fpcupdeluxe - 32bit or 64bit IDE, trunk or stable

[PascalDragon]

Yes, that is correct. What I explained is that it does not matter much since 80 bit precision is legacy anyway.

It does matter much, because FPC will use 80-bit precision on platforms that support it. So if you compile a program for e.g.i386 or x86_64 excluding Win64 then the compiler will make use of Extended for floating point code and thus it needs to be handled correctly and consistently which a compiler compiled for x86_64-win64 or any platform not supporting 80-bit floating point (e.g. aarch64) does not.