Design: Starlanes

Message

#76 Post by **Tyreth** » Thu Jul 31, 2003 1:10 am

mr_ed wrote:It's not so much an adjustable one with a #define.

The problem happens when we get engine tech that improves the speed of off-road mid-game. That's what we're discussing here..

From what I remember, the AI group was going to set their heuristics based on an unchangeable (mid-game) ratio.

We probably won't define offroad speed in a #define, but in an xml file.

What tsev said makes sense, the calculation of offroad travel is trivial.

Nightfish · #77 Post by **Nightfish** » Thu Jul 31, 2003 8:37 am

Alright, if we don't have to keep that ratio constant for the sake of the AI, do we want to keep it that way for the sake of gameplay? I'd prefer it that way because it gives you something definite. Offroading always sucks at the same level so I don't have to reevalute the situation with every new tech I get. I won't have to check my enemies techs every other turn to see if starlanes still act as choke points or if they are more or less worthless already. The main idea behind starlanes was to create some sort of "terrain", do we want to remove that in mid- to late game?

Thumper · #78 Post by **Thumper** » Thu Jul 31, 2003 10:01 am

I just had a thought... or two.

1. Are starelanes created automatically or only after a scout has successfully mapped the route between two stars?

2. Do starlanes exist for a race that hasn't mapped the route yet even tho it has been mapped by another race? This should add some real complexity to the starlane coding if the answer is NO.

3. Lets say that race X is mapping the route from G to H. If there is a force in system H that repelles the scout. Has X successfully mapped the starlane?

Thumper

Nightfish · #79 Post by **Nightfish** » Thu Jul 31, 2003 11:09 am

1. Automatically.

2 and 3 don't apply. If you could map the routes yourself the terrain effect we're trying to create would be completely negated and we might as well not have starlanes.

Thumper · #80 Post by **Thumper** » Thu Jul 31, 2003 6:40 pm

Nightfish wrote:1. Automatically.

2 and 3 don't apply. If you could map the routes yourself the terrain effect we're trying to create would be completely negated and we might as well not have starlanes.

I like the idea of explore before use. But its not a show stopper.

Thumper

Nightfish · #81 Post by **Nightfish** » Thu Jul 31, 2003 6:47 pm

It simply makes no sense. If I can map space myself and thus create my own starlanes I can essentially connect every star to all other stars. This is a) annoying micromanagement and b) it destroys the purpose of starlanes.

tsev · #82 Post by **tsev** » Thu Jul 31, 2003 6:56 pm

We went over the fact that starlanes would be permanent in the very early days of freeorion...i think mainly because we wanted to give terrain and shape to the galaxy. Think of starlanes as like mountains and deserts in Civ, they're always there, so live with it. (note I said "Civ" and not Civ 2 or 3....you can change terrain in those versions a la engineers and workers)

Staying with the Civ theme....Don't think of starlanes as "roads"

Nightfish · #83 Post by **Nightfish** » Thu Jul 31, 2003 6:58 pm

Two more things we need to know about Starlanes, then I guess this one is ready for public review:

How long are Starlanes? Can they span the entire galaxy or do they connect to the nearest stars first and then the next nearest, and so on.

Is the number of Starlanes per system completely random, meaning that a system with only one starlane can occur everywhere and not just in the outer rim?

If you can think of any other things we need to know about starlanes, feel free to add them.

Thumper · #84 Post by **Thumper** » Thu Jul 31, 2003 7:01 pm

Nightfish wrote:It simply makes no sense. If I can map space myself and thus create my own starlanes I can essentially connect every star to all other stars. This is a) annoying micromanagement and b) it destroys the purpose of starlanes.

Which brings me full circle... I, personally don't like the idea of starlanes in the first place. I think they impose an artifical restriction on the game play.

I'm a point and go type of guy. And seeing the descussion about having an option to turn off starlanes is right up my alley.

As long as speed is no longer affected by not having the starlanes.

However if they, starlanes, are a discovery during a"mapping" a new route it adds to the game in both depth and complexity.

Thumper

krum · #85 Post by **krum** » Thu Jul 31, 2003 9:17 pm

If starlanes connect only neighbouring stars, that would considerably decrease their importance as terrain, since it is easier to go to neighbouring stars anyway. I think both the length of starlanes should be bell-curved.

