1 hour ago, lwt1971 said:

The clouds degradation in live weather started with METAR integration, so I'd say that's the main issue here. But they're improving this over multiple SU releases with SU10 particularly being a great step forward.

I see no reason to see why METAR integration has anything directly to do with cloud depiction because you see this degradation between METAR zones when there is no METAR global weather has to interact with.

Integration algorithm has to take into consideration the MSA for the airfield, since METAR is restricted, in terms of observable water meteors and their bases, up to that reference altitude.

In some places this altitude can be rather low, so you can have an overcast layer above it and no clouds reported in the METAR.

You should be able to "glimpse" that layer from the GRIB data that comes from Meteoblue, but sometimes it's not that easy...

1 hour ago, Noel said:

I see no reason to see why METAR integration has anything directly to do with cloud depiction because you see this degradation between METAR zones when there is no METAR global weather has to interact with.

Ok, I was talking specifically about the clouds variety issues in live weather which I thought only came with the SU that introduced METAR integration. Don't remember what other weather-related changes they did in that SU.

I couldn't ask for much more...

Here are the links to two shots taken from a flight today between Copenhagen Kastrup and Lisbon LPPT, taken from the Fenix A320 flying at FL370 just entering the Biscay golf :

https://drive.google.com/file/d/15F9wWWv0nN-TYYoJUh0DHOiPDpLfC9ce/view?usp=sharing

https://drive.google.com/file/d/1i2-La-MjaucYkHb-L_jz1bdsp2p1cWgA/view?usp=sharing

Default online weather ( as well as ATC and AI )... 

I am testing DX11 + DLSS  - so far so good, although I never had problems with DX12 + DLSS and will revert to it afterwards...

P.S.: Well - CTD nonetheless 😕

After 3+ hrs I had a CTD 😞

I'de blame DX11 because I never had it in SU 10 beta with DX12... but I really don't know:

Faulting application name: FlightSimulator.exe, version: 1.27.16.0, time stamp: 0x00000000
Faulting module name: FlightSimulator.exe, version: 1.27.16.0, time stamp: 0x00000000
Exception code: 0xc0000005
Fault offset: 0x000000000227f467
Faulting process ID: 0x3c4
Faulting application start time: 0x01d8b54056ca5143
Faulting application path: C:\Program Files\WindowsApps\Microsoft.FlightSimulator_1.27.16.0_x64__8wekyb3d8bbwe\FlightSimulator.exe
Faulting module path: C:\Program Files\WindowsApps\Microsoft.FlightSimulator_1.27.16.0_x64__8wekyb3d8bbwe\FlightSimulator.exe
Report ID: 60a435fb-91fd-4320-8f7f-c51e2bef5856
Faulting package full name: Microsoft.FlightSimulator_1.27.16.0_x64__8wekyb3d8bbwe
Faulting package-relative application ID: App

Edited August 21, 20223 yr by jcomm

Just had an interesting phenomenon climbing out of KSUX Sioux City. METAR:  

Wind calm. Visibility 0.25sm. RVR runway 31 from 1200 to 1400ft. Fog. Vertical visibility 200ft. Temperature 14°C, dew point 14°C. Altimeter 30.13inHg. 

The ATIS was pretty much spot on in terms of wind, visibility (it reported no clouds though), temperature and altimeter. And the actual conditions at the airport seemed to correspond well: 

Above the haze layer at 6000ft: 

This got me thinking about what Seb was saying about the weather model in the recent QA. That METAR data is in effect below 1000ft and within 2km from the airport, IIRC. Obviously there's some interpolation/blending with the Meteoblue data in the region. This is SU9 by the way. 

For me the biggest issue with the weather so far (using the latest beta and Live-weather) is that most of the time it doesnt depict broken or overcast clouds when metar says so. There are a lot of blue sky and a few cumulus clouds. For me having correct cloud coverage above airports according to metars is more important than having just « nice looking clouds ». Because when metars say BKN006 and we can clearly see the runway from an altitude of 2000-3000ft, it’s not realistic. And even when parked at an airport that reports broken/overcast clouds it doesn’t depict it.

Edited August 29, 20223 yr by Franz007

Is the visibility issue fixed?
Because my last 2 flights on SU9 have not been great. Arrived in Agadir the metar tells me visibility 6000 m, I had a CAVOK up to 500 feet and suddenly, it puts me 6000m visi! Same for another Rabat flight 5000m of visibility and once again just before touchdown, I go from 9999 to 5000 all of a sudden! Not great for immersion (I fly in real weather.)

