New answers tagged

30

The documented way to detect an x87 FPU is to attempt to initialise it, and then read its control word (FPU_STATUS must be set to some non-zero value first): FNINIT FNSTSW WORD PTR [FPU_STATUS] This uses the non-waiting variants, otherwise the CPU will wait for a non-existent FPU to respond. If the status word is not 0 after this, no FPU is installed — the ...


1

The D language supports the maximum precision a platform provides. It has defined these 3 base floating point type float for 32 bits floats, double for 64 bits floats and real for maximum of platform. In the case of x86, it's the 80 bits x87 type. The unusual feature is that the compiler still supports x87 even when generating x86_64 code. In 64 bit mode, ...


0

I don't think the no value of 80 bit is correct. At the time of SSE however, large computational problems were not solved on x86, but rather on risc platforms that provided 64bit addressing capability, alpha, sparc, IBM power, SGI etc. There was also a vast performance gap between the P3 and the risc platforms of the time. All of these provided 64bit doubles,...


Top 50 recent answers are included