Re: [gpsd-dev] SemPiTernal - Bounding PPS uncertainty

[Eric S. Raymond]

Hal Murray <address@hidden>:
> 
> > Connecting a TTL signal directly to a RS-232 modem control input pin ends up
> >  triggering on 
> 
> Rats.  Sorry for poking send too soon.

OK, ignore previous complaint. :-)

> Connecting a TTL signal directly to a RS-232 modem control input pin ends up 
> with assert being the leading edge which is what ntpd uses by default.
> 
> (It also works in the normal case even through it doesn't meet specs.  It 
> won't work as well if you have long cables.)
> 
> Aside from wrong-edge, the other problem you can get into is pulses too 
> narrow for software to notice.  GPSDOs tend to have 10-20 microsecond pulses. 
>  I think I have added counters to ntpd's clockstats for the PPS/ATOM driver 
> and the PPS handling in the NMEA driver.

Clearly this is an area where my knowledge is deficient. I will learn.
-- 
                <a href="http://www.catb.org/~esr/";>Eric S. Raymond</a>