Board index FlightGear Development Weather

Discontinuity in the weather

Everything related to weather simulation, visuals should be discussed in the shader subforum.

Re: Discontinuity in the weather

Postby Thorsten » Fri Mar 09, 2018 2:07 pm

Just my cents on this - I run AW all the time and the altitude periodically goes crazy, even upsetting the autopilot.


I guess someone needs to re-visit these autopilots - the behavior you describe seems way too fragile. Any turbulent updraft will give you way more of a perturbation than a pressure change.
Thorsten
 
Posts: 12490
Joined: Mon Nov 02, 2009 9:33 am

Re: Discontinuity in the weather

Postby Parnikkapore » Fri Mar 09, 2018 2:53 pm

Let me correct - the altimeter goes crazy, and it would make sense for the autopilot to freak out trying to gain the same numbers back. But as I said, it's not a big deal.
There are free alternatives to (almost) every program you encounter. You just have to find them.
Parnikkapore
 
Posts: 929
Joined: Thu Oct 29, 2015 11:16 am
Callsign: HS-FGS
Version: next
OS: Kubuntu

Re: Discontinuity in the weather

Postby Thorsten » Fri Mar 09, 2018 5:09 pm

I also don't understand why the altimeter would go 'crazy' - if you get a change in pressure by a mbar or two, you see a corresponding change in barometric altitude - that's not crazy, that's how a barometric altimeter works.
Thorsten
 
Posts: 12490
Joined: Mon Nov 02, 2009 9:33 am

Re: Discontinuity in the weather

Postby V12 » Fri Mar 09, 2018 6:38 pm

Crazy is steep pressure change. Should be interpolated between From - To METAR stations based on distance between that station.
Fly high, fly fast - fly Concorde !
V12
 
Posts: 2757
Joined: Thu Jan 12, 2017 5:27 pm
Location: LZIB
Callsign: BAWV12

Re: Discontinuity in the weather

Postby Thorsten » Fri Mar 09, 2018 7:20 pm

We've just established that AW does just that, so I'm not sure I grasp where the conversation is going here :?:
Thorsten
 
Posts: 12490
Joined: Mon Nov 02, 2009 9:33 am

Re: Discontinuity in the weather

Postby V12 » Fri Mar 09, 2018 10:01 pm

Why not BW too ? Is it big problem ? Because if BW steep baro pressure change will disengage AP of some aircrafts, it should be considered as bug.
Fly high, fly fast - fly Concorde !
V12
 
Posts: 2757
Joined: Thu Jan 12, 2017 5:27 pm
Location: LZIB
Callsign: BAWV12

Re: Discontinuity in the weather

Postby Thorsten » Fri Mar 09, 2018 10:09 pm

Why not BW too ? Is it big problem ?


I suppose if you make BW generate weather in localized chunks and let it interpolate, you're basically... getting AW, because that's what the difference is all about.

Because if BW steep baro pressure change will disengage AP of some aircrafts, it should be considered as bug.


Yes - but I think a bug of the AP, it should be able to handle such things easily. There's a flaw in reasoning 'we must carefully constrain the environment because my plane's AP can't handle it otherwise'.
Thorsten
 
Posts: 12490
Joined: Mon Nov 02, 2009 9:33 am

Re: Discontinuity in the weather

Postby curt » Fri Mar 09, 2018 11:35 pm

It would be interesting for the original poster to log and plot the pressure data. BW does smoothly interpolate some values (like wind) when METAR changes. In this case the interpolation is over time from previous value to new value, it's not interpolated over distance.
Aerospace Engineering and Mechanics
University of Minnesota
curt
Administrator
 
Posts: 1168
Joined: Thu Jan 01, 1970 1:00 am
Location: Minneapolis, MN

Re: Discontinuity in the weather

Postby V12 » Sat Mar 10, 2018 6:57 am

I will try log some wind and pressure data. But I don't know how setup that log.

Check my fgtracker logs (callsign BAWV12), all small altitude changes are results of suddenly jump of baro pressure.
Fly high, fly fast - fly Concorde !
V12
 
Posts: 2757
Joined: Thu Jan 12, 2017 5:27 pm
Location: LZIB
Callsign: BAWV12

Re: Discontinuity in the weather

Postby V12 » Fri Mar 16, 2018 7:49 pm

With Concorde I performed subsonic test flight LSGS-LZIB (https://fgtracker.ml/modules/fgtracker/ ... ID=6997819) with advanced weather and recorded some data (https://ulozto.net/!avKjq81v71Dc/lsgs-lzib-ods) with 1 second interval. I analyzed that data and found 2 discontinuity :
1. pressure inches Hg at sea level, when METAR string doesn't contains pressure data, AW "interpolate" last known pressure to 29.92 inHg in 2 seconds (yellow rows 3205-3207). All other small pressure changes has been perfectly interpolated
2. wind speed and wind direction are not interpolated, I'm using Live data, tile selection METAR and wind model Aloft Waypoint. That model should interpolate wind speed and direction
aloft waypoints is like the above, but lets you specify many positions (called here "waypoints") and AW will interpolate them in the 3D space. In METAR mode it works automatically and, instead of using the user data, it guesses the vertical distribution of winds on its own based on the reported ground wind, especially using as waypoints the live METAR stations.

