> amd64, ia32 emulation on an amd64 and ppc32 produce the same -- mostly > correct -- result for r/w/t now.
Yay. I've committed this fix to my tree. Thanx. > But things like still appear (they > also happen with 1.2.4 on i386): (ele should be zero/undefined) > > <wpt lat="47.370760795" lon="8.535247492"> > <ele>9999999562023526247432192.000000</ele> > <name>DB ZH</name> > <cmt>DB ZH</cmt> > <desc>DB ZH</desc> > <sym>Waypoint</sym> > </wpt> If I'm guessing this problem correctly, it only impacts the read of waypoints when the alt is unknown, right? That's a known outtage in the jeeps library that's about second or third on my list of stuff to deal with next. The core problem is that GPSBabel has a (bad) concept of "unknown alt" and the Garmin protocols have about three different concepts of "unknown alt" and the interface between the two has no concept of unknown alt, so it's not at all uncommon to see wacky results for that. Yeah, it's really lame, but it's bulky to fix becuase I have to change 23 different formats at the same time to "fix" this. Watch the GPSBabel-code list for a fix on this one. RJL -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
