README - For non-programmers with great ideas

[aciddose]

There are a lot of different techniques, and when it comes to cabinet models things are constantly evolving. You might look into a non-linear FIR impulse response system although I think that's an aging technology at this point. I have no idea what's being developed currently.

Don't say it's nothing new, it is entirely a new thing if existing tools do not exist. Sometimes the only way to get what you want is to learn to write the code yourself. That's what I've done for my own projects and I'll continue to do as long as I'm alive. If products offered by others do not satisfy me, the harsh reality I've been forced to face up to is that either I become better than everyone else or I accept what I am capable of achieving on my own. One of those is a little less realistic than the other.

Synthesizers like Xhip or sets like Xhip Effects are not often sought after. They're extremely esoteric, so much so that it could be that as the author they only serve solely me in a complete way and are not ideal for anyone else. Part of that is as you've said, an issue of control over what may seem like minor design choices. So clearly there is no market for the product I'm looking for, I may be one customer out of billions of these humans on this huge rock and it seems hopeless to me that everyone should bow before me and worship only what I deem appropriate. So in such a case, such software is not commercially marketable or practical outside a hobby project. And, as you may know or may come to know, doing it yourself is almost never cheaper than having someone else do it. The expense of developing a skill is far higher than the output from that skill itself unless you're willing to bend and bow and exploit the capability you develop far beyond what you may consider acceptable.

Unfortunately, the structure of the software marketplaces today still isn't open-source or as egalitarian as might be hoped. The reality is that programmers are a special sort of person, found far and few between, and multiplying many factors together makes it ever more infinitesimal. You might be the only person in the whole world who wants your software to do exactly what you want, so who else do you expect to build that?

[Music Engineer]

So,I have an idea for a plugin. Nothing new, the available products don't have the features I want. Some come close but certain things need to be expanded on. [...] Where/who do I submit my idea to ?

If some product already comes close to what you have in mind, you could file your idea as a feature request to the developer(s) of that product. And/or you could just post your idea here in this forum and see what sort of responses you get.

[aciddose]

You might look into a non-linear FIR impulse response system ...

https://en.wikipedia.org/wiki/Nonlinear ... ed_systems
This is called a Hammerstein-model, as one example which is built by combining non-linear and linear blocks to approximate a non-linear system's behavior. Likewise, all modern digital filters use similar techniques to approximate the behavior of a non-linear circuit to varied degrees.

With a sophisticated enough model for the non-linear behavior of the amplifier, the cabinet and speaker model become less important and can be approximated well by being split into multiple stages of linear and non-linear filter models rather than attempting to produce a model of a non-linear amplifier and speaker/cabinet in combination.

Modern techniques utilize Fourier transforms to apply convolutions and FIR filters dynamically. If you consider for example that the dynamic properties of a frequency domain signal are frequency-discrete, and that there is no reason the filter applied (simple multiplication) need be static but rather could be generated by some low-order polynomial approximation functions.