I thought I'd write a small update about what I'm working on now, why its rather extensive and difficult, and where this leaves the project. Since TX16Wx 2.0 Free graduated to released, there has been very few bug reports, which I hope is a sign that it mostly works as advertised. On the other hand, I've been rather busy with more logistical stuff, PR, business talks and such. And, to my shame, also some real, non-TX work, since I do have to support myself too. In short, and without the whining, not all my time has gone towards feature programming on the TX.
Still, I have been able to get quite a bit done on the "Wave matrix" feature of TX16Wx Pro. I have also realized that it is a much, much bigger feature than I originally envisioned. Simply put, it adds a completely new dimension to the TX16Wx. And I don't just mean the work flow. Since this adds a completely new way of mapping samples, it also impacts almost every part of the sound engine and the UI. In fact, it pretty much doubles the complexity of the sound architecture, and not always in a programmatically nice way. I've been on the verge of scrapping the feature quite a few times. Phew.
The feature is really cool though. For you who have not read my previous forum and mailing list posts about it, here is a quick recap; The wave matrix is pretty much what it sounds like, a matrix of wave mappings. (n x m). Each axis can define either simple round-robin stepping, or any modulator as input, and this value on note on determines which sample will trigger.
Now, this can obviously be achieved by having similar trigger conditions on a set of groups. In fact, I am planning to add this as well to the TX16Wx. But the usage is a bit dissimilar. One advantage of the matrix is that it can be saved as a regular sound item along with a sample set, then re-used by simply mapping one or more of them into a program/group(s). Another is the easier overview (less cluttering in the mapping screen).
I've been playing with the feature a little, and having multiple samples trigger at random or controlled by a modulator in a riff makes for some whacky sounds indeed. Of course, one more serious feature use case will be setting up matrix definitions for drums for example. Having one axis trigger on velocity and another on random values to make a more dynamic multi-sample drum sound. Or the same, but using for example an external controller to select articulations or attack variations of a string sound.
Now, think about combining this with modulation based group switching as well. ZOMG! Soo kewl!
The bad news is of course that this feature has taken way longer than I had hoped. We're hitting December soon, and I think there is quite a bit more to do. Alas, I am indeed behind schedule on the TX16Wx Pro. Sorry about that. Have no fear though, at the very least this feature will be available reasonably soon, though you might have to wait a little more for the other promised features.
Next week I'll be in Tokyo, visiting family and friends, so if you are a Japanese TX16Wx lover and want to buy me a Kirin, send me a message. If you are a non-Japanese TX16Wx lover and want to do the same, pre-order a TX16Wx license. ;-)
Enough blabber for now. Back to the programming for me, and I hope you take some time to make Christmas songs with the TX16Wx.