V3.09 (non-beta) Ready

[prbflight]

Other than the polar chart spikes, everything else seems fine on this new version.

I agree with Tarbat.  This newest beta is working fine for me other then the syncing of aircraft history to aircraft that I have mentioned in previous reports.  As I'm not in a high traffic area where one flight bumps the next's history onto MyFlights I have accepted this normal also.  I haven't seen anyone else on the general forum complain of this issue.

Paul@cyyb

[RodBearden]

Here's my spikey polar, generated over the 20-ish hours since I downloaded the latest exe.

EDIT - My RB's only actually been running for about 3 hours since I downloaded the exe - I turn it off when I'm asleep or at work.

Rod

[tarbat]

Rod, was that with just local hardware aircraft, or with network aircraft as well?   I only get a spikey polar if I have real-time network turned on.

[RodBearden]

I always have the network on. Except when I took this screenshot of course- I turned all flights off to make it clearer :-)

Rod

[abrad41]

For what its worth, I downloaded the "exe" this morning and the computer has been running all day since about 08.00.  Everything is working fine here, I include a screenshot of my polar and everything is normal there, but I do not use the network, so this could be a network problem.

Andy

[pjm]

fwiw these are the area's I'm interested in

I've done a little logging this morning and analysed the results, and found approx 16% error rate - i.e. invalid Longtitude. The areas involved confirm my suspicions of the 2 blackholes in the diagram above.

fwiw of 14502 MSG3's captured in about 80 minutes, 2312 have invalid Longtitude. My logging so far shows no errors in the Latitude.

Latitude (±90°) range -32.5117 <-> -35.1491
Longtitude (±180°) valid range 147.8562 <-> 152.4303
Longtitude (<>180°) invalid range -208.2465 <-> -211.9795

Raw data is here for those interested:
http://www.petermcgrath.info/AirNav/30003.xls

I must say, I'm looking forward to getting this bug fixed - an additional 16% of aircraft on the radar screen will be great!

[pjm]

Very interesting comparison from SBS plotter and RB Polar diagram.

Seems SBS plotter is taking the invalid Longtitude's from RB and adding the 360° offset and displaying the aircraft.

In the example here you can see SBS plotter has tracked JST444 for the past 5 minutes and displayed it on its diagram, while it had JUST come out of the blackhole and started to be displayed by RB...

[tarbat]

pjm, a conclusive test would be to reset your polar chart, and clear the data in sbsplotter, then run both for a day.

You can then save an OUT file from sbsplotter, put that into your outlines folder, and then display the outline of the "true" polar chart from sbsplotter in ANRB.

This will also be a useful technique to test the fix when Airnav finally fix this bug.

[pjm]

Without even resetting my existing polar diagram its pretty conclusive - this is with just approx 8 hours today worth of sbsplotter

[AirNav Development]

We are still working on the blind rectangle problem. We sent a new .exe to 5 different users a few minutes ago and we are waiting for their feedback.

[pjm]

looking good :)

[AirNav Development]

Taking in account the feedback from 2 users (Australia and NZ) we have the blind rectangle problem finally corrected. Only one problem remains: spikes in the polar diagram. We have contacted the users that reported it.

[DVDW]

I reset my polar diagram yesterday afertoon and this morning it showed like this. (network not active). I can't observe any spikes.

Regards,

Dirk.

[tarbat]

I reset my polar diagram yesterday afertoon and this morning it showed like this. (network not active). I can't observe any spikes.

Yes, I'm pretty sure these spikes are caused by network aircraft that have transitioned fromlocal hardware aircraft.  Airnav, let me know if there's any testing you want me to do.

[pjm]

Yes, I'm pretty sure these spikes are caused by network aircraft that have transitioned fromlocal hardware aircraft.  Airnav, let me know if there's any testing you want me to do.

I think its more complex than that, I don't usually have the network on, but I left it running last night (mine is the delayed one) and recorded port 30003, there were over 100 "spikes" last night from all around the globe (as far away as Africa on the other side of the planet - much like the UK :) )

So there would have been no transitions there. Likely the ones people are seeing on the polar diagram are just the ones under the "limit" before they get discarded as obviously fallacious.

Actually after looking at the data again, it looks like it may be corrupted local data after all - not sure why it only happens when the network is on ?