MIDIbox SID can be fully programmed using the most minimal control surface (LCD, a few buttons, 1 encoder). All the additional control surface controls make it faster to get to certain functions. Programming patches becomes faster and more intuitive, and the knobs can be fun if you want to tweak the sound as you play.
If you just want to play preset SID sounds (the included presets are pretty good), you don’t need any of it. I still recommend the minimal control surface (“step B”) though, so you can see patch names.
You can also create SID patches with your computer using editor software available around here (I don’t use it personally), which may negate the requirement for a lot of controls.
As far as the bassline, lead, drum engines … if you want to play monophonic 3-oscillator SID sounds, most of your time will be spent with the lead engine. The other engines are optimized for different things. Drum and bassline have built-in mini-sequencers and a few parameter optimizations for creating some pretty damn cool little grooves. But it is all available to you with the most basic MIDIbox SID.
In MIDIbox SID, there are two different uses of the word “engine”. The software uses “engine” to describe the programming modes, such as “drum” or “lead”. A single PIC supports all of these engines (or modes) - they are like resident programs.
“Engine” is also used on the MB-6582 (the 8-SID monster synth you’ve no doubt seen in pictures floating around here) to describe a Core+SID set. The MB-6582 has up to 4 SID pairs controlled by 4 PICs, which can be selected for control from the front panel using the “SID Engine” buttons.
Let’s put it another way: a SID Engine is simply the chosen parameter set for a given patch, all handled (albeit not simultaneously) by a single PIC. In a multi-SID setup, a SID Engine could also refer to one of its Core+SID modules, each of which has its own selectable software “engine”.