Live weather?

[Orlaam 306]

I've only been on this sim for a few days, but the weather is so amazing.  My only gripe is with the turbulence and no cirrus clouds.  Turbulence will be a setting now which appears that the low setting will be much improved for smaller aircraft.  When you see the rains and storm graphics, it'll be mind boggling real.  It's different seeing it in person.

[Sky_Pilot071 580]

Cirrus clouds will be icing on the cake so to speak.  😊

sp

[MarcG 2,993]

8 hours ago, Stearmandriver said:

This is an interesting take.  Pre SU7, weather at airports often bore no resemblance at all to actual conditions, as the 12 hour model runs would often blow the timing of events like frontal passage by whole hours. This completely negated the use of any real-world flight planning tools, and badly broke the experience of networked flying as well (ie Vatsim.). 

The integration of METAR into the depiction of airport weather is probably the single greatest advancement this sim has made in weather accuracy.  It can be debated whether the clouds are more or less "pretty" now (I think they look good), but conditions are certainly more accurate than they ever were before.

The introduction of METAR has caused these METAR "bubbles" which look horrendous, still in the Sim from SU7 to now (let's see if they fixed it in SU12 today), plus it was widely regarded that general live weather was better before that METAR addition. As I said they've gradually got it going back to what it was like pre-SU7, but those bubbles are still very evident and incredibly immersion breaking.

[Stearmandriver 2,297]

What I found far more "immersion breaking" was arriving at an airport and having the airport IMC when it was supposed to be VMC (or vice versa.). You couldn't use Foreflight, Jepp, WSI, or even Navigraph weather tools at all.  Network pilots using sims with accurate weather were all operating happily in sync with the real weather at that airport, and then those folks in MSFS were trying to fit in as able, even though our conditions were often completely different.  In other words, everyone else was operating in conditions in sync with the real world... And then there was us.  😁

I don't personally see a problem with flying through a weather transition to get into the conditions that are known to exist in the vicinity of an airport.  It happens in real life all the time.  

[MarcG 2,993]

14 minutes ago, Stearmandriver said:

What I found far more "immersion breaking" was arriving at an airport and having the airport IMC when it was supposed to be VMC (or vice versa.). You couldn't use Foreflight, Jepp, WSI, or even Navigraph weather tools at all.  Network pilots using sims with accurate weather were all operating happily in sync with the real weather at that airport, and then those folks in MSFS were trying to fit in as able, even though our conditions were often completely different.  In other words, everyone else was operating in conditions in sync with the real world... And then there was us.  😁

I don't personally see a problem with flying through a weather transition to get into the conditions that are known to exist in the vicinity of an airport.  It happens in real life all the time.  

Yeah think you've missed the point about METAR bubbles, they're not transitions, they're direct issues with the rushed integration of METAR in SU7 that is still yet to be fully implemented/finished/fixed.

Unfortunately it's a prime example of whereby a voting system doesn't work very well; votes were harvested, Asobo rushed the implementation to meet demand, which clearly needed more time to be worked on and we've had a half-baked "feature" ever since.

Go through the mountainous threads on the official forum since SU7, tons of examples to be seen there 🙂

[Tom_L 533]

1 hour ago, MarcG said:

Go through the mountainous threads on the official forum since SU7, tons of examples to be seen there 🙂

Showing one example here for convenience: Where might the airport be?🤔

Although to be fair, it wasn't that noticeable from the cockpit, but I've seen other examples of huge TCU's in a circle right above an airport in otherwise relatively fair surrounding weather. If you're flying airliners with Live weather, you'll come across these bubbles regularly, and oftentimes they are immersion breaking.

To the OP: Live weather is obviously under constant development, and it is improving. It has gained accuracy in terms of where and at which altitude clouds are displayed, but at the cost of lost variety and immersion - e.g. stratiform cloudscapes were stellar in the release version, now it's mostly the "puffy" cloudtypes in varying thickness, multilayered cloudscapes are reduced to mostly one layer only, thus creating towering cauliflower-type of clouds. It is obvious that METAR synthesis does not sit well with the weather engine, but they are trying. For the most part, I consider it reasonable. Had I not gotten used to the immersive Live weather in the release version, I probably wouldn't miss much.

