Too small distances between thermals suitable for gliding

:wave: Thank you using the Bug section, using templates provided will greatly help the team reproducing the issue and ease the process of fixing it.

Are you using Developer Mode or made changes in it?

No for testing, yes during reporting for position

Have you disabled/removed all your mods and addons?


Brief description of the issue:

Significant problem that discourages a lot of people from using MSFS for gliding is too low spacing/distances between thermals and that those thermals typically cover too huge area - this makes realistic cross country soaring impossible and discourages glider pilots or people wanting to learn soaring from using MSFS for this purpose - for example classic MacCready “speed to fly” cross country strategy between clouds cannot be executed. People try to circumvent current shortcoming by setting wind speed to 0 thus completly disabling “ground thermals” and leaving only the “cloud” component - that says a lot about current incorrect state of thermals.
I am considering here the most common case of typical flatlands soaring with single Cu clouds layer - very commonly occuring case in Europe. Please also note I am concentrating only on “distances” aspect of thermals in this report, other incorrect aspects of current thermal model should/will be reported in other bug reports.
Expected changes in relation to distances between thermals:

  1. Distances between usable for gliders thermals should be the same magnitude as distances between Cu clouds visualized in given flight , as you should find stronger lift under every Cu cloud (unless it is an old, dissipating one), currently in MSFS strong, wide thermals can be spotted “densely” - almost everywhere, unrelated to Cu clouds.
  2. Near the ground airflow should converge more horizontally towards the thermal trigger, stealing warm air from larger area, thus reducing chances of another nearby thermal, currently in MSFS according to 3d thermal visualization it looks like air is being “produced out of nowhere” by the ground and rises parallely over huge areas without any convergence near the ground - not forming narrower column of air.
    Current MSFS model doesn’t seem to account that thermal starts to rise at a specific relatively small thermal trigger and this fastest rising airflow leaves lower pressure area behind, that sucks air near the ground from all sides as depicted on this symbolic picture (taken from G. Dale Presentation Part 1 - YouTube ):

    Please note the arrows of airflow near the ground pointing mostly horizontally towards the center of the starting thermal - this way thermal is “stealing” warming air from over large surrounding ground area - so typically there is single strong thermal (it may sometimes be joined by some weaker ones) and also because of this pressure gradient towards thermal core - thermals are not horizontally wide - typical gliders do wide circling of about 500m in diameter, and tight circling is around 200m (or even less).
    But I don’t want to digress too much into another topic of incorrect thermal shape and lift distribution - this should be another bug report.
  3. In spaces between Cu clouds you should mostly meet calm air, some very gently descending air, some turbulence or if you’re lucky a thermal that is on it’s way towards condensation height where it should form Cu cloud in few minutes (the time it takes with given lift speed for this airmass to reach condensation height), if you are very unlucky you can meet downdraft, but it shouldn’t be that often and rather way less strong and wider then thermals (I am not talking here about vortexes caused near rising air bubble, that can be narrow and strong).
  4. With current weather settings it is not possible to create a preset with very spaced out Cu clouds - in real life, especially in the evening, Cu clouds tend to be well developed (over the best heat sources), but quite scattered - with distances between Cu clouds as high as 20km-30km - currently it is impossible to reproduce such scenario - in MSFS even with SCATTER parameter set to 100% we can do maximum spacing of less than 10km.
  5. There shouldn’t be 2 completly unrelated phenomena (“ground thermals” & “cloud thermals”) as currently seem it is modelled in MSFS - lift under the cloud should simply be the same ground thermal bubble that rised up to the condensation height and formed the cloud, clouds can generate some additional lift very near cloud bottom due to latent heat, but the largest lift component should be this air rising from ground - possibly drifted away from ground heat source with the wind - just like is depicted symbolically on picture below (from the same presentation):

Provide Screenshot(s)/video(s) of the issue encountered:

EDLO, time and weather as depicted on later picture:

I crossed out places that according to my RL glider experience shouldn’t have significant thermal updraft as visualized by 3d thermal visualization:

Areas crossed out with pink color have totally unreasonable lift - top of clouds should denote maximum lift height, if there are such strong thermals they should form clouds starting from surrounding Cu cloud bottoms height (condensation height).
Areas crossed out with red are lift that is unreasonable in spaces between clouds - there shouldn’t be strong, usable for gliders lift between clouds (such lift would form cloud), please also note that incorrectly there is no 3d thermal lift visible under most of the clouds - or it misses their cloud bottoms (which is also incorrect).
Area crossed out with yellow is a huge updraft of air rising over urban area - while indeed urban areas are commonly generating lift this lift concentrates in relatively narrow column over thermal trigger - like a largest/hottest factory roof, most of other warm air is being sucked toward such thermal trigger thus contributing to this relatively narrow column of air, not “whole city” rising as visualized.
What I, as RL glider pilot would expect in this case would be:

Please note that I only symbolically mark with green lines places where I would expect cone-shaped thermal usable for gliders, in spaces between clouds I wouldn’t expect any lift strong and durable enough to be able to rise there circling with a glider.
Another picture (same location) but from the side:

I crossed out huge strong lift on the left side of image that is “going nowhere”, that is doesn’t form any cloud at the condensation height.
In yellow I marked a smaller thermal that although near the cloud seems to be missing it’s bottom completly, please also note that in places marked with question marks 3d thermal visualization doesn’t show any significant lift.
Red crosses on the right side mark some incorrect lift that is too near to be the cause of those clouds.

