I want to control a 7 x 46 segment LED matrix. (this used to be 7x23 but my needs have now changed) in my other thread they said I could get the two PIC's to communicate to each other to scroll text across the entire display. So now I just need to know if this is really involved or not
You didn't mention what PIC chip you plan to use. However, if you proceed with using a PIC16F877, you can daisychain them by connecting their serial ports as:
As soon as the character scrolls off on one display, the ascii code for that is passed to the next PIC. Like passing the baton. Just a suggestion. There are other ways of course.
You didn't mention what PIC chip you plan to use. However, if you proceed with using a PIC16F877, you can daisychain them by connecting their serial ports as:
As soon as the character scrolls off on one display, the ascii code for that is passed to the next PIC. Like passing the baton. Just a suggestion. There are other ways of course.
Yes, there are other ways, but this way is fast, simple, and easy (and requires hardly any code to do) - there doesn't seem much point in getting involved in more complicated schemes, unless there's other considerations.
With chips (like the 16F877) which have a hardware USART you can simply use that, which makes live very easy, and has very little overhead - but you can also do it with a software UART on any PIC.
No it doesn't, I don't see as there's any difference between the different speed chips - all seem to work at over 20MHz without any problems. I can't see as they run three different production lines for the same chip, and deliberately make poorer ones on two of them?. As I see it ALL the chips come off the same production line, and they mark them at random with the different speeds.
Even assuming they DO actually speed test them, can you see a 20MHz engineered chip failing so badly it gets down rated to 4MHz?.
I've never heard of a 4MHz chip behaving any differently at 20MHz, and I've also run an old 4MHz 16C84 (which were only available as 4MHz!) at 10MHz in place of a 16F84 - it behaved perfectly (and still does), it was a design for a colour bar generator.
If that were the case then why wouldn't they mark them all for 20MHz?
I'm sure it's a quality control issue...The samples failing tests at 20MHz cause the whole batch to get marked for slower speeds. Or perhaps they can tell ahead of time by the quality of the silicon that a batch is not going to be as good? Dunno.
Their tests are probably more rigorous than ours. We may not be testing every aspect of the chip at 20MHz.
cool I'll just order the 4mhz then it's cheaper, and I don't think speed is going to be an issue here.
I've got another question that probably should be in the other forum but I'll ask anyway. on the matrix i'm creating what type of transistors should I use. I'm going to use this basic schematic (of course adjusted to 7x64), but it's kind of dated, so i just wanted to see if the 2N2222's were still the best choice.
The chips have work over their entire temperature range. High/low temps might cause problems. Embedded chip companies do tend to be very conservative about their specs though.
They also might not test the 4MHz chips. They know that they will run at least at 4MHz with very little tesing. If they ran a full test on the ones labled 20MHz to guarantee that speed it would add a bit of cost.