 

Edited March 21, 2023 by Tom_L

[JRBarrett 3,079]

1 hour ago, MarcG said:

Yeah think you've missed the point about METAR bubbles, they're not transitions, they're direct issues with the rushed integration of METAR in SU7 that is still yet to be fully implemented/finished/fixed.

The problem with so-called “METAR bubbles” (which I rarely see) has nothing to do with how fast or slow Asobo has been about implementing METAR. It is caused when there is a significant disparity between the predicted clouds and visibility from the MeteoBlue model, vs. the actual clouds and visibility derived from current real-time observations. The MB forecasted conditions in many cases are from a model prediction that was made many hours before, while the METAR is usually no more than an hour old.

Several months ago LiveWeather went though a spate of problems where the server would get “stuck” providing MB model data that was hours or even days old - which was very evident by comparing the worldwide cloud patterns on the opening screen with a current real-time satellite photo. That would greatly exacerbate the “bubble” problem when (for instance) a low pressure system would be over New York City with low clouds and rain (reported by METARS from the NYC airports) while the MB model still had the storm system 300 miles west of NYC, back over Detroit.

I haven’t seen that particular issue for a long time.

The only solution would be to use forecast model data that is based on a much higher resolution grid, and is updated more often. That would lead to a much higher probability that the model predictions would more closely match the actual METARS (for clouds and visibility) more often. 

Such model data is available in the US - the HRRR (High Resolution Rapid Refresh) model, which is run every hour, and is model that local NWS offices use to create the TAF (terminal forecast) for airports in their jurisdiction. 

But, such models are not available everywhere, and MSFS has to cover the entire world. The MB NEMSGLOBAL model that MSFS uses is the only current weather model that has worldwide coverage and contains forecast data for each hour in a 24 hour period. The NOAA GFS model also has worldwide coverage, and a higher-resolution grid than NEMS, but only contains predictions at three-hour intervals - not hour by hour.

Things that could be done to supplement the accuracy of the MB model on a synoptic scale would include incorporation of r/t weather satellite data, and r/t NEXRAD weather radar data - at least in those places where it is available. (The US and Canada, Western Europe, Japan and Australia).

Edited March 21, 2023 by JRBarrett

[MarcG 2,993]

40 minutes ago, Tom_L said:

Showing one example here for convenience: Where might the airport be?

Depends where you look, I'm at work so can't log in right now to show you. I see the bubbles a lot (but certainly not all) of the time, personally I simply preferred live weather pre METAR integration but can understand why it was requested by users. I just wish Asobo took the time to implement it properly before rushing release and inserting catch up ever since, I live features to be near complete when released.

[Stearmandriver 2,297]

8 hours ago, JRBarrett said:

The problem with so-called “METAR bubbles” (which I rarely see) has nothing to do with how fast or slow Asobo has been about implementing METAR. It is caused when there is a significant disparity between the predicted clouds and visibility from the MeteoBlue model, vs. the actual clouds and visibility derived from current real-time observations.

Yup, this is what folks fail to understand: an obvious METAR bubble, if it's there, isn't showing you a problem with METAR integration - it's highlighting the problem with the Meteoblue weather.  This is the very problem that requires METAR integration to fix in the first place.

Like I say, even if it's a sharp transition, that's far better than having inaccurate weather at the airport, the place where it matters most.  Sharp transitions and "bubbles" do happen in real life too, but even if the implementation in a particular case were entirely unrealistic, it's still better than having terminal weather be completely wrong.

 

 

[fppilot 4,371]

On 3/20/2023 at 5:49 AM, mattloveday1987 said:

Hi All

Juat beginning the mass install of MSFS as I type… what is everyone’s view on the live weather? Is it accurate or are there any add ons that are better the default setup?

Thanks

Matt

I have had life weather be very good at times and absent or abhorrent at other times.  I keep REX Weather Force installed and ready.  And a key suggestion: pay little or no attention to MSFS ATIS/AWOS.  Along with ATC it is one of the most major flaws in MSFS.  Ever see humidity above 100%?  During the cooler or colder periods of the year ATIS reports dew points higher than temps.  Embarassing.  And at times ATIS reported winds are the winds aloft for the altitude I am at.  ATIS reported winds should always be surface winds.

Edited March 21, 2023 by fppilot