In my work I frequently reuse the same hardware. All the LCD's I frequently use, and which have the typical 14-pin dual-row connector (the character types) use the HD44780, though they came from different sources. That enables me to avoid some of the pitfalls of having to buy LCD's without clear device-specific datasheets.

What I have found helpful is to examine the controller IC (which, as has been pointed out, may not be accessible) in order to trace the interface connector pins back to the controller IC.

I have had to deal with LCD's that used the KS0066U controller, and found that they behave identically to the ones with the HD44780. Now, that's likely to be a coincidence, but the controller datasheets don't indicate a significant difference, IIRC.

I, too, have encountered 16-pin interface connectors that had backlight connections on them, and I have also encountered 16-pin connectors that had two E signals, one for each controller.

It's been a long time since I dwelt over the 5-cm-thick Hitachi LCD controller databook. Timing differences do appear to be caused by different driver strategies. More driver IC's seem to lead to different cycle times than fewer, and operating the HD44780 in "open-loop" mode seems only to work when the HD44780 is the only player. If one monitors the "busy" flag, the timing issues seem to go away.

Now, I don't know anything about a negative-going 'E' strobe, not having encountered it in any specification.

RE