Why not a subtractive synth with LFOs that go to much higher rates?

VST, AU, AAX, etc. plug-in Virtual Instruments discussion
User avatar
Unaspected
KVRian
1186 posts since 4 May, 2012

Post Tue Jan 21, 2020 7:42 am

Acid Mitch wrote:
Tue Jan 21, 2020 6:06 am
DJ Warmonger wrote:
Tue Jan 21, 2020 4:41 am
...or maybe there are some internet haters that prevent low frequency oscillators from being fast, there are surely not rational or technical limitations.
There actually are technical limitations and rational decisions but it has nothing to do with weather you hate the internet or not.
Indeed, control rate signals having lower resolution creates a more general approximation and any kind of smoothing is going to then make it impossible to exceed a certain frequency.

Audio rate modulation is more taxing for the CPU so it has only become a realistic option more recently. I think The Drop might have been the first time I encountered audio rate modulation in a commercial plugin.

Acid Mitch wrote:
Tue Jan 21, 2020 6:06 am
empphryio wrote:
Tue Jan 21, 2020 4:30 am
...and fm synths are really phase modulation synths.
Yes,Yamaha DX,etc use phase modulation but there’s still alot of actual fm out there.
Absolutely. FM is definitely FM and PM definitely is PM. It depends on the destination and location of the modulation source.

I don't know exactly how the DX uses PM but if it is modulating the scan rate via PM then this can be viewed as FM in the same manner that reading through an array of values at a different rate changes the pitch.

User avatar
Halonmusic
KVRAF
3335 posts since 13 Nov, 2015 from Norway

Re: Why not a subtractive synth with LFOs that go to much higher rates?

Post Tue Jan 21, 2020 8:24 am

FB 3100/3200/3300 can do it aswell
Don't be a Sheep. :borg: | Opinions matters :evil:

empphryio
KVRist
74 posts since 14 Apr, 2019

Re: Why not a subtractive synth with LFOs that go to much higher rates?

Post Tue Jan 21, 2020 8:28 am

Thanks to those who gave intelligent responses.

Elektronisch
KVRAF
3870 posts since 3 Feb, 2010

Re: Why not a subtractive synth with LFOs that go to much higher rates?

Post Tue Jan 21, 2020 8:34 am

empphryio wrote:
Mon Jan 20, 2020 7:37 am
So basically subtractive plus fm. I'm sure there are some. But I wonder why it's not just common. Anyone know of some?
I mean an LFO that can go at least 20 times higher than the carrier wave. Seems like a very slight change that could really open up potential.
To tell you the truth synths rate can very from each other, some offer very high some, as example Repro, IK Uno at highest rate doesnt feel high at all. BUT on something like Sylenth1, Virus, Massive X, Dune, Serum high rate does feel like a higher rate. But now here is the thing, if it would go higher, what would purpose it would serve for mentioned high rate lfo synths.

You would probably just be hearing some odd things in high frequency range on generic waves and on s&h something like a white noise.

User avatar
Halonmusic
KVRAF
3335 posts since 13 Nov, 2015 from Norway

Re: Why not a subtractive synth with LFOs that go to much higher rates?

Post Tue Jan 21, 2020 8:37 am

empphryio wrote:
Tue Jan 21, 2020 8:28 am
Thanks to those who gave intelligent responses.
I'll just leave these links as im a huge fan of Björns stuff. These (lfos) go way into audio rate:

https://www.fullbucket.de/music/fb3300.html

https://www.fullbucket.de/music/fb3200.html

https://www.fullbucket.de/music/fb3100.html

Be sure to check the rest of his portfolio aswell :tu:
Don't be a Sheep. :borg: | Opinions matters :evil:

Acid Mitch
KVRAF
2010 posts since 5 Oct, 2005

Re: Why not a subtractive synth with LFOs that go to much higher rates?

Post Tue Jan 21, 2020 9:03 am

Are you looking for software or hardware, ?

User avatar
pdxindy
KVRAF
16319 posts since 3 Feb, 2005 from in the wilds

Re: Why not a subtractive synth with LFOs that go to much higher rates?

Post Tue Jan 21, 2020 10:15 am

empphryio wrote:
Mon Jan 20, 2020 7:37 am
So basically subtractive plus fm. I'm sure there are some. But I wonder why it's not just common. Anyone know of some?
I mean an LFO that can go at least 20 times higher than the carrier wave. Seems like a very slight change that could really open up potential.
Most synths are controlling the LFOs at a reduced sample rate compared to the Oscillators. Uses less CPU and works fine for the usual slow LFO speeds.

So if you want to use LFO's that go to a much higher rate, you need a synth that handles that.

