|
SpaceEngine TODO List - Updated 14th Dec 2015
|
|
| Tzion | Date: Wednesday, 15.06.2011, 22:17 | Message # 16 |
 Observer
Group: Newbies
United States
Messages: 8
Status: Offline
| What is the schedule for the next release?
|
| |
| |
| SpaceEngineer | Date: Thursday, 16.06.2011, 09:36 | Message # 17 |
 Author of Space Engine
Group: Administrators
Russian Federation
Messages: 4800
Status: Offline
| I know that it is possible to model galaxies rotation and collision on GPU at interactive frame-rates. It is possible for one or two galaxies. But SE has thousands of galaxies... Maybe it is still possible to model its interaction:
1) Model moving of galaxies in expanding universe with N-body simulation, assuming that galaxies are dots (not accurate due to scattered dark matter).
2) Model rotating of isolated galaxies with analytic equations (well known in astronomy).
3) Detect galaxies that approach each other and start to colliding. In these galaxies, model colliding process with another N-body simulation.
But I am unsure that this is possible with interactive frame-rates at every frame. Maybe compute this one frame at 10 or 100, but then we must wait for weeks to see galaxies moving. The problem with N-body modelling is that it is impossible to launch program at any point of simulation time or to rewind time to any point. We must begin with some starting conditions and run thousands of simulation steps to reach desired point of time. Therefore, it is impossible to model several galaxies (as seen on the screen), we must model all of them simultaneously (except isolated galaxies that are modeled by analytic equations).
And then I implement procedural galaxies in near infinite space, modelling it really becomes impossible...
*
|
| |
| |
| SpaceEngineer | Date: Thursday, 16.06.2011, 09:41 | Message # 18 |
|
Author of Space Engine
Group: Administrators
Russian Federation
Messages: 4800
Status: Offline
| Tzion,
I do not think I will publish a new release date, because every time I do this I am late in issuing a stable release, I made many changes to code in a rush in past versions that resulted in many strange bugs. Do you remember the instability of 0.93 version and the new bugs as black screen in 0.94? So I think it's better for me release the next version when I am ready to do so 
*
|
| |
| |
| Deltawing777 | Date: Friday, 17.06.2011, 14:08 | Message # 19 |
 Observer
Group: Users
United States
Messages: 11
Status: Offline
| I am curious how will you overcome the Acceleration of game versus time over a network for each client? Example: The game is networked, I grab a ship and goto the moon but I accelerate the game to get there faster, Another client is on earth watching and does not accelerate the game. Where will the position of the moon be on earth for each client one accelerating the game and the other that is not? This was a issue that was never solved in the Orbiter 2010 simulator. How will you deal with one person in one time (accelerated) and the other clients in normal time
Edited by Deltawing777 - Friday, 17.06.2011, 14:18 |
| |
| |
| Tzion | Date: Friday, 17.06.2011, 14:44 | Message # 20 |
|
Observer
Group: Newbies
United States
Messages: 8
Status: Offline
| From what I understand, motion is going to be Newtonian, rather than relativistic, and long distance travel will be solved with warp jumps.
|
| |
| |
| Deltawing777 | Date: Friday, 17.06.2011, 15:19 | Message # 21 |
|
Observer
Group: Users
United States
Messages: 11
Status: Offline
| Quote (Tzion) From what I understand, motion is going to be Newtonian, rather than relativistic, and long distance travel will be solved with warp jumps.
I understand....But in regard to the planets and moon rotating how can the moon for example be in one spot for me? and in another spot for others? or in different spots for everybody that jumps at different times? How will a server keep up with the universe and all planets and moons etc.... and their locations and rotations?
|
| |
| |
| SpaceEngineer | Date: Friday, 17.06.2011, 15:30 | Message # 22 |
|
Author of Space Engine
Group: Administrators
Russian Federation
Messages: 4800
Status: Offline
| It is impossible for players to use different time scales in online game. So it will be in real-time for each player. Acelleration of time will be not necessary as there will be hyper-drive to cover long distances.
