I read the ISA reference and I am clear that the 2 instructions differ in the type of value they load (integer vs single precision float). What I don't understand is that the effect of the load is same regardless of the datatype, it should just copy the values from source to destination.
My question is, are there some differences that I am overlooking which necessitate the different types of instructions? Would using VMOVDQA to move some data from memory to a register and then performing float operations on it create some issues?