Aurora Inspired Game

[iemfi (OP)]

I found out about Aurora about 2 weeks back and was absolutely amazed by the game.   It inspired me to make a game using some of the mechanics of Aurora with a ship building concept I did for Ludum Dare a few months ago.   Thought I'd show some of the progress so far, hope this is the right place for that.   Would love any criticism, feedback, suggestions, or even a name to call the damn game.  Images at hxxp: imgur. com/tpUXRJ2

System View



The idea is to have fully Newtonian movement and combat.   Starting with more or less what we do now with hohmann transfers and gravity slingshots in the early game to more direct movement as engines get more powerful.   The planets and stuff are orbiting nicely but I'm still working on the fleet movement aspect of it (orbital mechanics is a bitch).   Same thing with interstellar travel, the idea is to only have slower than light interstellar travel for the early game so that early expansion will be a seriously costly undertaking.   The game plays in real time, with up to 1 million times faster than real time time warp.  

Industry


Pretty similar to Aurora here, the difference is I want to focus on the generation of energy.   Solar array construction around the sun, fusion plants, geothermal reserves from planets.   Massive amounts would be required for interstellar travel.  

Ship/Module Design


Ships and modules will be collections of tiles.   Damage to the ship will damage/destroy tiles based on where the projectile hits.   Right now the design only cares about whether your crew can get from place to place and whether the various rooms are closed off from space, and the total count of each tile.   Would like for the various modules to have their characteristics rely on the specific placement of tiles.   Haven't been able to think of a proper system though.  

Research


The idea is that you can only divide your research time between various categories.   A roll will be made for any techs within that category.   Each time a successful roll is made the player gets to choose whether they prefer more work in that sub-category.  

[3_14159]

That looks really, really awesome!

Some short questions, since I'm going to be on the train soon (more to come later):
- Which language did you use?
- Especially, which framework for the UI?
- Did you really make that in under two weeks?
- Is your concept 'realistic' physics aside from FTL or physics plus?
- What's your plan for the development?
- Did you see the Pulsar4x project? http://aurora2.pentarch.org/index.php/board,169.0.html

[iemfi (OP)]

- Which language did you use?
C# / XNA
- Especially, which framework for the UI?
UI is from scratch (which is probably not the best of decisions but I wanted to try my hand at making a UI)

- Did you really make that in under two weeks?
Yup, 2 weeks of full time work on it.   

- Is your concept 'realistic' physics aside from FTL or physics plus?
Realistic physics as much as possible.    I'm thinking of doing without even FTL at the early game.    Voyages at 50% lightspeed which take a year to build up the energy required to launch and a decade of travel time.    Sort of something which could be done in real life given physics as we know it.    I think Aurora makes interstellar travel too easy.   

For interplanetary travel I'm hoping to have a robust Newtonion movement system.    Today I just got the hohmann transfer working, so you give a fleet a target planet, the system calculates the time to initiate the transfer and plots the path then follows the path exactly.    I was hoping to simulate it properly but orbital mechanics is proving to be way too damn difficult (and probably too resource intensive).    The target is something close enough but not have fleets crash into planets or get lost in space for no reason (but still happen after certain events).   

- What's your plan for the development?
To get a playable version ASAP.    But really I don't have a plan at all, which is part of the reason why I made this thread.   

- Did you see the Pulsar4x project? hxxp: aurora2.   pentarch.   org/index.   php/board,169.   0.   html
Yup! Couldn't even access the github page though, broken link.   
EDIT: Just read through more of the Pulsar4x forum, I didn't even know it was still in development.  I think the main difference is that Pulsar4x seems to be following Aurora very closely.  I'm hoping to be more different, for example I don't want anything to do with the whole time pulse business.  Unless absolutely impossible my game does try to actually update anything until the player wants to look at it.

[alex_brunius]

Keep up the good work!

And let us know as soon as you have a playable version and need testers 



I actually had a dream of a game just like this a few weeks ago. I had played lot's of Aurora and Lots of Kerbal space program, and in my dream I was playing a game as complex/deep as Aurora, but still modelling the extreme difficulties and energy cost of early space travel the way that KSP does.

In such a game you can get really scared at the prospect of any simple threat, because of the massive investments in energy and time needed to establish even basic colonies off Earth, not to mention in other systems. And you can also understand how isolated such colonies would be before your science advanced considerably.

[alex_brunius]

BTW any thoughts on multiplayer?

