Re: [NTLK] Newton (Fast) Serial Docking for newbies

From: RAParker (RAParker_at_Quadzilla.NET)
Date: Sat Jan 29 2005 - 12:16:38 PST


On Friday, January 28, 2005, at 01:36 PM, Adriano wrote:

> Did you try to set the Keyspan speed in the Keyspan serial assistant
> 'Advanced settings' panel?

Yes. I tried to fiddle with the settings to get a 230400 connection
with the 230400 enabler on the Newton, to no success. The Keyspan
Serial Assistant diagnostics always reports a connection of 76800.
That, of course, results in a mismatched connection and the Newton just
times out waiting for the connection.

I have only been successful (and easily so) with a 115200 connection.

>
> The Keyspan adapter is a RS422 device, it should be faster and more
> stable than a RS232 device.

The Keyspan does seem very stable but appears to only be able to handle
a 115200 serial connection. No complaints though. Even at 3 times the
(theoretical) speed, it probably too fast for the Newton to keep up,
anyhow. I've done some preliminary tests and I'm seeing only about 2x
speed boost. Cutting backup times nearly in half.

I've watch the bytes counts come across and the transmit/receive's are
definitely faster. But...the pauses between blocks of transmissions are
also much longer. With the MP120, I'd call it a processing bottleneck.
I'm sure 2x00's would do a whole magnitude better than a 120.

The good news is a full backup of my Newton (8MB) can now be done in
30-40 minutes instead of 60-90 minutes. I'll do some byte/second
calculations when I can devote several hours to testing, timing and
calculating.

I'm also going to pickup one of those RS-232 to USB adapters and try
230400 again without the limitations of the Keyspan device.

>
> Take a look on the Advanced settings panel of yor keyspan software and
> try to configure it for the 230K bitrate. If you need detailed info
> about the settings of the keyspan assistant, just drop me a line, and
> i will post it later.
>

I'm currently experimenting with those settings, only now to simply
reduce the H/W overruns to a minimum. I'll contribute to the wiki page
when I've finalize some basic results. Thanks again for all the
valuable information, the results thus far are very encouraging.

Ron

--
---
--------------------------------------------- ThinkApple
low|               5% use a Mac                    |\/|\
end|                                               |/-|/
mac|            20% wish they did                  |\ | @ Quadzilla.net
-------------------------------------------------- A LEM Bumper Snicker
---
--
-- 
This is the NewtonTalk list - http://www.newtontalk.net/ for all inquiries
Official Newton FAQ: http://www.chuma.org/newton/faq/
WikiWikiNewt for all kinds of articles: http://tools.unna.org/wikiwikinewt/


This archive was generated by hypermail 2.1.5 : Sat Jan 29 2005 - 14:00:01 PST