On picture above is a symbolical depiction what I would expect to see on the visualization in relation to the 3 nearest clouds (please note I only mark thermal locations I do not try to depict those thermal shape)

Provide coordinates (DevMode > Options > Display position)


Provide time & date of the observed Weather issue

EDLO - Oerlinghausen, 10 JULY 2022

Detailed steps to reproduce the issue encountered:

Use time & position as pictured above, turn on 3d thermals visualization

PC specs and/or peripheral set up if relevant:

does not apply

Build Version # when you first started experiencing this issue:

Can be visualized since Su11beta, experienced even earlier

:loudspeaker: For anyone who wants to contribute on this issue, Click on the button below to use this template:

Do you have the same issue if you follow the OP’s steps to reproduce it?

Provide extra information to complete the original description of the issue:

If relevant, provide additional screenshots/video:

i recall reading a book a long time ago. It was like Call of the Wild meets Into the Wild or something. I don’t recall the title but one of the major themes in the book was actually gliding. It was my only experience with gliding since i recall it being fairly descriptive in a way that a reader could assume the author piloted gliders. I think it was paragliding or hang gliding actually but the way he described thermals felt authentic (at least in a fictional adventure title maybe based on a true story).

Even just from that, it definitely felt there was something off about gliding. Thanks for explaining what it was. Very informative. Voted!

The relationship between thermal height and strength influences how far the are apart. which could be looked at also.

Another query while on the topic , each region has the same building blocks for thermals , although Australia , Namibia and the desert regions of the USA do NOT have the same strength which is considered as to strong in europe or the uk .

Please consider this when implementing thermal models across the globe .

Thanks for the product so far though.


The main problem is that they decided to use the easy arcade thermal version instead of the more realistic one, which already exists according to a statement from the MSFS twitch stream.

So if they don’t want 100 bug reports about how unrealistic the thermals are they should consider to update to the existing more realistic version. Then it would actually make sense to test everything.


Agree, i bet when they release the more advanced/realistic air/CFD model we will see complains again it may be overdone or something even if it’s more realistic because many is not used to the new model. Good they will add options for turbulence though.


‘‘a lot of people said there are too many thermals, it’s too easy and they are unrealistic… that’s true and that’s by design’’

the realistic model lost against the currently implemented arcade version, because of…
you guessed it: complaints it’s too hard, they don’t understand it.

The good thing is they are at least considering to implement the realistic version in the next update.
But how realistic that version is, is yet to be seen (if they really implement it at all…).


Do you have the same issue if you follow the OP’s steps to reproduce it?

Different user reports huge lift in live weather over urban area allowing him to glide in a helicopter

Provide extra information to complete the original description of the issue:

It is an example of too huge (instead of converging - pt 2 of this bug report) thermal over urban area - example of “whole city rising” case as seen in 3d thermal visualization

If relevant, provide additional screenshots/video:

Helicopters Are Gliders Too (Massive Updraft)

So sad, really. We as users were sceptical if Asobo would make a realistic simulator before release 2020. Now when they try to make it as realistic as it gets. They get complains it’s too hard even if the air behaves realistic.

Even arcade games we want to be really hard and not dumbed down. It’s many games out there that is intended to be hard. But this simulator that tries to recreate real world physics some complains the real world physics is too hard. The real pilots can complain but the physics will never change. Pilots needs to adapt and learn how the physics works. Now we learn this dumbed down version instead of the more realistic version. I bet it will be even more complains when they release the more realistic version. Why not implement the realistic version instantly instead of changing systems all the time?

Have i bought a simulator or not?

Same things with that METAR implementation. They changed weather simulation into a fixed weather that instantly changing depending on those METAR. Can’t create realistic behaving weather with that mindset to have weather fixed to those. We try to use a cloud to use as a thermal guidance but because the weather needs to be fixed to a METAR the cloud suddenly dissapear. Weather doesn’t work like that. Asobo had the correct mindset before release of this sim with the weather simulation. We as users has changed their mind set. We as users begs to have less of a simulated weather in the flight simulator because we are used to simple weather simulations since older flight simulators. That is really bad in my opinion. This sim atmosphere can’t behave realistic because it needs to behave same as older simulators.

I beg you Asobo, go back to pre release 2020 mind set :slight_smile: If you get complains it’s too hard i bet you are making good progress and don’t listen to those complains. Just telling it’s too hard is not a constructive criticism anyway.


