Well I would probably report it as a bug (on the main forum not the beta) and this will give people a chance to test it and vote for it if they are experiencing the same thing
6 Likes
I don’t know why people are looking for another explanation (nefarious or otherwise) as to why there would be a rough transition at certain hours when we have a pretty good one already from Asobo. According to Asobo, there should be four every day as they only get six hours worth of Meteoblue weather data at a time. That six hours of weather data is smooth and fluid from hour to hour, but it would not be smooth between data sets. Hour 6 of the old data and hour 1 of the new data could be quite different because forecasts can change as you get closer to the hour, and METAR data is blended in by Meteoblue themselves according to Asobo, which would only make sense for the first hour. So if the last Q&A was accurate, there would necessarily be a METAR-style rough transition across a large percentage of the world every six hours, and I doubt there’s much that can reasonbly be done about that. They can slow it down, but only to a point since the weather would also still have to transition to hour 2 of the new data set and stay in sync. It’s not an easy problem (if it works as Seb described).
4 Likes
Then we could see that hard transitions 07:06Z, 13:06Z, 19:06Z and 01:06Z. And then add METAR fog transitions every 30 minutes. 48 multiplied by every single airport that reports local METAR+4 global meteoblue transitions. If we are located at one airport all the time for 24 hours that means we would have 52 transitions during that 24 hour period of time. Pre su7 we had only 2 global transitions per day nothing else. I never noticed those 2 transitions though.
The only things that blends what i understand from that q&a is the things that we can’t see. Winds, pressure and temp? But i’ve noticed hard transitions when a METAR report completely different winds. For example if the old METAR reports 4KTS and the next reports 20KTS it changes instantly to 20KTS when the METAR updates.
Even if the 12hour data they used before su7 may have been a bit more inaccurate i found that weather 100% better/more realistic than what i have after su7. It felt like a fluid without hard transitions all the time.
Actually i could see more of the weather i had outside my window back then than what i see now. Now it feels forced into a state with a fixed type of clouds (comulus/CB).
One thing i wonder is, they mention they blend in METAR. How can they blend in METAR when they can’t know 6 hours before what kind of METAR will be reported? Meteoblue sends out only data every 6 hours right?
Martial even said the new METARs in between those 6 hours meteoblue updates is on top of Meteoblue. A completely different system. Meteoblue has always used METAR data to calculate their data with those supercomputers they have. But that is old METARs together with data from sattelites and other data hey use as initial conditions for those calculated forecasts. The new METARs is added on top of the Meteoblue data to have more precission Martial said. But more precission/frequency in updates of weather also means more hard transitions and fixed weather. That we have seen after su7 both in clouds, visibility, pressure and winds.
Correct, thats how it works now but it were not how it was advertised or released and i can’t see the improvement over pre su7 weather system.
4 Likes
We’re talking about cloud transitions, right?
Complaining about clouds suddenly changing from “beautiful” stratus to " ugly" cumulus?
METAR is not used for clouds, so there are 4 transitions a day in this context. Either way, reducing the weather updates from four a day to two a day will cause the same people complaining about transitions to complain that the weather doesn’t match meteoblue or real life because a 12 hour forecast is not as reliable as a 6 hour forecast. Kinda like how the same people who claim to not want METAR blending are complaining that the weather doesn’t match what they see outside of their window. And the same people that claim to not want the weather to be messed with artificially are calling for Asobo to artificially generate more stratus clouds because “they’re beautiful” .
I think it’s very useful to try to understand the technology and make suggesting based on the underlying limitations, instead of just complaining constantly even after the system has dramatically improved.
3 Likes
Well, i have tried to do that. My suggestion is to have a realistic fluid weather system without METAR interfaring. Then if it’s 24 hour old data who cares. The real weather we have could be caused by a thing that happened 2 days ago on another continent. But in the real world we can’t complain the weather isn’t what we thought it should be when we arrive at the location we predicted a complete different weather before we started our trip. We need to accept the weather didn’t match our prediction. To me it’s more unrealistic when the weather is matching METAR spot on. Because then i know it’s not a dynamic weather. It’s fixed.
Well, hope they nail the weather. I just tried to pinpoint an issue i’ve seen 07:06Z. It’s an issue to me because those transitions destroys immersion.
I now need to plan to avoid those transitions.
5 Likes
Weather updating multiple times a day is one thing, however cloud types are also changing during a certain time. The clouds seem to stay in place but are broken up and less cohesive. I don’t think this is due to a weather update from Meteoblue.
As users we can only report our observations. Even though Seb recently explained how the weather technology works, even he admits to not fully understanding.
11 Likes
This is just wrong. Look at how the cirrus cloud changes to puffy cumuliform nonsense.
Something is definitely wrong and it needs an attention from the devs immediately.
12 Likes
I don’t care too much about that. SIM simulated weather will never be exactly like real life weather . Many times Weather forecasts are also different from real weather,as we all know, despite the immense computing power dedicated to them. My desire is that the sim could show all the different type of clouds as much similar as possible to the real ones including thunderstorms with thunders and lightnings in order to have a more realistic flight feeling when going through them.
1 Like
But for the first time it actually felt like real weather in a sim when this got released without hard transitions. If they simulate the atmosphere to be like a fluid then they have as real weather as it gets. If they fix the weather as METAR says they have a solid state like a road. Weather is not solid.
The hard transitions they introduced in su7 is far from real.
@ZodiaLight That clip shows it perfect. After that switch the clouds stays broken apart. We can see in that clip that the weather has not changed only the forms of the clouds to be split up. That must be a bug.
4 Likes
i’m amazed by how little attention this gets, not only by the devs but the users aswell, i guess that’s what you get and what MS/Asobo is heading for, gamifying a sim.
to this day the “no dumbing down of the sim for the xbox release” is outright a lie
- lod quality.
- texture size for planes reduced
- obvious checkerboard pattern (it may only be smaller than before) of clouds, water and vegetation
this is on live weather
and all this, and i checked fps with the same hardware since release, it doesn’t even run any better to justify the downgrade.
8 Likes
I can I move thispost to have it both here and into the bug section of the forum? This topic needs the attention of the developers it might be a bug thatif solved could fix all the clouds issues we had since su 7
3 Likes
If I was you I’d start a fresh new thread in the General Bug forum, not in the beta area. As I highlighted in this post below the beta threads get archived and if they don’t get logged they get forgotten about, which may well happen to this thread;
https://forums.flightsimulator.com/t/suggestion-regression-bugs-slipping-through-beta-periods-move-the-threads-dont-archive/
Yes but then we will lose all the videos snd photos and we’ll have to start all over again…a moderator could move it ?
4 Likes
Copying my post from another thread:
Can confirm that this happens exactly like described. I also witnessed this bug at a different time of day, close to LKPR, if I remember correctly at around 13:00 Zulu.
This would fit with the guess of the old thread that this behaviour has something to do with the 6 Hour Meteoblue update cycle.
I swear OP, if you single-handedly bring back pre SU7 clouds by noticing an obscure bug I will come to your home and hug your feet.
8 Likes
I’m glad that’s where that went. 
1 Like
It’s like an entirely different algorithm has been used to display the exact same weather.
5 Likes
I hope this thread will be quickly classified as bug-logged as this is maybe the root cause of the main complaints we are reporting since SU7.
6 Likes
Agree, i get the best looking weather pre 07:06Z that means the good algorithm and data starts at 01:06Z? The whole night when i working is the best weather LOL and most of us in europe is sleeping at 03 local time here in Sweden. Can’t even plan to use the good weather. Those living in US are happy though.
7 Likes
