Weather transition is not smooth after SU7

It’s beginning to dawn on me why this is done - how else are you going to show the “transition” - you can’t put stratus cloud bands (you can of course) over the METAR data because it wouldn’t fit - that means when metar f.e. reports SCT 3/8 (which would also be possible with stratus but MSFS can’t) and you put any “cloud bands” over it, then it is not SCT 3/8 but overcast - (which of course also occurs in nature) - so they try to represent these puffy clouds according to METAR - and in order not to have a complete “mix” they interpolate the METEOBLUE data with a similar shape so that it fits together halfway, but unfortunately it doesn’t or only sporadically ! and I guess that over metar stations the “high level” clouds will be partly or even hidden - that would give this “look” - but that’s just an idea !

Unfortunately, in my eyes it really looks like “FSX” only with 3D clouds!

4 Likes

My suggestion is:

Remove METAR based weather then improve ATIS to report correct things for us in the sim and then generate METAR based of the weather in the sim. Then use that METAR in VATSIM instead of real METAR. Then we need to know what type of forecast model is in use to be able to plan our flights enroute. Then they can improve the flight planning in the sim and we are able to plan our flights in the sim instead.

1 Like

Alright, let’s try another perspective to view this whole thing and another attempt to convince the “pro-METAR injection” community that it’s not the only solution for a multi-platform online network.

Let’s assume there is a world where each individual can only speak the language they were born into and rely on translations to communicate with one another. The universal language is English.
Three people receive tasks that need to be done, but each one of them speak a different language. Spanish, Cantonese and Mandarin. The English speaking server relays messages. The spanish speaking person uses a translator that prefers grammar over vocabulary. While the two Chinese speaking individuals receive a vocabulary-only translation. At the same time they need to communicate with one another. It won’t go well, because while the two asian individuals would ace any vocabulary test, the spanish guy has a better understanding about the task that needs to be done. And all three of them have to communicate in a language foreign to all of them.
So while the task is still given in English, why not let the communication between the three individuals be in spanish? The spanish fellow has a far better understanding what needs to be done and can much better relay the message in his own way to maximize the team-effort. The chinese folks might still only be able to use translation via vocabulary, but the results are still the same, it doesn’t matter if the source is English or Spanish. And the two chinese fellows will still be able to communicate with each other and the spanish person.

If you have all three relying on a vocabulary only method, or be vocabulary only in one sentence and grammar based in another for the spaniard, it will not only be a mess for himself, but may still not be able to communicate with the chinese. So a loss-loss situation really.

I am not going to spoil what you need to fill in for all the metaphors. I am pretty sure you get the picture. I want the more grammar-based translation back. Not a random word spaghetti. The spaniards english was perfectly fine for me. Still a heavy accent, but very well understandable. Just needs more work on the translation engine.

EDIT: Let me go a step further and also suggest the task not be given by voice, but in writing, giving the Spanish guy a chance to perform his task even more efficiently and correctly. This would be the equivalent of a 48-hour lag to the real world weather.
You can’t simulate an “exploding” thunderstorm formation by live METAR, but you sure could by a history of METARs dating back 48 hours.
Not to mention the heavily METAR based engine of P3D could easily profit from this by allowing for a more dynamic way of injecting weather as it doesn’t have you rely on a single METAR, where the next could significantly change, but rely on a string of METARs to produce a “moving picture” like a movie.

If MSFS produces a 250/4 wind it will produce a METAR that reads 250/4, relaying that 250/4 wind for P3D and XP engines to use, thus everyone online flying in a 250/4 wind. The same goes for visiblity, cloud cover, pressure, temperature and dewpoint.
So why do people insist MSFS or dated data cannot be used online?

1 Like

At least, it could be an option user could choose.

2 Likes

Exactly… But I don’t think that’s going to happen. they won’t cook an extra soup for everyone because that doesn’t make any profit.

MSFS is intended for the masses. This can be seen in air races as an example. But such players are not interested in a meteoroligically correct weather representation. That’s okay, everyone’s own. But I think they should at least stick to their announcements. And there was no word of metar-based weather but of a completely new dynamic weather engine. When I fly with my airliners, I want to experience natural skyscapes, as it was in the beginning. And not unlovingly clapped cumulus clouds everywhere. That’s what fascinates me about flying with MSFS: Believable and natural looking weather.

