2 hours ago, MrBitstFlyer said:

Rex uses a global weather model and some simmers don't like the transitions between areas.

Rex doesn’t use any kind of “model”. It takes the METAR closest to the aircraft’s position and sets the weather globally to those conditions. It does seem capable of a greater variety of cloud depictions, but it is using the default MSFS cloud generation engine to do it.

Edited August 17, 20223 yr by JRBarrett

1 hour ago, JRBarrett said:

Rex doesn’t use any kind of “model”. It takes the METAR closest to the aircraft’s position and sets the weather globally to those conditions. It does seem capable of a greater variety of cloud depictions, but it is using the default MSFS cloud generation engine to do it.

Rex is able to use the cloud variety available in presets, to populate live weather.  They have no access to a weather SDK so the weather is global in nature and needs to transition between areas - a clever solution.  I use the term weather 'model' to explain the process they have used.

REX does not use presets to generate weather within the simulator.  We created our own Weather SDK to access key aspects of the weather system.

This allows us to generate and transition weather within the simulator. We accomplish this two ways:

1.) Via real weather provided by Metar data nearest the aircraft. That we can provide the most accurate and live weather possible within the simulator. We transition this real weather during flight and provide various means to transition instantly or over a period of time.

2.) The second way is if you choose to utilize dynamic weather presets - which are not the same as the presets within the simulator. These are groupings of weather scenarios that provides dynamic, transitioning, changing weather during flight. For example, if you select Approaching Cold Front in Weather Force you will experience the effects of an approaching cold front over time as you fly.

For more information here is our website dedicated to the product and its features:

http://rexsimulations.com/weatherforce.html
 

1 hour ago, rstough said:

REX does not use presets to generate weather within the simulator.  We created our own Weather SDK to access key aspects of the weather system.

Thanks for explaining how its done 🙂  Great Weather addon!

6 hours ago, rstough said:

REX does not use presets to generate weather within the simulator.  We created our own Weather SDK to access key aspects of the weather system.

This allows us to generate and transition weather within the simulator. We accomplish this two ways:

1.) Via real weather provided by Metar data nearest the aircraft. That we can provide the most accurate and live weather possible within the simulator. We transition this real weather during flight and provide various means to transition instantly or over a period of time.

2.) The second way is if you choose to utilize dynamic weather presets - which are not the same as the presets within the simulator. These are groupings of weather scenarios that provides dynamic, transitioning, changing weather during flight. For example, if you select Approaching Cold Front in Weather Force you will experience the effects of an approaching cold front over time as you fly.

For more information here is our website dedicated to the product and its features:

http://rexsimulations.com/weatherforce.html
 

Very nice. What happens in METAR-sparse areas, by the way? Thinking of Africa specifically now. Is it METAR interpolation or do you base the "uncovered" weather off of precipitation/satellite maps at all?

Edit: By precipitation/satellite maps I'm referring to a combination of these (from Windy, in this case):

(Thunderstorm data)

(Cloud bases)

(Cloud tops)

(Rain/thunder)

 

And so on. I'm not sure how feasible it might be to query external weather data and form a weather "environment" based on the above data, though.

 

Edited August 18, 20223 yr by aktorsyl

These are examples of Live Weather from 'before' and whatever they have to do to get back to this, I'd like them to do it.

 

 

 

 

On 8/17/2022 at 8:45 PM, SubtotalGuide said:

Don't worry, I won't be bothering with this any more either. Not worth putting up with a product that keeps getting worse waiting for better times with "the crew" happily defending the sacrifices we are forced to make.

Moving on to more pleasant and productive things in life.

https://www.twitch.tv/videos/1565642019?t=00h37m06s

 

I'm a meteorologist and GA pilot. Cloud depiction has definitely deteriorated since SU7... absolutely no doubt. We had incredible cloud layers and types which were incredibly realistic. Now we don't... comparatively.

I still can't believe that proper depiction of atmospheric visibility (aerosol density) is not a higher priority for people. Live weather mode still does not utilise any aerosol data from meteoblue. The result is an atmosphere where visibility throughout the atmosphere never changes. It's too clean and clear. 

Edited August 19, 20223 yr by Falconjet112

47 minutes ago, Falconjet112 said:

I'm a meteorologist and GA pilot. Cloud depiction has definitely deteriorated since SU7... absolutely no doubt. We had incredible cloud layers and types which were incredibly realistic. Now we don't...simple. 

I still can't believe that proper depiction of atmospheric visibility (aerosol density) is not a higher priority for people. Live weather mode still does not utilise any aerosol data from meteoblue. The result is an atmosphere where visibility throughout the atmosphere never changes. It's too clean and clear. 

