The problem with master keying in this way is that with a single access lock, there is one pattern that opens it. If you add a master wafer to one pin, there are now two patterns that open it. If you add a master wafer to two pins, there are now FOUR patterns. In your example with all pins master wafered, there are 64 (!) patterns that now open the lock. MUCH less secure than before. "So why not only master wafer one pin?" Aha! The OTHER problem with this is that a rogue can decode the master key with a few blanks, some files, and some time alone with the lock. I'm writing this from memory, so I might be wrong, but I believe the way you do it is you grind a blank to match your personal key, except for one notch, which you leave as a zero cut. Try it in ther door. Does it open? Then the master key has a zero cut for that notch. Does it not open? Grind that notch down one height, and try it again. Does it open? The master key has a one cut for that notch. If you try all the heights and it only opens when it's the same as your personal key, then your personal key and the master key have the same cut for that pin. Now do the same again for the next pin with another blank. The reason this attack is effective is because you can ask the lock "do I have _this specific cut_ guessed correctly?" And it will answer. If, as the door owner, you instead install two locks, both of which open the door, but where one of them is keyed alike for your whole security domain, and one is personally keyed, this attack doesn't work at all, because the attacker would have to guess ALL of the cuts correctly before the lock opens. I'd love for you to do a video on lock picking.