serial port speed

Archiver

Stampers,
I need to read an Elo touch screen (serial output), process the packets,
and send the converted results to serial DACs. I've done this in the past
with no problem, but now I am being asked to do it at a higher throughput.
There has been recent discussion about the Stamp's reliability at 9600 baud.
Is this something that can be overcome by using off-board RS-232 circuitry?
Are the newer p series Stamps or the SX a better choice?
The BX-24 indicates that it's serial port can be run at 460 Kbaud! Is it
time to try one?
I'd appreciate benefiting from your experiences.
Thanks, in advance,
Mike

_________________________________
Mike Walsh
walsh@i...

Archiver

[font=arial,helvetica]In a message dated 4/12/01 8:41:00 PM Central Daylight Time,
walsh@caltech.edu writes:

The BX-24 indicates that it's serial port can be run at 460 Kbaud! ·Is it
time to try one?

More hype than hope. ·Read those BX-24 specs very carefully.[/font]

Archiver

At 6:37 PM -0700 4/12/01, MIKE WALSH wrote:
>Stampers,
>I need to read an Elo touch screen (serial output), process the packets,
>and send the converted results to serial DACs. I've done this in the past
>with no problem, but now I am being asked to do it at a higher throughput.
>There has been recent discussion about the Stamp's reliability at 9600 baud.
>Is this something that can be overcome by using off-board RS-232 circuitry?
>Are the newer p series Stamps or the SX a better choice?
>The BX-24 indicates that it's serial port can be run at 460 Kbaud! Is it
>time to try one?
>I'd appreciate benefiting from your experiences.
>Thanks, in advance,
>Mike

Hi Mike,

The issue is that bytes arrive at the serial port, head to heel, so
fast that the stamp does not have enough processing time between
bytes to do its otherwise very effective WAIT or SKIP commands. That
is at 9600 baud and above for the BS2 and BS2e. The BS2SX and the
BS2P are faster, and they can handle 9600 but start having trouble at
19200 baud and above.

The BS2P has a modifier that allows you to capture long serial input
sentences (up to 127 bytes) very fast directly into the scratchpad
RAM:
'capture 127 bytes to scratchpad with timeout
SERIN dpin,baudmode,100,nodata,[noparse][[/noparse]SPSTR/127]
I don't know what the speed limitation of that is, but I guess it
would be pretty fast.

-- regards,
Tracy Allen
electronically monitored ecosystems
http://www.emesystems.com

Archiver

Tracy,
I've never tried to use that modifier, but it looks like it could help. It
would seem that I should be able to use one stamp as a source, creating
sequential character values, while reading it with another stamp. I should
be able to see how much is lost at different speeds and with using the
modifier vs no modifier.
Thanks for the help!
Mike


At 08:37 AM 4/13/2001 -0700, you wrote:
>Hi Mike,
>
>The issue is that bytes arrive at the serial port, head to heel, so
>fast that the stamp does not have enough processing time between
>bytes to do its otherwise very effective WAIT or SKIP commands. That
>is at 9600 baud and above for the BS2 and BS2e. The BS2SX and the
>BS2P are faster, and they can handle 9600 but start having trouble at
>19200 baud and above.
>
>The BS2P has a modifier that allows you to capture long serial input
>sentences (up to 127 bytes) very fast directly into the scratchpad
>RAM:
> 'capture 127 bytes to scratchpad with timeout
> SERIN dpin,baudmode,100,nodata,[noparse][[/noparse]SPSTR/127]
>I don't know what the speed limitation of that is, but I guess it
>would be pretty fast.
>
> -- regards,
> Tracy Allen

_________________________________
Mike Walsh
walsh@i...