Sorry your browser is not supported!

You are using an outdated browser that does not support modern web technologies, in order to use this site please update to a new browser.

Browsers supported include Chrome, FireFox, Safari, Opera, Internet Explorer 10+ or Microsoft Edge.

Work in Progress / MPL3D Solar System

Author
Message
Cloggy
19
Years of Service
User Offline
Joined: 31st Oct 2004
Location: Rayleigh, Essex
Posted: 2nd Apr 2008 21:40
Fantastic work! Over 2 years in the making and it shows.

I agree with everything already said and think it shows just what can be acheived if you put in the effort and just keep at it.

Well done.

D3DFunc - Superfast Antialiased text & much much more
Lazlazlaz 1
18
Years of Service
User Offline
Joined: 18th Sep 2005
Location: England
Posted: 3rd Apr 2008 03:10
This is by far the most impressive thing I've seen done with DB. I got a bit lost though, it turns out the universe is really rather large I'll have a gander at the manual I think.

As with most people I jump around which projects I work on and which I drop for a while.
Currently I'm back working on Sioux, a Hollywood Western RTS.
Morcilla
21
Years of Service
User Offline
Joined: 1st Dec 2002
Location: Spain
Posted: 4th Apr 2008 20:35
Thanks to everyone for your comments

Lazlazlaz, it wouldn't be bad to have some more feedback about the manual or about if you still are (or anyone is) 'lost in space'
Plystire
21
Years of Service
User Offline
Joined: 18th Feb 2003
Location: Staring into the digital ether
Posted: 5th Apr 2008 08:59
Just had a quick go with the demo.

I found navigating to be a little... difficult to get to where you want to go.

The main problem was with the fact that the engine would take you through everything that is subcatagorized under the thing you wanted to go through first! I wanted to look at "The Milky Way"... and I never got to see it zoomed out. The engine wanted me to look through everything that was subcatagorized under the Milky Way FIRST!!! Gah! That could take me FOREVER! Clicking to go the opposite direction in the list of "things to visit" didn't help either.

I think having a checkbox in the settings for this would be great. Like if it was checked then it would go through subcategorized stuff before getting to the actual item. If it was unchecked then it would only go through the items that got returned in the search.


Another thought for potential updates would be to allow the user to compile a list of places to visit. This list could contain anything like planets, stars, star clusters, constellations, galaxies, nebulae, etc. Then the user could save the list and load it up later, and that way they could go on tours of areas that interest them without being required to go other places as well.


Aside from features that would be nice to have, I also ran into problems with zooming. Sometimes the zoom would go EXTREMELY slow, or it would go too fast. And sometimes (This occurred when I was looking at the Saggitarius A*) when you go to a small object after looking at ginormous object, it would still be super zoomed out, and you would have to zoom back in. This became bothersome when I was zoomed out by around 100 ly and went to a small object, on which the zooming goes extremely slow.


Other than the aforementioned stuff, it was a great experience which I will most likely share with friends and family.


The one and only,
~PlystirE~

Morcilla
21
Years of Service
User Offline
Joined: 1st Dec 2002
Location: Spain
Posted: 5th Apr 2008 13:53 Edited at: 5th Apr 2008 13:54
Great feedback, please let me help you:

* How to get to the Milky Way: Go to the Sun (F1) and press the right mouse button (reverse advance). Alternatively, you can press 'right control' key (next kind of object) until you get to the galaxies, then you are five left clicks away from the Milky Way (as 1st galaxy is Andromeda).

I agree it would take forever to advance sequentially, that's what the 'right control' key (next kind of object) is for.
Another way to orbit the Milky Way or any object, is by selecting it visually:

1-Enter free camera ('F' or mouse wheel button)
2-Show the mouse pointer with the right mouse button
3-Move the mouse pointer over the center of the Milky Way (if you can't find the center, turning names on ('N' key) will help you.
4-Click at the center of the Milky Way to orbit it. (Milky Way name will appear in red before clicking.)



* About the zoom: Bear in mind that the orbital camera always tries to leave you at the same distance that you were orbiting the previous body. So if you go from a big object to a small one, you can appreciate the size magnitude difference.

So it is not really that the zoom is 'out', it is that you are really far away, and you are not zooming in, but travelling to the object, that's why it gives the impresion of taking so long.

To travel quickly to the small object, use the travelling shortcuts, as 'Page Up' key. Alternatively, you can press 'Shift' while advancing to move faster.

Auto-zoom only takes place when you cannot be any closer than you reached. To tell if you are zooming or moving, turn on the data (key 'N' twice) and read the 'observing distance'. If it changes, then you are moving, if it doesn't then you are zooming.

Long explanation, I hope it helps.

Other than that, great ideas for the future, specially the user custom list. Thanks

Attachments

Login to view attachments
Morcilla
21
Years of Service
User Offline
Joined: 1st Dec 2002
Location: Spain
Posted: 8th May 2008 13:37 Edited at: 8th May 2008 13:50
Hi all,
I guess you'd like to watch the 'release' video:

MPL3D Solar System - Galaxy tour

It shows many of the current features of the simulation. And I say 'current' because I'm already making plans for the 1st update, pointing to add more stars, etc.
I'll tell you how it goes from time to time
KISTech
16
Years of Service
User Offline
Joined: 8th Feb 2008
Location: Aloha, Oregon
Posted: 8th May 2008 19:04
Very cool stuff. It keeps me inspired to press on wtih the development of my game knowing that someone was able to make something that looks that good.

Gunslinger
16
Years of Service
User Offline
Joined: 29th May 2007
Location:
Posted: 8th May 2008 20:59
Awesome. Nothing to say, just awesome.
qwe
20
Years of Service
User Offline
Joined: 3rd Sep 2003
Location: place
Posted: 2nd Jun 2008 00:06
hey. im curious how you did the plasma emanating from the stars? not the line/bar particles flying out linearly, but the gas-like floaty stuff?

for my space project i was thinking of using cellular automata for that. on a flat rectangle, then wrap the rectangle around the star, point it at player

also, what sort of shaders did you use? i thought my star was looking pretty good but holy crap. it looks like maybe cartoon shading while manipulating texture UV coordinates (i made my plasma effect by having each UV coordinate circle around its original coordinate, creating a liquid surface effect)
Morcilla
21
Years of Service
User Offline
Joined: 1st Dec 2002
Location: Spain
Posted: 2nd Jun 2008 19:54
Hey thanks for the comments.

Quote: "how you did the plasma emanating from the stars?"


These are made with a set of ghosted planes. Use different textures for them, but with the same fashion.
To obtain the effect, fade them out at the same time that they are scaled up
I think this fade-out / scale-up technique can be used for many effects like smoke, fire, explosions and things like that.

Quote: "what sort of shaders did you use?"

Well, depite that many tries are needed for a good result, you are not on a wrong way so far. It is a pixel shader that calculates the final colour based on the original texture and a noise texture. Playing with UV coordinates can also give you nice surprises sometimes. Bear in mind that the textures applied usually change dramatically the final result.

So far I'd like to improve those 'line/bar particles flying out linearly' but I think it is hard to make them more realistic and keep a nice performance. I'm open to any ideas, suggestions are welcome.
qwe
20
Years of Service
User Offline
Joined: 3rd Sep 2003
Location: place
Posted: 3rd Jun 2008 06:54
just a suggestion, i think it'd look more realistic if those 'linear bars' were big transparent gaseous clouds of plasma emanating

you have plasma emanating, but only on that expanding plane. so the user cant actually 'fly through' the plasma coming out, and teh plasma only comes out of a ring of teh star. would look nice IMO if plasma was also coming out all around the surface of teh star
Morcilla
21
Years of Service
User Offline
Joined: 1st Dec 2002
Location: Spain
Posted: 3rd Jun 2008 13:56
Quote: "big transparent gaseous clouds of plasma emanating"


I agree that would be better but, how to achieve a nice result?

Any custom mesh for this would be repetitive (or many different custom meshes would be needed) not to say that they could look artificial too. Also I've tried different textures, but they didn't look better to me...

Quote: "'fly through' the plasma coming out"

Yep, that is a nice idea. Still, it depends on the implementation of the idea above, so not really sure about which would be the best approach

I've thought about using some sort of 'expanding spheres', but as I said, I think these ideas might look ugly if they are not perfectly implemented. Also I expect a big performance impact, as usually many objects are needed, and the poly count goes on...
Silvester
18
Years of Service
User Offline
Joined: 7th Dec 2005
Location: Netherlands
Posted: 3rd Jun 2008 17:36
I think it looks pretty realistic and nice already if I may be honost.
Morcilla
21
Years of Service
User Offline
Joined: 1st Dec 2002
Location: Spain
Posted: 5th Jun 2008 15:06
Sure, I appreciate that as well. Someone told me that particles look like 'blobs' and make no sense. Oh well I guess they haven't tried... Nevertheless I'd like to improve them if I can, I think I'm taking a look into DarkPhysics particles if I have the time.
Morcilla
21
Years of Service
User Offline
Joined: 1st Dec 2002
Location: Spain
Posted: 27th Jun 2008 21:02 Edited at: 27th Jun 2008 21:27
Hi all,
I'm working on the new version, "MPL3D Solar System v1.1" from now on.
Before starting other changes, I've focused on the user interface, as it is a very important improvable area, here is what has come out so far:



As you can see, there is a new menu, and a new coordinates (Rigth Ascension/Declination) dome, which are mostly finshed.

Other improvements made:

- Handling improvements, like the new visual controls for time and camera.
- Antilaliased lines for constellations, orbits and R.A./Dec. dome.
- User can set color for data output.

Still pending to do for this version:

- Put additional scientific data for stars and planets.
- Menu translation (to spanish).
- Other little improvements.

Also, I posted here new space images from the simulation:

Stars, planets and space backgrounds

That's all for now

[Edit: I forgot to thank Sephnroth, who provided an excellent tutorial to use windows menus with DGDK. Thanks a lot.]

Attachments

Login to view attachments
Morcilla
21
Years of Service
User Offline
Joined: 1st Dec 2002
Location: Spain
Posted: 8th Aug 2008 19:35 Edited at: 8th Aug 2008 19:36
MPL3D Solar System 1.1 is now available, please download it from here:

http://www.mpl3d.com/downloads.htm

These are the main new features:

-New user interface, Windows type menu and icons for visual control, customizable data output colour.
-Local selector, allows to select any body within a star system.
-Celestial coordinates grid, Right Ascension and Declination data output.
-Additional scientific data for stars and planets.
-Windowed mode, 16:10 and 16:9 resolutions.
-User tips, handling improvements, internal improvements.
-Check for updates option.

Here is a pic of the new 'Local selector' feature:



Also, the final appearance of the user gui has been improved a little bit compared with the last picture shown in the previous post.
I hope you find its handling much more improved. All feedback is welcome

Attachments

Login to view attachments
CuCuMBeR
21
Years of Service
User Offline
Joined: 11th Jan 2003
Location: Turkey
Posted: 23rd Aug 2008 19:03
Hey Morcilla,

I was wondering if you have an excel sheet of the solar system data, like the planet and known asteroids (maybe comets too) properties(mass, density, orbit time...etc)

If you have such a detailed excel sheet or something smiliar, please share it.

There is always one more imbecile than you counted on.
Morcilla
21
Years of Service
User Offline
Joined: 1st Dec 2002
Location: Spain
Posted: 29th Aug 2008 13:35 Edited at: 29th Aug 2008 13:36
Hi CuCuMBeR, I see you still have no email...

Well, I'm afraid that, believe it or not, I don't have that excel
At the beginning there was only the Solar System itself, so there was no need for data load, and the data was embedded inside the code. One of the pendings to-do things is to normalize this situation.

However the web is full of this kind of tables, you may start around here:

Solar System Parameters:
http://www2.jpl.nasa.gov/galileo/sepo/education/plansat/table.html
http://www.planetsalive.com/?planet=Main&tab=E

If you want to dig further I'd recommend JPL Solar System Dynamics:
http://ssd.jpl.nasa.gov/

It has tables with physical and orbital data, as well as small-bodies data, diagrams, etc.

I hope that helps
CuCuMBeR
21
Years of Service
User Offline
Joined: 11th Jan 2003
Location: Turkey
Posted: 31st Aug 2008 18:15
Okay thats what i needed, thank you.

Although now i need to know how you do that planet position finding according to the date supplied.

I read somethin like kesslers equation and somethin like that all around the web but i just dont understand the formula.

Could you supply some basic code to show how to do this?

PS. i sent you an e-mail, probably you ignored or its gone to junk mail.

There is always one more imbecile than you counted on.
Morcilla
21
Years of Service
User Offline
Joined: 1st Dec 2002
Location: Spain
Posted: 1st Sep 2008 12:19
Well, I recommend you the examples at the codebase. There is a very good application there posted by Dark.Five:

Accurate 3D Planet Simulation (no media required)

That code is nicely written, and it is by far accurate for any gaming purposes.

Other useful examples that may help (more simple):

Solar System Recreation (No Media)
Simple solar system model - DBPro
Advanced Day/Night Simulation

Yep, I can't find your e-mail
Please send it again at my profile address or to mpl3d@wanadoo.es_nospam
Morcilla
21
Years of Service
User Offline
Joined: 1st Dec 2002
Location: Spain
Posted: 28th Nov 2008 16:47 Edited at: 28th Nov 2008 16:48
I'm currently working in the new version v1.2, it includes so far:

- New logarithmic brightness system for the stars and the starfield. It is based on the real apparent magnitude of the stars, that fits the logarithmic perception of the human eye, giving a more natural sensation to the star rendering.

- Exponential camera zoom for planets and moons. Right now, it allows to get 10,000 times closer to these bodies, enhancing the experience of large magnitudes in sizes and distances.

- New camera speed algorithm, inspired by Dark Coder's planetary patches source code. The camera speed is now directly proportional to the distance to the surface of the closest body. In this way, the closer the camera gets to the surface, the slower it goes, allowing a much better control.

- Camera collision system for the solid bodies. This prevents the camera from entering the planets interior (as well as stars, etc.) For example, the orbiting bodies will 'push' the camera if the camera lies in its orbital path.

- Many other little changes, including a new star flare system, affected by the 'solar filter' option, as well as new textures to be applied accordingly. These are samples of the Sun and its new looks:



Upper pics show solar filter on, while right most ones show solar flare on. Maximum brightness (and standard setup) is the bottom-right picture.

I would like to include some other features, like:

- Dark Coder's atmosphere shader, for planets with atmosphere.

- Better textures for the main planets.

Also, the changes always bring in new cases, and little bugs here and there, that must be tested and fixed, so that's another thing in the to-do list

Attachments

Login to view attachments
sindore
19
Years of Service
User Offline
Joined: 2nd Jul 2004
Location: Bedfordshire, UK
Posted: 28th Nov 2008 20:41
this looks amazing can't wait to have a look around are local solasystem.

soul sucking devils, twisted body of the damed, slivering slim drips from every poor, sin licking at your ears, and the smell stinging your eyes, and if you don't like it, get out of my kitchen!
Morcilla
21
Years of Service
User Offline
Joined: 1st Dec 2002
Location: Spain
Posted: 20th Dec 2008 17:42 Edited at: 20th Dec 2008 17:44
Well, it seems that, due to the exponential zoom feature, standard in-built particles are not suitable anymore, or in most cases, for the project, as they cannot be scaled once they are created.

And so, I've been preparing a new particle system that is based on excellent Philip's black hole code snippet.

Here is a video where these new particles can be seen in action:

http://www.4shared.com/file/76823841/a1b1649f/MPL3D_Solar_System_-_WIP_stellar_particles_v12.html

They can be seen affected by the different parameters. By now the future use would be the one shown at the very beginning of the video, although other uses could be studied.

Please feel free to comment. All feedback is appreciated.
Morcilla
21
Years of Service
User Offline
Joined: 1st Dec 2002
Location: Spain
Posted: 15th Jan 2009 19:40
Here is a video showing the new particles as they will appear for our Sun:

http://www.4shared.com/file/80790887/caa4851a/MPL3D_Solar_System_-_Stellar_particles_v12__WIP_.html

Each particle is a now a little bit more complex than before, having a new 'photon' look. Also, its trajectory system is more refinated and advanced.
Morcilla
21
Years of Service
User Offline
Joined: 1st Dec 2002
Location: Spain
Posted: 5th Mar 2009 10:45 Edited at: 16th Mar 2009 19:40
New video published in YouTube:

Winter Constellations, an educational video in line with the spirit of the International Year of Astronomy 2009.

The video has been loaded in High Definition, so make sure that you choose that version, if your local YouTube allows you to do so.

Also, here is a report with the current feature status for the new version v1.2.
As a summary:

-There are 20 main areas of features and improvements.
-Menu options count up to 134, being 21 of them new for this release.
Morcilla
21
Years of Service
User Offline
Joined: 1st Dec 2002
Location: Spain
Posted: 14th Mar 2009 18:20
There are two new Work In Progress (WIP) videos available:

Betelgeuse video:

Shows the surface animation for such a big star like Betelgeuse, that it is not able to maintain a completely spherical shape. Video is in time speed +1. Also, the 3d star sound can be heard.

Gravitatory particles video:

Here can be seen two close stars in the Algieba star system (Eta Orionis).
These stars are very massive, and so they are their gravity forces.
The stars are not close enough to produce a direct mass transfer. However, solar winds and big Coronal Mass Ejections (CME) can be captured by close stars, specially if the eruption heads towards the companion star. 3d star sound has been recorded too.

These videos have been recorded in 16:10 (panoramic). If you see the videos stretched in the web page, download and play them with the Windows Media Player or any other player of your choice.
Garion
16
Years of Service
User Offline
Joined: 7th Dec 2007
Location: Poland
Posted: 15th Mar 2009 01:49
Sweet
lucifer 1101
15
Years of Service
User Offline
Joined: 26th Jan 2009
Location: Melbourne, Australia
Posted: 15th Mar 2009 02:38
hi Morcilla

i just am wanting to ask if you could go over my program and make a few suggestions of what i can do to make it better (mostly graphical)

i have been watching this thread and what you have been making is incredible..

heres the link http://forum.thegamecreators.com/?m=forum_view&t=144856&b=8

btw i love th new interface...
Morcilla
21
Years of Service
User Offline
Joined: 1st Dec 2002
Location: Spain
Posted: 15th Mar 2009 12:47
Thanks Garion

lucifer 1101, thanks also for the comments. I have posted my suggestions in your two related threads:

Galaxy Generation Program
Solar Exudus (WIP)

I hope you find them useful
lucifer 1101
15
Years of Service
User Offline
Joined: 26th Jan 2009
Location: Melbourne, Australia
Posted: 15th Mar 2009 12:57
Thankyou for all of your help so far..

I cannot wait to see what you do with this next...
Morcilla
21
Years of Service
User Offline
Joined: 1st Dec 2002
Location: Spain
Posted: 29th Apr 2009 11:41 Edited at: 29th Apr 2009 11:42
I've posted a new picture of the Earth in the TGC Gallery:



This is using Dark Coder's atmosphere shader, and Green Gandalf's normal map shader.
Many thanks to both of them for their kind collaboration and support.

I'm applying the atmosphere shader to all bodies with atmospheres, and the normal map shader to the rocky bodies. Venus and Mars will have both, in the same way as the Earth does in the pic.

Attachments

Login to view attachments
Serial Velocity
15
Years of Service
User Offline
Joined: 24th Aug 2008
Location:
Posted: 29th Apr 2009 19:00
That atmosphere looks stunning! I've tried implementing the shader into my space sim but I failed miserably

Also, in MPL3d Solar System. What kind of scale are you using? What represents 1 DBPro Unit? I always wondered how you managed such a massive scale.

Morcilla
21
Years of Service
User Offline
Joined: 1st Dec 2002
Location: Spain
Posted: 30th Apr 2009 12:10
Hey thanks!
Well, the basic scale system has two main scales, one for open space and another for 'inside a system'.
For open space, 1 DBPro Unit is 1 Light Year.

If you get close enough to a star (that is, a 'system'), then scales are expanded. This would be the main second scale used.
This one is variable, depending on the size of the star (curiously, I made this inversely proportional to that size, the bigger the star, the lower the dynamic scale).
And so, it results that, for example, inside the solar system, 1 DBPro Unit is around 625 km. One Astronomic Unit (AU) is around 240,000 DBPro units, and the Earth results to be around 10 DBPro units in size.

It is a complex subject, I think that this current way is becoming kind of legacy, and that I should review it at some point. Having the Earth so small makes it conflict with the camera range is very near from it. Not to say to represent smaller things like a spaceship.
I'm starting to cover this issues with the exponential camera zoom. It introduces a third scale, that comes in place when you are close enough to a planet or any other coordinates. It is similar to dinamically increasing the previous commented second scale, the closer to the planet, the bigger the scale, so the planet is scaled up, and positioned away accordingly
Serial Velocity
15
Years of Service
User Offline
Joined: 24th Aug 2008
Location:
Posted: 2nd May 2009 20:18
I see now, never thought about doing it like that in my game

Cant wait for more progress.

KeithC
Senior Moderator
18
Years of Service
User Offline
Joined: 27th Oct 2005
Location: Michigan
Posted: 4th May 2009 08:30
Great program! I may have to grab a copy for my kids one of these days.

You wouldn't be able to point me in the right direction for planetary textures (of our solar system); would you? Working on our own project that needs them.

Morcilla
21
Years of Service
User Offline
Joined: 1st Dec 2002
Location: Spain
Posted: 4th May 2009 14:02 Edited at: 5th Jun 2009 21:25
Quote: "...never thought about doing it like that in my game"

Well, if I'm not mistaken, you should do some 'trick' (or special computing ) if the ships are going to be near the planets and not just in the background. Otherwise I'm afraid that those ships will look very big.

Quote: "Great program! I may have to grab a copy for my kids one of these days."

Hey thanks, I think the kids will find this great, entertainment and learning at once
Although it would be great if you purchased the program, please do so only if you want to support the project, that would help. However, you don't have to do so. Here is a 100% discount coupon code, with 25 licenses, for you and the community:



It has to be entered at the program itself, thru 'Buy full program' option. Enter and validate. Price should be set to 0

Quote: "...point me in the right direction for planetary textures (of our solar system)"

Sure, here are some links to planetary textures.

These ones are license free to use in games, giving proper credit, but they are copyrighted for distribution:

http://planetpixelemporium.com/planets.html
http://www.mmedia.is/~bjj/planetary_maps.html

These ones are more serious stuff, usually free for any use, giving proper credit:

http://visibleearth.nasa.gov/
http://www.mapaplanet.org/

Use these ones if you are not going commercial, otherwise you must ask the authors:

http://www.johnstonsarchive.net/spaceart/cylmaps.html
http://laps.noaa.gov/albers/sos/sos.html
http://www.celestiamotherlode.net/catalog/earth.php
http://evildrganymede.net/art/maps.htm

Thanks all for your comments

Attachments

Login to view attachments
KeithC
Senior Moderator
18
Years of Service
User Offline
Joined: 27th Oct 2005
Location: Michigan
Posted: 4th May 2009 19:00
Outstanding; thanks! I'll still buy a copy to support you one of these days (when I get the kids their own computer.

As for the links, they are exactly what I was looking for...and the few that are paid textures are well worth it!

Thanks again, you saved me a lot of searching.

-Keith

Morcilla
21
Years of Service
User Offline
Joined: 1st Dec 2002
Location: Spain
Posted: 5th Jun 2009 21:29 Edited at: 5th Jun 2009 21:35
KeithC, no problem, glad you find them useful

All solar system planets heve been updated or revised, you may want to take a look to all of them here:

http://www.mpl3d.com/index2.htm

Here is Mars, as a sweet preview, uploaded to the TGC gallery (click to go there):


The great martian canyon, seen from an orbit around 600 km of altitude. Field of view 55º.

Also one user, ludovisuis, made a couple of videos with the current v1.1 release version:

Different types of gaseous planets (part one)
Different types of gaseous planets (part two)

A very nice effort that helps us very much to show more insights of the simulation, as well as being educative all along
Serial Velocity
15
Years of Service
User Offline
Joined: 24th Aug 2008
Location:
Posted: 5th Jun 2009 22:09 Edited at: 5th Jun 2009 22:10
Wow, the new planets look awesome! I would recommend using an alpha mapped ring system instead of ghosting, it seems to give a more realistic feel to it and it has more volume.

I especially love that earth you've got there! How did you manage to do the specular reflection? I tried in a shader but it completely failed

Morcilla
21
Years of Service
User Offline
Joined: 1st Dec 2002
Location: Spain
Posted: 6th Jun 2009 18:00
Quote: "I would recommend using an alpha mapped ring system instead of ghosting, it seems to give a more realistic feel to it and it has more volume."

Yeah well I have to ponder somethings about this. As you will have noticed, it has no shadows so far either. I'm looking for a solid shadow system that can be used along with the other shaders

Quote: "I especially love that earth you've got there! How did you manage to do the specular reflection?"

Thanks a lot, well it is the same than the one posted above, just that I turned the oceans a little bit bluer. So the specularity is included in the Green Gandalf's normal map shader.
Just texture the stages with diffuse, normal and specular maps, apply the effect and set, at least, the light direction
Serial Velocity
15
Years of Service
User Offline
Joined: 24th Aug 2008
Location:
Posted: 6th Jun 2009 18:40 Edited at: 8th Jun 2009 18:37
Quote: "I'm looking for a solid shadow system that can be used along with the other shaders"

Here is the one I use, darkcoder made it a few years ago and it was really easy to implement.

Shader:


DBPro Code (shouldnt be too hard to port to GDK):


Morcilla
21
Years of Service
User Offline
Joined: 1st Dec 2002
Location: Spain
Posted: 8th Jun 2009 12:50
Hey thanks Serial Velocity.
Well, yes. I have pending to try Dark Coder shadows, although the shader that he posted in your thread seems a little bit improved than the one you post. Anyway thanks for the DBPro code as well.

I've been worried about if it would cast the shadows of the ring over the planet as well.
Also, at the same time, it has to cast the shadows of the moons over the planet and vice-versa... So still not totally sure about the best solution
Serial Velocity
15
Years of Service
User Offline
Joined: 24th Aug 2008
Location:
Posted: 8th Jun 2009 18:27 Edited at: 8th Jun 2009 19:25
I havent managed to be able to put his improved one into my game yet, not sure why

The shader only affects the ring really. I imagine you could put the same shader on the moons aswell because when I had a look at it, it looked like it could work, ill try it and edit the post.

EDIT: It works kinda but theres no diffuse lighing so im going to edit the shader to add this, heres a screenshot of the same shader on a huge moon.



Attachments

Login to view attachments
Morcilla
21
Years of Service
User Offline
Joined: 1st Dec 2002
Location: Spain
Posted: 13th Jun 2009 12:10
Allright, thanks for the input and showing the test results. Great to know that it cast the whole shadow over the moon object

I haven't seen much differences for implementation purposes between the two shaders. After examining them with more detail, I realize that:

- A shadow shader for a planet/moon must contain the calculus for all the other planets/moons around the common barycenter (that is, any near object), except itself.
(This should include any other object around casting shadows, like the ring.)
For example a planet with 3 moons should use a shader that calculates the shadows for those 3 moons, while each moon should have a shader calculating the shadow of the planet, as well as the one coming from the other two moons.

As far as I understand, the only way to do this in a parametrized way, is to include a 'for' loop inside the shader, to tell how many shadows are around, and I think that implies using 3.0 shaders...

- The shadow shader code itself should be incorporated into any existing shader to be useful. Moons could have a normal map shader, while planets can have an atmospheric effect shader, and so on. By just applying the basic shadow shader it is not enough, or the planets and moons would look 'effect-less'. This could have some complications as well...

- Rings objects are not clear. I'd say that the shader is designed only for spherical objects...I haven't really tested it yet.

To summarize, I haven't decided so far.
Dark Coder's shadows could look great, but they imply further shader development and testing.
On the other side, if the stencil shadow testings perform as they do now in the 1.1 version (that is, to compile the 1.2 release version using Visual Studio 2005 instead of VS 2008), it could save a lot of time, allowing a faster release. Of course they do not look the same...
qwe
20
Years of Service
User Offline
Joined: 3rd Sep 2003
Location: place
Posted: 14th Jun 2009 18:47
what if you just point the ring system away from the star in realtime, then the shadow could be put into the image (whether ghost or alpha) texture and there's no need to change the texture in realtime, or use shaders etc. just have the shadow at the correct angle
Morcilla
21
Years of Service
User Offline
Joined: 1st Dec 2002
Location: Spain
Posted: 15th Jun 2009 14:42
Uh, sorry, I do not completely understand you, but thanks for trying to help

Quote: "there's no need to change the texture in realtime"

Well, if you mean using something like pre-rendered shadow maps, I'm afraid that they only work fine on static scenaries...
qwe
20
Years of Service
User Offline
Joined: 3rd Sep 2003
Location: place
Posted: 18th Jun 2009 12:50 Edited at: 18th Jun 2009 12:51
unless you rotate the object each loop

or do you put all your objects into a shadow engine?
Morcilla
21
Years of Service
User Offline
Joined: 1st Dec 2002
Location: Spain
Posted: 18th Jun 2009 20:34
Quote: "unless you rotate the object each loop"

Sorry I'm afraid I still do not fully understand you

I currently use standard in-built stencil shadows. Since they are buggy, I'm pondering to use Dark Coder's shadows instead.
Morcilla
21
Years of Service
User Offline
Joined: 1st Dec 2002
Location: Spain
Posted: 10th Jul 2009 19:38 Edited at: 10th Jul 2009 19:41
Atmospheric shader has been parametrized for extrasolar planets.

Every planet has his own touch, although many atmospheres are similar, each one is different.

This should provide new nice views, just go to
http://www.mpl3d.com/index2.htm (news date 25th June 2009)
to see some of them

Still, just stepping towards the perfect concept, and so, some places do not behave as nice as they could, but definitively it is an improvement.

Other new improvements so far:


Base Features

- Visual magnitude revised for variable stars: Corrected magnitude (based on General Catalogue of Variable Stars) for stars like Betelgeuse.

- Visual magnitude calculation and output: Apparent magnitude from the point of view is calculated and can be read when 'Scientific data' is ON ("d" key).

- Star colour based on spectral class. A new shader has been integrated and it will base star colour on its spectral class (when available). This results in 380 different star colours, assigned to more than 1000 different spectral classes.


Secondary features

- Dynamic Level of Detal System (LOD): Level of detail system dynamically changes the quality of objects and textures based on their angular size.

- Text output system revamped to a solid and super fast angular size method.

- Extradet objects (probably to be renamed to '3d topological' or something like that) integrated into the shader and LOD systems. Also, their vertex have been relaxed around 50%, so they do not look so sharpy.


Menu options

- Freeze Time: Freezes the simulation time.

- Atmospheric shader: Option to turn On/Off independently the atmospheric shader for planets with atmospheres.

- Relief shader: Option to turn On/Off independently the normal map shader for planets with solid surface.

- Level of Detail ->Max: Fixed maximum detail for planetary objects and textures.

- Level of Detail ->Max: Dynamic detail for planetary objects and textures.

- Level of Detail ->Min: Fixed minimum detail for planetary objects and textures.


As usual, the full report: http://www.mpl3d.com/Featuresv12.htm

Have another pic of the new looking of the atmospheric shader in this clarified jovian:



All comments welcome
qwe
20
Years of Service
User Offline
Joined: 3rd Sep 2003
Location: place
Posted: 11th Jul 2009 16:59 Edited at: 11th Jul 2009 17:21
wooow! please help me, when i start to learn shaders when i start working on the graphics of my solar system that wont be for a long time though. but would you perhaps consider showing me how you do things like have a night/day side, how you apply those amazing shaders, etc? this would be in a few months, as i am planning on making the engine before applying the graphics

i really enjoy watching the cannibal stars, and the neutron stars, etc. some really nice videos. your program has come a really long way over, idk, its been like a year now since it was released right?

im planning on recreating a lot of my space exploration game. im going to have a procedurally generated universe, and my plan is "BIG", as in "VERY VERY BIG"

morcilla, do you think you could spare 10 minutes and help me out with constructing my orbit equations? i assume you use a X(theta)= and Y(theta)= and Z(theta)= ? im having trouble figuring out how to put the orbital parameters into the X=Y=Z= equation in order to have the orbits be accurate. right now my orbits are just circles, with two parameters, the semi major (which makes the circle radius) and the orbital period

Login to post a reply

Server time is: 2024-04-18 23:41:36
Your offset time is: 2024-04-18 23:41:36