Based on the discontinuity in 2) there are steep changes in the KIAS and ofcourse TAS and GS (not displayed in the graphs). Autothrottle can handle slow wind speed and direction changes as visible in the graph, but not in the case of discontinual change. And when that change is the source of the overspeed, AP should be disengaged in very realistic simulated aircrafts like IDG A3xx. Duration of the one overspeed 10 kts above selected speed was 7 seconds (green rows 1222-1229)

I will try collect another data from longer flight over many METAR stations across Europe.
Fly high, fly fast - fly Concorde !
V12
 
Posts: 2757
Joined: Thu Jan 12, 2017 5:27 pm
Location: LZIB
Callsign: BAWV12

Re: Discontinuity in the weather

Postby Thorsten » Fri Mar 16, 2018 8:17 pm

That model should interpolate wind speed and direction


-> see $FGData/Nasal/local_weather/local_weather.nas

especially the functions

wind_altitude_interpolation
wind_interpolation
set_wind_ipoint_metar

- this has interpolated fine in every test I've performed so far.
Thorsten
 
Posts: 12490
Joined: Mon Nov 02, 2009 9:33 am

Re: Discontinuity in the weather

Postby V12 » Fri Mar 16, 2018 10:03 pm

Nice, but somewhere is something wrong. My "blackbox" protocol :

Code: Select all
<?xml version="1.0"?>
<PropertyList>
 <generic>
  <output>
   <line_separator>newline</line_separator>
   <var_separator>:</var_separator>
   <chunk>
    <name>Baro alt</name>
    <type>float</type>
    <format>%f</format>
    <node>/instrumentation/altimeter[0]/indicated-altitude-ft</node>
   </chunk>
   <chunk>
    <name>Pressure inhg</name>
    <type>float</type>
    <format>%f</format>
    <node>/environment/pressure-inhg</node>
   </chunk>
   <chunk>
    <name>Pressure sea level ingh</name>
    <type>float</type>
    <format>%f</format>
    <node>/environment/pressure-sea-level-inhg</node>
   </chunk>
   <chunk>
    <name>KIAS</name>
    <type>float</type>
    <format>%f</format>
    <node>/instrumentation/airspeed-indicator[0]/indicated-speed-kt</node>
   </chunk>
   <chunk>
    <name>TAS </name>
    <type>float</type>
    <format>%f</format>
    <node>instrumentation/tas-indicator/indicated-speed-fps</node>
    <factor>0.592484</factor>
   </chunk>
   <chunk>
    <name>GS </name>
    <type>float</type>
    <format>%f</format>
    <node>/velocities/groundspeed-kt</node>
   </chunk>
   <chunk>
    <name>wind direction</name>
    <type>float</type>
    <format>%f</format>
    <node>/environment/wind-from-heading-deg</node>
   </chunk>
   <chunk>
    <name>wind speed</name>
    <type>float</type>
    <format>%f</format>
    <node>/environment/wind-speed-kt</node>
   </chunk>
  </output>
 </generic>
</PropertyList>


Especially, wind direction and wind speed are standard FG properties and should be correct. TAS is Concorde specific instrument, I'm not sure about KIAS and barometric altitude.
Fly high, fly fast - fly Concorde !
V12
 
Posts: 2757
Joined: Thu Jan 12, 2017 5:27 pm
Location: LZIB
Callsign: BAWV12

Re: Discontinuity in the weather

Postby V12 » Sat Mar 17, 2018 7:35 pm

Another example of the non interpolated wind speed and wind direction :
Image

Source file : https://ulozto.net/!mvx97lSsip5j/lsgg-egll-ods

This is my AW setup :

Image
Fly high, fly fast - fly Concorde !
V12
 
Posts: 2757
Joined: Thu Jan 12, 2017 5:27 pm
Location: LZIB
Callsign: BAWV12

Re: Discontinuity in the weather

Postby wkitty42 » Sun Mar 18, 2018 4:30 pm

does enabling "Terrain Presampling" make any difference?
"You get more air close to the ground," said Angalo. "I read that in a book. You get lots of air low down, and not much when you go up."
"Why not?" said Gurder.
"Dunno. It's frightened of heights, I guess."
User avatar
wkitty42
 
Posts: 9148
Joined: Fri Feb 20, 2015 4:46 pm
Location: central NC, USA
Callsign: wk42
Version: git next
OS: Kubuntu 20.04

Re: Discontinuity in the weather

Postby V12 » Sun Mar 18, 2018 8:00 pm

Not. But with terrain presample and terrain effects kill my FG session in 30 minutes on memory consumption 7GB physical and 10 GB virtual.
I recorded another data and found, that all discontinuity are after METAR changes. After 3-5 seconds interpolation again works well.
Fly high, fly fast - fly Concorde !
V12
 
Posts: 2757
Joined: Thu Jan 12, 2017 5:27 pm
Location: LZIB
Callsign: BAWV12

PreviousNext

Return to Weather

Who is online

Users browsing this forum: No registered users and 2 guests