Re: TI-H: ti-calc schematics...


[Prev][Next][Index][Thread]

Re: TI-H: ti-calc schematics...




Yes, you can control the TI link port by accessing port 7. However,
the TI calculators do not have an accurate timebase (they use an RC
pair rather than a crystal) so do not expect any kind of accurate
serial communication. Of course you could probably manage to do
RS-232 by having a loop measure the baud rate of an incoming CHR$(13)
-- I was just reading on article on how to do this on a C52 -- and
then you can send/rec stuff at that baud rate just knowing how long
to wait between bits. There is a pretty decent chance that the
8X won't be able to accurately or consistantly match that baud
rate though, it may be off a few percent. Over about 2.5% timing
error and you'll start to see errors. I recommend using something
like the graphlink if you really must have serial communication.
The E2 will also give you some extra interfacing options (4 I/O
and bi-dir half-duplex serial)

The TI Protocol is asynchronous, meaning that it does not rely
on knowing the speed the other calculator is sending (at least
not in theory, there are some electrical limits that keep the
baud rate fairly low -- capacitance and inductance both being
factors to some extent). Basically, the sending calculator
takes one of the two lines active (the lines are active low),
then the other calculator acknowledges by taking the other line
low, then the sending calculator takes its line high again
and waits for the receiving calculator to do the same. Its
simple, fairly error-tolerant, but it is slow.

I believe I discussed why TI most likely chose an RC pair
over a crystal or resonator in a past message, you may want
to check the archives.

Hopefully those guys over in Europe have the two frayed,
rusted wires connecting them to the US fixed by now. ;-)
Otherwise, this message will probably take a while to
get to the list.

Regards,

Bryan

Eric M. Fiterman wrote:
> 
> What exactly is the TIP?  Can it be changed at the assembly level?
> 
> > No.  Its TIP (TI PROTOCALL)
> >
> >
> >

-- 
Bryan Rittmeyer
mailto:bryanr@flash.net
http://www.flash.net/~bryanr/


Follow-Ups: References: