Pilots rely on TAFS more than they would METARS when you compare departure to arrival phases of flight. Departing within 30-45 minutes, METARS. Having a 3+ hour flight, they’d use TAFS to get an idea and then once within range of say about 30-40 miles would catch the ATIS, which would only have about a 20 minute window of actual use. They all have their own jobs to do. But METAR is just what we’ve always used for weather before and so it makes sense people would see no use for TAFS unless they are planning moderately long trips.
1 Like
I think it would have been more useful/realistic to only show TAF in the sim. But not change the weather to fit those TAF. Or maybe have those reports i showed here above that we could scroll forward in time inside the sim.
Agreed. We know changes are most likely coming soon enough. No sense in beating a dead horse.
1 Like
I really hope you are right 
Actually i found this information at Meteoblue page. They have more frequent updates of Meteobluedata since su 7 right. Every 6 hour compared to every 12 hours pre su7.
- Interim updates (06:00 and 18:00 UTC) tend to be less complete due to fewer observational data, so that the forecast quality of the model runs based on the 06:00 and 18:00 UTC assimilations has not been significantly better than those from the 00:00 and 12:00 UTC assimilations.
scroll down to quality of model runs
Maybe explains the less quality in weather after su7? And the less quality after 07:00z switch that has been reported?
Maybe all of this discussion is the more frequently updates we get from Meteoblue.
In that case none of the issues we have reported since release of su7 is a bug. Only maybe a bit more accurate but less quality because of those 2 interim updates they added. I bet they also integrate the messured data every hour that also interfare with the model and makes it less quality.
I think we that liked the weather pre su7 wants the 12 hour raw model data back again. That is the thing we are missing right now. Accurate or not, i don’t care actually.
And a question, How can Asobo improve the data they get 06:00 and 18:00? Thats Meteoblue data they get. Only Meteoblue can improve that data. But Meteoblue also says this.
What matters is not how often , but how well we update ! because they know the more updates the less quality it gets. I think that describes the weather in the sim well after su7.
Asobo only renders the data they get, the output of the rendered data will be the quality of the data they input.
Well i’ve learned since this sim got released, the quality is not important as long as it matches those METAR. I preffer quality though.
Source? Just curious as I don’t know what the update cycle actually looks like.
Adding 6z and 18z model data shouldn’t decrease the simulator weather accuracy or quality. It should improve it actually unless there’s something wrong with the Meteoblue products.
The 5 day forecast from a 6z model run has typically less skill than the 5 day forecast from a 0z run because it’s missing stuff like upper air observations from weather balloons.
But these differences are a lot less apparent in the short range of a few hours, and that’s not even what we’re comparing when talking about Flight Simulator Live Weather. For example, the 22z simulator Live Weather would be a 10 hour forecast if you only had the 12z and 0z model runs available. But it’s only a 4 hour forecast if you have the 18z model run. Despite having less data, the 18z 4 hour forecast should be more accurate than the 12z 10 hour forecast.
1 Like
Even though several thousand posts have now been dedicated to explaining every nuanced bit of this issue in graphic detail, I’m worried that Asobo is going to have the same reaction if prodded about it in the next Q&A: “These guys wanted METAR based weather, and now they don’t!?”
Martial explains it here.
Also in the q&a right before su7 were released Seb explained they had more frequent updates coming from Meteoblue in su7.
It’s not me that explains that. It’s on their page when they explain the difference in quality. They simulate weather, adding in some new data into that simulation i bet will interfare with the already calculated/simulated model. Maybe they describe a problem with their product. But i think it’s more a limitation they describe. Either we have a more accurate model or a more detailed model.
Well, i do not expect they will add the old 12 hour model as an optional feature. I think they will stick with this new model. But to me it would be awesome if they did 
For me it’s not only the graphics. For me it’s the whole weather experience. I know that weather doesn’t behave like that. I know for sure real weather doesn’t need to be set in stone because a new METAR report has been published.
He means graphic detail as in being very closely looked at and picked apart. Nothing to do with actual graphics. Just a different way to say precise.
If I asked you to describe in graphic detail what a computer looks like, I am asking for all of the information you could tell me.
1 Like
Thanx for explaining Now i see what he meant. But i’ve seen comments like those that wants the weather to behave like weahter only wants “eye candy” before. In some parts they are correct, but it’s not the whole aspect. We want the weather to behave and look realistic because we have the most detailed terrain to fly in that both looks and feels realistic to fly in. To me a “simplified” weather model that needs to be accurate and set in stone to METAR doesn’t fit anymore. To me it’s acceptable in those older sims that didn’t have the volumetric technology. Now the weather should be formed over time dynamically instead. If not, the volumetric clouds are pointless. They could insert more realistic looking cloud-textures instead if the dynamic aspect of weather is not important in flight simulators.
