Abrupt Weather Changes?

[Rudiii 257]

Not seen a thread or much discussion on this so hoping to get some insight.

Loaded up the sim for an hour after SU7 and mostly things went well. I was trying out the FA/18 over China Lake and out towards LA.

As I was approaching LA, the haze and fog looked great in the distance but unfortunately when I reached the haze/fog it really suddenly changed and the entire area was foggy including the area I had just come from which had no fog previously.

I then looped around and headed for LAX and once on approach, an abrupt change happened again; this time from low visibility/fog/haze to perfect clear skies.

Is anyone else experiencing this? Could it be due to the servers being completely rammed? Or maybe due to me flying such a fast aircraft between weather stations/airports/areas?

I really hope abrupt weather changes are not the latest issue to contend with.

[vcarlo 529]

 

3 minutes ago, Rudiii said:

Not seen a thread or much discussion on this so hoping to get some insight.

Loaded up the sim for an hour after SU7 and mostly things went well. I was trying out the FA/18 over China Lake and out towards LA.

As I was approaching LA, the haze and fog looked great in the distance but unfortunately when I reached the haze/fog it really suddenly changed and the entire area was foggy including the area I had just come from which had no fog previously.

I then looped around and headed for LAX and once on approach, an abrupt change happened again; this time from low visibility/fog/haze to perfect clear skies.

Is anyone else experiencing this? Could it be due to the servers being completely rammed? Or maybe due to me flying such a fast aircraft between weather stations/airports/areas?

I really hope abrupt weather changes are not the latest issue to contend with.

https://forums.flightsimulator.com/t/weather-transition-is-not-smooth-after-su7/472467/14

[Rudiii 257]

20 minutes ago, vcarlo said:

 

https://forums.flightsimulator.com/t/weather-transition-is-not-smooth-after-su7/472467/14

Thanks, search only brought up threads from August last year for me, somehow.

Less people discussing it than I expected for something as immersion breaking as this. Especially as the 10SM issue seems to be far more discussed and I wouldn't have even realised that one!

 

[vcarlo 529]

33 minutes ago, Rudiii said:

Less people discussing it than I expected for something as immersion breaking as this. Especially as the 10SM issue seems to be far more discussed and I wouldn't have even realised that one!

 

Probably because several similar threads are scattered everywhere.  Here's another >> https://forums.flightsimulator.com/t/u-s-metar-visibility-incorrect-at-most-airports/472609

[michaelray 236]

I did a short hop from LFBO to LFBD in the CRJ at FL180.

Fog was forecast for both airports. All good at LFBO and nice to climb up through the fog layer to on top.

However cruising along areas of fog form almost instantly right in front of you. You fly over one area and then another pops in. I didn’t check but assume they will be near airports with METARs.

Same on approach to LFBD. It was certainly more abrupt then what I was expecting. 

I think it will be a amazing once Asobo have had more time to fine tune it but for now it is jarring.

At least they gave us a heads up that it still had warts on it.

[RabH 43]

I had the same approaching LSZH (Zurich) this morning, the change was very abrupt! It is better than before though and will hopefully improve again with the next update, the haze is certainly a huge improvement!

[scotchegg 4,311]

Same doing a circuit from and back to Lugano, sudden transition into dense haze then back to clear with few in the distance, then dense haze popped in on approach.

On the other hand the haze was really well done, best I’ve ever seen in a flight sim. Just needs work smoothing…

[Rudiii 257]

It absolutely looks better visually. I really do hope they can iron out the abrupt changes specifically.

I do wish there was a way to have dynamic weather presets for the time being as opposed to static presets. I'll have to see how much of a pain this will be over the coming months.

[Farlis 2,352]

This is what happens when you re-introduce the old station based weather system without figuring out how to interpolate between the forecast model and the METAR reports.

I consider this a work in progress.

Edited November 19, 2021 by Farlis

[jrw4 627]

2 minutes ago, Farlis said:

This is what happens when you re-introduced the old station based weather system without figuring out how to interpolate between the forecast model and the METAR reports.

I consider this a work in progress.

This ^

The challenge with interpolation is how to do it, without bogging down our computers. This is a nontrivial computational burden that involves both distance and time. It's not only flying between stations, but also watching what happens when new weather observations become available. I have been jumping around various airports that are shown as being IFR or LIFR on Skyvector. The weather depiction is simply outstanding. If Asobo stuck with the original "smooth" weather model provided by Meteoblue, the interpolation problem is obviated both in time and space, but we all wanted METAR-based weather. So they gave it to us. This introduces transitions. Resolving that one will take a lot of additional effort.

 

[mwilk 1,109]

This may be a case of, "Be careful what you wish for, you might get it."

[Gilandred 732]

25 minutes ago, jrw4 said:

we all wanted METAR-based weather.

 

I don’t believe that’s true at all.  Personally I only needed a reliable way to read the existing predictive weather model conditions at any given airport.  Whether or not that corresponded 100% to real world METAR conditions was irrelevant to me.  I think Asobo misinterpreted the underlying intent of what was being requested, and thereby made the solution way more complex than it needed to be.

[Farlis 2,352]

I'm also not sure if "we all" wanted METAR weather.

What I wanted was somewhat reliable depiction of current real world conditions, which are reported in METAR reports hourly but also through various predictive and observational weather models.

It's about what data to use, how to interpret it and then how to inject it into the sim. 

The most important thing that needed, and probably still needs to be adressed, is proper visibility depiction. Whether it is humidity, fog or clouds, smog or smoke that influences it, the most important thing is that the sim shows the real world current conditions and not some approximation where the visibility is completely different and rain might or might not happen. 

I don't need every raincell to be reflected down to the kilometer, but, considering that radar can show where precipitation is live, globally, I expect something similar in the sim, if Meteoblue provides such data.

How this is achieved is up to ASOBO. But it is important that it is achieved.

[WestAir 2,532]

11 minutes ago, Gilandred said:

I don’t believe that’s true at all.  Personally I only needed a reliable way to read the existing predictive weather model conditions at any given airport.  Whether or not that corresponded 100% to real world METAR conditions was irrelevant to me.  I think Asobo misinterpreted the underlying intent of what was being requested, and thereby made the solution way more complex than it needed to be.

Sadly, you (we) are in the minority. The vast majority of devoted simmers wanted METAR accurate WX. Many for Vatsim, others for VNAV and WX reports, and so if winds were reported as 090/09, many had a problem if it were 085/05.

And so the solution is what we got, but in my opinion it came at the price of realism. I didn't care if winds didn't match the real world, I cared about the realism of the system, and in my opinion we lost that. 

Weather popping is here to stay, because the alternative is inaccurate ATIS reports.

[Rudiii 257]

36 minutes ago, mwilk said:

This may be a case of, "Be careful what you wish for, you might get it."

I wish I had been a louder voice in pushing for a switch to turn METAR on and off. 

It's obviously popular and needed for the people who plan ahead and use things like VATSIM so the feature is very welcome.

But my personal taste makes me wish I could just use the forecast model while others still have access to the METAR injection model.