Active Sky faced difficulties in bringing the meteorological

[Krakin 5,726]

6 hours ago, MrBitstFlyer said:

Cirrus clouds have been missing from both sims, but a clever addon developer has managed to create pretty good cirrus/stratus type clouds for XP12.  No idea if he will do something for MSFS, but now somebody has proved it can be done, I don't suppose it will be long until better cirrus depiction makes an appearence in MSFS.

Screenshot is from XP12 with a weather plugin.

That does NOT look good

[Ricardo41 5,935]

8 hours ago, MrBitstFlyer said:

 

Looks terrible. 

[sanh 823]

I think MSFS are not focused on a weather API because it will be a lot of work and they are focused on adding more api/sdk features for scenery and aircraft designers.

I think the vast majority of MSFS simmers are happy enough with the weather so they probably don't have it down as an area to focus on. 

I think that's fair enough. That being said - Active Sky is the best 3rd party weather addon ever made for any flightsim and I do wonder what they could produce with a fully featured weather api.

[Ricardo41 5,935]

I fail to grasp why people won't let this go and move on. Again, is there any evidence that Microsoft plans to open up the weather to third party developers? I'm not seeing it. 

Still, people cling to this hope like Leonardo di Caprio holding on for dear life to Kate Winslet's life boat. 

Maybe it will happen, maybe it won't.

So why not move on and develop other addons? 

Look at the developers of Chaseplane. I'm sure they spent no more than a New York minute lamenting the closed camera system - they moved on and developed Flow - a brilliant addon that for me, in the professional version, has become part of every flight. 

With regards to MSFS 2020, maybe it's time for Active Sky to pivot?

[Daube 348]

1 hour ago, Ricardo41 said:

So why not move on and develop other addons? 

You should do that and let us continue to discuss, maybe.

[KERNEL32 493]

4 hours ago, Ricardo41 said:

Looks terrible. 

The "chunk" of cumulus blob in top right looks bad, yes (out of place) but the rest as far as the high clouds on left and toward the horizon to me look plausable and much more varied than what MSFS gives us now. 

I always wondered, and not sure of the technicalities behind it, but why couldn't they "trick" the system and use "airborne scenery" that is thin like cirrus but still can keep the "crispness" and have the scenery draw as either opaque or transparent depending on the cirrus type, and turn off collision of said "scenery". One thing with cirrus is that unlike cumulus and other types of clouds, cirrus tend to keep their "shape" much longer so wonder if this could work as a workaround?

Edited March 11, 2023 by KERNEL32

[GoranM 3,762]

12 hours ago, MrBitstFlyer said:

Cirrus clouds have been missing from both sims, but a clever addon developer has managed to create pretty good cirrus/stratus type clouds for XP12.  No idea if he will do something for MSFS, but now somebody has proved it can be done, I don't suppose it will be long until better cirrus depiction makes an appearence in MSFS.

Screenshot is from XP12 with a weather plugin.

The fact this is the first time cirrus clouds have been created for a commercial flight sim (X-Plane), due to an open SDK, and this is the first iteration of said cirrus clouds, while also comparing them to real cirrus clouds, these look great!  Everything has to start somewhere, and they can only improve with every version or update.

This could be a great start for things to come for MSFS.  So to those people bashing the image, and the work the developer has done, just reel it in a bit, and appreciate the fact this is literally V1.0 of new clouds that we haven't seen before.  Give it time, and try to be a little more positive with some constructive criticism thrown in, and maybe the developer will be encouraged to continue working on improvements.  And then follow it up with cirrus clouds for MSFS.  

Edited March 11, 2023 by GoranM

[F737MAX 5,492]

6 minutes ago, KERNEL32 said:

[...] but the rest as far as the high clouds on left and toward the horizon to me look plausable and much more varied than what MSFS gives us now. 

I always wondered, and not sure of the technicalities behind it, but why couldn't they "trick" the system and use "airborne scenery" that is thin like cirrus but still can keep the "crispness" and have the scenery draw as either opaque or transparent depending on the cirrus type, and turn off collision of said "scenery". One thing with cirrus is that unlike cumulus and other types of clouds, cirrus tend to keep their "shape" much longer so wonder if this could work as a workaround?

The issues with cirrus clouds as a 'scenery' are that like landclass textures for ground scenery, textures may abruptly stop at the boundaries and are incongruous with their surrounds. Also, unlike other cloud layers, there would be no dynamic quality to cirrus if it were implemented as such. Additionally, given how cirrus can start as low as 20,000 feet A.S.L., passing through the 'scenery' layer may appear odd.

All that being said, the middle third of that earlier screenshot showing high level clouds is far more realistic than Live Weather currently is in MSFS.
A 'scenery layer' of cirrus clouds may be an acceptable temporary solution for anyone who would like to use it until better design or stronger performing hardware allows MS/Asobo to replace it.

[Noel 4,483]

1 hour ago, KERNEL32 said:

I always wondered, and not sure of the technicalities behind it, but why couldn't they "trick" the system and use "airborne scenery" that is thin like cirrus but still can keep the "crispness" and have the scenery draw as either opaque or transparent depending on the cirrus type, and turn off collision of said "scenery". One thing with cirrus is that unlike cumulus and other types of clouds, cirrus tend to keep their "shape" much longer so wonder if this could work as a workaround?

From HiFi:

On 3/9/2023 at 11:22 PM, Damian Clark said:

A hybrid approach where in the near-range distances a 2d rotating billboard sprite of a volumetric cloud structure (ala http://niniane.org/clouds/ from FS9) is utilized, synchronized with and complementary to the longer range clouds, is one way things can improve, increasing near resolution while keeping performance and full 3d volumetric appearance.  We dabbled in this a bit a few years ago with mixed results that did not make it to market. 

