Grains, Stacks On Stacks...
From the original Analog Industries archive. Some links and images from that era no longer resolve.
Well, that took a minute, but we are hugely pleased to announce that Quanta, our hybrid granular / subtractive softsynth, is now available for purchase in the Audio Damage store. What seemed like a moderately complicated project back in October of last year turned in to an eight-month exercise in applying the Kübler-Ross model to plugin development. Nominally, we try to have a bit of restraint in our designs, and hope to apply the K.I.S.S. principle to most things we make, within reason. But for Quanta, we took a more subtractive approach, and stuck everything up to and including several kitchen sinks in it, then carved away until a firm foundation remained.
Obvs, you can get the highlights, bullet points, and marketing hyperbole in spades over on the AD site, so I’ll use this space to add some color commentary about the process. This all started when we were shopping for a new product idea, and I pondered that many (most?) granular synths for desktop were fairly complex and un-fun – not to disparage them at all, mind; it is a complicated process – and it might be an “interesting exercise” and furthermore “pretty fun” to make something more like Borderlands, which doesn’t really exist outside of the iOS environment. Adam concurred, and we initially thought about using the grain engine we had built years ago for Discord. It became rapidly obvious that wasn’t up to the task of a big polysynth, so Adam went in to his DSP Fortress of Solitude, and came out a month or so later with a full granular oscillator that you’re using pretty much as-is in Quanta.
Now that we had that, it was largely a matter of wrapping things around it to make it in to a fun instrument to use. After getting several requests for adding MPE functionality to some of our existing products, I dropped Roger Linn a line and got us a couple Linnstruments to work with (detailed in an earlier post, if you recall) and decided that, since there aren’t any MPE granular synths, it may be an interesting addition. This exploded the modulation sources to proportions big enough that it required a mod matrix. And since we have a mod matrix anyhow, might as well throw an Imperial fuckton (1.3 metric fucktons) of other mod in the thing. Which is how it ended up with four 99-step arbitrary function/breakpoint generators.
So, with a mod matrix and a pretty good grain engine, and some excellent new mod sources, it became a design project, winnowing all that down to the most usable feature set and parameter ranges, and making it all look pretty. I’m not the best judge of such things, being somewhat too close to it to actually see if there’s any forest all these trees are sitting in, but I do know that I love and will definitely use the result, and I can’t say that about all of our products.
The one other real departure from our normal modus operandi is that we have decided to add a demo version. We have gotten on fine over the years not having demos, and frankly, any lost sales were more than made up for by the fact that demos are a pain in the ass to make and support. But since we spent so much time on this (about three times as long as a normal product) it necessarily has to cost a bit more than our normal run, and we can’t, in good conscience, ask you guys to buy something that expensive on blind faith. Hence, a demo. It is fully functional, but can not save (neither presets, nor its state in a project), and stops making sound after 20 minutes.
Anyhow, head over and give it a spin, and let me know if you have any questions about the decisions we’ve made, or comments on its overall usefulness. This is, by far, the most complex software product we’ve made, and we’re sweating a bit. It’s a large slab of work for a 2.5 person company. Quanta is $79 introductory price, and will go up to $99 on August 1.
14 comments
Just noticed, is this the first ad product to have a demo? Seems like it. Also iirc the first to introduce introductory pricing.
sounds great, makes me miss devine-machine's lucifer less even though they're different beasts...
I suppose I should have a more insightful, articulate comment to offer than "FUCK YES!!"
- c
@fabi: Our very first product had a demo. It was cracked and saw wide distribution about 6 hours after we put it up, and we instantly started getting support requests from people using the crack. The way I knew this was that we had only sold like 50 units, so I knew every customer.
After that, we were like "fuck everyone. Burn it all down."
-CR
This might sound weird but I was thrilled to find a proper, useful sample and hold implementation on this instrument. Input sources, key triggering, smoothing...it's all there. Thank you!
@silent5 I don't understand why more instruments don't have that, to be honest. It seems so obvious in retrospect.
Did we really start back in October? I was thinking December, but you're probably right. No wonder it seems like we've been working it for awhile.
The sample & hold was a relatively late addition, as in post-SuperBooth. It may have been suggested by Simon-Claudius. (I just tried to search our Slack archive, but Slack is stupid about searches and generates a zillion hits on "sample" as you might imagine.) The code is simple, and it was easy enough to patch it in to the mod matrix, and I've always been fond of generating mod signals by running LFOs through sample & holds, so in it went.
Writing the granulator was probably the most-fun part of the project for me, although the FEGs were a close second because I've wanted to build envelope generators like that for a long time and was waiting for the right context.
@beauty pill: I'll take it.
Yeah sample and hold is one of those simple but useful things that often gets overlooked, or relegated to the role of "random". It's rare to find it implemented this way outside of modular stuff.
I'm completely taken with the demo and will be buying this ASAP. Congrats on another great release!
@silent5: now that you mention it, I recall that the desire for more "random" was partly why we decided to put in the sample & hold. There wasn't any randomness in the modulation sources before that, except for the triggered-random source.
Demo! Fabulous. This is gonna be lovely on the ipad too. You guys thinking of iphone as well?
I'm with Beauty Pill! Oh, and dig the waveform display.
I guess I'll write my first comment after many years of lurking.
Congrats on the release again! I really think you are the first to do the whole simple granular thing right. You hit the sweet spot between simplicity and flexibility really well.
I'd love to see a simple 4OP FM synth following the same design principle from you guys. There really aren't all that many alternatives out there. I'd love a modern FM synth that doesn't try to emulate the DX-7 or some other nostalgic bullshit, something that isn't held back by trying to replicate a 30 years old feature set.
Keep up the great work!
@atlastop No, definitely not. I don't like shoehorning this stuff on to the iPhone; it is very difficult to do, since it is simply not designed for the context, and there's not a big enough income gain to warrant the incredible amount of trouble. (I don't even like putting the effects on there, to be honest, but those aren't actually any trouble, so I do it.)
@Delta Sign: we have further plans for this platform, as it gets refined, but I don't think we've specifically discussed a 4-op. To be honest, that's a more difficult prospect, given that the platform we've created is essentially identical to FM8, so a 4-op would be a somewhat dumbed-down version of one of the most omnipresent synth plugins on the planet. That just doesn't seem like a very good idea to me. (And I just know the second we'd release it, NI would announce FM9. That's the way these things seem to work.)
We are definitely going to make at least one more synth using a somewhat altered version of this platform; that is for sure. We have too much invested in it to let it languish. But I don't think it will be FM specifically.
-CR
Fair enough! I'm looking forward to your next synth!
Fair enough.... Just wondering. I figured it wasnt easy to do and it explains why there are so few great synths for the iphone.
I just wish a developer would make a sample based synth which used the 5d touch. But i can only find seaboard
Comments are closed on this post.