I have a very hard time forgiving games for their bad AI (all complex games tend to have "bad" AI) unless they support an option to play against a real intelligence (Multiplayer)

The more complex, the harder to code a challenging AI that doesn't cheat, and the more important support multiplayer, at least if you ask me.

[3_14159]

- Especially, which framework for the UI?
UI is from scratch (which is probably not the best of decisions but I wanted to try my hand at making a UI)

On the risk of repeating myself: The UI looks really, really neat.

- Did you really make that in under two weeks?
Yup, 2 weeks of full time work on it.   

Just wow. You're certainly programming very quick. What's currently working?

- Is your concept 'realistic' physics aside from FTL or physics plus?
Realistic physics as much as possible.    I'm thinking of doing without even FTL at the early game.    Voyages at 50% lightspeed which take a year to build up the energy required to launch and a decade of travel time.    Sort of something which could be done in real life given physics as we know it.    I think Aurora makes interstellar travel too easy.   

Well, it's kind of a design philosophy. With Aurora's resources, interstellar expansion is critical. By making resources more abundant (for example, by ignoring all of the TN minerals and instead using 'real' materials), that would be a whole other game feeling, I think.

For interplanetary travel I'm hoping to have a robust Newtonion movement system.    Today I just got the hohmann transfer working, so you give a fleet a target planet, the system calculates the time to initiate the transfer and plots the path then follows the path exactly.    I was hoping to simulate it properly but orbital mechanics is proving to be way too damn difficult (and probably too resource intensive).    The target is something close enough but not have fleets crash into planets or get lost in space for no reason (but still happen after certain events).   

What happens if you change your mind mid way through and burn the other way? Do you calculate a new orbit? What happens with projectiles?

[iemfi (OP)]

BTW any thoughts on multiplayer?

That is definitely at the back of my mind when coding but I don't want to get too far ahead of myself.  I fear a game like that is going to be hard to balance for PvP as well.  But as you say an intelligent AI is probably going to be much harder than that.  I'm personally ok with games with cheating AIs though, especially if it's well hidden.  In a game like this there are RP justifications for big mismatches in starting resources too. 

Just wow.  You're certainly programming very quick.  What's currently working?

The system map is mostly working, just got the moving of fleets around done.  Building installations/fleets/mining is working.  Designing/saving basic modules and ships is working, but as I said I'm still at a loss on the ship tiles designing gameplay.  Right now it simply adds up all tiles of a certain type. 

What happens if you change your mind mid way through and burn the other way? Do you calculate a new orbit? What happens with projectiles?

Well you're going to have a very bad time trying to burn the other way mid way if your engines are anything similar to modern space travel tech.  Yes, the idea is to calculate a new orbit.  I think there needs to be some hand waving though.  I'm having problems with things not matching up exactly.  I think for projectiles gravity is not going to have a significant effect at all (a few meters a second of acceleration is nothing compared to a projectile going at 10% c).  I'm thinking similar system to aurora, chance to hit based on ship acceleration, projectile speed, fire control.  Haven't put much thought into combat yet.

[Konisforce]

On the Newtonian part of it, I think the twin specters are inertia and gravity.  Gravity seems (I'm sure you've thought about this, so apologies if I'm coming at it from a rudimentary perspective) the easier one to handle, since there is basically a 'right answer' for any given orbit.  So if you did want to change direction, the time to burn to zero and then accelerate to the new orbit could be hand-waved out, at least on a large time scale.  For tighter time scales / actual battles, seems that might make more difference.

Also, if there's any balance / design things I can help with, I'd love to help out.  I do a lot of homebrew strategy game design.

[iemfi (OP)]

For tighter time scales / actual battles, seems that might make more difference.   

I'm thinking for combat acceleration of the ship could just be an attribute which affects hit chances and not affect moving around.    Gravity on the other hand would have no impact on combat.   

Distance difference if ship accelerated = (0.   5 * acceleration * (projectile flight time))
So a 10m long ship with 100 m/s^2 of acceleration could be 50m away from it's expected location if the projectile took 1 second to reach it.     Meaning a 20% hit chance from evasive maneuvering.   

Hit chance = (target cross section (m)) / (Distance difference if ship accelerated)   *  (tracking speed (rad/s)) / (target angular velocity) * (target cross section (m)) / (targeting system resolution(m))

Active sensor and fire control would be combined into one unit.    Maximum range would be determined by the maximum range of the fire control, tracking speed would be the minimum of the weapon system and the sensor.    The cross section would be the longest length of the target ship.    Problem with this is that ships could end up all being circles.    Is this all too complicated for gameplay?

Damage I think is where the big departure from real physics would be.   Damage would ignore speed and only depend on the size of the projectile.   Because in real life a ravioli going fast enough could probably blow up a battleship, which wouldn't be that great for game play I imagine.  .  .   Another reason would be to preserve the game play decision between whether a player wants more damage or more accuracy.   The damage is going to be simulated tile by tile.   For example a laser would hit the ship from a random angle and start destroying tiles in a straight line. 

[alex_brunius]

Distance difference if ship accelerated = (0.   5 * acceleration * (projectile flight time))
So a 10m long ship with 100 m/s^2 of acceleration could be 50m away from it's expected location if the projectile took 1 second to reach it.     Meaning a 20% hit chance from evasive maneuvering.    

Hit chance = (target cross section (m)) / (Distance difference if ship accelerated)   *  (tracking speed (rad/s)) / (target angular velocity) * (target cross section (m)) / (targeting system resolution(m))

Actually if your going for full realism you would want to use the target cross section in m^2 (for example 10x10 = 100m^2) / area of circle of possible locations 50m^2*3.14 = 7854 m^2.
This would give you a hit chance of around 1.3%.

[iemfi (OP)]

Hmm, problem is that all ships are flat like pancakes since the ship design is purely 2D.  I guess we could just have all the ships 1m tall and calculate hit chance based on area of circle but I think that would make the numbers come up all strange?

[iemfi (OP)]

Anyone good at orbital mechanics? I have hohmann transfer working between planets with circular orbits.  The problem it does not work with planets/asteroids with highly eccentric orbits.  Google doesn't seem to be much help.  What method(s) are used for transferring between planets with eccentric orbits?

[Nathan_]

Anyone good at orbital mechanics? I have hohmann transfer working between planets with circular orbits.  The problem it does not work with planets/asteroids with highly eccentric orbits.  Google doesn't seem to be much help.  What method(s) are used for transferring between planets with eccentric orbits?

While there is probably a way to do the math, you could simply fluff it that the player needs more powerful engines to travel to such objects.

[iemfi (OP)]

While there is probably a way to do the math, you could simply fluff it that the player needs more powerful engines to travel to such objects. 

Hmm, yes I guess that's always an option.   I have the orbit transfers working for the more circular orbits working nicely though (they're plotted by just adding the correct velocity then simulating the position based on interaction with gravity), so it seems like such a pity . 

I've uploaded the game in it's current state.   I'd be very grateful if you guys could take a look and help shape it.   You'll need XNA hxxp: www.  microsoft.  com/en-us/download/details.  aspx?id=20914 if you don't already have it installed.   

hxxp: www. mediafire. com/?khbu9g646h5ebjd

Whats working:
System map: Can move fleet around (click on target planet), give auto survey order to fleet (auto survey should automatically refuel ship when required but I think it's not working right).   Hold down right click to pan, mouse wheel to zoom. 
Industry: Mining and production are working, can only do one thing for shipyards (build ship). 
Ship / Module design: Mostly working.   Need ideas on game play here though. 
I'll use a railgun module as an example:
Projectile damage: Number of railgun core modules. 
Projectile Speed: Number of accelerator modules. 
I'm thinking terminals would provide a 100% increase on each of the surrounding tiles.   Terminals need to be manned by crew, and crew need armour between them and the outside of the ship.   This isn't implemented yet.   I'm worried it won't lead to very fun game play.   Any ideas on a fun system which requires thought in design while still allowing room for creativity would be much appreciated.   So far this whole aspect has been very disappointing.   Also I'm drowning in a huge mess of numbers, I think I really need to plan it out in excel first or something.   

Events: Show contact found/loss and combat messages.   I put ships on all the other planets to test the hit chance calculation, ships don't actually take damage yet though.   The locations/velocities of ships are calculated down to the 100 nano-second interval.   It's all pretty pointless now though since the ships just orbit in a circle.   I'm thinking there needs to be a tactical view of battles or something. 

Default ship/planet/system names are from text files in the Data folder (right now they're all set to a list of star names I found).   If anyone likes to muck around with names I'd be grateful for any lists to use. 

[Nathan_]

Hmm, it doesn't like my 7800 GTX card, so I can't really help see how it works beyond that.