Flight Question Atmospheric wind effects do not feel right

[Thorsten]

I again say that regarding that IMO there should be a global climate model with seasonality change, added to some local variation and some random super-local effects

I'm not sure what you're after.

Compared to the delta-v requirements of launching into a different inclination or from a different latitude, the requirements to compensate for wind drift assuming average winds are negligible. If you're flying by hand, the errors you make will be far larger than wind drift, and any automated system will eat such errors for breakfast.

Similarly, stratosphere winds are not a concern at all for Shuttle entry trajectories. Windspeeds are just too slow compared with the velocities involved. The ranging errors they introduce are negligible compared with other effects and can be compensated easily (and again are much smaller than the ranging errors I get piloting by hand).

In aviation weather, if you're after immersion the experience is not so much in the average, but in singular events far from the average. Jet streams in the stratosphere, weather in the troposphere. But that's not really captured by an average plus some variation. If you're after the experience during approach and landing, focus on getting the altitude dependence of tropospheric winds and the boundary layer physics correct - that's where it can actually be felt.

A global averaged climate model is nice, but you'll have to log and compare trajectories with analysis software to see what it did.

 

[fred18]

I'm not sure what you're after.

Compared to the delta-v requirements of launching into a different inclination or from a different latitude, the requirements to compensate for wind drift assuming average winds are negligible. If you're flying by hand, the errors you make will be far larger than wind drift, and any automated system will eat such errors for breakfast.

Similarly, stratosphere winds are not a concern at all for Shuttle entry trajectories. Windspeeds are just too slow compared with the velocities involved. The ranging errors they introduce are negligible compared with other effects and can be compensated easily (and again are much smaller than the ranging errors I get piloting by hand).

In aviation weather, if you're after immersion the experience is not so much in the average, but in singular events far from the average. Jet streams in the stratosphere, weather in the troposphere. But that's not really captured by an average plus some variation. If you're after the experience during approach and landing, focus on getting the altitude dependence of tropospheric winds and the boundary layer physics correct - that's where it can actually be felt.

A global averaged climate model is nice, but you'll have to log and compare trajectories with analysis software to see what it did.

I don't know if i expressed myself not clearly, but what you're saying is basically the same of what I'm stating.

I'm saying that, since Dr Martin is willing to introduce a wind model inside the sim, a good global /seasonal model is enough, that local definitions may be added and that gusts may be random. That the system should have some options so users may customize it if they want to (see reproducibility for debug, OMP or other needings)

And I'm saying that random events reproducibility IMHO is not mandatory at all, since once the global/seasonal model is fixed you'll already have general reproducibility granted.

basically I'm the guy who's saying that it's enough to keep things "simple" for this :lol:

 

[Face]

that would happen anyway if users choose different wind options (no wind, realistic wind, custom wind). I think it would be enough just to agree among users to set gusts to 0 for realistic, to agree on the custom settings if custom, or simply agree on no wind.

Sure, settings I can synchronize between clients. As I said, random gusts would be the first thing I have to turn off, with no option to turn it on, because it is not reproducible and just... well... random.

Anyway, the possibility to synchronize two physics engines is hardly what the average user will want to have, so it is a weak argument. My vote is for the reproducible system, though.

 

[jedidia]

What about procedural techniques, where the seed of the random generator is saved?

That was pretty much my first thought when I heard "reproducability".

Doesn't exclude a global model either. Just have local variations as a function of date and seed.

 

[fatcat]

From a novice point of view, I have turned off the atmospheric wind effects parameter simply to get the DGIV to run straight down KSC 33 on take off.
Could I change the wind direction to headwind???

 

[boogabooga]

I wish that we could toggle between a surface-relative and an air-relative flight path marker in surface HUD.

 

[Abdullah Radwan]

Why not use realistic weather from airports (Called METAR) and upper wind and cloud data to make realistic weather data?

Some links:

http://metars.com/

http://weather.uwyo.edu/upperair/sounding.html

https://www.weather.gov/jetstream/cloudchart

https://www.simbrief.com/system/dbquery.php

 

[Urwumpe]

Why not use realistic weather from airports (Called METAR) and upper wind and cloud data to make realistic weather data?

Because it would not be deterministic. This means, depending on your timewarp or going backwards in time, you would get different wind speeds.

Also, you can only get present wind and not historic wind data.

EDIT: And worse, because of the update intervals of the weather services, you can expect 3 hours of the same wind, even if it is completely different in reality because of the local wind field. Do you know a bit about sailing, especially sailing on lakes?

