Cytomic "The Drop" Resonant Filter

[andy-cytomic]

use the sensitivity?

That still does not allow you to modulate the full range.

I'm sure its supposed to be this way but I'm interested in the reason.


Also is there a way to save oversampling settings? Saving preset as default does not save the setting & means i have to select higher oversampling every time i open the plugin.

The full modulation depth is +-96 semitones, which is +-8 octaves, which allows the knob to have a decent range and not be too twitchy. So if you have the low pass filter barely letting through audio at 100 hz then at full modualation the LFO or Env will open the filter up to 100*2^8 = 25 khz, which is plenty enough range to get what you need done without making the knob too sensitive to dial in something reasonable.

If you reduce the cutoff to 0hz then 0*2^8 = 0, so this is a convenient way to clamp down the filter to block audio without having to also reduce the modulation amounts as well.

When you choose "save as default" then the oversampling should also be saved. There may be a problem with the permissions on the settings file stopping the defaults from being saved, for example if you don't have write permission on the file. The settings file is located just one folder back from the preset folder, so if you choose "Show Presets Folder" then you can navigate and find the settings file. This is a plain text "xml" file that you can easily edit with any text editor:

Code: Select all

<?xml version="1.0" encoding="UTF-8"?>
<Cytomic Product="The Drop" Version="0.9.18">
  <Settings Version="0.9.18">
    <Tuple Key="SingleProgram" Value="true"/>
    <Tuple Key="ExternalSidechain" Value="true"/>
    <Tuple Key="CheckIo" Value="false"/>
    <Tuple Key="ShowPeakNeedle" Value="false"/>
    <Tuple Key="OverSampleRealtime" Value="Off"/>
    <Tuple Key="OverSampleRender" Value="Off"/>
    <Tuple Key="CurrentPath" Value="/Users/andy/Library/Application Support/Cytomic/The Drop/Presets/test a.tdp"/>
  </Settings>
</Cytomic>

Change the lines:

Code: Select all

    <Tuple Key="OverSampleRealtime" Value="Off"/>
    <Tuple Key="OverSampleRender" Value="Off"/>

to:

Code: Select all

    <Tuple Key="OverSampleRealtime" Value="x2"/>
    <Tuple Key="OverSampleRender" Value="x8"/>

or whatever other multiple of 2 you want to have the oversampling set as.

I need to check again with what Apple and Microsoft are up to with their latest operating system with permissions, they keep on changing things so sorry for the hassle!

[homunculus]

Hey Andy, anything new? Not talking about the Drop, just any other plugin in general the Drop gets used in every single project of mine, from 10 to 20 of them doing various things, the Drive in this is just amazing, nothing even close in the plugin world. Its just pure joy how this makes everything sound better.

[MFXxx]

Hi Andy know you're busy as usual

Any updates you care to Drop in on us...please

[andy-cytomic]

Hey Andy, anything new? Not talking about the Drop, just any other plugin in general the Drop gets used in every single project of mine, from 10 to 20 of them doing various things, the Drive in this is just amazing, nothing even close in the plugin world. Its just pure joy how this makes everything sound better.

10 to 20? You must have a huge machine!

I have been working on supporting signed binaries so I can release the AAX versions of The Glue and The Drop, I work 4 days per week on that and then 2 days per week on finishing The Drop and other R&D stuff if I'm going with the flow.

Ages ago I implemented a basic model of the SEM 1-A SVF but I didn't like the sound so decided not to include it, but more recently I saw the schematic of the OSCar SVF filter which on paper looked much better than the SEM 1-A, more drive and self oscillation. Yesterday I got a chance to implement an initial version of this and have confirmed it sounds great so I'm adding it to the list of filters

[sqigls]

smashing
you flipping rock, with or without socks on.

[MFXxx]

I have been working on supporting signed binaries so I can release the AAX versions of The Glue and The Drop, I work 4 days per week on that and then 2 days per week on finishing The Drop and other R&D stuff if I'm going with the flow.

Ages ago I implemented a basic model of the SEM 1-A SVF but I didn't like the sound so decided not to include it, but more recently I saw the schematic of the OSCar SVF filter which on paper looked much better than the SEM 1-A, more drive and self oscillation. Yesterday I got a chance to implement an initial version of this and have confirmed it sounds great so I'm adding it to the list of filters

As always a busy man and trust you and family are and continue to reap the rewards of your 6 day weeks and endeavours.

The OSCar SVF filter, is this similar to what GeFroce worked on and will this be a twin parallel filter? Regardless I can't wait and really look forward to the working result.

The Drop is still (one of) the best and smoothest filters in the collection.

Any updates on the cascade Hp as I'm experiencing ASIO spiking regardless of 2/4 pole. Not a gripe as I know it's all WIP.

[andy-cytomic]

...
The OSCar SVF filter, is this similar to what GeFroce worked on and will this be a twin parallel filter? Regardless I can't wait and really look forward to the working result.

I don't know what GeForce have done so I can't comment on that.

I'm modelling the 2 pole SVF of the OSCar, not the surrounding signal routing in that synth. It will appear in The Drop as both a high pass and a low pass which are processed serially.

The Drop is still (one of) the best and smoothest filters in the collection.

Any updates on the cascade Hp as I'm experiencing ASIO spiking regardless of 2/4 pole. Not a gripe as I know it's all WIP.

Thanks! By the very nature of the modelling there will be spikes, which I try to minimise but I cannot completely control. Audio quality comes at a price: at each sample there are 1 to 8 iterations of the algorithm to converge on a solution. I could remove the cpu spikes by always using all 8 iterations, but then the cpu load would be huge all the time, so you probably wouldn't even be able to run a single filter that way if you are getting spikes. More iterations will be needed if you have any of the following: high frequencies in your input signal, high cutoff, high resonance, high drive.

