[Xastir] Serial KISS interface - long delays

[J. Lance Cotton]

Curt Mills, WE7U wrote:

> Same here.  It turned out to be that the parameters changed from
> 10ms steps to 1ms steps in the kernel, and an SuSE upgrade broke my
> AX.25 because of that.  That's actually the reason that I wrote the
> serial KISS interface code, so that I could operate again.  Turns
> out cranking up my numbers by 10 in AX.25 configs would have got me
> running a bit more quickly.  ;-)

What kind of errors would be seen by this kind of change? I got some error 
about bad port configuration, which sounds like what you describe. I wonder if 
I should try adjusting my settings. Thanks!

> That's it.  For serial TNC interfaces we do character pacing, so as
> not to overrun slow TNC's (like an old dual-port PicoPacket that I
> have).
> 
> It shouldn't really hurt your RF operations in any way.

It doesn't hurt. I am just being impatient :-) I am testing a new digi for the 
local club and when I "Transmit Now" I have to wait a few seconds. It also 
means that, since I have relay digipeating on in Xastir, when Xastir hears a 
packet destined to be RELAY digi'd by me, it doesn't get TX'd until 2-3 
seconds later and by that time perhaps some other station has relayed it on. 
That's the way it's supposed to work, except Xastir has already begun sending 
the packet to the TNC for digipeating and so a duplicate occurs. So a few 
relays of a packet could come in while Xastir is twiddling its thumbs to 
accommodate the slowest TNC. Could we make the delay a configurable parameter 
(after the code freeze is over)? Maybe "hidden" like Mozilla's prefs; it could 
be in the config file, but have no gui interface to change it. A power-user pref.



-- 
J. Lance Cotton, KJ5O
http://map.findu.com/kj5o-14
joe at lightningflash.net