Splitting is almost there!

The splitting is mostly working as desired now, although the land mass shapes and layout are a bit too regular currently.  The irregular (perturbed) splits work in both roughly-horizontal and roughly-vertical directions, the shifting (translation) and regular rotation of the split plates works, and the "hinge rotate" of adjacent plates is also working.

The world started with an initial supercontinent plate.

In the first iteration of splitting, the supercontinent plate was split along a perturbed near-vertical line, and the two resulting plates were shifted to left and right.  In the second iteration, the resulting, the plate on the left was split along a perturbed near-horizontal line, and "hinge rotate" was performed, rotating each plate in opposite directions

Like the left plate, the right plate had been split, but instead of a hinge rotate, the plates were simply displaced.  The bright green plate at top was shifted toward the top and rotated slightly, while the olive drab plate on the bottom of the split was shifted toward the bottom and rotated.

After two iterations, there are three landmasses made up of four plates.

You may be wonder what "hinge rotate" is.  Simply put, it is rotation of two adjacent plates in opposite directions, with the point of rotation being the point where the split line dividing the two plates crosses from water to land.  It can create narrow bodies of water similar to the Red Sea and the Persian Gulf.

Only two iterations are currently being applied.  Once greater irregularity is implemented, I will experiment with a larger number of iterations.  Right now, the results are not great, as can be seen below.

Three iterations of splitting, shifting, and rotating resulted in this map.

The next steps will be to increase irregularity of the distribution of the supercontinent, and the irregularity of the shifting, which is currently too regular.  With two iterations total, no hinge rotates, and no split of the right plate during the second iteration, the result can be rather boring.

Rather boring.

Once more interesting results are obtained, the code for the rest of the ultra-simplified tectonics can be added.  This will involve splitting the rest of the oceanic portion of the map into plates, as well, assigning motion vectors to each plate, then doing some approximations to see where uplift due to collisions or depressions due to rifting should be applied.

I don't have to get the split/shift/rotate results perfect - the tectonics code will likely improve the desirable irregularity somewhat.  But I think I need a bit more irregularity before I'm quite ready to move on.

And that shall be it for tonight.