I feel like I might get flamed a bit for this, because I’m going to be presenting a case for why I think the things some people ask for are wrong. I do recognize that this is just my opinion and don’t mean to be come across as super arrogant, but at the same time I do believe my opinion is right ☺.
(Also, I posted a snarky reply in the thread of the person who wanted to be able to draw his own waves – this post is me trying to explain that snark, post something productive instead, and also apologize. Wasn’t very cool of me.)
So as the thread title says, this is about feature requests. Now, I spend a lot of time on the forums of various software and hardware manufacturers – all the gear I own and a few that I don’t. And on all of them (except maybe the Moog forums), the site is full of feature requests where people seem to be posting things along the lines of “wouldn’t it be cool if this could do X” where X stands for EVERYTHING I CAN THINK OF. There seems to be a push for each piece of software and hardware to be able to do everything that every other piece of software and hardware can do. They want to mutate everything into uber-gear, and I guess then only need one piece of software/hardware.
And I get that from a theoretical perspective: it would be cool to have one synth that can do everything that any synth can do and thus never need another instrument.
But there are a bunch of real-world considerations that make that idea wrong, when you think about it in more detail.
(As a special aside here, I think we need to address the “I’ve got this idea that sounds cool in my head but doesn’t actually have interesting real world applications” type of idea separately. This is super common, actually. An example from these forums that springs to mind is when someone was asking for fractal LFOs. I’m not talking about the recent post with the diagrams of new waveshapes with one described as “fractal”, but rather an older thread where it sounded like someone put together the idea that “fractals look cool” with “I wonder what that would sound like”. The answer is: not very interesting in any way. The interesting thing about fractals, visually, is how the detail on smaller scales mirrors the detail on larger scales. That doesn’t translate to anything interesting in the audio realm because the nature of our modulation targets means that we would only maybe hear two or three orders of magnitude, tops, of similar structure – and the higher, smaller ones, would be producing tiny tiny modulation effects. It doesn’t really make sense as an idea. These kinds of ideas are all over every forum, and since they’re not even practically applicable, I feel that what I’m writing below doesn’t even apply. It’s even possible that I’m wrong on this particular example, that there’s some synth that uses fractal LFOs and they sound amazing, etc. but even if I’m wrong on criticizing that specific example, my general point still stands. I’m not an audio engineer or a mathematician, so I recognize that I don’t really know what I'm talking about.)
The problem is this: every object I have which I have any real fondness for is the product of a fairly strict design philosophy. This is a set of ideas about what that device’s purpose is, and how it’s used.
I really love my Virus. I mean I seriously, deeply adore it. I think it would actually be my last possession I would give up. And I think that part of the reason that it is, to me, a profoundly beautiful object, is because Access has a very well-designed and implemented design philosophy. It’s a tightly focused thing.
The issue with a lot of these suggestions can be understood with that in mind: are they in accordance with the design philosophy? Do they fit the sonic goals of the Virus, or would they be a broadening of those goals, and if so, would that be a loss of focus? Do they fit the flow of the hardware controller, or would shoehorning such-and-such a feature in be a really awkward fit?
Let me talk about a really simple example of a possible design philosophy issue: loopable envelopes. It would be really simple to implement loopable envelopes, on a software level, and I do hear it requested. There are serious issues to implementing it on the hardware side: there is no dedicated switch for it and no indicator available to show whether looping is turned on. No dedicated switch necessitates menu diving – and that’s a design philosophy issue for the Virus, right there. Here’s a piece of the Virus philosophy: menu diving should be kept to an absolute minimum. I think the indicator is probably also a top-level philosophy issue: first order sound design features (which is a term I just made up to apply to “key” traits of a sound, as opposed to “subtle” traits of a sound) should have an indicator on the controller that doesn’t necessitate peering at the LCD screen. I think there are issues with the idea that are opposed to the philosophy of the Virus, and that implementing those ideas would move the Virus away from being the tightly-focused thing that I love.
Now I’m going to posit something further about the Virus design philosophy. This is more of a guess whereas I’m more confident in my affirmations in the previous paragraph, and it’s a bit more subtle. I’m going to guess that part of the design philosophy says: “LFOs take care of repeating modulations, and envelopes are for non-repeating modulations”. So loopable envelopes directly contradicts this axiom. Why would we have such an axiom? Because it keeps the Virus clean, conceptually. It lets a user make assumptions about a sound – when we hear something, it lets us guess that the LFOs are the source of a certain type of modulations. (Yes, LFOs in envelope mode add a bit of complexity to this idea. Wouldn’t it be cleaner, conceptually, to say also “one shot modulations will be the envelopes only?” Well, yes, but two things: one shot LFOs are part of the tradition of the analog synths that inspired the Virus, and there’s always a trade-off. The gain of having envelope mode for LFOs is substantial, whereas I don’t see a similar gain for loopable envelopes.)
So that’s a dead simple example. The cost to design philosophy increases exponentially with some of the examples people suggest. A third envelope is super popular, and does have obvious practical applications. But how do you deal with it on the hardware side? Right now the envelopes have a top level control priority. Implementing a third one would hurt that. (I personally find having slope relegated to a shift function bad enough). They also hurt the intuitive aspect in that now you’ve got to do more wondering about what control source is producing which modulation you hear. Right now, we have a knob for each of the eight main envelope functions, and an LED for each LFO level, so we have a top-level “at a glance” view of potential modulation sources. A third envelope, while useful, mars that “at a glance” concept – it contradicts that aspect of design philosophy.
These are basic, basic ideas – loopable envelopes or third envelopes – that don’t even touch on introducing new sonic concepts. Design philosophy issues become much more threatened by things like user-drawn waves. (I personally object to user drawn waves because I feel like they’re something that sounds much more exciting on paper than the practical uses of it, but I recognize that’s a bit different than design issues). Right now, the only thing you can’t do on the hardware is design user arpeggiator patterns. I think Access should be (and I think Access is) very cautious about expanding that list. That’s another top-level design philosophy issue (and I bet there are engineers at Access who are bothered by the fact that they couldn’t figure out a practical way to do it on the hardware).
But more than control issues: that’s a serious broadening of what the Virus is. That’s a bigger change, I think, than sample playback would be, to give some reference of scope. And it throws the Virus hat into the ring with products that are really good at user-drawn waves, like Absynth and Zebra. I feel like the idea has implications, such as “if you allow user drawn waves I think that design philosophy suggests that you also need Absynth-style complex envelopes”. And that blows the concept of the Virus wide open, and it becomes something unfocused. It is no longer a tightly designed object.
None of this even touches on another really Big and Important concept: the Virus is an instrument. Instruments need to have character. I don’t want an instrument that does everything I can think of. A beautiful, well-made tool is for a precise purpose, not a broad purpose. A computer is a broad-purpose device, and that’s why they’re not satisfying as instruments (I think this can be accommodated for with careful consideration and intelligent design of controllers but I’m trying to illustrate a point, if you see what I mean).
(Another aside: this is why the “new LFO shapes” post is actually a good suggestion. Do those shapes have practical applications? Yes – I can look at the pictures and hear the effects. Does it have a design cost? Not that I can see, in that it would simply add a layer to the “waves” section of the LFO menu which already exists. “Not that I can see” of course doesn’t mean that I’m not overlooking something.)