 

[Abdullah Radwan]

Because it would not be deterministic. This means, depending on your timewarp or going backward in time, you would get different wind speeds.

Also, you can only get present wind and not historic wind data.

EDIT: And worse, because of the update intervals of the weather services, you can expect 3 hours of the same wind, even if it is completely different in reality because of the local wind field. Do you know a bit about sailing, especially sailing on lakes?

Good point about time wrap, I didn't really think about it.

These links are from a quick Google search, so you can get a better result. As a flight simmer, one of the available weather engines allows historical dates up to 6 years. While there is a lot of limitations (Like you need to connect to the internet), it remains a lot better than the current implementation. Maybe experts had a better approach.

I don't know anything about sailing, but these services are the same which real aviation and flight sims relay on it. If we want a reasonable accuracy, Orbiter should do it the same way the flight simulator does.

 

[jedidia]

Do you know a bit about sailing, especially sailing on lakes?

Yeah, that usually looks like "there's no wind". "There's still no wind!". "I think there will never be wind!" :lol:

 

[Urwumpe]

Yeah, that usually looks like "there's no wind". "There's still no wind!". "I think there will never be wind!" :lol:

Or
"Why is the wind now coming from starboard?"
"Why did the boom just wack me at the head?"
"Where did this gust just come from?"
"I hate swimming!"

 

[Ripley]

... If we want a reasonable accuracy, Orbiter should do it the same way the flight simulator does.

Yes, but some Orbiter scenarios / manoeuvres / flightplans / situations / require the user change date dramatically at times, well beyond those 6 years.

What if you want to replay Voyager's flight? What wind was blowing that day?

 

[Urwumpe]

At least we can tell what wind was blowing on November 9th, 1989.

The wind of change. :rofl:

 

[Abdullah Radwan]

Yes, but some Orbiter scenarios / manoeuvres / flightplans / situations / require the user change date dramatically at times, well beyond those 6 years.

What if you want to replay Voyager's flight? What wind was blowing that day?

No need to know what wind was blowing that day. You can use the current wind data for every date. It's not super realistic but 'it works' and it's way better than the current implementation. At least not have some insane wind blowing off your beautiful landing :lol:

In regards to the internet connection, maybe the program comes with the wind file which will update automatically when the user is connected to the internet.

The idea is very limited however it might give some heads-up for new ideas.

 

[GLS]

No need to know what wind was blowing that day. You can use the current wind data for every date. It's not super realistic but 'it works' and it's way better than the current implementation.

... and what website should we visit to get the winds at, lets say, 10247ft above Easter Island on January 26, 2024?

 

[Urwumpe]

At least for the Space Shuttle landings, this could get interesting... should people stop playing Orbiter when a Hurricane strikes the Cape?

 

[Abdullah Radwan]

... and what website should we visit to get the winds at, let's say, 10247ft above Easter Island on January 26, 2024?

As I said, we could use the current wind data for the future and past dates. So I am going to use today as a reference. It's not totally realistic but It's okay.

Visit this website and choose the closest time and altitude, I chose 1800z (= 18:00 UTC) and 10000 ft.. You can see the Upper Wind and Temperature chart below. Locate the closest arrow to Easter Island (Honestly, I didn't know where it was, I went to Google maps :rofl then you can read this PDF to understand the arrow data. If I located the correct arrow, The wind is 330 degrees at 35 kt.

Here is another website provides this data in a raw format, but it supports the US only.

 

[Urwumpe]

Again: Would it be possible to land a spacecraft in any weather?

 

[n122vu]

Agreed. This could definitely get interesting.

"Flight, we have a typhoon warning at the edge of the prime recovery zone..."

...I am totally sure that wind doesn't cause a plane to side-slip.

Clearly you've never taken off in a Cessna 150 in a strong quartering crosswind. The moment the mains leave the ground the plane weather-vanes into the wind and you're side-slipping down the runway.

 

[Marijn]

Clearly you've never taken off in a Cessna 150 in a strong quartering crosswind. The moment the mains leave the ground the plane weather-vanes into the wind and you're side-slipping down the runway.

No. I never did. But that is an edge case. I was discussing the experience in Orbiter in steady flight.

Actually, I doubt there is a weather-vane effect when the mains leave the ground. The weather-vane effect is in operation as long as you actually touch the ground imho. It will vanish as soon as you fly. I assume you're talking about the time it takes to make the course correction to correct for the wind.

Edit: I do have many launches in a glider behind a towplane in all sort of conditions. So I know what it is like to fly centimeters above the ground behind a plane which is not airborne yet.