Update NumLock comment now that we have sync
diff --git a/unix/xserver/hw/vnc/Input.c b/unix/xserver/hw/vnc/Input.c
index 5c458f5..228b382 100644
--- a/unix/xserver/hw/vnc/Input.c
+++ b/unix/xserver/hw/vnc/Input.c
@@ -537,9 +537,9 @@
* get confused when we do a fake shift to get the same effect
* that having NumLock active would produce.
*
- * Until we have proper NumLock synchronisation (so we can
- * avoid faking shift), we try to avoid the fake shifts if we
- * can use an alternative keysym.
+ * Not all clients have proper NumLock synchronisation (so we
+ * can avoid faking shift) so we try to avoid the fake shifts
+ * if we can use an alternative keysym.
*/
if (((state & ShiftMask) != (new_state & ShiftMask)) &&
vncGetAvoidShiftNumLock() && vncIsAffectedByNumLock(keycode)) {