BTW, in real-time some interesting astronomic events (solar eclipses, asteroid impacts, supernovae explosions, etc) will be much more interesting to see, because players must wait for them for some years
*
|
| |
| |
| Neon | Date: Friday, 17.06.2011, 15:31 | Message # 23 |
|
Explorer
Group: SE team
Australia
Messages: 208
Status: Offline
| Firstly, the game doesn't alter time do move, it just alters
speed, and in a newtonian system, theres no time dilation so alls good
I can see your problem. Player A agrees to meet player B on such n such
a planet at some location. Player A's computer the planet is at x and
Player B's is at y, n they are a different coordinate, so how to have both
players be at same same place at same time.
I think there's only one way to handle that. On connection, players
universe will sync with the current players on the server. It's the
easiest, and quickest solution.
|
| |
| |
| SpaceEngineer | Date: Friday, 17.06.2011, 15:44 | Message # 24 |
|
Author of Space Engine
Group: Administrators
Russian Federation
Messages: 4800
Status: Offline
| Syncronisation with server is needed, because a players' PC clock might show different times. Sync must be done each minute. Universe time for all players is the same, co-ordinates of all moving planets will be the same for all players too.
*
|
| |
| |
| DevonX | Date: Monday, 20.06.2011, 01:46 | Message # 25 |
 Space Pilot
Group: Users
Norway
Messages: 113
Status: Offline
| What are your plans for your next release? what of the points will you have?
|
| |
| |
| SpaceEngineer | Date: Monday, 20.06.2011, 10:05 | Message # 26 |
|
Author of Space Engine
Group: Administrators
Russian Federation
Messages: 4800
Status: Offline
| Wow, I forgot to update TODO list... I plan to make procedural galaxies, loading/generating all stuff in parallel threads, and will probably make water and 3D clouds.
*
|
| |
| |
| j0KeR | Date: Monday, 20.06.2011, 15:04 | Message # 27 |
 Explorer
Group: Users
Greece
Messages: 178
Status: Offline
| Quote (SpaceEngineer) Lights of the night side (hot planet, lava, cities)
Happy to see lights in the night and lava planets are next . Cant wait !! :))
Quote (SpaceEngineer) The correct rendering of dust sprites
What exactly is this ?
"Novus Ordo Seclorum"
Intel Core(E6850) 2 Duo 3.00 GHz \ Geforce GTX560 1024MB \ 4GB 667(MHz) 2 Channel RAM \ Windows 7 (32bit) \ Samsung SA700D 120hz (3D)
Edited by j0KeR - Monday, 20.06.2011, 15:07 |
| |
| |
| DevonX | Date: Monday, 20.06.2011, 16:56 | Message # 28 |
|
Space Pilot
Group: Users
Norway
Messages: 113
Status: Offline
| Quote (SpaceEngineer) water and 3D clouds.
Omg this makes me very happy :3 how much will you do with the clouds? Just the couds or the dynamic variation shadowng and self shadowing? And will they be dynamic ?
|
| |
| |
| SpaceEngineer | Date: Monday, 20.06.2011, 18:09 | Message # 29 |
|
Author of Space Engine
Group: Administrators
Russian Federation
Messages: 4800
Status: Offline
| Quote (j0KeR) What exactly is this ?
This is a dark dust cloud visible on Milky Way. They must appear brown instead of grey, with a small tuning of galaxy shader I hope.
Quote (DevonX) Omg this makes me very happy :3 how much will you do with the clouds? Just the couds or the dynamic variation shadowng and self shadowing? And will they be dynamic ?
Dynamic clouds are impossible for now. Look at cloud layer around a planet, it's static and simply rotates around the planet. Having dynamic cloud layers means continiously updating of textures, but it lowers FPS. So 3D clouds will be static because they're generated based on clouds layer texture. They will simply be moving around planet with the cloud layer. But lighting effect will of course be dymanic. I can't really say about shadows, god rays and other stuff at this time - I must try implementing them first.
*
|
| |
| |
| DevonX | Date: Monday, 20.06.2011, 22:43 | Message # 30 |
|
Space Pilot
Group: Users
Norway
Messages: 113
Status: Offline
| Quote (SpaceEngineer)
Dynamic clouds are impossible for now.
Well i wouldnt say impossible... http://www.youtube.com/watch?v=pLfHDul5XGw
*
|
| |
| |
ENG
New site
United States 
Russian Federation 
Australia 
Norway 
Greece