The downdraft is too strong.
Even on calm days there are lots of heavy downdrafts of -5 m/s or more, making it very difficult to gain altitude by soaring.
Even if you struggle to gain 500ft, you frequently lose what you gained in an instant, so you lose motivation.


This is a really good description of the problems with the current implementation of thermals in the sim. The sim has options for realism in lots of its features. Maybe a simple ‘realism’ option for the thermal model would satisfy both the arcade people and the simmers.

It would be great if they could make a few simple changes:

  1. Make the probability of a cumulus cloud being rendered high over the areas of lift. Low probability over areas without lift. Not making a thermal guaranteed below a cloud would be realistic, possibly easier to implement, and in my mind more fun.

  2. Use a reasonable relationship between ridge height, wind speed and the height of ridge lift.

  3. Use the ambient temperature and an estimate of lapse rate to calculate the thermal height and strength.

I reckon they have done well making the terrain influence the placement of thermals. All they need to do is draw the clouds above them and make the thermal strength, size, height and distance in between a bit more realistic and it would be a huge improvement. Too please everyone, they could just add an option to turn off thermal realism or just leave it as it currently stands - enormous lift everywhere.

1 Like

How, in this case, will the wind affect the clouds? Will they move with the wind along with the terrain?

1 Like

If we want to have it realistic they need the atmosphere to behave and be simulated realistic. Thats a simple fact. Then how they will make that happen and at the same time have it accurate to the real world to please those needs to be able to use real world planning tools, i don’t really know. The issue to make a realistic atmosphere in flight simulators has always been the need to have it 100% accurate to real life weather that is happening at this moment. That thing limits devs to make it as realistic as possible i bet.

The thermals forms clouds above those thermals, the thermals is more laminar airflow vertically and the clouds are the thermal that has been turbulent flow that has gotten unpredictable. Thats why it’s better to be under those clouds. not inside.

They need to simulated thermals like this to feel realistic:

Then the question is, how to make a simulation and at the same time make it static to those METARS that may completely change the weather at any time without need to behave like weather as long as it matches those to be able to plan realistic.

1 Like

There is no wind in these experiments. What would it all look like if there was wind?

But where the air moves up (thermals) there is no horizontal air movement because it moves up. When the air cools off it starts to move horizontally again and then moves towards colder places where the pressure is higher. Air (in between those high and lower pressure zones) where it’s really high velocity in horizontal winds there is no thermals formed. Seb explained they will implement full CFD simulations soon with air that moves up needs to come down somewhere. It’s in the R&D stage. I think that will make the air feel better. But those clouds? What are the plans for those? I would like them to be formed because of the atmospheric condition. Not fixed in place. Some says just place those clouds over the thermals. But that is not how it works as you said. Those clouds/air flows dynamically to replace the hot air that rises. This happening both locally and globally. That experiement were really local. But it’s same behaviour in the atmosphere.

The simplest way to achieve this would be to just have the thermal drift at the same horizontal velocity as the wind - and then draw the cloud above where the thermal ends up at the calculated cloud base.

As you know this is not entirely realistic, because thermals have upward momentum and thus tend to drift with the wind a bit slower than the prevailing wind - the reason it’s much easier to fall out of the downwind side of the thermal and why the occasional extension of a circle upwind is a good idea.

But, for the sake of making it possible, rather than perfect - just trigger thermals from the terrain, have them drift with the wind and then draw the clouds at the top of thermals. This will mean the clouds will drift with the wind.

Yes, that may be a start. But that will not change how much distance it is between those thermals right. It’s a really hard thing to simulate propperly i bet. It needs to simulate colder air too. As it is now i think it only simulates the hot air that moves up. Correct me if i’m wrong about that.

Your right @Perrry, the distance between them is a big problem now. I was just replying to Arni’s question about clouds drifting with the wind. Problem with the distance between thermals is that in mountains they are over the peaks, but in flat country they are randomly spread around. But the rule of thumb is that they are around 3 times the distance apart compared to their height.

As for the colder air bit: I am finding sink as well as lift in the sim in live weather.

1 Like

yeah, maybe you are correct the cold air is simulated. I’m not 100% sure about that. To me it feels more like the uneven terrain causes those downdrafts more than cold air. Well we all agree the distance between the thermals are not correct and to me those thermals should cover less area. It should have less cylindric from. Or is it even simulated to have cylindric from?

To me it feels like those thermals should be able to form the volumetric clouds we now have dynamic. But maybe volumetric clouds can’t know where a thermal is formed. I bet it’s hard to connect those clouds to know where a thermal is formed.

Hope they find a thing that makes the glider experience more realistic. I heard they already had a more realistic system in place but didn’t implement it yet because they needed to have a simplified version to make it easier for those that are new to gliders.

In my opinion it’s always better to learn how it really is to begin with. I think we will see more complains when they suddenly add that other more realistic version of code when they learned the easy and less realistic system.

Maybe they add it as an optional feature.