Maybe also the number of starlanes per system - this way it would both be easier on the AI maybe? What do the AI guys think? The length should definately be bell-curved, if non-neighbour stars starlanes won't be too much trouble for the AI.

tsev · #86 Post by **tsev** » Thu Jul 31, 2003 11:35 pm

Thumper wrote:
Nightfish wrote:It simply makes no sense. If I can map space myself and thus create my own starlanes I can essentially connect every star to all other stars. This is a) annoying micromanagement and b) it destroys the purpose of starlanes.
Which brings me full circle... I, personally don't like the idea of starlanes in the first place. I think they impose an artifical restriction on the game play.

I'm a point and go type of guy. And seeing the descussion about having an option to turn off starlanes is right up my alley. As long as speed is no longer affected by not having the starlanes.

However if they, starlanes, are a discovery during a"mapping" a new route it adds to the game in both depth and complexity.

Thumper

Then by all means, uncheck "Starlanes" in the options menu and play without them.

NightFish wrote: Two more things we need to know about Starlanes, then I guess this one is ready for public review:

How long are Starlanes? Can they span the entire galaxy or do they connect to the nearest stars first and then the next nearest, and so on.

I think we can make this a configuration option. MOO3 has a similar option where you specify starlane length and the number of them.

NightFish wrote: Is the number of Starlanes per system completely random, meaning that a system with only one starlane can occur everywhere and not just in the outer rim?

If you can think of any other things we need to know about starlanes, feel free to add them.

I think the answer to this depends on your first question. If we have starlanes only connecting close planets, then one would assume that you'd have only outer rim systems with one starlane. If they're fairly random, then I don't see why we couldn't have "dead end" systems in the core. I personally favor this second idea.

We definitely should limit the maximum length of a starlane, though. I think it would just clutter up the UI if we had starlanes going from one end of the galaxy to the other.

Thumper · #87 Post by **Thumper** » Fri Aug 01, 2003 4:20 am

tsev wrote:We definitely should limit the maximum length of a starlane, though. I think it would just clutter up the UI if we had starlanes going from one end of the galaxy to the other.

Which brings us to Worm Holes.

Should WHs be long links within a single galaxy -or- should they connect to other galaxies only. As I just thought of this multi galaxy angle I thought it might be interesting to toss it out for comment.

IFF WHs link galaxies then:

1. There should be no more than two per galaxy.
2. They should be considered prime real estate.
3. They should give the controlling race a very large financial boost. (by controlling I mean the race that controlls both ends of the WH.
4. They should be given a very high protect or capture rating by the AI.

Thumper

Nightfish · #88 Post by **Nightfish** » Fri Aug 01, 2003 6:36 am

The idea of setting an upper limit for starlanes is very good. If we made it purely random things would look really weird. One more thing I'm worried about is this: What if a starlane links A to B but C is in between those two but not linked to either of those?

Can we somehow head off weird looking combinations? How does MoO3 handle this?

Thumper · #89 Post by **Thumper** » Fri Aug 01, 2003 6:44 am

Nightfish wrote:Can we somehow head off weird looking combinations? How does MoO3 handle this?

Very randomly... and very dorkish! Two adjacent stars may have over a dozen starlanes of travel between them

Not to mention a worm hole or two.

Thumper

tsev · #90 Post by **tsev** » Fri Aug 01, 2003 11:53 am

Nightfish wrote:The idea of setting an upper limit for starlanes is very good. If we made it purely random things would look really weird. One more thing I'm worried about is this: What if a starlane links A to B but C is in between those two but not linked to either of those?

Can we somehow head off weird looking combinations? How does MoO3 handle this?

Well, to answer your second question first, the MOO3 map is 3D so if you've got a starlane that overlaps other systems, just rotate the map and you'll alleviate the problem. I thought this was too much work to see exactly the geography of the galaxy and I'm glad we decided to go with 2D for FO.

About the A to B, whilst intersecting C.......good question. Here are some thoughts:

- We can simply not allow this to happen during universe creation. We do a check to ensure that each starlane intersects with two and only two systems.

- We represent said starlanes as beziers, and curve them around the systems in question (not recommdended)

- We can represent these starlanes as a line that darkens or lightens based on how close it is to the intersected planet. Example: suppose A connects to B, but the starlane must be drawn so that it goes through C.

Please excuse this crude drawing

Code: Select all

A------------------C-----------------B
 ^              ^     ^            ^
 |              |     |            |
 -> Line starts |     | Line ends <-
    out white   |     | up white
                |     | again
   As the line <-     |
   approaches         -> As the line
   system B, it          moves away from
   begins to fade        system B, it
   to a semi-            gradually gets
   transparent           more opaque.
   color.

The first suggestion might be the best.