From: George on
Jim Brain says...

> Nice to see you in here, George.

Thanks, Jim. It's nice to be here.


From: AgentFriday on
George <gh424NO824SPAM(a)cox.net> wrote:

> I agree. Something else is going on. Maybe not opening the
> port correctly. Or something. An awful lot of people have
> used RS232 at 300 bps, and even 1200 bps, with good results.

Well, to make the world seem right, this is what I tell myself:
- Most terminals came with their own routines
- Anybody typing outside of a terminal program may have
chalked up the occasional error to line noise (not uncommon
back then...)
- Actually we tried putting a delay between each byte sent, and
almost never saw an error. It was back-to-back that screwed
up.
- People transmitting data via modem (up/downloads) most likely
used a xfer protocol that corrected the occasional error.

Thanks everyone for your input. And Thanks to George for the
rock-solid code. :)

AgentFriday <my_nic_here(a)hotmail.com>
From: christianlott1 on
On Mar 23, 10:31 pm, AgentFriday <my_nic_h...(a)hotmail.com> wrote:

> - Actually we tried putting a delay between each byte sent, and
>    almost never saw an error.  It was back-to-back that screwed
>    up.

I used the swiftlib routines at:
http://www.csbruce.com/~csbruce/cbm/swiftlib/

on an rs232 cart of jim brains design.

The delay on the server side was necessary. I didn't time it but it
wasn't as fast as I wanted, certainly not as fast as the uIEC which I
later purchased. It also wasnt worth the hassle of getting the serial
port reconfigured every time I rebooted my xp machine (or some problem
like that).

From: George on
AgentFriday says...

> Thanks everyone for your input. And Thanks to George
> for the rock-solid code. :)

You're welcome. But I still would suspect things aren't set
up right using the stock routines, like both ends not using
8N1, or different baud rates, or something like that.


From: AgentFriday on
On Wed, 24 Mar 2010 13:34:04 -0700 (PDT), christianlott1
>I used the swiftlib routines at:
>http://www.csbruce.com/~csbruce/cbm/swiftlib/

There's that csbruce guy again!

>on an rs232 cart of jim brains design.

I assume it was a basic transciever circuit that plugged
into the user port?

>The delay on the server side was necessary.

So even with replacement routines you couldn't receive
back to back? Do you recall what baud?

It'll be interesting to compare swiftlib against George's.


AgentFriday
<my_nic_here(a)hotmail.com>