> Maybe I was speaking loosely.
> In the original spec, +3V is a zero, and -3V is a 1
> Everything inbetween is undefined.

I have had a lengthy discussion about this on (I think) com.arch.embedded.
RS-232 specifies
- mandatory behaviour for a sender
- mandatory behaviour for a receiver
These two sets are NOT the same, and beside the mandatory behaviour there is
a great deal of freedom for the implementor of both the sender and the
receiver.

The 1488, 1489, MAX232 etc. implement the mandatory behaviour, and each has
made his own choices within the allowed freedom. But each compliant sender
can still talk to a compliant receiver. RS-232 might be outdated, but it is
(was) still a good spec.

The trouble starts when people think like: this is how a 1488/1489 behaves,
so this is what I can expect from an RS-232 interface. If I drive it with
0..+5V it will work perfectly. That is NOT true for all RS-232 interfaces:
you can only expect what is stated in the RS-232 spec. The trouble is that
the RS-232 spec is not freely available, an the specs of the schips are....

Wouter

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics