Regarding e,q,d as exponents:

q and d are used in some worlds to specify extended precision floats, so
maybe it makes sense to keep them.

the original problem was with radix literals where the radix was greater
than 10.  I see nothing wrong with 2r110e3 being equal to 48, but 16r1e3
being 4096 is flat-out wrong.  Removing d as an exponent designator does
nothing to help that.

So I see the question of supporting q and d as orthogonal to the radix
question.  Clearly exponents cannot be used for radices greater than 10,
and some might argue that they shouldn't be supported at all for bases
other than 10, although I find them convenient.

So I suggest fixing the exponent/radix bug, and leaving q and d options
as I can't see any damage caused by them.

../Dave