33 minutes ago, bogdansrb said:

As a comparison, clouds in MSFS have much better definition without affecting performance that much (in fact I get better performance there).

Even MSFS users have said MSFS clouds look like a volcanic eruption.  But let's avoid the comparisons, if possible.  Better definition doesn't mean better looking.  

33 minutes ago, bogdansrb said:

Every area that is covered with snow this time of year shows only patches of snow in the sim

Not true.  Depending on the METAR, snow cover can be sparse, moderate or full cover.  I've seen cases for all.

33 minutes ago, bogdansrb said:

I read that x-plane shows snow on the ground only if it's snowing at that moment, not overall snow coverage.

I would like to know where you read this.  The screenshot I posted showed snow, but snowfall/precipitation was set to zero.  Here's another one. No rain.  Full snow cover. 

27 minutes ago, bogdansrb said:

I don't really know how to explain this better. You're flying and it suddenly starts raining

And this is not something you are aware can happen when flying from one area to another?  In my flying days, this happened many times.  It also happens when I'm driving.

This can also happen in the sim because of a new METAR that has just been downloaded.  The weather can't change dynamically as it's downloading.  It has to download first, then needs to be executed.  

27 minutes ago, bogdansrb said:

I've also seen that it almost always rains inside the clouds even if it doesn't on the ground. 

This is 100% accurate.  Clouds contain moisture.  Obviously it depends on the type of clouds and the amount of moisture, but it's the air pressure that keeps the moisture inside the clouds.  When the rain drops are too heavy for the air pressure to keep them in the clouds, they fall.  So this is absolutely correct behavior, and if LR have done it this way, then it's a big step in the right direction.

Edited December 30, 20223 yr by GoranM

Quote

Even MSFS users have said MSFS clouds look like a volcanic eruption.  But let's avoid the comparisons, if possible. 

What does this have to do with what I said?

Quote

Better definition doesn't mean better looking. 

It means it's more realistic looking

Quote

Not true.  Depending on the METAR, snow cover can be sparse, moderate or full cover.  I've seen cases for all.

I'll investigate more on this. But so far I've flown in areas like Canada or Iceland and have seen sparse snow coverage.

Quote

And this is not something you are aware can happen when flying from one area to another? 

It seems to happen too often and sometimes randomly.

Quote

This can also happen in the sim because of a new METAR that has just been downloaded.

I thought this was an XP11 thing and will be fixed in XP12.

4 minutes ago, bogdansrb said:

realistic looking

who did when. 

8 minutes ago, bogdansrb said:

XP11 thing and will be fixed in XP12.

XP12 weather runs on what the weather was, what the weather will be and extrapolates between the two.

XP11 had only instanious weather with no changes and no facility to change it.

 

 

Weather in XP12 is a considerable step fwd from previous versions but in order for users to better understand how it is being done now LR should spend some time writing about it or producing one of those videos where someone from LR explains how this or that is done.

A week ago maybe, there was a very good thread at the .org XP12 Weather forum where the discrepancies between what was being depicted in XP12 with Real World Weather and "reality" or what other tools depicted based on forecasts. It looks like there was some problem either with their servers or the way forecasts downloaded from NOAA were getting used to build the weather for XP12's World. It appears to have been fixed but I can't tell for sure because I haven't been using XP12 lately.

The weather in Xp12 is based on observation ( METAR ) and forecast ( GRIB data from the NOAA GFS model ). This is exactly what Active Sky has been doing since ages for FSX, P3D and also in some way XP11, and what MFS does. MFS uses Meteoblue as it's feed, and it has 60 levels of data. XP12 uses 10 levels and goes up only to FL380 if I'm not mistaken, and from there up extrapolates based on the (wrong) assumption that it can be done successfully using ISA lapse rates. This creates discrepancies in temperature and pressure levels above that FL, and most airliners and other aircraft cruise at or above that level. The rest of the troposphere down to ground level is done based on 10 levels though, and interpolated in between.

