Thermals, Up & Downdrafts - More Realism Updates

Official Microsoft Flight Simulator Wishlist
, , , ,
204

Any ‘weather related’ development is promising but for ‘thermals’ it’s crucial to distinguish between:

  1. air going up (or down) as it moves over a slope.

  2. air moving up and down totally unrelated to the clouds

  3. air moving mostly up under likely-looking Cu or CuNim clouds (with the latter sometimes having air moving violently down)

  4. Hard-coded design to put vertically-moving air in certain locations.

MSFS does a passable job at (1) but it really is just the first five lines of code, i.e. lift is simulated if the plane is above a wind-facing slope but the formula applied (lift rising to 20,000 feet from a 200-foot slope) isn’t ideal. There’s a trick we use to work around that by setting weather ‘AGL’ and killing the wind above something like 1500 feet AGL. FWIW it is trivially simple to radically improve the simulation by sampling maybe 5 points below the plane instead of just one and having a better ‘lift AGL formula’, but again the rookie error is to think a pivot from a single sample to a full-scale CFD is the answer (it never is).

For ‘thermals’ (i.e. 2,3,4) the rookie error in flightsims is to lean heavily on (2) and hope sim pilots don’t know the difference, and that’s mostly what we currently have in MSFS (we call it ‘ground lift’ as the lift is mapped to the ground surface type e.g. urban vs desert rather than the clouds).

To Asobo’s credit they have added lift under some of the clouds i.e. (3), but it’s not common and offset somewhat from the cloud. But it’s very much a step in the right direction.

So for Thermals Up & Downdrafts the MOST important thing is to have the vertical air movement pretty much derived from the cloud map (assuming Cu clouds). Hopefully one day we’ll get this.

IF we have thermals/clouds connected, then it’ll make sense to think about what else could be done for vertical air movement.

The most common incremental effect is ‘streeting’, i.e. the action of thermals forming under Cu’s on a windy day can have the effect of those thermals (and hence the CLOUDS) forming ‘streets’ - these are quite important for soaring as you can follow a ‘street’ and make good progress into-wind that otherwise would be difficult.

I’d suggest the next thing after ‘streeting’ would be ‘wave’, i.e. on a windy day with a ‘stable’ atmosphere, the air can form very widespread cross-wind ‘ripples’ with the air rising up and down. Often those rising areas of wind remain fixed over the geography, as the wave in the first place is triggered by some high ground upwind. Wave lift often causes stationary ‘lenticular’ clouds to form and for glider pilots soaring the front edge of those clouds is a bit like ridge soaring but without the rocks.

In a nutshell, it’d be nice to get some of these incremental features but linking the lift primarily to the clouds (in addition to the ridge lift) is the most important immersive feature.

9 Likes
205

I haven’t seen evidence the sim can generate cloud streets. The pattern they apply to cloud generation seems to have a more fractal geometry. I’m not sure the model they use is granular enough to detect the formation of streets. So will thermals applied to the fractal cloud map be close enough to mimic realistic behavior?

Same applies to gravity waves - we’re talking a visual application in the sim based on what? Satellite observations? This versus an often-invisible synoptic pattern revealed only by numerical models and occasional, periodic cloud formation (this is where the analysis comes in - not that clouds are present, but why they’re present). I’d be curious to know if they can tie it all together to produce the behavior.

Haven’t seen a single lenticular cloud in the sim, and I fly around the back side of the Sierras a lot.

4 Likes
206

yeah absolutely - MSFS isn’t even close to doing this, nor is it capable of modelling waves / lenticular clouds, because the cloud layer model was originally designed purely as a fractal 3D pattern drifting with the wind. Nice graphics but no physics. Asobo have added some lift near the biggest clouds in the main cloud layer but it’s very basic and tends to get hidden in their ‘original’ more widespread lift mapped to each ground type.

Apologies if my explanation in the previous post was confusing. My intent was to emphasise the most important thing for MSFS in terms of thermals is to tailor vertical air movement to match that cloud layer 3d model (even though this may sound a modest goal to many). The rationale is if Asobo can’t do that then any idea of modelling more advanced lift/cloud effects like streeting or wave are a pipe-dream, at best we’d get the graphics but no physics.

6 Likes
207

No, I understand the physics of the weather (or lack thereof in the sim) - it’s nice to converse about it. How it pertains specifically to soaring is not my wheelhouse, but is interesting and I appreciate the insight. I think the trick remains in the implementation - how to fill the gaps in the ingested data with useful-enough modeling. Weather in the sim is a fairly universal issue and I don’t think there’s a lot of common ground between reality, sim-verse, dev promises, consumer expectations, and the understanding of what’s going on behind the scenes that should tie it all together.

4 Likes
208

Almost a year has passed since I wrote my last post here and gave free rein to my hopes.
Even though many things have improved (I’m writing ‘improved’ now), nothing has changed in terms of weather and gliding conditions. :sob:

Real weather and its simulation are not just a few functions for games. You have to really want it to be good and put a lot of work into it. You can’t just do it in between.

There are several posts in the wish lists that actually have the same background, just with different headings.

How do we manage to collect all these topics under the main heading “Real Weather”?

Surely it is good to discuss the topics sorted into different posts. But in doing so, we lose the entire number of votes that basically stand for the same thing.

I would like to see a main topic that doesn’t need to be discussed at all. But everyone who votes in a sub-area of weather improvement should also cast their vote there, so that the extent of the need can be fully recognised.

There was once a ‘bad guy’ who said: ‘If it can’t be done this way, it can be done another way.’
Perhaps this will help to motivate certain decision-makers to make an improvement possible.

Regards

209

Ms needs to get theor gliding right if they want to even slightly be respected in this kind of event.

4 Likes
210

That is correct.
However, you have to ask whether they even want it.
Condor has always been a gliding simulator and nothing more than a gliding simulator.
The MS flight simulator can be used universally.

However, with the start of ‘24, the proportion of games has increased dramatically, which means that the seriousness of a simluation must be questioned at the same time.
(Every developer who concentrates in a certain direction does a better job.)
If I enter MS204 on YT, I get videos that focus on computer-technical comparisons or a game application like Pilotwings64.

This looks completely different with Condor.
So developer always have to ask themself what they want.
I have often asked myself what the purpose of such a forum is.
Surely you want to benefit the general majority.
But the gamers are always in the majority, unlike the serious sim pilots.

This already indicates what happens when you ask for the majority.
So it is logical to see where the FS is heading.
And the question about the majority of the money, or the initial thought of creating a good flight simulator, is justified.

Does a serious simulation pilot need alpacas running around on a mountain ridge in the Andes? Probably not. Now you could say, it’s nice if they are there. And then I say: yes, I agree, but you can include them if everything else is running correctly. I’d rather spend my computing power on simulation than on animals running across the steppe in Africa. This is a flight simulator!

Nothing against the game part, but please do it in a different App. I don’t want to wait for my real weather simulation until the game world has been optimised in FS.

So is the focus on simulation at the moment? clearly NOT!

If this direction is not corrected, then FS will never play in this league.

Technically, the Condor offers me much more that I expect from a simulator. Even in the interior view with the animated pilot and his hands, the FS can’t keep up.

I can’t even think about it any more, otherwise I’ll be inclined to get the latest version of Condor and change my mind. Then I can save myself a lot of time here and not spend so much time trying to push the FS in the right direction again.

Sounds frustrating?

And it is!

Regards

4 Likes
211

(post deleted by author)