EA : rounded clouds horizon and weird haze

[Nemo 1,303]

Thanks @mpo910, I'll try that too.

Once again, the community with its talented and enthusiastic individuals shows its power. That's one of the outstanding side effects what makes our hobby so great.

Edited June 12, 2021 by Nemo

[mpo910 2,111]

44 minutes ago, Nemo said:

Thanks @mpo910, I'll try that too.

Once again, the community with its talented and enthusiastic individuals shows its power. That's one of the outstanding side effects what makes our hobby so great.

Thanks. You are right. But this is only possible at P3D! Because they allow us to do this (not hard coded but many cfg possibilities) .

But LM has to do some further enhancements!

Regards

Marcus

[Afterburner 373]

Just out of curiosity, I am currently 10km southeast of UUEE. I took a picture from my window, which is directed to the north. Then I fired up P3Dv4 (with ASP3D, ASCA, Envtex), MSFS using live weather, and placed my airplane at runway 24R and directed the camera to the northwest. Here are the screenshots taken at 21:23 local time:

Reality:

https://imgur.com/7DWoSix

P3Dv4:

https://imgur.com/cjbYjjS

MSFS:

https://imgur.com/znCp6cY

Could anyone using P3D 5.2 take a screenshot from the same spot at the above indicated time with EA enabled (preferably using ASP3D historical weather)?  I would like to provide a basis to compare sunset colors and coulds.

Edited June 12, 2021 by Afterburner

[B777ER 2,474]

The weather transitions using latest version of as3pd and EA with volumetric clouds on are just awful. I hope there is improvement somehow.by HiFi on this. Until then, it's EA lighting only with ASCA for me.

Edited June 12, 2021 by B777ER

[Doug47 574]

The yellow light at sunrise/sunset still hasnt been addressed either. It looks ridicules. I’m assuming, mostly likely correctly, that no one from LM has been outside? 

Edited June 12, 2021 by Doug47

[coastaldriver 915]

GSalden and mpo910 - Yes those mods are an improvement from the default config. Thanks. Interesting insights and yes those alterations to the config file are an improvement.

When you read the True Sky Developer information or SDK (LM would have these tools but not us) it is clear that there is a lot that can be done - however - LM cannot just dump this into the sim as released either - it is programmer level work because of its ramifications across the platform. I think the game producers are doing much better with this product but the games do not have the dynamic variables that real weather reproduction is going to throw up. I think LM have more work to do here seeing they have the tool kit not us!.The volumetric cloud slider I think is an example. The other issue is of course the lighting and sky colouring. This is a very hard one because AS and REX both use a shader texture but you have to choose a place or area for the light effect. The light you get in a Northern Latitude say Norway is not the light or sky you get in a Equatorial region or in a Southern Latitude and again it depends on the time of the year as well!. 

The sky or atmospheric colour and clarity is a function of latitude, time of year and atmospheric particles. After all the very fact we see a blue sky is a result of the way light and the black of outer space is seen from the earth's surface. Where you are different blue or hue of blue!. I think the current EA has captured this reasonably well and I get different results in shade and hue now in different locations. I move the sim around checking from a Southern latitude (The Falklands is a good one) to Mid latitude - the KRND default area and then to a tropical area. The outcome of sky colour is not the same. Again if a weather engine is being used the outcome will be different again. Look we even now get moon light and moon shadows replicated for full and bright moons - that is a significant advance and improvement,. And let not us forget the real wave motion etc (Alas only if you have an NVIDIA card) to see the waves breaking dynamically with a big fat thunderstorm at a Pacific tropical airfield (using the metar and weather P3DWX engine) the other day was simply amazing!

A point about weather engine and metar and airep interpolation. All weather is reported in a standard format half hourly across the globe. Forecasts are produced generally for a 12 hour or 24 hour period or if they are TTFs will be 3 hours. Now meterologist do not report all cloud nor do they make a distinction of the cloud itself. They do not report lesser amounts only the major or dominant amount going from bottom to top, so you may have 4 ocktas of cumulus at 3000 ft then say 2 ocktas of cumulus at 4500ft but the lesser amount will not be reported at all, it may be given in the forecast but not the metar. The lack of definition of the cirrus cloud is an issue for me but then Cirrus like all cloud (by definition " a visible aggregate of atmospheric moisture" ) but cirrrus is always ice particles not water vapour which is why it gets that very flat wispy look, the swirls are just the movement of air following a temperature and pressure gradient at high altitudes where it is is depends on where the tropopause is and this depends on whether your in an equatorial regions where it can be as high as F500 or mid latitude F400 or southern or northern latitude (F300 of less sometimes) all cloud formation stops at the tropopause which is why a thunderstorm top hits a limit and is flat.

I am just mentioning this so we can appreciate what we are asking a simulation of cloud to do! in other words it probably cannot but it can produce an approximation. As for the True Sky and cirrus we are not yet at a good approximation, not sure how to get that really flat wispy whoorl appearance (not yet anyway). 

 

 

 

[coastaldriver 915]

The following are a few shots of default airport and default fair weather loaded only with GSaldens mod implemented. All I am now doing is changing the volumetric cloud slider.

