Well, I found a bug in the Morph Dry ↔ Wet modes when spread is non-zero and in mono output mode (right output not patched). The additive 100% Wet mono mode was mildly different than the un-morphed output. The subtractive mono mode was wildly different.
Also the Notch output was sometimes inverted depending on the selected filter slope.
All has been fixed in v2.14.2dev2
@cubistguitar had a great idea for a BP ↔ Notch Morph mode, so I added it.
Note I inserted the new mode before the Dry/Wet Morph modes, so dry/wet patches made with the previous beta version will be broken.
I also added a slope detector mode to the WinComp module via a “B normalled to A -1 sample” context menu option.
Here is the relevant updated WinComp documentation:
The binaries for v2.14.2dev3 are available below
I plan to submit all the enhancements to the library this weekend. So time is running out to report any bugs or request any more enhancements. After the 2.14.2 release I will put aside further development of the free Venom modules for a while so I can work on my next commercial plugin.
@DaveVenom Had a thought re: different curve outputs for Multimode filter. Right now, there’s Low, BP, and High outputs. What if the outputs could be different slopes instead? For example, output 1 could be Low 12db, next output could be Low 24db, and third could be BP, etc through a user selected right click menu. Morph could still follow the different outputs, but configued to a combo of 1, 2, and 3, rather than just modes. This could be similar to Joranalogue Filter 8 which has multiple slope outputs, but with the added magic of Morph: Filter 8 – Joranalogue Audio Design
Will your next premium plugin be part of the Venom Oscillations or will it be a separate plugin entirely? Wallet’s all drained this year so I can’t get more modules for a while unless you have an intro sale and it’s just as good as Sofia’s Daughter or Spice Factory… if you wanna make another emulation of a hardware module expanded, do make noise maths, we don’t have a direct emulation of that one that I know of and it’d be an insta-buy for a lot of people… look forward to your next project and hope it takes a while so we know we get some gold to keep supporting your development
I think that was done already and couldn’t be added to the library for legal reasons.
@DaveVenom the multimode filter is great for pinging, but the spread control seems too sensitive. Moving it slightly goes way out of tune and makes it hard to adjust. Is that by design?
Nano Modules Arc is close ain’t it?
Anyway, my perception from reading some older convos, if I recall correctly, is that a 1:1 Maths clone is A) not something that Make Noise would be that happy with and B) maybe not as great idea as it seems at first glance since the multi-purpose utility that is its forte is not as quintessential in VCV as it is in hardware. The ability to instantly add in whatever slew limiter/function generator/EG/VCO you need makes it less important to fit all those things in one box so to speak. Venom’s recent slew limiter is a great example of something that can cover some of what maths does in arguably a more useful way. Though I do get the appeal of a clone for folks who have used Maths for years.
Anyway, Dave will decide for himself what to do next of course.
The Joranalogue Filter 8 is an amazing piece of kit! I have a lot of respect for their design, both visually and functionally. Also their documentation rocks!
I thought about providing a mechanism for simultaneous outputs at different slopes - I agree it would be nice to have. But then I rejected it. Partly out of fear of making the filter more complicated to use, and partly due to the impact it would have on the morph computations. The different output types have different phase shifts that can lead to phase cancellation when cross-fading between them. I am already inverting certain outputs while morphing to compensate. Introducing variable slopes adds even more phase discrepancies, which opens up a whole can of worms in the morph that I didn’t want to address.
I might revisit this idea in the future, most likely as a different module. But that is a very low priority. I have a number of totally different modules in mind to “complete” my collection that I want to create first.
This will be a new plugin - tentatively named “Chaos Boxes”. To start it will contain my digital take on Rob Hordijk’s Benjolin and Blippoo Box. It is something I have always wanted to do, but until I learned how to create the MultiMode Filter, it was beyond my capability. The resonant character and “pingability” of the MultiMode Filter make that code base ideal for this project.
I already have nice VCV implementations of both using free modules, so anyone can try it out for free. But there is something to be said for having each in one self contained module. As much as I love playing with my existing emulators, I get frustrated when I accidently drag part of the PatchMaster UI out of position. And there is a CPU and perhaps sound quality price to pay using multiple modules in the emulation.
Sign me up, I am just now getting into the chaotic systems kind of patching. I love the intent to create self organizing control systems based on the inputs and outputs of the patch. Really lovely concepts to work with in composition and sound design. New ideas for me, and fun fun fun.
Definitely by design. The primary purpose of the Spread is to generate two different resonant peaks.
It just so happens that the mono subtractive mode lends itself to another use case when the spread difference is miniscule and the slope is relatively steep - the filter can become a highly selective band pass filter using LP, BP, or HP outputs, which is what I did with the Venom Chorale patch. But I agree, the useful range of the spread knob is miniscule for this technique. I find it easier to type in a value.
The filter crossfading and morphing are very cool and rare features. It does up the complexity level quite a bit as you pointed out @DaveVenom. Just a thought, and I don’t know if this would be applicable to a state variable filter, but could you decouple the filter “definition” from the “implementation”? Let me explain…
I did a pole mixing filter recently (and yeah that is a different architecture) where the filter accepts VCA levels for each pole contribution. I chose not to include the myriad of filter types in the filter itself, just an input jack for the VCA levels. Then I added a second module for the filter definition, with presets covering common filter types. To use it one instantiates a Pole Dancer filter and then give the Pole Dancer a Personality to set the filter type. The Personality is just a static set of control voltages for the pole mixing levels. That solves setting a single filter type.
But we want morphing, right? So instantiate multiple Personality modules and use regular VCV Rack modules to morph between filter types. Since each Personality is just control voltages for a mixer this is bread & butter for VCV Rack modules. And shout out to the Cross Fade 3D as one way to do morphing. Here’s an example of using Cross Fade 3D to morph between 8 different filter types. The Personality modules are on the left and the Pole Dancer on the right with morph control elements stuck inbetween.
Pole mixing is a different arch than state variable, but thought I might throw that out there as an idea for decoupling some of the morphing complexity.
You have put a lot more thought into this than I have, but yes, I had vague similar thoughts, especially after re-acquainting myself with the Filter 8 documentation. If I ever go down this line I am sure it will be using some topology other than state variable. But I have a lot more to learn before I can do that. As I said, it is low priority right now, and odds are it may never happen.
Oh that filter looks cool, are you going to release that to library?
I’d love to release it to the library, but it’s going to be a long road to get there if it ever does. This is a DIY hardware project with no end in sight
Yeah, I think Arc may be a good alternative for our purposes and we could just add modules to do everything maths can do since we don’t have the limitations of hardware and money for as many instances of modules as we need. There’s enough options for certain parts that we can add in the plugin if we need instead of the whole module.
So it’s more of a combination module that expands upon some of your existing ones? More runglers RrRrRrunglers Have RrRrRridges! Good to know more of what’s coming, and if you already have a free alternative then it might help my wallet a bit more if it’s gotta wait
