Home World Forum
Stars! AutoHost web forums

Jump to Stars! AutoHost


 
 
Home » Stars! Clones, Extensions, Modding » Stars! Nova - Open Discussion and Help » Nova - How should it be different to Stars!
Re: Nova - How should it be different to Stars! Thu, 25 January 2007 00:01 Go to previous messageGo to previous message
Iconian is currently offline Iconian

 
Officer Cadet 2nd Year

Messages: 233
Registered: January 2006
Location: Nevada, USA
Here's something from the RGCS I found a couple weeks ago, from here. Jim Lane explains what he plans on doing with his version of Stars!

Quote:

>> >We all wish you get a Stars3beta out somewhere "soon"...
>> Me, too. But I'm not the only person who could write something like
>I'm curious: Just what kind of improvements can we expect from the new
>version? More design slots? Bigger universes? Revamped battleboard? Pop
>redistributing itself to colonies?


Since it isn't done, the feature set isn't fixed. But as for general
directions...

No required limits on number of game objects (fleets, ship, ship
designs, mine fields, races per game, start systems, etc.), but the
ability to place limits on a particular game during set up, if
desired. Stars! 2 is limited by the Win16 API. This version assumes a
32 bit (or better) CPU.


Deeper race design, using something similar (but more flexible than)
SN's idea of physiologies and social models.


Hull designs are not specified by the game engine. Like other limits,
they can be fixed by the scenario, or it can be made possible to
design new hulls during the game.


More variation in galaxy creation, similar to SN.


More variation in, and more meaningful, winning conditions. Doing this
really means setting up a fairly sophisticated system for triggering
game events in general, with one possible action for a trigger being
to declare a victor. A key point that goes along with this is that it
is possible to create non-zero-sum games.


The possibility for more than one universe per game, possibly with
different physics. Might not be fully implemented at first, but the
game engine does not preclude implementing it later (I hope).


Completely revamped tech. Separate FTL and normal space engines, very
different types of offense and defense, etc. All of this is
implemented in terms of lower-level primitives that can be recombined
in a variety of ways. For example, weapons inherently have certain
characteristics: range, damage, mass, various costs, interactions with
other tech (i.e. defenses),rate of fire, accuracy, velocity,
maneuverability, local / remote tactical intelligence, etc. At a game
engine level, a "beam weapon" is a manufactured device that can
perform an action (create a beam) at a given rate while consuming
defined resources. The beam itself, once it has been fired, is a
separate game object that has a constant velocity, a constant
direction of travel (inherited from the projector that created it), no
tactical AI, and has particular effectiveness formulas vs. different
kinds of defenses. Each aspect of the projector and the beam can be
separately influenced by the technological knowledge of the
projector's builders, and by it's operators. All of these
interrelationships are defined by RDL, and can be very simple, very
complex, or anywhere in between.


A more sophisticated version of SN's RDL, more completely integrated
into the game engine.


The ability to play at more than one level of abstraction. In a
beginner's game, many different decisions are coupled together for
simplicity. . An advanced game gives the players direct control over
the underlying game controls. This is essentially the same idea as
"governors" used in many other games, but the scripts are in RDL 2 and
are available for modding. There will be at least two levels of
abstraction, and perhaps three.


A deeper way to specify orders to units, based more directly on a
real-life military model. Units get doctrines that are triggered by
various circumstances. The doctrines, the triggers, and the actions
taken to implement a doctrine are all in RDL, and can be modded. The
tactical AI is based on a flocking model; advanced players have
control over the flocking rules.


The battle engine is similar in concept to SN's: distances are
measured in meters, not "squares", movement is Newtonian.


A more complete economics model, allowing for trading without
wolf-lamb, ship scraping, etc. I'm not sure I've figured out how to do
it yet, but I'd like it to be possible to defeat an opponent without
ever firing a shot, by economic warfare.


Better game security, based on published crypto techniques.


Better support for Autohost'ing, although not as complete as SN had
planned.


---


There are some things that I have no plans to change:


One species per stellar object (planetary system, etc.).


The basic concepts of game set up, race design, ship design, fleet
design, etc. Players must have enough degrees of freedom to support a
Stars! like complexity of game play.


The basic idea of interacting compound interest formulas, although the
details are different.


No game-enforced contracts between players. Backstabbing will always
be possible.


The AIs have exactly the same information available to them as the
players. It is not possible for them to cheat. (As a result, there
will be a handicapping system, which can be applied to any player in
the game.)


One of the key, and very difficult, ideas is that there should never
be any single winning formula. Stars! 2 was not quite perfect at this,
but it was extraordinarily good. The intent is to create
sissors-paper-stone loops in every area of the game.


2D, not 3D. There is no gain in depth of gameplay in a 3D universe,
and the UI is much harder.


Turn based 4X. Jeff and I had discussed the idea of a variable
simulation period; a game could be set up to play at different amounts
of game time per turn, possibly varying it over the course of the
game. Using a very short simulation clock, one could approximate an
RTS, but the game engine remains fundamentally turn based.


---


I'm sure I've left out a lot. I'm willing to consider alternate ideas,
but please understand that I have a vision for this game, and there
are some popular ideas that don't fit that vision. I'm working on the
game that I want to play, and hope that others will enjoy it, too.


Jim Lane



That's a long list, and I could also add a lot of my own ideas. However, if I could put my finger on the one single improvement to Stars! that woudl best for the game, it would be: Micro Management. From what I understand, Stars! has lost more players for the huge amount of MM than for any other reason. I'd suggest finding ways to script certain behaviors into the game. Take a look at Space Empires. It had an OK system. If you want to know what I think would be best in terms of MM though, then take a look at Black and White. In that game you get a "creature" that you are able to train in certain behaviors. Basically, the game facilitates a path for the player to create an AI, easily. So, just set up a user interface where the player can attach certain "weights" to different behaviors, like Black and White does. It means making big improvements to the AI. It means that you put the player in charge of AI's. But really, how is that much different from Stars! in its current form? The player has access to all sorts of mini AI's, or programs, or scripts: production queues, waypoint orders, research, ship controls, etc.

I could go on, but that should probaby do.


[Updated on: Thu, 25 January 2007 00:02]




Yeah, bread too.

Don't Let the Stars! Fade Away

Report message to a moderator

 
Read Message icon5.gif
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message icon9.gif
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Next Topic: Nova
Goto Forum:
  


Current Time: Wed May 01 14:15:59 EDT 2024