Board index FlightGear Development Weather

A unified weather system?

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

Re: A unified weather system?

Postby gluon » Sun Jun 17, 2012 7:37 am

I think what polly is looking for is the dialog for placing clouds, cloud layers etc. which is kind of hidden under Debug->Local Weather (Test) ?

Improvements:
  • What I find a small bit annoying with the current advanced weather dialog is the "Local weather is already running" message.
    Intuitively, when I change something with the weather dialog, I would expect it to use my new settings and/or restart (that is clear/end) automatically when I press OK. The three buttons are a bit confusing and since I never looked into the documentation (who really does?) I'm never sure what the difference between OK and clear/end is.
    So I propose only two buttons: [OK] and [Cancel] where "OK" automatically uses my settings.
  • I don't know if there is support for icing in FG.
  • How about wind shear or gusts/blasts? So far I know the weather is pretty constant and this would make it a bit more lively with sudden changes.
  • GUI: A lot of commercial weather engines like REX, have a screen describing the weather situation. That's practical for flight planning. They even have a radar included.

As far as I am concerned I would second Gijs and Bjoern. Plus having about 15-20 scenarios, maybe with increasing weather intensity (from fair weather to severe thunderstorms).
Stuarts suggestion sounds good, too. But having too many screens may confuse the users if the screen layout is not designed cautiously.

For the more advanced settings, the question is more: What could be done? As i see it this isn't clear due to the limited understanding of advanced weather and then people expect a Basic Weather++, just like you would have it in commercial sims.
gluon
 
Posts: 68
Joined: Tue Aug 18, 2009 7:43 am

Re: A unified weather system?

Postby Thorsten » Sun Jun 17, 2012 10:58 am

Some design layout ideas based on the discussion so far:

* one menu item containing the METAR dialog - allows to specify if METAR is fetched online or to allow to enter a string manually, also contains the aloft winds online fetching option (if that gets implemented) - after 'Okay' on this dialog, METAR gets fetched if applicable, parsed and stored in the tree and the current METAR string is always shown here.

* one menu item containing the weather master configuration. This has three main dropdown boxes:

- initial weather scenario (like 'High Pressure', eventually with a text box containing a more detailed description of what conditions to expect), this also has the option 'METAR' if a valid METAR string is available for the system

- how weather is generated ('effects configured by user', that's the current Basic Weather setup, 'atmosphere physics based', that's the Advanced Weather setup)

- what to do beyond the initial weather setup ('never change scene' as in current Basic Weather, 'change whole scene with METAR' as in current Basic Weather, 'keep scenario' (as in 'repeat tile', i.e. variability in cloud configurations and weather, but no real weather evolution), 'offline weather evolution (i.e. use Advanced Weather's offline engine), 'evolve based on METAR' (as in Advanced Weather's METAR mode).

This menu is structured as a launcher, i.e. 'Okay' clears all current weather simulation and restarts based on set parameters, 'Close' just closes without restarting. An additional button allows to show configuration details. What is shown in the detailed menu depends on the selection of dropbox 2.

- If 'effects configured by user' is selected, it brings up something like the current 'Basic Weather' dialog which allows to specify all cloud layers and winds and (as Polly suggests) allows to review and fine-tune parsed METAR before running the system

- If 'atmosphere physics based' is selected, it brings up the various offsets and the wind model selection of Advanced Weather

* a third menu item contains all runtime-options for weather and environemnt shader control with can be set without restarting the system (for instance the sliders for ground fog, snowline, air pollution, a turbulence scale factor,...)

It'd also be possible to run hybrid modes. For instance, one could use the layer wrapping of Basic Weather, but set up the cloud configuration inside the layer not via the current xml-config but via Advanced Weather's cloud placement routines which generate a variety of patterns like streaks, patches, cloud banks or undulatus. Or to use Advanced Weather's offline weather simulation but feed the results to a Basic Weather type rendering.

To answer Stuart's question, currently Advanced Weather has no support for 2d clouds, but I think it can be made to use them if needed. I don't think one can easily get into a hybrid where 2d and 3d clouds are mixed, but use either 2d or 3d clouds should be possible... the problem is - what's the point? You can't have synchronized thermals and cap clouds when using the 2d layers, you can't have cloud-terrain interactions, you can't have cloud drift in the wind,... so it makes more sense to enable Advanced Weather functionality only of the user has 3d clouds activated.


How about wind shear or gusts/blasts? So far I know the weather is pretty constant and this would make it a bit more lively with sudden changes.


We have it :-) - just use the sliders for gust frequency, strength and directionality and you can generate pretty nasty conditions.

What I find a small bit annoying with the current advanced weather dialog is the "Local weather is already running" message.
Intuitively, when I change something with the weather dialog, I would expect it to use my new settings and/or restart (that is clear/end) automatically when I press OK.


Point taken - I agree that this can be changed and okay should lead to a restart based on the changed parameters.

I don't know if there is support for icing in FG.


I think that needs to be implemented on the aircraft side - the weather system can only tell that it's cold outside. It can't know how the FDM changes due to icing. I think the SenecaII has icing implemented.

GUI: A lot of commercial weather engines like REX, have a screen describing the weather situation. That's practical for flight planning. They even have a radar included.


You mean like the larger-scale situation showing the distribution of high and low pressure areas in a 3000 km radius, or the precipitation radar? We can't easily do this offline, and in online mode it would depend if one could fetch that info for free from somewhere...
Thorsten
 
Posts: 12016
Joined: Mon Nov 02, 2009 8:33 am

Re: A unified weather system?

Postby kyokoyama » Sun Jun 17, 2012 7:36 pm

Speaking of gusts and blasts, could jet blasts from the wake of a flying airplane also be controlled with a unified weather system? (or can turbulence etc. only be based on a static location?)
Look for "B-BIRD" "N127KY" or "AVA0004" -that's me.

Despite having over 1700 posts here, I am not even close to being the most skilled guy here... I'm just words and bad drawing, not experience. :P
kyokoyama
 
Posts: 1981
Joined: Sun May 03, 2009 2:16 am
Location: Earth
Callsign: B-BIRD, N127KY
Version: 2.12.1
OS: Windows Vista

Re: A unified weather system?

Postby Johan G » Sun Jun 17, 2012 8:52 pm

Thorsten wrote in Sun Jun 17, 2012 10:58 am:
I don't know if there is support for icing in FG.


I think that needs to be implemented on the aircraft side - the weather system can only tell that it's cold outside. It can't know how the FDM changes due to icing. I think the SenecaII has icing implemented.

If I recall correctly only one FDM has native support for icing, and that's the old UIUC (FG Wiki). I assume it can be done in nasal as well, but it is, as Thorsten say, at the individual aircraft level.

kyokoyama wrote in Sun Jun 17, 2012 7:36 pm:Speaking of gusts and blasts, could jet blasts from the wake of a flying airplane also be controlled with a unified weather system? (or can turbulence etc. only be based on a static location?)

I think that better is done through multiplayer and nasal. Otherwise the weather code would have to handle all the multiplayer, and AI, aircraft as well as the weather.

While I do not know how it can be done, I do not doubt that it can be done.
Low-level flying — It's all fun and games till someone looses an engine. (Paraphrased from a YouTube video)
Improving the Dassault Mirage F1 (Wiki, Forum, GitLab. Work in slow progress)
Johan G
Moderator
 
Posts: 6277
Joined: Fri Aug 06, 2010 5:33 pm
Location: Sweden
Callsign: SE-JG
IRC name: Johan_G
Version: 2020.3.4
OS: Windows 10, 64 bit

Re: A unified weather system?

Postby Thorsten » Mon Jun 18, 2012 6:50 am

I think that better is done through multiplayer and nasal. Otherwise the weather code would have to handle all the multiplayer, and AI, aircraft as well as the weather.


You can't have any second subsystem setting environment parameters. In the best case, they are overwritten by weather, in the worst case you get two systems fighting for control and blocking each other. The way to go is that MP/AI calls the Advanced Weather function to create an effect volume (so the effect is properly registered with weather) and takes care to update its position regularly by setting the variables in the effect hash.

Weather then does the relative position comparison and executes the effect when applicable just like any other effect. It should well be doable, given that we can have several hundred thermals in the scene.
Thorsten
 
Posts: 12016
Joined: Mon Nov 02, 2009 8:33 am

Re: A unified weather system?

Postby Hooray » Mon Jun 18, 2012 8:21 am

(Note: I didn't yet read any of the postings here)

Is there anything else (besides better framerates, we already know that one...) on your wishlist?


I already mentioned this a while ago, but as you brought up the UI complexity, I am posting this again:

I think, it would probably be more intuitive to replace the current "Start/End" buttons and just add a "terminate/disable" item to the tile selection (combo box). That way, the whole "Start/End" thing could be renamed "Apply" and automatically restart the local weather system, using the selection from the combo box. This is something that I ran into at the beginning of the year, when I was working with your code. While your GUI message is helpful, to many people this may seem unnecessarily complicated and non-straightforward.
Please don't send support requests by PM, instead post your questions on the forum so that all users can contribute and benefit
Thanks & all the best,
Hooray
Help write next month's newsletter !
pui2canvas | MapStructure | Canvas Development | Programming resources
Hooray
 
Posts: 12186
Joined: Tue Mar 25, 2008 8:40 am

Re: A unified weather system?

Postby Hooray » Mon Jun 18, 2012 8:35 am

Thorsten wrote:What level of detail would you like to control? For instance, wind speed and direction are in general different at different locations as well as different in altitude - how would you like to set this? Should the system come with its own wind model, is there a need for the user to set up specific situations, ...?


Note that this is not a feature request, just some brainstorming (you asked for it!):

A while ago I got my hands on "professional" (i.e. FAA certified) flight simulation software, and the one thing that I found really interesting and helpful regarding weather was the ability to set up specific weather situations/scenarios, based on geographic positions (airport, runway, navaid, fix or offsets from these) and time.

Basically, you could set up some weather phenomenon for certain positions (and time), so that flying a real life approach can be made pretty challenging by adding some interesting twists (visibility, turbulence).

With the recent work that Scott & James have done on exposing the navaid, fix and airport data to Nasal, this would now also become feasible using Nasal:

- cat3 conditions at run way threshold
- windshear during missed
- turbulence in hold (i.e. at navaid/fix)
... etc

For flight instruction purposes (but also for missions/adventures or Linux Tag), such scenario support would be pretty cool

From a UI point of view, this could be made to work like the current route manager dialog: a listbox with items added via buttons, this could be some simple markup (ATPOS, ATTIME)
Please don't send support requests by PM, instead post your questions on the forum so that all users can contribute and benefit
Thanks & all the best,
Hooray
Help write next month's newsletter !
pui2canvas | MapStructure | Canvas Development | Programming resources
Hooray
 
Posts: 12186
Joined: Tue Mar 25, 2008 8:40 am

Re: A unified weather system?

Postby Thorsten » Mon Jun 18, 2012 9:34 am

and the one thing that I found really interesting and helpful regarding weather was the ability to set up specific weather situations/scenarios, based on geographic positions (airport, runway, navaid, fix or offsets from these) and time.


That'd be fairly trivial to add to the METAR dialog - simply specify a METAR string for a position, switch off METAR fetching, write a short control loop that checks if we are at the position, let that dump the METAR string into the tree and trigger a parsing, run weather in METAR mode and voila - the parsed METAR gets there right where it should be.
Thorsten
 
Posts: 12016
Joined: Mon Nov 02, 2009 8:33 am

Re: A unified weather system?

Postby Hooray » Tue Jun 19, 2012 5:15 pm

thorsten wrote:The way to go is that MP/AI calls the Advanced Weather function to create an effect volume (so the effect is properly registered with weather) and takes care to update its position regularly by setting the variables in the effect hash.
Weather then does the relative position comparison and executes the effect when applicable just like any other effect. It should well be doable, given that we can have several hundred thermals in the scene.


That's exactly what I was going to suggest, too - Basically, this is only relevant whenever multiple aircraft fly around, i.e. either AI traffic or multiplayer.
Preferably, each aircraft would expose a "wake-turbulence-strength" factor - which could be dynamically adjusted (power setting)-
The LW system could then traverse all active AI/MP aircraft and pick up the "wake-turbulence-strength" property, sample the current position and compute a sensible turbulence dimension/position - after a configurable delay, the turbulence would "expire".

It would probably be possible to keep track of a handful of properties for each aircraft/engine, such as for example:
  • last-wake-turbulence-strength
  • last-wake-turbulence-lat-deg
  • last-wake-turbulence-lon-deg
  • last-wake-turbulence-timestamp
  • last-wake-turbulence-course
  • last-wake-turbulence-tas


A Nasal control loop could then update all properties for each aircraft/engine, so that the LW system could pick up values to model aircraft-induced turbulence.

In the beginning, it would probably be easier to add such properties to all AI aircraft, this could be done with a simple Nasal script - supporting MP aircraft, would probably require using the MP channels (mp_broadcast.nas)...

There's plenty of information available on line about wake turbulence, so that modeling some heuristics would seem doable: http://en.wikipedia.org/wiki/Wake_turbulence

It would be interesting to see if we could come up with an algorithm to animate cloud textures accordingly, so that wake turbulence is actually visualized:
Image
Image

There are tons of technical/research papers on simulating/modeling and visualizing wake turbulence: http://www.personal.psu.edu/lnl/papers/vecpar.pdf
Last edited by Hooray on Tue Jun 19, 2012 6:56 pm, edited 1 time in total.
Please don't send support requests by PM, instead post your questions on the forum so that all users can contribute and benefit
Thanks & all the best,
Hooray
Help write next month's newsletter !
pui2canvas | MapStructure | Canvas Development | Programming resources
Hooray
 
Posts: 12186
Joined: Tue Mar 25, 2008 8:40 am

Re: A unified weather system?

Postby Hooray » Tue Jun 19, 2012 6:43 pm

Regarding the vortex paper that I posted, here's the relevant extract citing the vortex computation formula, they use:

Image

Note that there's been a discussion on simulating this as part of the FDM over at the JSBSim tracker (6 years ago): http://sourceforge.net/tracker/?func=de ... tid=369399
Please don't send support requests by PM, instead post your questions on the forum so that all users can contribute and benefit
Thanks & all the best,
Hooray
Help write next month's newsletter !
pui2canvas | MapStructure | Canvas Development | Programming resources
Hooray
 
Posts: 12186
Joined: Tue Mar 25, 2008 8:40 am

Re: A unified weather system?

Postby Toanphuc » Wed Jul 11, 2012 3:44 pm

I hope we could have some unique weather effect like Tornadoes,Micro Burst and some strong hurricanes,i think it could add some challenges to my flying experiences :D
A good flight sim is not to make people feel fly easier but to make them feel fly is as real as it get.
Other Free Flight Sims viewtopic.php?f=3&t=8847
Toanphuc
 
Posts: 180
Joined: Sun Jun 20, 2010 5:26 am

Re: A unified weather system?

Postby stuart » Sat Jul 21, 2012 8:54 pm

Hi Thorsten,

I'm looking seriously at how we can unify the weather.

One aspect I'd like to make common is selecting the weather scenario.

For Basic Weather, there is a one-to-one mapping from scenario to METAR string.

For Local Weather I think the nearest equivalent is the weather scenario tile. Is that correct?

From looking at the code, as far as I can make out, the weather scenario tile maps to a set of METAR definitions, and currently uses a random number generator to select between them.

Ideally I'd like to have a single set of scenarios used by both Basic and Local weather modes. For Basic Weather that just maps to a METAR string. For Local Weather this either maps to a weather scenario tile, or if a weather scenario tile doesn't exist, falls back to a METAR string.

Given this, we could then have a Weather Scenario dialog that would allow the user to select METAR, or any of the appropriate scenarios for either Basic or Local weather

Does this sound reasonable?
G-MWLX
User avatar
stuart
Moderator
 
Posts: 1589
Joined: Wed Nov 29, 2006 9:56 am
Location: Edinburgh
Callsign: G-MWLX

Re: A unified weather system?

Postby statto » Sat Jul 21, 2012 10:33 pm

I don't care what the final answer is but I just want the ability to easily turn "off" weather.

I understand that's a rather strange statement as weather can never truly be "off", but I would like a generic, clear skies, variable visibility*, no winds at any level, throughout the entire earth, simply for ease of scenery testing and possibly for plane testing purposes.

Sorry that's not necessarily helpful to the discussion, but I feel it is important.
Custom Scenery available from http://www.stattosoftware.com/flightgear
statto
 
Posts: 2106
Joined: Fri Jan 25, 2008 9:57 pm

Re: A unified weather system?

Postby someguy » Sun Jul 22, 2012 1:37 am

@statto: that's easy enough, just code what you want into the preferences.xml file, and don't select any other weather.
User avatar
someguy
 
Posts: 1650
Joined: Tue Nov 25, 2008 5:54 am
Location: USA
Version: 2019.1.1
OS: Mac OS X 10.11.6

Re: A unified weather system?

Postby stuart » Fri Aug 17, 2012 4:48 pm

FYI, there is now a unified weather UI in git, so setting up weather should now be a much more consistent experience.

Bugs can be raised through the issue tracker as normal.

-Stuart
G-MWLX
User avatar
stuart
Moderator
 
Posts: 1589
Joined: Wed Nov 29, 2006 9:56 am
Location: Edinburgh
Callsign: G-MWLX

PreviousNext

Return to Weather

Who is online

Users browsing this forum: No registered users and 2 guests