Live Weather Does Not Match

[CharlieFox00]

Three times tonight I had a nice blend of the generated weather and METAR that included low layers and visibility that otherwise weren’t being picked up by the mesoscale generation outside of the METAR area. Bubble schmubble, it looked good and behaved as expected based on all the forecast products. The blend worked and made for both a beautiful and practical flight.

The only way it would have been discernible as a bubble is if I were looking at the in-game radar depiction (which is more inaccurate outside the bubbles, anyway) or if it was clear skies everywhere else except the bubble.

The only thing that “got” me was a METAR-induced altitude drop, however it dropped my altimeter to the correct altitude (high to low, look out below), so the model wasn’t behaving correctly there.

@SkipTalbot

I’ve seen the model nail frontal boundaries both spatially and temporally.

Only if you don’t mind it being off (as in ahead or behind) by 30-50 miles as the model ages. When playing around the periphery of marginal weather (as often is the case this time of year), that can make a huge difference.

[SkipTalbot]

@CharlieFox00

Only if you don’t mind it being off (as in ahead or behind) by 30-50 miles as the model ages.

Sure, it can and probably will diverge, but if the run’s forecast for each hour is accurate, it’s totally possible the sim will draw the front right where it should be, and continue to do so, even if the model run is twelve hours old.

Meanwhile the weather the METAR is reporting could also easily get displaced 30 or 50 miles before it updates. One of those (probably less common) times when the forecast model is more accurate and current than the latest METAR observation.

[ConsularBuckle4]

Should be added that all RW planning tools which are not based on terminal information use precisely the same type of weather forecasts as the one which is the source of weather in MSFS. So if I plan a VFR flight from EKBI to EKAH and use the official wind data for the planning I’m just as likely to be off as if I used Meteoblue. And doing it in the simulator helps me prepare to dealing with it in RW.

[Perrry]

The only improvment i can see that would actually improve the simulation and accuracy at the same time is to have the “icon 13” model. When looking at those weather models i have found that model the most detailed and accurate. Meteoblue even use that as default model when we check clouds forecast on their weather map page. But now we had this NEMS model and that i found realistic enought. But try make that fit those METAR sources that is something completely else is really bad in my opinion.

[CharlieFox00]

@SkipTalbot

Meanwhile the weather the METAR is reporting could also easily get displaced 30 or 50 miles before it updates.

If it’s a significant change, SPECIs are a thing. Some ASOS stations will report every 10-20 minutes during weather events. In fact, here’s KBLU from yesterday:

KBLU 280913Z AUTO 18008KT 1 3/4SM BR BKN004 OVC010 M01/M01 A2987 RMK AO2 VIS 1V5 T10061011

KBLU 280852Z AUTO 19006KT 4SM BR BKN004 OVC010 00/00 A2988 RMK AO2 UPB0759E01SNE0759B01E11 SLP115 P0001 60007 I1000 I3000 T00000000 50003`

KBLU 280842Z AUTO 22004KT 5SM BR BKN002 OVC010 00/00 A2989 RMK AO2 UPB0759E01SNE0759B01E11 P0001 I1000 T00000000`

@ConsularBuckle4

Should be added that all RW planning tools which are not based on terminal information use precisely the same type of weather forecasts as the one which is the source of weather in MSFS.

There’s truth to this. The sim is fairly accurate on a synoptic scale, and those forecasts IRL are taken with a fairly hefty grain of salt for the enroute phase. We literally go into flying with the understanding that different products get stale - it’s part and parcel of the game. Where it’s causing friction in the sim is in the terminal environment because that’s where traffic convergence, instrument approaches, landing performance, etc all come into play.

It’s interesting, I’ve found myself having to argue much harder than I thought I would in favor of the current system because of all the groupthink that seems to be based on a casual or artisanal approach to simming (and a hefty dose of misconception from some as to how things actually work IRL). I’d prefer everything to be a unitary source as well, but not the direction for which y’all seem to be advocating.