In Bazille, the Osc's can modulate anything and they can be set anywhere from 0 to well above hearing... they can track the keyboard or not, etc. The Bazille LFO's are control rate, not audio rate so you would use one or two of the 4 Osc's as LFO's for audio rate modulation. You can also use one of the filters with resonance up full as a sine modulator at audio rate and adjusting gain or modulating cutoff changes the sine to a more complex waveform.

In ACE, the LFO's are also capable Osc's and are at audio rate.

Some synths have fast LFOs, but does not automatically mean it is calculating at audio rate... it may still be control rate. In Dune 3 for example, that can be set by the user. Setting it to calculate at audio rate uses noticeably more cpu, but the sonic difference is clear.

Whywhy
KVRian
507 posts since 11 Aug, 2018

Re: Why not a subtractive synth with LFOs that go to much higher rates?

Post Tue Jan 21, 2020 10:34 am

I would add :
Lfo or osc at audio rate are nice as source modulation but the mod destination needs to be able to receive audio rate modulation (filter needs to be great or they can " explode ", zipper noise......). Audio rate modulation can bring some nice texture to a lot of modules ( filter, phase,chorus, phaser....)welcome to the modular world:)

User avatar
gentleclockdivider
KVRAF
2437 posts since 22 Mar, 2009 from gent

Re: Why not a subtractive synth with LFOs that go to much higher rates?

Post Tue Jan 21, 2020 11:36 am

Unaspected wrote:
Tue Jan 21, 2020 7:42 am
I think The Drop might have been the first time I encountered audio rate modulation in a commercial plugin.


Uh , no not really
Reaktor could do audio rate modulation since the beginning , all filters that had an FM input ( processed at sample rate ) , and since version 5.0 ( introduction of core framework ) you could code anything that is capable of audio rate modulation .
Blocks ( reaktor 6 ) is the perfect example of that

However , the drop is arguably the best dedicated filter plugin ever created :tu:
Eyeball exchanging
Soul calibrating ..frequencies

Acid Mitch
KVRAF
2010 posts since 5 Oct, 2005

Re: Why not a subtractive synth with LFOs that go to much higher rates?

Post Tue Jan 21, 2020 11:58 am

^ Vaz modular, Timewarp 2600, Arturia Minimoog, synthsquad and several others also preceded The Drop.

cron
KVRAF
3289 posts since 27 Dec, 2002 from North East England

Re: Why not a subtractive synth with LFOs that go to much higher rates?

Post Tue Jan 21, 2020 11:58 am

Unaspected wrote:
Tue Jan 21, 2020 7:42 am
Indeed, control rate signals having lower resolution creates a more general approximation and any kind of smoothing is going to then make it impossible to exceed a certain frequency.

Audio rate modulation is more taxing for the CPU so it has only become a realistic option more recently. I think The Drop might have been the first time I encountered audio rate modulation in a commercial plugin.
This is the core problem AFAIK. It's cheaper in terms of computation when you optimise for the L in LFO. I think a few people were complaining that Massive X's LFOs alias when you push them into audio rate as an example.

There's not really a fixed definition of what "modular synthesis" is, but for me one of the key things is that there's no distinction between control rate and audio rate signals. An oscillator can be an LFO, and an LFO can be an oscillator. I imagine some of the more exotic modules that have popped up in the recent modular boom are challenging that idea though.

User avatar
gentleclockdivider
KVRAF
2437 posts since 22 Mar, 2009 from gent

Re: Why not a subtractive synth with LFOs that go to much higher rates?

Post Tue Jan 21, 2020 12:08 pm

The only difference between an osc and an lfo in a real modular
oscilllator = alternating current
lfo = direct current
Both have the same infinite resolution , and luckily we 've been spoiled the last few years to get exactly that in a software environment , be it reaktor ,cherry , softube, vcv rack etc...all good stuff , all processed at sample rate
good times
Eyeball exchanging
Soul calibrating ..frequencies

HunterKiller
KVRAF
2104 posts since 10 Mar, 2006

Re: Why not a subtractive synth with LFOs that go to much higher rates?

Post Tue Jan 21, 2020 11:40 pm

Acid Mitch wrote:
Tue Jan 21, 2020 11:58 am
^ Vaz modular, Timewarp 2600, Arturia Minimoog, synthsquad and several others also preceded The Drop.
Yes and Timewarp2600 is particularly good at it, out of that early generation of more serious emulations.
"The educated person is one who knows how to find out what he does not know" - George Simmel
“It's what you learn after you know it all that counts.” - John Wooden

User avatar
Unaspected
KVRian
1186 posts since 4 May, 2012

Re: Why not a subtractive synth with LFOs that go to much higher rates?

Post Wed Jan 22, 2020 6:27 am

gentleclockdivider wrote:
Tue Jan 21, 2020 11:36 am
Unaspected wrote:
Tue Jan 21, 2020 7:42 am
I think The Drop might have been the first time I encountered audio rate modulation in a commercial plugin.


