Jump to content

[1.1.2][1-1-2] May 13-2016 EnvironmentalVisualEnhancements

rbray89

By rbray89
in KSP1 Mod Releases

 Share

Recommended Posts

Poodmund

Poodmund

Posted
Posted (edited)
3 minutes ago, pingopete said:

Ah man! The possibilities in RSS, I new it was possible with EVE but guess I had to see it in action to grasp its true awesomeness. Axially tilted auroras on planets like Uranus, Jupiters giant storm... This has me wanting to remake all of Earths clouds dammit! :D 

I try to figure out this kind of stuff so that the likes of yourself, Proot et al can make stuff that blows my mind. :D

EDIT: I've added a link to the post into my signature so its easy to find for later. :wink:

Edited by Poodmund
Link to comment
Share on other sites
pingopete

pingopete

Posted
Posted (edited)
Just now, Poodmund said:

I try to figure out this kind of stuff so that the likes of yourself, Proot et al can make stuff that blows my mind. :D

Well I do have an old 8k hurricane punched out with transparency lying around somewhere... :) But I'm thinking now like multiple hurricanes with several cloud layers move about and faded with each other like the Jupiter clouds could give earth a really fluid looking weather system

EDIT: to somewhat replicate this kind of motion

 

Edited by pingopete
Link to comment
Share on other sites
DasBananenbrot

DasBananenbrot

Posted
Posted
20 hours ago, Poodmund said:

@rbray89 I got it sorted.

Animation25a7a8.gif

Basically had to use the following:

  • Texture on the Yp or Yn cube face
  • Speed: 0,0,x (where x is speed of rotation of storm)
  • Offset: x,y,0 (where x is Latitude range 0 to 180 and y is Longitude range -180 to 180)
  • rotationAxis0: 1,0,0
  • rotationAxis1: 0,0,1
  • rotationAxis2: 0,1,0

Basically what I had been thinking after my original post and yours just with the speed modifier being the z value. Now proof of concept it sorted, just have to make it look good now. :D

Thanks for your help.

Nice pioneering there. Amazing what you can do with EVE.

1 hour ago, pingopete said:

Well I do have an old 8k hurricane punched out with transparency lying around somewhere... :) But I'm thinking now like multiple hurricanes with several cloud layers move about and faded with each other like the Jupiter clouds could give earth a really fluid looking weather system

EDIT: to somewhat replicate this kind of motion

 

I'm pretty sure not long in the future you and the likes, who are doing real modding magic, will make something like this possible. You only have to look what you have achieved so far to know, you can and will make this a reality :D 

Link to comment
Share on other sites
Poodmund

Poodmund

Posted
Posted (edited)

@rbray89 is it possible for the cube map system to be told not to render an image to a cube face? At the moment if you dont specify a texture...

  • Xp: 
  • Xn:
  • Yp: texturePath
  • Yn:
  • Zp: texturePath
  • Zn:

... for example, iit all falls down and weirdly when you try to save either a fully black .dds or a transparant png to act as an empty dummy texture to assign to the faces you want to be blank, EVE renders them as a solid colour as seen below:

qeoEr2Cl.png

So yeah, for whatever reason if the alpha map is blank i.e. all black, it seems to render an flat colour to it rather than being fully transparent.

EDIT: In the screenshot shown it was using a cube map on the Blue channel with the X and Z faces using the "blank" texture.

Edited by Poodmund
Link to comment
Share on other sites
Proot

Proot

Posted
Posted
Just now, Poodmund said:

@rbray89 is it possible for the cube map system to be told not to render an image to a cube face? At the moment if you dont specify a texture it all falls down and weirdly when you try to save either a fully black .dds or a transparant png to act as an empty dummy texture to assign to the faces you want to be blank, EVE renders them as a solid colour as seen below:

qeoEr2Cl.png

So yeah, for whatever reason if the alpha map is blank i.e. all black, it seems to render an flat colour to it rather than being fully transparent.

I'm using a 2x2 px empty texture for that. Seems to work nice.

Link to comment
Share on other sites
Rainbowd4sh

Rainbowd4sh

Posted
Posted

How exactly do you install? The instructions state to simply put the two zips into the game data folder, yet there is one zip,

Sorry if I am just being plain stupid, but I can't figure it out. I spent a good hour doing different things and nothing works. 

 

Link to comment
Share on other sites
Poodmund

Poodmund

Posted
Posted

Ah I think I may have found the cause. In the Texture Configs, if you specify to use the same file for two different cube maps, if doesn't render the ones higher on the entry list.

For example, the config used for the screenshot above was: 

OBJECT
	{
		name = PoodsOPMVO/Textures/aurora
		isCubeMap = True
		texXn = PoodsOPMVO/Textures/blankchannel
		texYn = PoodsOPMVO/Textures/auroraTest
		texZn = PoodsOPMVO/Textures/blankchannel
		texXp = PoodsOPMVO/Textures/blankchannel
		texYp = PoodsOPMVO/Textures/auroraTest
		texZp = PoodsOPMVO/Textures/blankchannel
	}
	OBJECT
	{
		name = PoodsOPMVO/Textures/storm1
		isCubeMap = True
		texXn = PoodsOPMVO/Textures/blankchannel
		texYn = PoodsOPMVO/Textures/blankchannel
		texZn = PoodsOPMVO/Textures/blankchannel
		texXp = PoodsOPMVO/Textures/blankchannel
		texYp = PoodsOPMVO/Textures/storm1
		texZp = PoodsOPMVO/Textures/blankchannel
	}

However, If I remove the storm1 entry, everything works fine. It seems that the duplicate use of the blankchannel image in multiple config entries seemed to have borked it.

Link to comment
Share on other sites
gkorgood

gkorgood

Posted
Posted (edited)
12 minutes ago, Rainbowd4sh said:

How exactly do you install? The instructions state to simply put the two zips into the game data folder, yet there is one zip,

Sorry if I am just being plain stupid, but I can't figure it out. I spent a good hour doing different things and nothing works. 

 

You never want to put the .zip files right into GameData. You want their contents in GameData. The reference to 2 .zip files is from just a few weeks ago, when there were 2. Now there's one, something like "AnyCPU-EVE.zip" I think.  there are 2 zip files. see edit. double click them, and you'll see a GameData folder in each. double click those as well, and drag their contents into gamedata

EDIT: sorry, I was mistaken, there are still 2 .zip files, "AnyCPU-Configs-Release.zip" and "AnyCPU-EVE-Release.zip". you need both if you're just using EVE. put the contents of both into GameData; they should be "BoulderCo" and "EnvironmentalVisualEnhancements"

Edited by gkorgood
zip confusion
Link to comment
Share on other sites
Axly Kerman

Axly Kerman

Posted
Posted
2 hours ago, fireblade274 said:

Yo guys so is this all working for you in 1.1.2? I see a lot of awesome pictures but I cant tell if your using 1.1.1 or 1.1.2; is it stable for you? In 64x?

2 CTDs within 6 minutes including restarting KSP. Both happened when exiting VAB. 64 bit.

rared log directories from the two crashes http://s000.tinyupload.com/index.php?file_id=07083539857338456640

If I should have included more than the two directories (2016-05-01_172138 and 2016-05-01_172712) I apologize.

Link to comment
Share on other sites
Rainbowd4sh

Rainbowd4sh

Posted
Posted
7 hours ago, gkorgood said:

You never want to put the .zip files right into GameData. You want their contents in GameData. The reference to 2 .zip files is from just a few weeks ago, when there were 2. Now there's one, something like "AnyCPU-EVE.zip" I think.  there are 2 zip files. see edit. double click them, and you'll see a GameData folder in each. double click those as well, and drag their contents into gamedata

EDIT: sorry, I was mistaken, there are still 2 .zip files, "AnyCPU-Configs-Release.zip" and "AnyCPU-EVE-Release.zip". you need both if you're just using EVE. put the contents of both into GameData; they should be "BoulderCo" and "EnvironmentalVisualEnhancements"

Thank you very much, it's working now! It looks amazing. 

Link to comment
Share on other sites
rbray89

rbray89

Posted
  • Author
Posted
2 hours ago, Axly Kerman said:

2 CTDs within 6 minutes including restarting KSP. Both happened when exiting VAB. 64 bit.

rared log directories from the two crashes http://s000.tinyupload.com/index.php?file_id=07083539857338456640

If I should have included more than the two directories (2016-05-01_172138 and 2016-05-01_172712) I apologize.

You are running out of memory. Looks like it might actually be caused by Kerbal Alarm Clock: 
  

5/1/2016 5:26:41 PM,KerbalAlarmClock,API Ready
 
(Filename: C:/buildslave/unity/build/artifacts/generated/common/runtime/UnityEngineDebugBindings.gen.cpp Line: 64)

DynamicHeapAllocator allocation probe 1 failed - Could not get memory for large allocation 673616.
DynamicHeapAllocator allocation probe 2 failed - Could not get memory for large allocation 673616.
DynamicHeapAllocator allocation probe 3 failed - Could not get memory for large allocation 673616.
DynamicHeapAllocator allocation probe 4 failed - Could not get memory for large allocation 673616.

 

Link to comment
Share on other sites
curtquarquesso

curtquarquesso

Posted
Posted
23 minutes ago, InfectedGrowth said:

Quick question: what is the hotkey for the UI editor? Did it change from alt+n?

12 minutes ago, rbray89 said:

ALT+0

This tends to get asked a lot. Any way that info could go in the OP?

 

 

 

Link to comment
Share on other sites
rbray89

rbray89

Posted
  • Author
Posted
3 minutes ago, curtquarquesso said:

This tends to get asked a lot. Any way that info could go in the OP?

I prefer a case-by case basis... Exposing the GUI directly would lead to (in my opinion) too many people messing with things they probably shouldn't mess with :)

Link to comment
Share on other sites
VonFrank

VonFrank

Posted
Posted

I'm a little curious about how the celestial shadows function works.

With the most recent versions of EVE and RSS, I notice that eclipse shadows on Jupiter seem to work properly when the config file is modified appropriately, but any planet farther than that fails to render the shadows. Is there a limit to the calculated distance from the sun for the celestial shadows to work properly?

Also, I've noticed that many shadows are comprised of 100% umbra and no visible penumbra at all. Every shadow (regardless of relative size and proximity) should have at least some penumbra, but not all should have an umbra. So is this an issue somehow with RSS conflicting, or is it a feature of realism not possible at the moment?

Link to comment
Share on other sites
rbray89

rbray89

Posted
  • Author
Posted (edited)
46 minutes ago, VonFrank said:

I'm a little curious about how the celestial shadows function works.

With the most recent versions of EVE and RSS, I notice that eclipse shadows on Jupiter seem to work properly when the config file is modified appropriately, but any planet farther than that fails to render the shadows. Is there a limit to the calculated distance from the sun for the celestial shadows to work properly?

Also, I've noticed that many shadows are comprised of 100% umbra and no visible penumbra at all. Every shadow (regardless of relative size and proximity) should have at least some penumbra, but not all should have an umbra. So is this an issue somehow with RSS conflicting, or is it a feature of realism not possible at the moment?

Shadow configs are required for them to work. What pack are you using? Logs? Screenshot? (Help me help you)

The celestial shadows *should* be fairly accurate with regards to umbra and penumbra. Not certain with RSS though. Things that could trip it up would be things like scaling issues and Kopernicus not setting radii correctly.

As distance increases from the Sun, penumbra becomes less and less. The only reason we have penumbra is due to the size of the light source. So if your sun is only a sliver of light, your penumbra will basically be non-existant. 

So yeah, in many circumstances (eg. Jool) the size of the sun isn't large enough to create MUCH of a penumbra (it is still  there if you look carefully). Kerbin-mun creates a sizable Penumbra on the other hand. Most of the images online demonstrating Umbra/Penumbra aren't good representations because the scale is WAY off.

EDIT: Added code for the curious. v is the point of interest, R is the Solar radius, r is the body radius, P is the Sun position p is the body position. 

inline half BodyShadow(float3 v, float R, float r, float3 P, float3 p)
	{
		
		float3 D = P - v;

		float a = PI*(r*r);
		float3 d = p - v;

		float tc = dot(d, normalize(D));
		float tc2 = (tc*tc);
		float L = sqrt(dot(d, d) - tc2);

		float scale = tc / length(D); //Scaled Sun Area to match plane of intersecting body
		float Rs = R * scale;
		float A = PI*(Rs*Rs);

		float s = saturate((r + Rs - L) / (2 * min(r, Rs)));
		s = (INV_PI*asin((2 * s) - 1)) + .5;
		return lerp(1, saturate((A - (s*a)) / A), step(r, tc)*saturate(a));
	}

 

Edited by rbray89
Link to comment
Share on other sites
VonFrank

VonFrank

Posted
Posted
55 minutes ago, rbray89 said:

Shadow configs are required for them to work. What pack are you using? Logs? Screenshot? (Help me help you)

The celestial shadows *should* be fairly accurate with regards to umbra and penumbra. Not certain with RSS though. Things that could trip it up would be things like scaling issues and Kopernicus not setting radii correctly.

As distance increases from the Sun, penumbra becomes less and less. The only reason we have penumbra is due to the size of the light source. So if your sun is only a sliver of light, your penumbra will basically be non-existant. 