3 hours ago, Axis3600 said:

Is the visibility issue fixed?
Because my last 2 flights on SU9 have not been great. Arrived in Agadir the metar tells me visibility 6000 m, I had a CAVOK up to 500 feet and suddenly, it puts me 6000m visi! Same for another Rabat flight 5000m of visibility and once again just before touchdown, I go from 9999 to 5000 all of a sudden! Not great for immersion (I fly in real weather.)

I've been reading this thread and honestly it looks like some of the observations, which I do not doubt are true and precise in what they aim at describing limitations in MSFS's weather, but truth is that my experience has been, at least since SU9 was released, quite the opposite.

I'd say that 99% of the time, and when I play chasing bad weather all around the World to go there and taste the feel of MSFS under bad  adverse weather coinditions, what I see rendered in the sim, and most of the effects on the aircraft ( less the turbulence and updratfs / downdrafts inside of convective clouds which are still under tame mode since a few releases due mostly to the user community asking for it... ) are as good as I could ask for ?

Are there limitations ? Of course there are - there are still so many weather phenomena not modelled, but heck - this is not a Meteorology Simuulator ....

Edited August 30, 20223 yr by jcomm

6 hours ago, jcomm said:

. . . , but truth is that my experience has been, at least since SU9 was released, quite the opposite.

I'd say that 99% of the time, . . . , what I see rendered in the sim, and most of the effects on the aircraft . . .  are as good as I could ask for . . .

The problem is that there are always people who want to have everything 100% like in reality. This problem is magnified by the fact that there are essentially two factions: those who demand a 100% accurate representation of METAR weather conditions and those who are satisfied with a generally believable weather representation without the weather having to be exactly as it appears when looking out the window.
A correct METAR display always requires a blending between airfield weather and weather outside the METAR range. To solve this task programmatically, so that there are no jumps in the weather display, seems to be extremely difficult, if not impossible.
We should give Asobo time to slowly approach this goal and not become impatient. i personally give Asobo at least another 2 years. However, I don't belong to the faction of METAR nerds (if I may say so) and therefore I basically don't care if Asobo succeeds with it or not.

Well I guess you can make all simmers happy some of the time and some of the simmers happy all the time, but you cannot make all of the simmers happy all the time. 

I think we're getting closer though. With the more realistic of low cloud layers and better interpolation between METAR and "non-METAR" areas in SU10, things will improve. 

11 hours ago, asanosho said:

However, I don't belong to the faction of METAR nerds

I guess you don't fly on Vatsim/Ivao then!

Let's put it in very clear words:

Why do VATSIM/IVAO users think they have the right to demand a true-to-life METAR representation if this results in the global weather representation, which was very decent until METAR incorporation was implemented, becoming worse for all users as a result?

In my opinion Asobo has capitulated here to a very loud but actually very small minority of users and tried too fast to build into a quite well working system that implements a revolutionary new technology for weather generation (visualization of a global weather model) something that is based on a completely different principle (discrete weather measurements).

That this would not work right away can actually be imagined by any sensible person.

Therefore, also and particularly VATSIM users should have the patience to wait one or two years longer for an improved METAR integration (if this is technically satisfactorily possible at all). I sometimes have the impression that this is not the case at least with some VATSIM users.

7 minutes ago, asanosho said:

Why do VATSIM/IVAO users think they have the right to demand a true-to-life METAR representation if this results in the global weather representation, which was very decent until METAR incorporation was implemented, becoming worse for all users as a result?

I think it's not unfair to ask this question. And it does seem that the METAR implementation was indeed rushed, and that the devs are still working on getting a decent blend between the global model and the METAR model. 

I don't use VATSIM because my wife will throttle me if I sit here all evening jabbering gibberish at my computer. But I still want METAR data to be incorporated into the sim. 

Especially given that the sim is capable of running different weather in different areas, there should be a way to properly integrate it with the "global" weather such that it isn't problematic. Hopefully that'll be addressed at some point. 

My biggest issue with the weather engine these days is that when it updates with fresh data, there's no smooth transition. That becomes a problem if the previous data is wrong, radically different, or not loaded.  There have been a few times where I'll be crossing the threshold in CAVU conditions and suddenly I'm in the middle of a thunderstorm and a 30kt crosswind.  

10 minutes ago, eslader said:

Hopefully that'll be addressed at some point.

It's an ongoing project and in the current beta they've made some more changes in order to hopefully get more smooth transitions. 

11 minutes ago, eslader said:

My biggest issue with the weather engine these days is that when it updates with fresh data, there's no smooth transition.

Agree.