MetOffice too here, active glider pilot since 42 years, and I have to disagree....

Cloud variety was never, IMO, a big plus since release, and I don't actually recall seeing it depicted better in the past. I do recall the "activity" inside of clouds, as well as other "features" which were actually more bugs than features, like air density calculation when within a cloud ( meanwhile fixes in one of the SUs )...

With SU10 beta things have gotten a bit better in as far as cloud depiction goes, and from the link I posted above it looks like their algorihms for cloud depiction are using METAR ( as they should near the corresponding stations ) and also some other models even appart from meteoblue's GRIB / GRIB2.

In as far as visibility goes, I was a great critic of the absence of a proper way of setting visibility based on observation reports. Yet, from SU to SU I have been noticing changes in the way visibility actually get's calculated when reported lower than CAVOK or 9999 in an observation, and when it is really nearing CAT I, II, III limits. Truth is the simulator does a very acceptable job in reproducing it.

I wonder if they changed their use of the aerosol density variables when they introduced the blend between METAR and GRIB / GRIB2 data depiction ?

Anyway, it's a dynamic process, it is surely NOT STAGNATED, and they're tackling so many fronts that I really have to respect their dynamism and will to do better.

Is it Perfect ?  No, of course far from that...  Is it good enough - surely, specially taking into consideration it's not closed ( quite on the contrary ) from further refinement...

 

P.S.: I don't want to place huge ecxpectaitons on their upcoming "soaring weather" in order not to get disappointed... I use MFS primarilly for soaring task preparation / debrief, because of the astounding secenery in the areas I usually fly over IRL. Gliders, even the best, perform far from realistically in MFS ( as a matter of fact also in X-Plane 11 ), and I would love to see them feel more like in Silentwings or even Condorsoaring. Soaring weather already got some first representations with SU9, and now SU10 beta, and the above linked portion of ther latest twitch suggests they're fine tunning it, but I honestly don't expect it to get to the level of Silentwings, or Condorsoaring, although I am sure that it can get to a level that makes it viable to be used to play a bit with what soaring flight is...

Edited August 19, 20223 yr by jcomm

16 hours ago, jcomm said:

https://www.twitch.tv/videos/1565642019?t=00h37m06s

What I heard from the Q&A:

• Xbox's memory limitations have been the primary reason for degradation in cloud depiction IQ at least that was my interpretation of some vague commentary.   No worries here because Xbox is headed for more memory in future releases.  In Aprill '22 the development Xbox Series X devkit revealed 40Gb of faster RAM--so the article said.  So for our "10 year" product service life I think there is a strong likelihood of being able to exploit more RAM for better cloud rendering going forward.
• WU XI is Canada!  Yeah I'm stoked about this!
• De-encrypting OEM planes to open them up to 3rd party enhancement.
• Helicopter support

I am reminded again of just how huge the project is, really compared to just about any other title I can think of.  

17 hours ago, jcomm said:

https://www.twitch.tv/videos/1565642019?t=00h37m06s

Here's what I heard from the Q&A:

Seb: "I can talk a little bit about it, even though I don't know how the entire system works"

That's all I needed to hear. And again, the question I have since SU7 immediately popped up: where the hell did the guy go who knew? Because if there's anyone left who knows how to master said systems - why not ask him when you know you gonna have to answer that particular question?

 

26 minutes ago, Tom_L said:

Seb: "I can talk a little bit about it, even though I don't know how the entire system works"

Lol, I also noticed that, but I wouldn't read too much into it. Similarly there was a lot of questions that Martial couldn't answer. 

39 minutes ago, Tom_L said:

Seb: "I can talk a little bit about it, even though I don't know how the entire system works"

I took that as Seb talking about the Meteoblue side of things, and not knowing all the intricate details of how their system works.

I also noticed that, but my interpretation was the same as that of lwt1971 - it's not Seb's beach....

You'de probably get the same type of answer if you asked about ATC or AI traffic ( ? )...

But even  not being his speciality he did mention a few important details regarding the migration from the initial forecast-only weather injection to the one used after one of the SUs where METAR data was introduced and the blending algorithm was first used.

Later he also mentioned that clouds take yet another approach.

I also noticed that, but my interpretation was the same as that of lwt1971 - it's not Seb's beach....

You'de probably get the same type of answer if you asked about ATC or AI traffic ( ? )...

But even  not being his speciality he did mention a few important details regarding the migration from the initial forecast-only weather injection to the one used after one of the SUs where METAR data was introduced and the blending algorithm was first used.

Later he also mentioned that clouds take yet another approach.

At least he confirmed more than once that Meteoblue feeds are still being used, contrarilly to what some suggested here at the forums sometime ago.

Edited August 19, 20223 yr by jcomm