Real Traffic Data Quality

[mmcmah 407]

I received this this morning from Balthasar:

"In other news, there's a good chance the traffic coverage has improved as of a few minutes ago, if you want to go take a look and let me know if it looks better now?"

Those of you who have reached out to him might want to give him feedback. I'm only now going to get the chance to test things. 

Edited April 14, 2023 by mmcmah

[kiek 1,215]

8 hours ago, mmcmah said:

Do you think historic will be restored in some fashion Nico? That's what allows us to time shift in order to accommodate when we can fly.

Ofcourse it will. But I guess at least starting from now on. Tomorrow you can go back to today without problems etc..

But maybe Balthasar from RT can repair the history, don't know.

[kiek 1,215]

7 hours ago, G-YMML1 said:

I have FB KDEN and today I got PSXT message KDEN.xml found but cannot be loaded. No document element found. Yesterday I did not have this problem. I re-ran airports scanning with PSXT and still receive the same message

Then the .xml file has a syntax error. There is a FAQ about that.

Restart with a fresh file.

Be careful: delete the corrupted file from the updated_airports folder! Otherwise you will continue to try to use the corrupted one instead if the newly generated file.

 

Edited April 14, 2023 by kiek

[kiek 1,215]

2 hours ago, somiller said:

 

I just loaded into KDEN with PSXT/RT and it read the xml fine - I just wanted to see if maybe PSXT was more critical than Learner regarding KDEN.xml

No, both programs share the exact same code. Learner is just a slimmer version of PSXT (created by conditional compilation)..

Edited April 14, 2023 by kiek

[G-YMML1 582]

I ran KBOS in the historic mode for 45 minutes this morning  - worked flawlessly. Testing KDEN now with 2 hours back (14:00 GMT). So far, so good

[mmcmah 407]

This is what I sent to Balthasar about 12 hours ago:

I tried KDEN, LAX and YSSY in realtime and found that there was a good amount of traffic moving on the taxiways and runways. The movement was still not as consistent as before, and there was some sideways movement, especially at YSSY. In addition, the takeoff rolls seemed to be very jerky and they took a long time to get up to speed and into the air.

Per Nico’s suggestion earlier in the week, I have changed PSXT to a 45 second display to smooth out some of the data anomalies, but it still seemed jerky at times.

I also tried EGLL on a time shift (looking back about 20 hours) and found that it looked very good as well. The takeoff rolls were jerky and sometimes too slow to get up to speed (it looked too interpolated at times), but the traffic was moving and generally looked very good. So, at least looking back less than a day, the historical traffic was looking good as well.

[somiller 694]

13 hours ago, kiek said:

No, both programs share the exact same code. Learner is just a slimmer version of PSXT (created by conditional compilation)..

Thanks Nico, nice to know!

[G-YMML1 582]

To my biggest regret, there is historical stream (as of today) for ENGM. Such a beautiful scenery by Aerosoft 😞 Live Mode is OK.

Edited April 14, 2023 by G-YMML1

[mmcmah 407]

2 hours ago, G-YMML1 said:

To my biggest regret, there is historical stream (as of today) for ENGM. Such a beautiful scenery by Aerosoft 😞 Live Mode is OK.

You regret that there is a historical stream for it?

[G-YMML1 582]

Sorry!!!! I forgot a word “lack”. Lack of historical stream! 

[vidha 91]

There is still some inconsistencies in live data:

Spotting at LPMA today with these observations:

 

LPMA 2023-04-15 TIME PERIOD UTC 08:00-08:40

EJU17MZ  OE-IVM    LIS->FNC		Landed perfectly, no lag
EDW284   HB-IJW    ZRH->FNC		Too high on approach
LGL71FN  LX-LGQ    LUX->FNC		Pause on downwind, reappear to high on short final
TAP1699  CS-TJR    LIS->FNC		Lag/pauses on approach, too high on short final
TRA7T    PH-HXA    AMS->FNC		Lag/pauses on approach, too low on short final

 

Edited April 15, 2023 by vhaaland

[kiek 1,215]

2 hours ago, vhaaland said:

There is still some inconsistencies in live data:

Spotting at LPMA today with these observations:

 

LPMA 2023-04-15 TIME PERIOD UTC 08:00-08:40

EJU17MZ  OE-IVM    LIS->FNC		Landed perfectly, no lag
EDW284   HB-IJW    ZRH->FNC		Too high on approach
LGL71FN  LX-LGQ    LUX->FNC		Pause on downwind, reappear to high on short final
TAP1699  CS-TJR    LIS->FNC		Lag/pauses on approach, too high on short final
TRA7T    PH-HXA    AMS->FNC		Lag/pauses on approach, too low on short final

 

Hi Vidar,

It would be interesting to know the QNH's. Normally these problems occur al very low QNHs.

Btw,  these problems also occurred in previous versions of PSXT+RT, it has never been 100% perefct at all times.

Hopefully in RT v10 this problem will be gone because v10 will provide real altitudes instead of altitudes at barometric pressure ..

[vidha 91]

3 hours ago, kiek said:

It would be interesting to know the QNH's. Normally these problems occur al very low QNHs.

 

I know it's not perfect everywhere. What I wonder, is why some aircraft land perfectly and the next one doesn't. It seems to me there are some strange issue in the data stream. I've reported it to Balthazar

This day, the QNH was 1023 at LPMA at the time I made the observations. I did notice the QNH turn red at some point, but after a restart the issue was still there.

This airport has been tricky before, but sometimes they land perfectly with PSXT, and when they do it's very immersive with that special approach 🙂

 

Edited April 15, 2023 by vhaaland

[G-YMML1 582]

Still some issues with historic data. 

Yesterdays' LEBL at 7:10GMT, BA320 landed, disappeared after leaving runaway 24R, then, 10 seconds later reappeared on a taxiway. The same situation occurred with other landing planes that followed.

[kiek 1,215]

On 4/15/2023 at 5:37 PM, vhaaland said:

This day, the QNH was 1023 at LPMA at the time I made the observations. I did notice the QNH turn red at some point, but after a restart the issue was still there.

Aha, then that cannot be the cause, strange... Must be irregularities in the data stream then.

Edited April 17, 2023 by kiek