First Low setting

 

Second Medium Setting

 

 

[coastaldriver 915]

Now High

 

Last Ultra

The shift from low to medium is signficant more cloud changed amount and definition. Medium and High are pretty well the same but higher level cloud now begins to be more obvious and Ultra reall gives you a full sky.

[coastaldriver 915]

Now same scenario I have changed to real weather via the P3DWX engine. KRND is reporting vis 10+ and Clear. Looks the way it should to me 

 

[mikeymike 356]

7 hours ago, B777ER said:

The weather transitions using latest version of as3pd and EA with volumetric clouds on are just awful. I hope there is improvement somehow.by HiFi on this. Until then, it's EA lighting only with ASCA for me.

This is what I like but I am hating  the dark cockpit textures with qw 787.

mike

[zubillagajose 2]

My cfg

[Global]
render_grid_x_km_low = 0.1
render_grid_x_km_med = 0.1
render_grid_x_km_high = 0.2
render_grid_x_km_ultra = 0.2
render_grid_z_km_low = 1.3
render_grid_z_km_med = 1.3
render_grid_z_km_high = 1.3
render_grid_z_km_ultra = 0.2
cloud_resolution_low = 512
cloud_resolution_med = 768
cloud_resolution_high = 1024
cloud_resolution_ultra = 1024
max_cloud_distance_km_low = 200
max_cloud_distance_km_med = 200
max_cloud_distance_km_high = 200
max_cloud_distance_km_ultra = 100
cloud_shadow_strength = 0.5
cloud_shadow_range_km_low = 100.0
cloud_shadow_range_km_med = 175.0
cloud_shadow_range_km_high = 200.0
cloud_shadow_range_km_ultra = 200.0
cloud_shadow_texture_size_low = 1024
cloud_shadow_texture_size_med = 1024
cloud_shadow_texture_size_high = 1024
cloud_shadow_texture_size_ultra = 2048
crepuscular_ray_strength = 1.0
edge_noise_wavelength_km = 2.0
worley_texture_size = 256
cell_noise_wavelength_km = 2.0
max_fractal_amplitude_km = 2.0
cloud_threshold_distance_km = 0.5

[Cirrus]
max_density_gm3 = 0.15
distribution_base_layer = 0.05
distribution_transition = 0.95
worley_noise = 0.55
worley_scale = 1.0
diffusivity = 1.0
persistence = 0.63
octaves = 3.0
upper_density = 0.95
simulation = 0.0
cloud_width_km = 30
cloudiness = 0.15
base_noise_factor = 0.95
edge_worley_noise = 0.0
fractal_amplitude = 1.5
edge_sharpness = 0.05
churn = 1.2

[Stratus]
max_density_gm3 = 0.75
distribution_base_layer = 0.2
distribution_transition = 0.5
worley_noise = 1.5
worley_scale = 3.5
diffusivity = 18.0 
persistence = 4.0 
octaves = 1.0
upper_density = 1.0
simulation = 0.0
cloud_width_km = 100
cloudiness = 0.15
base_noise_factor = 0.1
edge_worley_noise = 0.1
fractal_amplitude = 5.0
edge_sharpness = 1.0
churn = 200.0

[Cumulus]
max_density_gm3 = 0.75
distribution_base_layer = 1.0
distribution_transition = 0.2
worley_noise = 0.6   
worley_scale = 3.0   
diffusivity = 5.0
persistence = 0.63
octaves = 1.0
upper_density = 1.0 
simulation = 0.0
cloud_width_km = 100
cloudiness = 0.15     
base_noise_factor = 0.3 
edge_worley_noise = 0.1 
fractal_amplitude = 4.5 
edge_sharpness = 1.0
churn = 1.0

[Cumulonimbus]
max_density_gm3 = 1.0
distribution_base_layer = 0.45
distribution_transition = 0.30
worley_noise = 0.6
worley_scale = 5.0
diffusivity = 20.0
persistence = 0.63 
octaves = 1.0
upper_density = 1.7
simulation = 0.0
cloud_width_km = 20
cloudiness = 0.15
base_noise_factor = 0.25
edge_worley_noise = 0.15
fractal_amplitude = 4.0
edge_sharpness = 1.0
churn = 1.0
 

[GSalden 4,650]

4 hours ago, zubillagajose said:

My cfg

[Global]
render_grid_x_km_low = 0.1
render_grid_x_km_med = 0.1
render_grid_x_km_high = 0.2
render_grid_x_km_ultra = 0.2
render_grid_z_km_low = 1.3
render_grid_z_km_med = 1.3
render_grid_z_km_high = 1.3
render_grid_z_km_ultra = 0.2
cloud_resolution_low = 512
cloud_resolution_med = 768
cloud_resolution_high = 1024
cloud_resolution_ultra = 1024
max_cloud_distance_km_low = 200
max_cloud_distance_km_med = 200
max_cloud_distance_km_high = 200
max_cloud_distance_km_ultra = 100
cloud_shadow_strength = 0.5
cloud_shadow_range_km_low = 100.0
cloud_shadow_range_km_med = 175.0
cloud_shadow_range_km_high = 200.0
cloud_shadow_range_km_ultra = 200.0
cloud_shadow_texture_size_low = 1024
cloud_shadow_texture_size_med = 1024
cloud_shadow_texture_size_high = 1024
cloud_shadow_texture_size_ultra = 2048
crepuscular_ray_strength = 1.0
edge_noise_wavelength_km = 2.0
worley_texture_size = 256
cell_noise_wavelength_km = 2.0
max_fractal_amplitude_km = 2.0
cloud_threshold_distance_km = 0.5

