Live Weather Does Not Match

I honestly didn’t know that. I am shocked. If navblue’s weather modelling suggested thick stratus over the midlands on a day when the Surface pressure is pretty much 1027mb everywhere, they might want to use a better weather model.

I will go check the navblue data as you have suggested.

Is it this site?

Can’t find the data you are alluding to.

Please don’t be shocked, because that comment above is not true as has been discussed in multiple other threads now. Yet it keeps popping up.

It’s Meteoblue not NavBlue:

Their numerical forecast model is indeed “real world weather”. It’s initialized with observations and is intended to represent the current state of the actual atmosphere. It’s as real world as you can get when depicting the entire atmosphere.

Flight Simulator, on the other hand, has numerous bugs and limitations in how it digests this model data. Sometimes it gets stuck with old data. Sometimes it’s missing data. Sometimes it simply can’t draw the current conditions due to limitations in the weather engine.

3 Likes

Maybe you misunderstood me then. Of course meteoblue is based on real world observations. But it creates a simulation of the real world weather. And the sim weather is supposed to be the weather of the simulation. I’m sure the intent is that the simulation closely matches the real world and MSFS wants their weather to be as close to the real world as possible. But simulations are often not exact reproductions of the real world, even if they are based on real world observations. There may be differences between the simulation and what you see outside your window.

Plus the other limitations and issues you mentioned.

That’s not “not true”. All weather in flight sims are simulations of the real world. How can they be otherwise? Perhaps you don’t understand what I mean by the word “simulation”?

The developers claim the opposite:

Matching real world weather as closely as possible is indeed the intent. I think folks know that they’re not getting an exact reproduction of the real world and its physics when they play Flight Simulator. That’s not what most of us are talking about when we say stuff doesn’t match reality. We want a reasonable representation of the real world. Saying that this is a simulation rather than real life isn’t helpful here when these Live Weather issues are often correctable bugs on the Flight Simulator side that need to be fixed. Your statement misleads people into thinking that Flight Simulator weather shouldn’t even be expected to represent or approximate real world weather, let alone match it. As if we’re getting merely dynamic weather because the Meteoblue weather model doesn’t model reality, when more often than not, it’s a Flight Simulator issue. I keep pointing this out because I want the devs to live up to the expectations they set, and not have folks settle for mediocrity.

You’re misunderstanding my words then. I’m not implying that the sim weather shouldn’t represent real world weather. I was understanding the OP to be asking why things don’t match between the sim and the real world.

You can say “I think folks know that they’re not getting an exact reproduction of the real world” but I’ve seen many posts that state the opposite.

You know and I know that there will be differences between the sim and the real world. And you know and I know that the sim and MeteoBlue are trying to have a high verisimilitude weather model.

Ok, I apologize if that’s the case. I think OP is “shocked” because he may be misunderstanding too. The solution here is to probably check the Meteoblue site to see what weather it’s producing, check if that’s matching the current satellite observations, and then see if Flight Simulator is matching Meteoblue. Then we’d know if the weather model is bad, or if Flight Simulator is bad.

As this image clearly shows, the 00:00z run of the NEMS model for the UK indicates that Meteoblue/NEMS where not expecting any significant cloud across the UK for the period of the flight I have described. As indeed did any of the commonly respected weather models (GFS, HIRLAM, UKM, GEFS, ECM etc.) With the current high pressure over the UK, it’s a no brainer for weather modelling.

My question is not “Why is the weather being modelled incorrectly?” (because it wasn’t) it is "Why was flight simulator presenting me with pretty much total cloud cover over Northern England when it wasn’t predicted to be, or quite clearly wasn’t, like that.

I bought Rex Weather force a long time ago but find it a bit of a pain to run it every single flight and remember to put fs2020 weather to clear skies each time, to make room for REX to drop their weather into. However, in the case of today’s weather, REX, which uses METARS to produce its weather data would have almost certainly have shown the correct weather in the simulator today. Maybe it is worth handing over weather to this third party product but I wish MS/ASOBO could produce equally realistic results.

I disagree. It totally, 100%, is meant to be real live weather.

You’re confusing what it does with what it’s supposed to be doing. It is supposed to accurately reflect real world live weather. It does that by using a model.

That may be so, but the goal isn’t to reproduce a flawed model. The goal is to reproduce real world live weather. That’s like throwing a ball at a basketball hoop and saying “my goal is to miss the basket cause that’s what it often happens”.

It totally is not true, because you didn’t say that flight sim weather is a model of real weather. You said flight sim weather is supposed to be a reflection of a weather model, which is not true.

1 Like

Aren’t those essentially the same question? Either way, we’ve been asking the same question for a while now. The best anyone can figure is that there seems to be some sort of breakdown between the model generated by MeteoBlue and the weather you receive in your simulator. Like a server somewhere is caching data and passing old results.

Asobo say they will deliver an improvement to the weather in SU4, and should fix it in SU6. Fingers crossed.

2 Likes

You’re confused as to the difference between “simulation” and real world. If I go to my house’s location in the sim, you are saying that it should have the exact same weather, minute by minute as the real world? Every house in the sim in the entire sim world must correspond exactly to the real world? Because that is what “It totally, 100%, is meant to be real live weather” means.

Or do you mean, as I think, it’s meant to be a simulation of the live weather that is “good enough”.

I think we mean the same things, but you are misinterpreting my words.

Are you thinking that I’m saying that MSFS has a goal of creating weather that has no correspondence with reality? What would be the point of that?

No. Meteoblue’s modelling is correct i.e. no cloud. FS2020 served up loads of cloud. Incorrect.

Yes. That is the intention. As i said, you’re confusing what it does with what it’s supposed to be doing.

That is the practical limit of the sim. It is not the goal of the sim. The goal of the sim is real world live weather, and that is what it’s supposed to show.

Your words:

This argument isn’t going anywhere. What the sim is supposed to show and what it can physically show are two very different things, and i’ll leave it at that.

That’s never going to happen. At least not in my lifetime.

There was a bug previously where the simulator was drawing weather for a model forecast hour that was valid 12 to 18 hours earlier. In other words, the simulator was drawing the weather from 12 to 18 hours ago. I think it turned out to be a server caching problem. Perhaps that’s reared its head again.

Yeah maybe, but the UK has been under this high pressure for many days now and it’s been pretty much wall to wall blue skies during the middle of the day, for about a week. Something went wrong but i don’t feel I can trust Asobo’s “live weather” without checking it against rain radar, sat images etc.

Interestingly REX weatherforce weather isn’t a particularly good solution either as it presents weather based on the nearest and most recent METARS. This gives accuracy for an area around a given airport but lacks the ability to resolve larger weather systems like approaching fronts etc etc. An amalgamation of the two concepts might be the way forward IMHO.

Are the live weather servers delayed again?

The lower photo from MSFS appears to be the weather for about 1200Z for Thursday April 22nd, so yes, it is delayed rather severely.

2 Likes

I’ve tried these types of squall lines, fast moving storms etc before and my result has always been the same as yours.