So yeah, in many circumstances (eg. Jool) the size of the sun isn't large enough to create MUCH of a penumbra (it is still  there if you look carefully). Kerbin-mun creates a sizable Penumbra on the other hand. Most of the images online demonstrating Umbra/Penumbra aren't good representations because the scale is WAY off.

EDIT: Added code for the curious. v is the point of interest, R is the Solar radius, r is the body radius, P is the Sun position p is the body position. 


inline half BodyShadow(float3 v, float R, float r, float3 P, float3 p)
	{
		
		float3 D = P - v;

		float a = PI*(r*r);
		float3 d = p - v;

		float tc = dot(d, normalize(D));
		float tc2 = (tc*tc);
		float L = sqrt(dot(d, d) - tc2);

		float scale = tc / length(D); //Scaled Sun Area to match plane of intersecting body
		float Rs = R * scale;
		float A = PI*(Rs*Rs);

		float s = saturate((r + Rs - L) / (2 * min(r, Rs)));
		s = (INV_PI*asin((2 * s) - 1)) + .5;
		return lerp(1, saturate((A - (s*a)) / A), step(r, tc)*saturate(a));
	}

 

Thanks for the reply.

I am using the latest version of RSS with a bunch of extra moons I've added on my own to fill in a lot of missing celestial objects. Saturn, however, and all of its moons are default in RSS so that has nothing to do with my additions. Here is the shadow config file: 

EVE_SHADOWS
{
	OBJECT
	{
		body = Earth
		caster = Moon
	}
	OBJECT
	{
		body = Moon
		caster = Earth
	}
	OBJECT
	{
		body = Mars
		caster = Phobos
		caster = Deimos
	}
	OBJECT
	{
		body = Phobos
		caster = Mars
		caster = Deimos
	}
	OBJECT
	{
		body = Deimos
		caster = Mars
		caster = Phobos
	}
	OBJECT
	{
		body = Jupiter
		caster = Io
		caster = Europa
		caster = Ganymede
		caster = Callisto
	}
	OBJECT
	{
		body = Io
		caster = Jupiter
		caster = Europa
		caster = Ganymede
		caster = Callisto
	}
	OBJECT
	{
		body = Europa
		caster = Jupiter
		caster = Io
		caster = Ganymede
		caster = Callisto
	}
	OBJECT
	{
		body = Ganymede
		caster = Jupiter
		caster = Io
		caster = Europa
		caster = Callisto
	}
	OBJECT
	{
		body = Callisto
		caster = Jupiter
		caster = Io
		caster = Europa
		caster = Ganymede
	}
	OBJECT
	{
		body = Saturn
		caster = Mimas
		caster = Enceladus
		caster = Tethys
		caster = Dione
		caster = Rhea
		caster = Titan
		caster = Iapetus
	}
	OBJECT
	{
		body = Mimas
		caster = Saturn
		caster = Enceladus
		caster = Tethys
		caster = Dione
		caster = Rhea
		caster = Titan
		caster = Iapetus
	}
	OBJECT
	{
		body = Enceladus
		caster = Saturn
		caster = Mimas
		caster = Tethys
		caster = Dione
		caster = Rhea
		caster = Titan
		caster = Iapetus
	}
	OBJECT
	{
		body = Tethys
		caster = Saturn
		caster = Mimas
		caster = Enceladus
		caster = Dione
		caster = Rhea
		caster = Titan
		caster = Iapetus
	}
	OBJECT
	{
		body = Dione
		caster = Saturn
		caster = Mimas
		caster = Enceladus
		caster = Tethys
		caster = Rhea
		caster = Titan
		caster = Iapetus
	}
	OBJECT
	{
		body = Rhea
		caster = Saturn
		caster = Mimas
		caster = Enceladus
		caster = Tethys
		caster = Dione
		caster = Titan
		caster = Iapetus
	}
	OBJECT
	{
		body = Titan
		caster = Saturn
		caster = Mimas
		caster = Enceladus
		caster = Tethys
		caster = Dione
		caster = Rhea
		caster = Iapetus
	}
	OBJECT
	{
		body = Iapetus
		caster = Saturn
		caster = Mimas
		caster = Enceladus
		caster = Tethys
		caster = Dione
		caster = Rhea
		caster = Titan
	}
	OBJECT
	{
		body = Uranus
		caster = Miranda
		caster = Ariel
		caster = Umbriel
		caster = Titania
		caster = Oberon
	}
	OBJECT
	{
		body = Miranda
		caster = Uranus
		caster = Ariel
		caster = Umbriel
		caster = Titania
		caster = Oberon
	}
	OBJECT
	{
		body = Ariel
		caster = Uranus
		caster = Miranda
		caster = Umbriel
		caster = Titania
		caster = Oberon
	}
	OBJECT
	{
		body = Umbriel
		caster = Uranus
		caster = Miranda
		caster = Ariel
		caster = Titania
		caster = Oberon
	}
	OBJECT
	{
		body = Titania
		caster = Uranus
		caster = Miranda
		caster = Ariel
		caster = Umbriel
		caster = Oberon
	}
	OBJECT
	{
		body = Oberon
		caster = Uranus
		caster = Miranda
		caster = Ariel
		caster = Umbriel
		caster = Titania
	}
	OBJECT
	{
		body = Neptune
		caster = Proteus
		caster = Triton
		caster = Nereid
	}
	OBJECT
	{
		body = Proteus
		caster = Neptune
		caster = Triton
		caster = Nereid
	}
	OBJECT
	{
		body = Triton
		caster = Neptune
		caster = Proteus
		caster = Nereid
	}
	OBJECT
	{
		body = Nereid
		caster = Neptune
		caster = Proteus
		caster = Triton
	}
	OBJECT
	{
		body = Pluto
		caster = Charon
	}
	OBJECT
	{
		body = Charon
		caster = Pluto
	}
	OBJECT
	{
		body = Eris
		caster = Dysnomia
	}
	OBJECT
	{
		body = Dysnomia
		caster = Eris
	}
}