[Cirrus]
max_density_gm3 = 0.15
distribution_base_layer = 0.05
distribution_transition = 0.95
worley_noise = 0.55
worley_scale = 1.0
diffusivity = 1.0
persistence = 0.63
octaves = 3.0
upper_density = 0.95
simulation = 0.0
cloud_width_km = 30
cloudiness = 0.15
base_noise_factor = 0.95
edge_worley_noise = 0.0
fractal_amplitude = 1.5
edge_sharpness = 0.05
churn = 1.2

[Stratus]
max_density_gm3 = 0.75
distribution_base_layer = 0.2
distribution_transition = 0.5
worley_noise = 1.5
worley_scale = 3.5
diffusivity = 18.0 
persistence = 4.0 
octaves = 1.0
upper_density = 1.0
simulation = 0.0
cloud_width_km = 100
cloudiness = 0.15
base_noise_factor = 0.1
edge_worley_noise = 0.1
fractal_amplitude = 5.0
edge_sharpness = 1.0
churn = 200.0

[Cumulus]
max_density_gm3 = 0.75
distribution_base_layer = 1.0
distribution_transition = 0.2
worley_noise = 0.6   
worley_scale = 3.0   
diffusivity = 5.0
persistence = 0.63
octaves = 1.0
upper_density = 1.0 
simulation = 0.0
cloud_width_km = 100
cloudiness = 0.15     
base_noise_factor = 0.3 
edge_worley_noise = 0.1 
fractal_amplitude = 4.5 
edge_sharpness = 1.0
churn = 1.0

[Cumulonimbus]
max_density_gm3 = 1.0
distribution_base_layer = 0.45
distribution_transition = 0.30
worley_noise = 0.6
worley_scale = 5.0
diffusivity = 20.0
persistence = 0.63 
octaves = 1.0
upper_density = 1.7
simulation = 0.0
cloud_width_km = 20
cloudiness = 0.15
base_noise_factor = 0.25
edge_worley_noise = 0.15
fractal_amplitude = 4.0
edge_sharpness = 1.0
churn = 1.0
 

Screenshots ?

Start with Fair Weather and change top to 12500. Set density to OC.

Take a screenshot of that and post it here.

Btw : stratus Churn 200 ?

Edited June 13, 2021 by GSalden

[coastaldriver 915]

Interesting outcome of test or check between real world sky/weather vis sim and true sky. I did a comparion test not just changing the weather in the sim to see what it looked like. I wanted to know how did the sim view or representation compare with the the real world. Using skycams for Australia I compared the cam shots (allegedly current) with the sim changing the location but keeping the time precise real time weather was provided by P3DWX (Downloading via NOAA and WMO). 

See the comparison shots here at this thread:   Clouds/Sky Comparison Reality vs. Simulators - The Prepar3d Forum - The AVSIM Community

Interesting - in only one location did the skycam match the metar report - why they are so different is uncertain except to say metars are only generated hourly or half hourly dependant on the airport category (airline or GA) so the metars are out of date automatically with real time views. Thee is also the issue of locations where they use automated equipment - there no cloud can be reported as the met people have yet to come up with a machine that will do that so you get NCD or no cloud discernible which of course is useless. Also in one location the metar did not match the weather on the skycam at all - Alice Springs was reporting OVC at F100 but you can see from the skycam it is not OVC at all but BKN at best in my intepreation is SCT cloud not OVC or BKN. Merimbula in NSW was probably the most accurate match. 

I guess it comes down to a large degree to the weather engine (AS or REX or P3DWX) and how P3D then interprets or reinterprets this data to produce a sky picture!. How this is done I do not know but I guess it is time to find out!. 

Oh and beware of CAVOK this implies no cloud below 5000 ft and vis greater than 10 km but you could have 4/8 cloud to the north on the hills above 5000ft but if CAVOK the sim will show no cloud and clear skies. 

So really the key now for me is not True Sky but which weather engine works the best or how do they work with the sim to produce weather. Another topic all on its own. 

 

Edited June 14, 2021 by coastaldriver

[SAS443 1,703]

6 hours ago, coastaldriver said:

Oh and beware of CAVOK this implies no cloud below 5000 ft and vis greater than 10 km but you could have 4/8 cloud to the north on the hills above 5000ft but if CAVOK the sim will show no cloud and clear skies

Indeed. 

Snapped this image last week. This is what CAVOK can look like. (Simulator would depict sky clear as you said)

Metreport: 21010KT CAVOK 22/14 Q1011

[zubillagajose 2]

nice

Edited June 14, 2021 by zubillagajose