In Windows' time.clock(), when QueryPerformanceFrequency() fails,
the C lib's clock() is used, but it must be divided by CLOCKS_PER_SEC
as for the POSIX implementation (thanks to #pypy).
diff --git a/Modules/timemodule.c b/Modules/timemodule.c
index 9ab2724..4f562c7 100644
--- a/Modules/timemodule.c
+++ b/Modules/timemodule.c
@@ -175,7 +175,8 @@
 		if (!QueryPerformanceFrequency(&freq) || freq.QuadPart == 0) {
 			/* Unlikely to happen - this works on all intel
 			   machines at least!  Revert to clock() */
-			return PyFloat_FromDouble(clock());
+			return PyFloat_FromDouble(((double)clock()) /
+						  CLOCKS_PER_SEC);
 		}
 		divisor = (double)freq.QuadPart;
 	}