will be the same basic arrangement as today. Also, Zebra 2 does not have audio rate modulations except in some specific cases. Maybe Zebra 3 will add some additional specific cases but I very much doubt there will be open ended audio rate modulations. CPU use would be massive.

About Bazille... it has a huge amount of modulation possibilities. 2 LFO's, 4 Env's, 2 Ramp Generators, 2 Mod Mappers, the Sequencer can be 4 individual modulators. The Quantizer, Rectifiers, Lag Generators and the 4 Multiples means you can combine signals into a vast array of possibilities. Osc's can also work as LFO with all sorts of unique shapes. The Env's loop and with rate and amp modulation can also make many complex shapes.

From what you are saying, seems like only a full modular environment would satisfy you. In that case Reaktor Blocks/Racks has fantastic sound quality.

If you want Polyphony and Modular, Bitwig's Grid is excellent.

Btw, one can use unlimited Bitwig modulators to control most parameters in Bazille. If you use multiple instances of Bazille, it can be polyphonic.