Agave modules

Hi All,

Yesterday, I stumbled across Fabian Esqueda’s Agave modules, and ended up taking the time to get them to compile with the latest version of the Rack API. I figured if there’s interest, I’d be happy to publish these modules, with the permission of the original author. You can try out the updated version by cloning my forked repo, and building from source.

I could also use some help with panel design, if it does look like there’s demand for these modules.

Thanks, Jatin

11 Likes

Alright, these dismissive remarks have got to stop. I don’t know if you have the social skills to realize that this isn’t okay, so let me spell it out for you: If someone puts work into a free/open-source project, they don’t want to hear your comments about the quality about their software unless they specifically ask you, end of story. On VCV communities, we expect all members to be positive and constructive. If you do this again, 1 week mute. If it happens after that, you’ll be banned from all VCV communities.

2 Likes

sorry.

2 Likes

Actually, there is! Rather than oversampling, Fabian uses a method (which he helped develop) called “Antiderivative Antiliasing” (sometimes abreviated as ADAA). For more information on how this is implemented in the code see here.

I haven’t done much analysis yet to see how much aliasing the waveshaper introduces, but I’ve found through my own experiments that it often takes some fine-tuning to get ADAA to work well. I have some ongoing experiments with ADAA happening here. In general, I’ve found that 2nd-order ADAA combined with 2x oversampling often supresses aliasing better than plain 8x oversampling (depending on the nonlinear function being analyzed).

I haven’t actually touched Fabian’s DSP code yet (just tweaking code related to the Rack API), but as I keep using these modules, if it seems like aliasing is a problem, I can look into this further.

4 Likes

Oh, that’s very cool! Thanks for correcting me on that.

2 Likes

Nice! The Metallic module seems like something I have been waiting for to appear in Rack.

Does anybody have a windows build yet?

Hi, interesting folder, thanks for porting!

Some minor observations…

  • a possible typo where “BLINK_LIGHT” is set to 1.0f no matter if the input is clipping or not.

  • no Makefile in the repo.

Here’s a Window’s build: https://drive.google.com/file/d/1v_oMsPyR4YuHvz5J7J3oMoxMVqMC_hRH/view?usp=sharing

Thanks for pointing this out. It looked like the author had put the Makefile under the .gitignore. It should be visible now in my forked repo.

Hmmm, yeah I had thought that line looked weird. Not totally sure what the author’s intentions were with that light, I’ll probably have to ask him.

Thanks, Jatin

I m available to help drawing knobs if you need, PM and let me know

Hi All,

Sorry it’s taken me a while, but I’ve been able to prepare these modules for an official release (with approval from the original author). @David did some fantastic work helping with the GUI components. Aside from some cosmetic changes, small runtime optimisations, and changes to be compatible with Rack v1, the codebase is largely unchanged.

If you’d like to try out a pre-release build, they can be downloaded from GitHub. Hoping to have these modules officially available in the library soon!

Thanks, Jatin

5 Likes

Awesome! I love MS20 filters. It would be neat to have an input to control the resonance on the MS20 filter; seems like the panel has plenty of space for that. I don’t know how interested/willing you are to change these though.