On Mar 1, 2006, at 9:13 PM, Bill & Pookie wrote:

> It will take you longer to send the 8 bits and parity than you
> receive the 8 bits, no parity.  If this keeps up, you will get
> behind and need to buffer the receive characters.If you can
> get the sender to send 2 stop bits, this would solve the major
> problem with speed.  But is the sender has a clock that is a
> little faster, then this difference in speed would build up...

Most systems that require parity will be wanting 7 data bits
plus a parity bit, and most that don't transmit parity will
send 8 data bits with a zero in the 8th bit, so that part is
unlikely to be a big problem.

The easy solution is to transmit much faster than you receive.
Something along these lines was my first HW project after getting
out of college; for newswire translation, I read 6bit codes at
75bps and transmitted 8bit codes at 300bps.  No problems.  (no
micro involved, though.  A couple 6402s, 555s, and a flipflop to
get the two clocks...)

Note that if you have a uart capable of doing different transmit
and receive clocking, you only need a single uart to accomplish
this - one receive channel and one transmit channel.  Alternately
you can receive with a SW uart and rely on the receive clocking
to simplify transmit (on HW uart) so you don't even need interrupts.

BillW
-- 
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist