Interresting findings with a logic analyzer (although I shouldn't have to be reverse engineer YOUR product initialization sequence...it should be properly documented in the first place).

The MOS product initializes the parallel LCD according to the 4-bit HD44780 standard and NOT the S6A0069 as is ...

By any chance, have you looked at the signals with a scope? I will try and see what I can catch with the code, but the most important thing is to follow the data sheet. I have always thought that the S6A0069 is fully compatible with the HD44780U, but you are right, Philip, they have quite a few ...

How fast is the instruction cycle on your ATmega? If it's running pretty fast, check to see if the setup time for data and\or RS is adequate. If not, add a little delay after setting the LCD data and RS. Your enable bit handling has sufficient delay.

Your initial initialization order (0x38, 0x38 ...

Also, you will need to have control of R/W because you will need to know if the display is ok to receive data (not busy).

Reading the busy flag is not a requirement, however, it is good practice. Its perfectly fine since he's using rediculously long delays. If the LCD stays busy for over 100ms ...

Hello silverstein,

I am looking at page 45 of this data sheet for the HD44780U .
I do not think that the code you have posted comply to it. Can you please check if ou can follow the suggested initialization?

Thanks,

I'm curious....why are you linking the HD44780 manual when all of the MOP-162A ...