A quick attempt at "rasterising" the invaders. Thicker lines ruin the shape and thinner blend horribly.
To me, it looks better without.
...although my partner says they look better with, if I darken the sides of the white invaders.
So here is a Half/Half comparison.
Wow, that's a slammin' effort there, matey... very interesting, and not even the idea I had in mind!
As for your current raster experiment on offer, I'm thinking two things: 1) the invaders would look better with uniform colors (red on top, red on the sides, etc); 2) you seem to be using a global mesh image for the tops of all invaders, but instead what if their basic shapes were broken up in to a series of horizontal lines? (on the thicker side would probably look better than too thin) Perhaps as few as six thick rows (drop walls) for each invader would be *just* enough to carry off the illusion without losing too much detail.
That said, doing it the above way sounds like a total PITA and really not worth your time. :/
Anyway, to reiterate my original idea-- I'd like to see a bit of glow around the shapes of the invaders, as with the old CRT's. So I'm thinking around each invader there'd be a popup "cloud" shape. The part touching and immediately around each invader would be more solid, and more diffuse with distance. Again, simulating natural raster / CRT glow. The global image would just have the same general "glow cloud" repeated for each invader position, a basic MS Paint job.
Also, I think lowering the top heights of the invaders would help blend the invader shapes in to their glow clouds. Perhaps one-third height would look about right without sacrificing the feeling of the invaders having height.
Hmm, you know... I've been working in the VP8 editor and script this past week, and I think my dumb ass might just be able to come up with a crude proof of concept of the above, if that helps.
OMG This looks awesome what a new take on space invaders.
Seriously, it's just about the best mashup of pinball and classic video games I've ever seen. Typically such efforts involve copying static images and sounds in to a table whilst sadly doing very little to actually achieve the spirit of the source video game. I do think there's a pretty huge amount of untapped potential in this genre, though.
For example, Frogger seems like it might work nicely-- the player having to hit the rights shots to dodge death, step by step getting the frog (represented by shapes / lights on the playfield) up to the top, thereby earning a chunky bonus. Even in very simple form I think it could be fun. Indeed, I reckon that most video games would still have to be dumbed-down quite a bit logistically for practical purposes, but I think what this table tells us is that much is possible!