As time goes by, there is one constant- one's memory sure does fade.
When I made this previous blog post, I thought about how far I have come with these bits and pieces of software and its quite interesting. I know I did some embedded code, well, an attempt at a cash register of sorts, with a graphic LCD and the only viable processor I had at the time, a PIC16C74A.
Now, as we know, that processor, is pretty ancient by today's standards - 4k of EPROM (yes, no such thing as FLASH) and 192 bytes of RAM. Now, to display graphics, and copious volumes of text there was just no way that was going to fly, but, I managed to make it work regardless. I did it by using an external EPROM and external memory that would be read to not only retrieve text, but also the fonts on the LCD.
Sadly I do not have any surviving photographs of that project (mainly due to not being able to afford a digital camera at the time) but, the code was developed between April and September 2001. All of it done, in DOS, on a very old IDT WinChip box that ran Windows 98.
The format of the fonts are the forerunner of what GraphLCD is today- indeed I am about to extract the fonts from that very old EPROM dump and incorporate them into the current library.
Not bad for an era where I had no idea how to code in any language except assembler. I had not even learnt C yet.
From the above it is clear no header byte was used, and this was done for three reasons:
1. My math in ASM was bad, hence why C remains the first choice of language for development of anything new, especially when it becomes formalized as an ISO standard.
2. That retarded PIC micro had abysmal math instructions, even adding numbers together was a pain in the arse as I recall.
3. Ease of readability. Back then I was not nearly as proficient as I am now. Remember at the time I did not even work in IT was mainly just postgrad and trying to find myself an engineering job.