Uh , no not really
Reaktor could do audio rate modulation since the beginning , all filters that had an FM input ( processed at sample rate ) , and since version 5.0 ( introduction of core framework ) you could code anything that is capable of audio rate modulation .
Blocks ( reaktor 6 ) is the perfect example of that

However , the drop is arguably the best dedicated filter plugin ever created :tu:
Agreed. The Drop is a perfect plugin with very well thought out design features that all plugins should have.

I did have Reaktor back then but as your comments support, you would have had to roll your own. I'm not sure if Reaktor boasted audio rate modulation until Blocks in Reaktor 6. As a commercial package, I mean. Because, of course, audio rate modulation was an option long before that - it just wasn't practical so it wasn't really seen in commercial plugins. Also talking from personal experience rather than as a plugin historian.

gentleclockdivider wrote:
Tue Jan 21, 2020 12:08 pm
The only difference between an osc and an lfo in a real modular
oscilllator = alternating current
lfo = direct current
Both have the same infinite resolution , and luckily we 've been spoiled the last few years to get exactly that in a software environment , be it reaktor ,cherry , softube, vcv rack etc...all good stuff , all processed at sample rate
good times
Maybe in a similar manner, control signals are scaled and shifted to operate between 0 and 1, opposed to an oscillator which commonly operates between values of -1 and +1. There is a distinction that can be made there - but there is no difference in operational principal - only in method of execution in some circumstances.

cron wrote:
Tue Jan 21, 2020 11:58 am
Unaspected wrote:
Tue Jan 21, 2020 7:42 am
Indeed, control rate signals having lower resolution creates a more general approximation and any kind of smoothing is going to then make it impossible to exceed a certain frequency.

Audio rate modulation is more taxing for the CPU so it has only become a realistic option more recently. I think The Drop might have been the first time I encountered audio rate modulation in a commercial plugin.
This is the core problem AFAIK. It's cheaper in terms of computation when you optimise for the L in LFO. I think a few people were complaining that Massive X's LFOs alias when you push them into audio rate as an example.

There's not really a fixed definition of what "modular synthesis" is, but for me one of the key things is that there's no distinction between control rate and audio rate signals. An oscillator can be an LFO, and an LFO can be an oscillator. I imagine some of the more exotic modules that have popped up in the recent modular boom are challenging that idea though.
Whilst I've set up various types of LFO using Pure Data and Reaktor, something I have not researched nor have any practical experience in measuring is how such aliasing might impact on the functionality of the LFO for modulation duties. I wouldn't think it should matter but that's just a presumption. Surely such an oscillator can be low passed much lower than an oscillator designed to be heard - Or it could be cross faded with a sine wave in the higher frequencies.

User avatar
gentleclockdivider
KVRAF
2437 posts since 22 Mar, 2009 from gent

Re: Why not a subtractive synth with LFOs that go to much higher rates?

Post Wed Jan 22, 2020 7:41 am

Unaspected wrote:
Wed Jan 22, 2020 6:27 am

I did have Reaktor back then but as your comments support, you would have had to roll your own. I'm not sure if Reaktor boasted audio rate modulation until Blocks in Reaktor 6. As a commercial package, I mean. Because, of course, audio rate modulation was an option long before that - it just wasn't practical so it wasn't really seen in commercial plugins. Also talking from personal experience rather than as a plugin historian.


LIke I said reaktor had audio rate modulation for certain primary modules
Red inputs were processed at control rate ( dropdown list from event rate rangin from 25hz to 3200hz ) , but you had an event to perm module that could be set at sample rate .
For example , patching the output of primary module oscillator into the P input of another primary module you had to insert an AUDIO to EVENT module , and the update speed of this computation was the control rate .., but when inserting an audio to event module perm module , the converted audio to event can be computated at a user defined speed , like sample rate
Image



Some primary modules also have modulation inputs that accepts audio inputs , the update of this port happens every sample tick iow , sample rate .
Lots of primary modules have these (ladder filter for filter fm , fm osc's etcc)..and this was all before reaktor 5 , I started using reaktor from version 3 and it was already there ( year 2001)

Reaktor 5 brought us the core system ( developed by vadim zavalishin ) of sync modular , basically reaktor's own programming language .
No limits whatsoever ( well no many )

You're just wrong thinking reaktor blocks saw the introduction of audio rate , all blocks are programmed in reaktor core , blocks could have been introduced with reaktor r 5 , but certain new core modules introduced in r6 mad it easier ( bundled and scope busses )
Monark does filter FM and was coded in reaktor 5 , isn't that audio rate enough for you ?
But yes , reaktor blocks is nice because it's so damn easy to patch high quality stuff..
Eyeball exchanging
Soul calibrating ..frequencies

Return to “Instruments”