As I am sure you are ware, the 'AAE' specifies the programmign of an internal ROM which generates the output code from the key matrix positions. So as yo've found, if you fit one with the wrong programming, it works, but gvies the worng keycodes. Not with this IC, but there are plenty of fake and mislabelled semiconductors around. 2 stories, one from a fried, one that happened to me. A friend found a supplier (in CHina, I thin) who had a vast list of ICs they claimed ot be able to source, including a number of HP custom parts. He asked about oen of the latter (I think a ROM for the HP41 calculator) and was asked 'what package and how many pins'. Given that said IC ony ever existed as an 8 pin DIP it is clear that said company would lable any package with any number. The one that happened ot me. A couple of weeks ago I bought what were claimed ot be 1A SCRs (thyristors, whatever). I fitted on into the cirucit I wwas working on and found it was 'on' all the tiem. Even with no input to the gate, the gate and cathode being linked by a resistor, it was still turnign on at switch-on. And it did not test shorted out-of-circuit. After much testing, I found I had, in fact, been supplied with PNP transistors (!). One had the correct number for the SCR on the can, the other 2 I had were not labeleld at all. If you think about the circuit, it turns out that the thing ends up with emitter and collector swapped round and a resistor between base and what is nwo the collector (most -ve electrode). OK, it's a poor transistor with minimal current gain but that reissotr supplied enough current to bias it on. So yes, mislabelled components are alas all too common now. You were lucky, I think, you got a keyboard encocder and not just some other 40 pin device. If I was being unkind I would guess that soembody had noticed the AAE version was useful for fixing Apple ][s, the AAD version was not a lot of use, and so they got re-labelled. They may not even be National Semiconductors devices.. -tony