The story of Vandal is a story of pure coincidence. What started out as a pretty laid-back design study, later turned into something that actually meant great fun to us. Even that much that we've put ourselves together and finally considered making a „product“ out of it.

Although guitar amp modeling finally managed to establish itself on the music gear market, we always missed a certain liveliness and interaction with the guitar or the playing style. Put simply, we missed „Rock 'n Roll“.

It was summer 2008, when, disillusioned, sceptical, but bursting with curiosity, we asked ourselves: how does one build a real good, realistic-sounding software amplifier? How impossible is it actually to develop algorithms for the amp and a cabinet simulation that do more than just apply a simple distortion curve and fit that up with subsequent filtering and convolution? How about going for discrete models of actual building blocks? Is it really necessary to render an entire cascade of class-A or A/B tube stages, in realtime?

Encouraged by our findings, we went on with the next idea: simulating a speaker cabinet by using a model that discretises the whole process in a similarly fine manner as with the amplifier before. We actually break the cabinet up into its individual components.

Typically, amp modelers use the principles of „convolution“. A real-world cabinet is measured/miked-up and the resulting signal is taken for „wrapping“ the actual guitar signal. This can result in a pretty faithful mirror image of the real cabinet, at least in terms of frequency and time domain. But, alas, there is no way to draw any conclusion about dynamic response and nonlinear processes inside the cabinet (and the microphone). In reality, it makes a great difference if you play at normal levels or if you treat your stuff real hard. But using convolution, the impulse response can always be just a single „snapshot“.

Other problems come up as well, such as the lack of (realtime) tweakability and added latency because of the underlying technology.

The speaker simulation of Vandal works indeed completely different. It calculates all components involved, plus their interaction, in realtime:

- Speaker (voice coil->magnetic field->mechanic mass/spring system, including nonlinear artifacts)
- Enclosure (size, resonances, damping)
- Feed-back effects from enclosure to speaker
- Feed-back effects from the entire cabinet to the power amplifier
- Recording room (size, resonances, damping)
- Microphones (inverse principles of a speaker)
- Positioning of microphones (close miking, on/off axis, ambience level, stereo width)

The advantage of this approach comes up in practical use: while an impulse response of an actual cabinet fully includes the mic, its position and the recording room, Vandal lets one choose components and proportions completely freely and „tweakable“. The model allows for vast freedom in sound design. American speakers in a brit cab? No problem. Or a 10-inch driver inside a 15-inch enclosure? A guitar speaker as the heart of a bass cab? Or the other way round? So what. That's what Rock 'n Roll is all about.

If you got already accustomed to impulse-response based cabinet modeling, you might have to re-calibrate yourself now. Instead of just applying a static acoustic „finger print“, Vandal's cabinet simulation will never sound twice the same. It will rather react very strongly to your playing style and highly interact with the signal coming from the amplifier.