toggle quoted messageShow quoted text
APRSIS32100211 seems to have fixed all my NMEA GPS (hard wired) problems. I will see if I can check Bluetooth later and with a bit of luck see if I can emulate a KISS TNC on a Netbook and try some RF at the weekend.
--- In firstname.lastname@example.org, "Lynn W. Deffenbaugh" <kj4erj@...> wrote:
There is a new release for both Win32 and Windows Mobile that fixes one
problem and optimizes screen redraws a bit.
If you have an NMEA GPS and don't see the time/date alternating in the
lower left corner even though you see satellite information across the
bottom, then this fix is for you!
If you'd like to see screen redraws be a bit peppier with less CPU load,
or would like the map dragging to go smoother (especially when taking a
maximized window back down to restored), then this release is also for you!
If you're not interested in either of those features, and have already
upgraded to 2010/02/09, then you'll probably be fine with that one.
As always, you can find the new release in the Files section of this
Yahoo Group. Just unzip the .EXE and drop it over the one that you've
been running and it's back to the races!
Lynn (D) - KJ4ERJ - Author of APRSISCE for Windows Mobile and Win32
The Background (if you have time to read)
This evening I was notified by N0ODV-9 that APRSIS32 wouldn't transmit
on his OpenTracker. Further diagnosis revealed that it wasn't actually
his OT at fault, but the fact that APRSIS32 didn't think he had a GPS
fix. It turns out that his GPS was putting more pieces into a $GPRMC
sentence than my client was expecting (I get really nit-picky when I
write code that interprets data according to a published protocol). The
problem was indicated when the client was showing satellite information
across the bottom, but not the GPS-based date and time in the lower
right corner. Even though it was claiming he had a 3D fix, I wasn't
getting the coordinate information. No coordinates, no position beaconing.
If you're trying to test RF transmissions, don't use the Transmit
button. All it does is attempt to force out a position update, but that
won't happen if you've got the min time set reasonably high or if the
client doesn't have a fix.
A more reliable way to force a transmission is to send a message to
someone. In the current version, APRSISCE/32 transmits outgoing
messages (and acks to incoming messages) via all available means. This
means it goes out APRS-IS, KISS, and AGW if they're all enabled. If you
send a message, you can expect to see PTT from your TNC.
My thanks goes out to N0ODV for his willingness and donation of time to
help me diagnose the NMEA GPS support issue.
Then, later this evening, I got a note from ZL3REW-8 that he wasn't
seeing any beacons on APRS-IS. A quick question confirmed that he was
also seeing satellites, but no times. I e-mailed him (her? Robin?) a
quick copy of the updated .EXE and the position packets were flying.
Another happy user.
Then I noticed some strange diagnostics in N0ODV's debug log file so I
got to looking around at the map painting code when coordinates were
changing or the map was dragged a small distance. I found that the
client was throwing away perfectly good map tiles (not really, just
swapping some pointers around) when it didn't need to. I implemented
better change detection in the loaded map tiles which results in
smoother map dragging and lower overhead / faster screen painting for
small map movements. This is especially visible if you maximize the
window, wait for it to catch up, and then restore it back to something
smaller. Dragging the map around in that window is MUCH smoother than
it was before.
Anyway, even N0ODV and ZL3REW should consider downloading this latest
copy. It's better than the one they got from me last night!
Lynn (D) - KJ4ERJ