When fetching model data to depict weather in a sim based on forecast the algorithm must be as coherent as possible. You will want to pick the closest run of the source model (NOAA in this case), and then from that run and taking into consideration the date-UTC of the flight, fetch data from the GRIB the closest possible to flight time interval.

If the flight starts and has a duration that is well within the time slice covered by the fetched forecast GRIBs then all can be done with those GRIB, otherwise if the flight takes place close to a new run of the model or is long enough to cross into it's time of reference, new GRIB data should be fetched and the time periods for the flight being done have to be used in the depiction cycle. 

In principle and in order to be able to create a time variation and not only a space variation, as your flight progresses from A to B the data regarding the forecast ahead of the current hour should be taken into consideration in order to start building a dynamic evolution of the weather based on space and time.

I don't really know how XP12 is doing it. My flights which were mostly performed before that quirk described in that .Org thread appeared to be acceptably close to what I expected, less those performed above  FL 380... Would be great to have some insight from LR.

In MFS the depiction, variation and integration with observation and even SIGMET data has been fine tuned some SU's away and is now very good and smooth.

For me the biggest problems now are the way turbulence is pushed into the scene, both it's form and it's intensity, way overdone most of the time and also not taking into consideration the type of turbulence - it's all the same, less wake turb.  And also the total absence of cloud depiction at night. Clouds don't exist at night in XP12 😕

 

 

Edited December 30, 20223 yr by cagarini

17 minutes ago, cagarini said:

being done now LR should spend sometime writing about it or one of those videos where someone from LR explains how this or that is done.

good point

 

Yes mSparks, but both are still superficial regarding the details of how the GRIB data and observation are actually dynamically updated... 

I did love Autin's "psychedelic" video :-), but would like to have further insight into how GRIB data and runs are used during a flight simulator session being flow in real time and with real weather...

couple of things i would add.

I like setting the weather via the instructor operator station (something like flight->show ios), 

there is also in the manual weather control the option to set how the weather changes, e.g. set clear and becoming worse or vice versa.

travelling for new year atm so cant grab screenshots.

1 hour ago, bogdansrb said:

It seems to happen too often and sometimes randomly.

 

I'm with you on this one, I think it's something to do with the Change Over Time weather setting (can't remember the exact setting name). It regularly happens to me even if it's blue skies above and immediately around me, at least I don't see any pattern to why it occurs (i.e. It's not when flying through or under clouds for me). 

About real weather which is the topic of this discussion, I loaded up a vanilla XP12 at a famous altiport, LFLJ which is covered in snow currently as can be seen from their live webcam.
XP set to live weather, depicts a scenario without any snow at all. Is this to be excepted because the METAR does not report snow on terrain? 

 

15 minutes ago, peroni said:

About real weather which is the topic of this discussion, I loaded up a vanilla XP12 at a famous altiport, LFLJ which is covered in snow currently as can be seen from their live webcam.
XP set to live weather, depicts a scenario without any snow at all. Is this to be excepted because the METAR does not report snow on terrain? 

 

That question has already been answered.

Edited December 30, 20223 yr by GoranM

Maybe it has been answered by some users I have on my blocked list because I cannot see a clear answer to this straightforward question.

Edited December 30, 20223 yr by peroni

9 minutes ago, peroni said:

Maybe it has been answered by some users I have on my blocked list because I cannot see a clear answer to this straightforward question.

No.  It was answered and you replied.

Unfortunately, some questions don't have definitive answers at the time they are asked.  It may be straightforward, but there is still no proper answer, yet.

Oh right, so this is one of the things are that is not implemented at the moment but will come with the big weather update some time in the future.

Cool, thanks
 

10 minutes ago, peroni said:

Oh right, so this is one of the things are that is not implemented at the moment but will come with the big weather update some time in the future.

Cool, thanks
 

Exactly (sort of).  Glad I could help.

Edited December 30, 20223 yr by GoranM

Questions relating to DVDs should not have been posted in this discussion. Stick to the subject please - the weather engine.