To all SSC Station occupants
Thank you for the donations over the past year (2024), it is much appreciated. I am still trying to figure out how to migrate the forums to another community software (probably phpbb) but in the meantime I have updated the forum software to the latest version. SSC has been around a while so their is some very long time members here still using the site, thanks for making SSC home and sorry I haven't been as vocal as I should be in the forums I will try to improve my posting frequency.
Thank you again to all of the members that do take the time to donate a little, it helps keep this station functioning on the outer reaches of space.
-D1-
This project is something I have been planning for the last couple of years. There has been a long build up and many discussion prior to this point about game mechanics and how the tech tree should operate. Over the last few months I have bee able to work out most of the technical details for the physical mechanics. The IRS (Inertial Reference system) that I created is network ready and tested. It took allot of trial and error but we can now synchronize high velocity (>1 km/s) objects over a network, which was the main multiplayer bearier up to this point.
So, I'm not experienced with video editing but I was able to throw together this demo video highlighting the major miles stones up to this point. This video is only concerned with the physical mechanics and some visual highlight, we have yet to release the details of the the tech tree. However, we are working on it (the tech tree) feverishly. We have allot of detail available on the tech tree and non-physical narrative. We are in the process of building a private forum where we can host discussion about it. If you are at all interested in participating please in touch me with.
Our near term goals is the launch of a kickstater/Patrion fundraising campaign and we are quickly getting to that point. I want to share latest information here so we can get your feedback.
Interesting. I sit comfortably and will follow it 🙂
Rendering : Is it an homegrown engine, or are you using an existing framework (Ogre/Unity/Unreal/other) ?
Physics : What 3rd part for physic, or is it home made physics engine ?
Do you plan to clone Frontier, E:D, or is it a pure sandbox game ?
Where is your forum ?
Rendering: I am using unity with custom shaders for the rendering effects and my own code base for the planetary quad trees
Physics: The physics system is a combination of physX and my own system to handle large scales and high velocities
The game is in many senses a sandbox game. The conquest/federation of your home planet and development of you solar system (launching satellites, mining, colonizing) will be your primary goal and means of advancing through the game.
The mechanics are heavily focused on population and economic modeling. You will advance your species from its early industrial era into what ever future you choose for them (based on the social structure you pursue). Currently there are three cultural paths (Cruelty,Mildness and Piety) and four species types (Sapien,Reptile,Grey,Octapoid). Allot of the work is being put into the way these paths will play out in terms of game play and narrative.
The forum is located here http://spaceplebs.freeforums.net/ it is still under construction but you are welcome to join.
Today's progress.
Updates to atmospheric shader. Objects in orbit visible from the ground are now colour corrected. Atmospheric alpha is correctly smoothed.
Ground colour effects dude to atmosphere


:good: Amazing
what is the planet size ? (km)
I have tested planet sizes up to a radius of 3000 km (earth size) with no problems. Both the atmospheric shaders and planet C-LOD are unaffected by scale. Since it's procedurally generated it's really easy to test various sizes quickly. The way the planets level of detail works makes it so that the tree will continue to subdivide until the required mesh density is reached (which generally depends on the collider precision that you require), regardless of size. For proper rendering at that size you just need to make a composite camera that breaks up the rendering based on z-depth (basically a near and far field camera), otherwise you can get some really bad z fighting.
Here's a gif from one of the early test I did with a large sized planet.
The way the planets level of detail works makes it so that the tree will continue to subdivide until the required mesh density is reached (which generally depends on the collider precision that you require), regardless of size. For proper rendering at that size you just need to make a composite camera that breaks up the rendering based on z-depth (basically a near and far field camera), otherwise you can get some really bad z fighting.
:good:
I'm using the same technique, more or less 😀
I would recommend looking into Logarthmic z-buffering as described:
http://outerra.blogspot.co.uk/2009/08/logarithmic-z-buffer.html
http://outerra.blogspot.co.uk/2013/07/logarithmic-depth-buffer-optimizations.html
Logarithmic z-buffer. Very cool. The only downside I can see is that it would require I modify every shader to use logarithmic z buffer. no?
Today's progress. So I've finally found a combination of noise that produces continents and mountain ranges. It is a combination of noise functions. There is one for the continents one for mountain ranges.
it has the form:
noiseC: Continent Height
noiseM: Mountain Noise
noiseD: Mountain Distribution Noise (optional, continent noise can also be used here)
Height = radius + |noiseM|*(noiseD) + noiseC
Edit: This was another version I was using, however, the coast lines did not look right.
Height = radius - |noiseM|*(1-noiseD) + noiseC
How do you plan to alternate Mountainous area and very flat plains on the planet surface (desertic zone for example) ? This is a problem on which I stumbled on the past
How do you plan to alternate Mountainous area and very flat plains on the planet surface (desertic zone for example) ? This is a problem on which I stumbled on the past
Today's progress:
Added basic terrain splats based on height.
Do you mean texturing or terrain topology
terrain topology
Well, I don't know if there is a simple answer to that. What I am doing seems to produce decent terrain some of which is flat and some mountainous. I have yet to add any terrain features that are not height based though, such as polar caps or deserts. Thats probably be what the two remaining color channels should be for.
Progress: Lost of work on optimization. Mesh operations now run completely in the background. Terrian tweaks. Shader terrain splat updates. Still a few artifacts at lowest LOD where water mesh meets terrain about 4k vertices (or 6 subdivisions) per terrain quad. Thinking about ditching the water mesh after a certain altitude. (not a z-fighting issue, really boils down to the number of verts in the lowest LOD. Can be solved by turning the subdivision levels up at the cost to CPU/Initialization time, however this does not effect the frame rate only the time it take the generate a terrain quad. With some i7 power you could easily )
Here the planet with the water mesh removed.
Hey, it's been a while sine I posted anything here.
Here's a vid I made this morning.
Time for an update. So there has been lots of work done recently on terrain blending and an object pooling system to populate the ground the flora and rocks and what not. The biggest breakthrough by though are the procedural terrain normals! The surface gradient is calculated analytically in the shader it's self, (using the same noise function as the CPU that generates the heights) on a per pixel basis. The result is are completely seamless normals across all levels of detail. Much better.
Here is a gif of the result. Forgive the harsh transition between the mountains plains. Normals are intentionaly a little overpowered for test purposes.
And one flying in
Keepin this short. Just wanted to send out an update.
Here are some random pics from throughout that last few weeks.
(Note: Images below do not picture procedural normals)
planet disco
Any news or update ?