Besides the reasons people have already mentioned:
- x86-16 had a rather strange memory addressing scheme which allowed a single memory location to be addressed in up to 4096 different ways, limited RAM to 1 MB, and forced programmers to deal with two different sizes of pointers. Fortunately, the move to 32-bit made this feature unnecessary, but x86 chips still carry the cruft of segment registers.
- While not a fault of x86 per se, x86 calling conventions weren't standardized like MIPS was (mostly because MS-DOS didn't come with any compilers), leaving us with the mess of
__cdecl, __stdcall, __fastcall, etc.
