diff --git a/libs/pbd/fpu.cc b/libs/pbd/fpu.cc
index c179def222..0c281a74e8 100644
--- a/libs/pbd/fpu.cc
+++ b/libs/pbd/fpu.cc
@@ -266,11 +266,21 @@ FPU::FPU ()
 			memset (*fxbuf, 0, 512);
 
 #ifdef COMPILER_MSVC
-			char *buf = *fxbuf;
+			char* buf = *fxbuf;
+#ifdef _WIN64
+			/* For 64-bit compilation, MSVC doesn't support inline assembly !!
+			   ( https://docs.microsoft.com/en-us/cpp/assembler/inline/inline-assembler?view=msvc-160 ) */
+
+			/* but instead, it uses something called 'x64 intrinsics'
+			   1: ( https://docs.microsoft.com/en-us/cpp/intrinsics/x64-amd64-intrinsics-list?view=msvc-160 )
+			   2: ( https://software.intel.com/sites/landingpage/IntrinsicsGuide/#text=_fxsave ) */
+			_fxsave (buf);
+#else
 			__asm {
 				mov eax, buf
 					fxsave   [eax]
 					};
+#endif
 #else
 			asm volatile (
 				"fxsave (%0)"