Double granularity about v86

[copy]

Yes, I am aware. The problem is mentioned in the Readme:

Calculations are done with JavaScript's double precision numbers (64 bit), so they are not as precise as calculations on a real FPU (80 bit).

When handling 64-bit integers, a few bits of precision will always be lost.

There's nothing we can do about it, except for a much more sophisticated FPU implementation with emulation of 80 bit floats.