Also with the cascade and ladder circuits the 2/4 pole switch won't make any difference, they are 4 pole filters that you just pick off the two pole output or the 4 pole output from.

Is it specifically the "Cascade SH" high pass you are having trouble with? If so then have you tried the "Cascade" high pass? The "Cascade" high pass uses less cpu while still offering an almost identical sound.

[KnickersDown]

I'm adding it to the list of filters

This is fantastic news, thank you so much. Can't wait to get my hands on it.

Andy, have you any idea when you might implement midi learn for the silver knobs? I'm desperate to get just the Low Pass Cutoff assigned to a control parameter.

The reason being, I've just discovered today I can do truly amazing things by controlling The Drop with another plugin. The types of sounds I've dreamed about for years are seriously within reach now.

[jeffswitzer]

The reason being, I've just discovered today I can do truly amazing things by controlling The Drop with another plugin. The types of sounds I've dreamed about for years are seriously within reach now.

Which other plugin? I was actually just wondering if you can control the drop via Cableguys MidiShaper or if there is a better alternative?

[KnickersDown]

Which other plugin? I was actually just wondering if you can control the drop via Cableguys MidiShaper or if there is a better alternative?

I'm not really familiar with Midishaper. I'm using Thesys by Sugarbytes with geForce's Oddity and The Drop. I've just spent the entire day playing with this combination. The oddity is a great softsynth with lovely oscillators but the filter compared to The Drop is lacking. But putting oddity through The Drop is incredible. I bypass the oddity filter by keeping it wide open all the time. Add Thesys to it and all this crazy rhythmic synth modulation goodness is going on and it's fantastic. Sounds like an old system 100 I had with a step sequencer many years ago. That Cascade SH filter is just magical. When the drop will accept midi control changes I'll be in heaven.

I figured out a way to run it in cubase now but unfortunately, cubase RAMPS the values out of Thesys. So if I open The Drop up to 127 and then suddenly snap it back to say, 05 via Thesys, Cubase stops it from snapping by adjusting the values to a sloppy ramp. Sorry if I'm not making sense. It's hard to explain. If I could assign a direct CC value to the cutoff it would be too excellent.

That and the OSCar filter. Ooooh, good times.

[sqigls]

I'm sure you can adjust the resolution of the distance between the automation nodes.

[andy-cytomic]

Which other plugin? I was actually just wondering if you can control the drop via Cableguys MidiShaper or if there is a better alternative?

I'm not really familiar with Midishaper. I'm using Thesys by Sugarbytes with geForce's Oddity and The Drop. I've just spent the entire day playing with this combination. The oddity is a great softsynth with lovely oscillators but the filter compared to The Drop is lacking. But putting oddity through The Drop is incredible. I bypass the oddity filter by keeping it wide open all the time. Add Thesys to it and all this crazy rhythmic synth modulation goodness is going on and it's fantastic. Sounds like an old system 100 I had with a step sequencer many years ago. That Cascade SH filter is just magical. When the drop will accept midi control changes I'll be in heaven.

I figured out a way to run it in cubase now but unfortunately, cubase RAMPS the values out of Thesys. So if I open The Drop up to 127 and then suddenly snap it back to say, 05 via Thesys, Cubase stops it from snapping by adjusting the values to a sloppy ramp. Sorry if I'm not making sense. It's hard to explain. If I could assign a direct CC value to the cutoff it would be too excellent.

That and the OSCar filter. Ooooh, good times.

Midi learn is still on the way, but it is fraught with problems. One is sync, you could be an audio buffer out if you send midi from one plugin to another so that your transitions don't happen when you want. Another problem is resolution, midi only has 128 values, so you will get zippering if you don't smooth it. Another problem is smoothing, you never know the original intent behind the incoming midi controler change, you may want a sharp step, but you may want a smooth ramp, but no smoothing algorithm is ever going to be able to respond exactly how you want all the time, it will always be a compromise. Another problem is the host interfering with the raw midi data and doing their own smoothing and ramping. All up it is a bit of a mess.

A much better solution is to use the sidechain input of The Drop, then you have a complete specification of what the modulation should be at every sample and most hosts can handle proper delay compensation on audio. Have a look at the expert sleepers silent way plugins. Another alternative would be to ask Sugarbytes or Cableguys to output audio from their plugins, then the mess of low resolution midi is completely bypassed.

[andy-cytomic]

Here are some filter sweeps so everyone can get a taste for what the upcoming Stave Var filter will sound like compared to the other circuit models. These examples have a lot of drive to try and highlight how each circuit model responds differently:

www.cytomic.com/files/thedrop-sweeps-lp2-cascade.mp3
www.cytomic.com/files/thedrop-sweeps-lp2-sallenkey.mp3
www.cytomic.com/files/thedrop-sweeps-lp2-statevar.mp3
www.cytomic.com/files/thedrop-sweeps-hp2-cascade.mp3
www.cytomic.com/files/thedrop-sweeps-hp2-sallenkey.mp3
www.cytomic.com/files/thedrop-sweeps-hp2-statevar.mp3

[sqigls]

[knowix]

A much better solution is to use the sidechain input of The Drop, then you have a complete specification of what the modulation should be at every sample and most hosts can handle proper delay compensation on audio. Have a look at the expert sleepers silent way plugins. Another alternative would be to ask Sugarbytes or Cableguys to output audio from their plugins, then the mess of low resolution midi is completely bypassed.

More devs should consider CV style input for their plugins