When your cube locks up or pops, you might be able to see what the problem is.
The thing that worries me about removing excess plastic is that some of it is there to keep the piece aligned. I saw a video where a guy cut the wings off of the edge pieces (the little ones, not the big ones), and I bet he gets more pops (but maybe fewer locks).
As soon as I figure whether I want a black or white cube I'll mod one.
I'm also planning on painting a cube. I'm going to replace yellow with black. Red will be darker and orange will be lighter and metallic. Green will be lighter while blue will be metallic. The idea is to increase the differences in the colors as sometimes I see the wrong color. And so on those four sides I'll have them divided into two groups of (adjacent) two by metallic vs non metallic, and dark vs light. The other division is of course blue/green and red/orange. The idea is to make it easier for my brain to recognize the different sides of the cube by making them visually distinctive in a pattern. I guess most people do this automatically with the given color scheme but I think little things will help...
The idea behind going with black is to incorporate the lack of color into the algorithms. It is a natural color for me to ignore as one must do during F2L. And when lining up the edges I will arrange for the last four edges to be the black ones. That way they'll be easy to spot. And I can think of them in terms of a single color rather than a color pair.
My usual now is to keep the edges in mind as "red/green", which is to say, the opposite orientation to "green/red". With the black edges, I will think of them as "orange" or "not orange" for the two orientations. I think this will take less brain power.
