Why hasn't the x87 FPU been deprecated? Wouldn't it be better to map those old opcodes to new and imroved FPU instructions? I'm no hardware engineer, but it may even be nice if a bit in a control register could determine which opcodes were available. I'm assuming something like this has never been done because it can't be done without adding latency?
C and C++ development systems for OS other than Windows use x87 to support long double data types. Also, it remains a good way to support mixed floating point scalar data types and the still popular traditional C double mode evaluation. Besides, have you noticed how many developers still use 32-bit Microsoft compilers, where x87 is the best mode, or even use x87 mode because it's there, with compilers which have excellent support for SSE?