Re: Slightly unmatched UART frequencies

GPG <[EMAIL PROTECTED]> wrote:
> MAXIMUM error is .5 bit over one frame. In your case frame = 10 bits.
> .5/10 = 5%

I remember the story at a former employee who was using an HP serial test
set at 38400 bps to debug a data stream that went to a slot machine
controller.  He kept having inexplicable data corruption errors until he
looked at the timing very closely on a 'scope and discovered that -- due to
an 'off by one' error in some register of the UART in the microcontroller
generating the data stream -- the actual bit rate was something like 10%
off.  It had gone undiscovered for years since the microcontroller at the
receiving end was using the same code!

---Joel Kolstad