[SealedVolcano50]

casual or artisanal approach

We poor peasants what do we know? Perahps from an aviation point of view you maybe right, but ultimately this is a game. What we ate asking is a weather system that behaves naturally, looks as realistic as it did on the past, both in visuals and effects! It’s a game, and while we want some degree or realism the sacrifice we are enduring for METAR is too mutch, knowing what yhe software is capable of. Asobo anyway need to listen to voth sides and give us the option for a toggle at least! Me too i never asked for this. I even both my xbox to play this game that now i hate when i could have gone for a ps5!!

[CharlieFox00]

@SealedVolcano50

We poor peasants what do we know?

We’re clearly dealing with two or more sets of expectations, thus I stand by my words, but they were not intended to insult, and I apologize if I insulted you. Things don’t always come across well in text.

My intent in joining this thread was to offer some real-world expertise into aviation weather products as well as an insight into how many people approach the sim, one that really hasn’t been discussed lately in the thread. I originally came here to advocate for more granularity and accuracy in timing of weather (specifically, the integration of real-time radar), and have provided evidence to back my observations and arguments. However, I also know when I don’t have expertise on a subject, which is why you don’t hear me talking about the weather’s effect on water and seaplane operations, for example. If I start making strong assertions in that field, please come thump me on the head.

Nevertheless, there are a lot of dominant, repeated assertions here (and in several similar threads) that things should be done a certain way, based on evidence that I‘d consider loose, again from that operational standpoint. Some folks are (perhaps unconsciously or unknowingly) advocating for less accuracy, especially as time passes from the origin of the model. That’s fine based on one approach to gaming, I get where you’re coming from, truly. If I were playing KSP, I wouldn’t want competing atmospheric datasets. However, I disagree with that approach from an operational aviation and real-world flight planning standpoint, which is the way I (and many others) approach a simulator that is purporting to offer the utmost in realism in so many ways.

Ironic that the thread is a bug report, but at this point, nearly 2800 posts later, it’s evident that it’s not so much a bug, but a question of theory, perception, and integration. In the end it’s become clear that our disagreements really come down to how one approaches the sim, or game, if you’d rather. Thus…

@SealedVolcano50

give us the option for a toggle at least

this sounds like this would probably be a best of both worlds solution!

[SealedVolcano50]

I want realisms too, but if the compromise to have a more real time weather means getting these kind of clouds

Then no, this is realisms breaking through and through. As i said i wouldn’t be to annoyed of sometimes getting METAR bubbles if at least the rest of the sky and the bubbles themselves lookd as good as they did in the past. I believe at the moment we have lost do much to just see the rain in the game at the same time as outside our windows and flight planning? …

[SkipTalbot]

@CharlieFox00

If it’s a significant change, SPECIs are a thing.

That’s the big disclaimer. “Insignificant” changes mean visibility could drop by miles, the ceiling by thousands of feet, or the sky conditions could go from overcast to clear without a SPECI being issued. Then we’ve got the same issue where the forecast model could be up to the moment correct, but the sim is drawing a bubble of hour old, completely different weather around the airport instead.

I see this a lot behind a frontal passage. Yeah, the METAR is going to update as the front itself passes due to the wind shift at least. But after that the conditions may continuously change while the METAR lags in response. The forecast models generally handle synoptically obvious systems quite well though, and can keep pace with these changes.

I’m not saying that’s always the norm, but there are situations where that’s routinely the case. And it highlights the problems with the current system in Flight Simulator where these two sources can be inherently incompatible with each other, like trying trying to jam the round peg of a point METAR observation into the square hole of the gridded numerical forecast model.

