>To all those who gave suggestions on my serial port reading problem. Thank
>you. It turns out that I had to insert a delay of 30 to 50 ms. before
>actually reading the port for data. I have learned a lot from the examples
>and have incorporated many of them in my routine. As a clarification to the
>problem, the speed is fast (38400), RS-422, they don't use handshaking, and
>the reads are 30 times a sec. There is no end of transmission character
>since the header that is sent embeds the length of the transmission. What I
>ended up doing is hooking up one computer to *see* the data and compared
>that with what I was getting from my routine. I've been reading serial ports
>for 10 years and still haven't perfected it, probably never will. One thing
>I did learn, there are many ways to do the same job. Thanks again.
>
>rod k.


rod:

The problem with that is that not all machines will require 30 to 50 ms. I
would guess that some may require more and others less. Normally, you open
the port, fill the buffer, and then see what's there. But, some external
devices require delays between sending and receiving data (Casio Boss units
for example) but those require very labor intensive study to figure out
what devices require what. Good luck.

tedd

http://sperling.com/
_______________________________________________________
Thinking about buying Jewelry?
Try looking at: http://earthstones.com