Upon further testing of my issue, it seems that only the Saturn and Uranus systems are having the issue of Shadows not functioning. Neptune, Pluto, and Eris all work as intended. Could it be an issue with having more than 5 bodies (1 main body and 4 casters) in the group of shadows?

 

For the penumbra, I see where you're coming from and that explanation is sound, but there is still a bit of an issue that is probably what caused me to think there was absolutely no penumbra whatsoever on objects that should at least still have a bit of one. The Jupiter system, for example, is one which has a varied selection of moon sizes, distances, and apparent angular diameters to test things on.

I took a screencap of a closeup of Ganymede's shadow on Jupiter and found this:

wYWCyLS.png

You can see that there is indeed a penumbra, but it is wobbly and undefined. In a non-static image, it shifts and warbles so much that your eyes tend to only notice the solid black area and it makes it seem that there is no penumbra.

This issue seems to scale with distance from the light source. Triton's shadow on Neptune is far worse due to it being at an extreme distance from the sun, yet if Neptune is hyper-edited to have an orbit as close as Mercury, and Triton is proportionally given a closer orbit to Neptune as to compensate for the Sun's larger apparent diameter at that distance, the Penumbra becomes solid and well defined.

Link to comment
Share on other sites
rbray89

rbray89

Posted
  • Author
Posted (edited)
1 hour ago, VonFrank said:

Thanks for the reply.

Upon further testing of my issue, it seems that only the Saturn and Uranus systems are having the issue of Shadows not functioning. Neptune, Pluto, and Eris all work as intended. Could it be an issue with having more than 5 bodies (1 main body and 4 casters) in the group of shadows? 

 

For the penumbra, I see where you're coming from and that explanation is sound, but there is still a bit of an issue that is probably what caused me to think there was absolutely no penumbra whatsoever on objects that should at least still have a bit of one. The Jupiter system, for example, is one which has a varied selection of moon sizes, distances, and apparent angular diameters to test things on.

I took a screencap of a closeup of Ganymede's shadow on Jupiter and found this:

You can see that there is indeed a penumbra, but it is wobbly and undefined. In a non-static image, it shifts and warbles so much that your eyes tend to only notice the solid black area and it makes it seem that there is no penumbra.

This issue seems to scale with distance from the light source. Triton's shadow on Neptune is far worse due to it being at an extreme distance from the sun, yet if Neptune is hyper-edited to have an orbit as close as Mercury, and Triton is proportionally given a closer orbit to Neptune as to compensate for the Sun's larger apparent diameter at that distance, the Penumbra becomes solid and well defined.

Ah... that is due to the scale of the universe... I can only use floats so even this scaled system will run into the floating point limit.

4 casters is the limit. So systems need to only use the closest/largest bodies in their shadow config.

Edited by rbray89
Link to comment
Share on other sites
Guest
This topic is now closed to further replies.
 Share
Go to topic listing
×
×
  • Create New...