1 Like
Every airfield with a weather station in SW England has had 0 height cloud for quite a few days now - the weather has been bad in the area but most definitely not that bad unless you’re at the one 800ft elevation field. METAR cloud isn’t working at all,anymore - this summer/autumn it wasn’t too bad - this is far too much. No ( legal ) VFR possible unless both ends don’t have any METAR cover… and the excessive short-period jerking around due to current turbulence affecting every a/c I’ve flown isn’t making it any fun either.
Again, I’d like to have wind & pressure match, at least - but please either just use METAR cloud reports to influence the global model slightly, or just don’t use METAR cloud at all. You can easily spot all the weather stations by their islands of low cloud.
5 Likes
I think the issue is that they can’t change the data they get. Either they get data that is not influenced at all or they get data that is influenced by newly reported METAR. The look of the weather is decided by Meteoblue how they manage the blending of METAR into their model. The blended/post processed data Asobo gets from Meteoblue Asobo has no control over only the rendering of that data. For example:
If Meteoblue sets clouds from METAR at 800FT to maybe fit the global model then we get the clouds at 800FT in the sim. Nothing Asobo can change
In my opinion they should have not changed the Meteoblue model and only injected the fog layer they handle on the client side or simply made the new weather model they recieve from Meteoblue optional. The more they add from those METAR the more we will notice we have METAR weather. Thats the limitiations of METAR weather. A METAR does not report the global weather. Only the local weather. But global and local weather is same thing in reality. Local and global weather is not two different things. It’s the same atmosphere. Thats why i preffer to only have Meteoblue model as source. It simulates an atmosphere that fits everywhere. It doesn’t need to be influenced by anything else that makes the atmosphere behave strange. It inputs initial conditions and then forms using those over time in those super computers Meteoblue has.
I know fog is handled on the client side because when azure weather servers are down we still get the fog layer injected.
Asobo can’t please all of their users without options.
Either they please those users that want the weather to be accurate to METAR or they please those users that want the weather to behave like weather. At release 2020 they made only those users happy that wanted weather to behave like weather. After su7 they pleased only those users that needed the weather to match METAR a bit more.
1 Like
So they’re getting a custom feed from MeteoBlue? because I can’t imagine there’s a generic data feed with METAR data mixed like that, who else is going to use it? If the feed is to their own specs then one might assume they can ask for changes to it. METAR cloud reports are hardly an accurate picture by themselves anyway.
Even if it’s not a personalised feed, they should be reporting issues to MeteoBlue anyway because this is most definitely a bug.
Hopefully when Asobo has their discussions with MB next month regarding API access or any other weather related topics we can see some discussion about it or a development roadmap in their next Dev update video. METAR access can be from any number of freely available sources that Asobo is using apart from MB, i.e. NOAA or any other open source. My hopes are for either an in built weather engine or the choice of choosing a weather source for the sim. These answers to these questions will be here sooner than we think. If Asobo has already somewhat officially announced plans to incorporate a turbulence API there’s definitely going to be more than that in the works probably.
1 Like
I bet their is some other customers who want that thing.
source:
I liked the global weather model we had at release even if it were less accurate. I know it had some issues like 225@3kts and the 12hour missing data. But that were bugs that should have been fixed instead of completely turn it into a METAR-weather.
1 Like
Well, under “Measurements” - which prresumably covers data used in METAR from airfield weather stations - they list satellites & radar for sources for cloud cover, and other observations ( maybe airfield cloud “sensors” might come under private sensors… then again so might airfield wind & temperature sensors ) as verification. I’m not disputing people wanting models using data from weather stations, I’m wondering who outside flight simulation customers is going to want METAR data specifically in the output mix- it’s not much use for en-route weather prediction ( not much use for anything but wanting to know the weather at the airfield right now, honestly )
Basic airfield cloud sensors basically just look straight up in a very narrow fov, right?
I’m not sure who wants it. I don’t want it to mess with the global simulation of atmosphere thats for sure. Those sources are completely different things.
Either we want the weather to be simulated or we want those messurements be set in stone. If they implement those messurements the weather is not simulated anymore because the weather needs to match those static sources.
I want the weather to be 100% simulated. But that thing we can’t have anymore since they started to implement things from METAR.
Please Asobo give us the option to choose the global weather model.
2 Likes
If they just took temperature / humidity / pressure from airfields just like any other weather station I think it might work fine as it is - I haven’t noticed any harsh transitions in air pressure or temperature at low level in a very long time. Quite why they have to mix airfield cloud “observations” in when they have perfectly good satellite radar/imagery I have no idea.
But it goes back to whether metteoblue offers custom data feeds and Asobo/MS went “hey, put METAR data in there”, or whether this is what they offer everyone, and I’m a bit doubtful of the latter. If it’s the former then Asobo can ask them to stop using the cloud readings.
1 Like
Agree, that they can ask, But the data they get is handled by Meteoblue. And whatever they choose to have in the sim it comes with limitations.
Global weather model only = least accurate but most dynamic
Messurements = Most accurate but least dynamic
And we as users preffer different things and thats why i only see options as the optimal fix to make most of the users happy.
1 Like