3 Likes

And did they tell us it would be accurate to whats happening IRL right now? No they didn’t! They told us it would be close to the weather we have. I’m totally fine with that as long as it is realistic. Now it’s not realistic or accurate.

The soup has already been cooked and was on the menu. That’s the mystery. Why take it from the menu when it IS popular among some virtual pilots (perrry you and me e.g.) and has already been (past tense) made.

Right, exactly that! I don’t care if it’s raining like at home or not? Where is it written that the weather must be exactly the same? If you want that, please go outside. :stuck_out_tongue_closed_eyes: This is a virtual world. Sorry, but slowly this really upsets me!

1 Like

Helpful would also be finally a comment from the official side. Maybe the whole thing is just buggy as hell at the moment. But none of them lets themselves be lowered to it. Three threeds are full of this for a week now.

2 Likes

That’s what I mean when I say why is it so important to have the exact same values (there is a difference between accurate weather and accurate weather ‘values’) in-sim as they are outside right now. Here in Vienna right now it’s pretty cold outside (4°C) and fair weather. If the simulator at the same time depicts weather around Vienna to be 10°C and be overcast why should I care. Those are not accurate weather ‘values’, but perfectly accurate weather for a November day here in Vienna.

2 Likes

MSFS airport surface winds, temperature and pressure have been coming directly from current METARS (not the MeteoBlue model) since December of 2020 — almost a full year now.

What is new in SU7 is trying to blend clouds and visibility between METAR and MeteoBlue. Asobo may eventually decide that it is simply too complex to work, but I am willing to give them some time to try to perfect it. There are some obvious bugs in this first attempt to do this - particularly discrepancies between METAR cloud heights and what is actually injected into the sim. Hopefully that will be addressed with a patch relatively soon.

3 Likes

I am not worried about bugs. I am worried about compromising a dynamic weather simulation. There shouldn’t be any need for “blending”. It should all “flow” seamlessly, like it did before. As I mentioned if the weather and the METAR (or any other crucial data) don’t match, the METAR (and any other curcial data) is what needs correcting, not the weather, if the weather itself is behaving realistically (not perfect but realistic)

When I think of live weather I think of it as being “alive” and imitating the real weather.
When I think of a live dinosaur in a video game I think of it as behaving realistically and not copy pasting a scene from Jurassic Park.

The frustrating part here is it was initially included that way. It’s not like it isn’t or wasn’t possible. It’s already been there. Many rough edges, but it has been there.

2 Likes

I thought it was integrated in SU4. Because that is when the wind got static at the METAR wind speed (Not realistic either). The wind should be around that value not at that value to be dynamic. Why strive to have everything static at a rate when thats not how winds or weather work IRL.

Well one hope is that they are currently relying on static weather data to get things right, because they are still working on figuring out how to do it the “Meteoblue” dynamic way. Like the wind bug they had with the same winds aloft all over the place. So they introduced the “static way” for realistic winds aloft.
It’s a huge speculation here, but if that is what they are doing I can very much live with it.
And if that really is the case then at least revert the clouds back for now.

+1, same weather almost everywhere



bland repeated clouds at ground level, always giant cumulous.

1 Like

Agree. I thought they should improve the winds to be more dynamic in this SU7 update. Now we are focusing on more things that got broken instead :rofl: I soon give up and it can be this static mess. I thought this sim weather had potential to be real great. But now i don’t know what to think about it anymore.

Besides it’s we the community that requested METAR in the sim because the weather didn’t match the current real time weather outside of our windows. No complaining to the devs they managed to implement it :man_shrugging: We as customers will never be happy. I think the devs needs to decide there own route for this.

2 Likes

Looking at the entire project I have to conclude they are now just playing around with seeing what can and can’t be done with weather, but that in due time they will get it very right–I would expect nothing less from Asobo.

They haven’t though, at least not in a way that works properly. All the time you have a metar that reports an overcast layer and all you get in the sim is blobs of cloud or scattered blobs then this isn’t doing it’s job correctly surely?

But should METAR inject the overcast or should meteoblue or if Meteoblue predicts overcast and not METAR what should inject what? It conflict with eachother. If metar says overcast and meteoblue does not it’s a problem.

Well exactly, your guess is as good as mine. But in its current state neither seem capable