My VFR flights on PilotEdge might make me a casual scrub to the virtual airline captains, but the introduction of METAR data in SU7 was breaking my PilotEdge flights and doing more harm than good. And those Vatsim and PilotEdge users were who this change was intended for. While trying to practice communications, I was annoying real air traffic controllers because of issues created by the game’s METAR injection. Like when I’m trying to follow pattern entry instructions and I can’t see a major class C airport when I’m just two miles out. Or I’m suddenly VFR into IMC because unlimited visibility went to 0 instantly. And it wasn’t because that’s what the actual weather was, but because the METAR transitions and implementations in Flight Simulator were all kinds of wrong. That situations has improved in subsequent updates, but I’m still seeing nasty transitions and bubbles. Until that stuff can be fixed, then as a user of an online network, I’d vote to actually keep only the model based weather. You could tell the controller you have “information Microsoft” and they’ll act accordingly.

[FarlisKaneebe]

@SealedVolcano50

but ultimately this is a game.

Ultimately this is a sim…

[SealedVolcano50]

Also…still a game too, not simulating very well the atmosphere, if at all…since even thunderstorms are gone…

[Perrry]

Well, no need to argue about that. It’s both at the same time. They always have said it’s
a sim for all of the sim pilots. That would mean we that liked pre su7 Meteoblue weather ”simulation” weather should be able to still use that. But we can’t we are here hoping to get that back instead of enjoying our sim/game we bought.

I will stop argue more of what should be in the sim. Neither of those sources will make it 100% accurate. Especially not by blending those sources. I will keep hope for Meteoblue only back as an option. While those enjoys the more planning friendly weather we have now. It will be hard to please us all with this blending. If options is not a thing then prove me wrong please.

[Perrry]

@CharlieFox00

operational aviation and real-world flight planning standpoint

I thought about a thing. Maybe they should name those weather options:

“operational aviation and real-world planning weather” For those that like the planning part to be realistic

and

“simulated atmospheric weather” for those like me that like the flight part to be realistic

Instead of one live-weather because we all know by now, it will be impossible to get 100% live? Besides to me that “live” means that the weather feels alive.

[mbraortns]

I want so much for the sim to be good and I’ve been burned several times before but really the weather (and sometimes the graphics) for my flights over the past several days has really been quite good. I even saw multiple cloud layers today between KPNS(Pensacola FL) and KMEI (Meridian Mississippi).
One of my pet peeves is cloud layers that follow me to the ground and that hasn’t happened once recently.
maybe its getting better?
I do wish the upper level translucent bowties would go away though.

[Perrry]

@mbraortns

maybe its getting better?

We can hope it does I’ve also seen a bit more varied weather. But i’m not completely sure. I still missing those thin lower stratus layers we had and the more realistic fog we had that were formed dynamically near ground. Not always where the fog were reported on those METARs but it felt real. Now i can only see those fog markers around airports.

I bet they have not worked on it over the holiday. I really hope su12 will be focused on weather to get it back as pre su7 standards at least.

[ArlindoCmdt]

How long will the MSFS development team ignore this issue? It has already been found that the real weather injection problem is related to the IPV6 protocol. When I fly I have to use a network connection with IPV6 turned off so that there is real weather injection. This solution is absurd, but it is what is working.

[Speedbrake22]

Still a problem, years after the first post was made.

On the way to mount cook from NZGT

[anon44786522]

Asobo is clearly focused on bringing weather changes to this next SU. Go back and watch the Q and A from December.

[mbraortns]

@mbraortns

One of my pet peeves is cloud layers that follow me to the ground and that hasn’t happened once recently.

and yesterday the clouds followed me down to the ground again on my flight from KMSY to KPNS which I ended at KMOB because the clouds following my altitude kinda ticked me off. But at least the clouds looked cool – not realistic when compared to the actual weather but cool nonetheless.

Weather doesnt seem to be weather – its just like “here’s some clouds…thought you’d enjoy them”
and now for something completely different…

[flybeast691]

Prime example of MSFS’s live weather model not being very accurate, precipitation disks around METARs and NO precipitation in between