[Tom_L 518]

Due to the current state of cloud depiction in Live weather I endorse the opening of the weather system for third parties.

However, the discussion should not omit the fact that - cirrus clouds aside - the cloud depiction from release until SU5 (changes in cloud rendering and lighting, SU7 - METAR integration - initially killed it entirely) was superior to all I have seen before (see screenshot).

It lacked accuracy in terms of cloud height, visibilty and overall correspondance to real world METAR reports, but it was on the other hand largely plausible. You could even see convective clouds forming leewards over the islands in the caribbean, and they managed stratiform cloudscapes quite well. I fear that the "ingestion" (I put it in parenthesis because I have no clue about the actual process) of METAR data is somewhat incompatible with the overall weather simulation process. And it seems obvious to me that it takes the developers quite some time to get the hang of it.  And that's where the experience of the "known suspects" could probably make a difference, as they always had to "blend" local weather areas into a greater picture.

 

[KERNEL32 493]

DCS is seeming to get it right, and that sim has voxel based clouds too correct? The clouds seem "more sure of themselves" like they want to be clouds (not sure of how else to describe MSFS clouds). There are thin layers and regular stratus clouds too, and cumulus. If MSFS could come up to this quality and variety (of course only if the conditions call for it) I would be very happy.

 

[Lucky38i 1,094]

Just now, KERNEL32 said:

DCS is seeming to get it right, and that sim has voxel based clouds too correct? The clouds seem "more sure of themselves" like they want to be clouds (not sure of how else to describe MSFS clouds). There are thin layers and regular stratus clouds too, and cumulus. If MSFS could come up to this quality and variety (of course only if the conditions call for it) I would be very happy.

DCS has amazing clouds, problem is they're not a natural result of a culmination of different conditions required to form clouds like MSFS's weather engine is. Not to mention it's mostly a static weather with no kind of dynamic changes. I know they're working on dynamic weather but I don't think it'll compare to an actual weather engine.

If you've used weather presets you can see the MSFS is capable of a larger variety of cloud types, the inherent issue is more with weather generation derived from meteoblue data. I'm speculating here but, Asobo probably hasn't further refined the I/O stream of meteoblue data to weather engine outputs to give us "more" since probably SU5. Most of the SUs since SU5 have been focused on the sim itself, hopefully after SU12 they'll dedicate more time to the weather engine.

[KERNEL32 493]

21 minutes ago, Tom_L said:

However, the discussion should not omit the fact that - cirrus clouds aside - the cloud depiction from release until SU5 (changes in cloud rendering and lighting, SU7 - METAR integration - initially killed it entirely) was superior to all I have seen before (see screenshot).

It lacked accuracy in terms of cloud height, visibilty and overall correspondance to real world METAR reports, but it was on the other hand largely plausible. You could even see convective clouds forming leewards over the islands in the caribbean, and they managed stratiform cloudscapes quite well. I fear that the "ingestion" (I put it in parenthesis because I have no clue about the actual process) of METAR data is somewhat incompatible with the overall weather simulation process. And it seems obvious to me that it takes the developers quite some time to get the hang of it.  And that's where the experience of the "known suspects" could probably make a difference, as they always had to "blend" local weather areas into a greater picture.

Absolutely 100% agree! And although many don't like when this is said, the devs obviously tuned it way back after Su5 to accommodate for the x-box version of MSFS. This is the very reason why you never heard anything after this hype from Jorg below in the Dev Q&A back in June of 2021 - the x-box has "capped" the capabilities of improving any graphic elements (including clouds) in MSFS (If they improve PC but NOT x-box then xbox sales go down, so they are kind of stuck). HOPEFULLY I am very wrong, and please - I hope they prove me wrong 😉

 

 

 

Edited March 11, 2023 by KERNEL32

[Lucky38i 1,094]

3 minutes ago, KERNEL32 said:

Absolutely 100% agree! And although many don't like when this is said, the devs obviously tuned it way back after Su5 to accommodate for the x-box version of MSFS. This is the very reason why you never heard anything after this hype from Jorg below in the Dev Q&A back in June of 2021 - the x-box has "capped" the capabilities of improving any graphic elements (including clouds) in MSFS. HOPEFULLY I am very wrong, and please - I hope they prove me wrong 😉

The problem with this statement, is that it ignores the differences between MSFS on Xbox and PC, and insinuates both variants are exactly the same. There's various graphical elements on PC not on Xbox, for example DLSS and FSR. Then there's the issue where it's practically impossible to run non-wasm/js (and atm WASM doesn't even work on Xbox) aircraft on xbox. To insinuate PC users are capped by Xbox users is ludicrous considering what PC users have access to that Xbox doesn't.

Does the weather need working, 110%, is MSFS forever capped, in terms of fidelity, because it supports Xbox? No, and there's no basis to suggest so. I mean just look at numerous ports of console games coming to PC and suddenly all the various technologies the PC variants use.

[Tom_L 518]

11 minutes ago, KERNEL32 said:

This is the very reason why you never heard anything after this hype from Jorg below in the Dev Q&A back in June of 2021

Silent as the grave ever since!

Since many of the weather related threads on the MS forums (go vote/participate!) have been tagged "bug logged" or "feedback logged" since the beginning of the year I have some hope left for a "WAU" (Weather and Atmosphere Update) one day. Would make sense after "Sim-", "World-" and "Aircraft-" updates.

Edited March 11, 2023 by Tom_L