Game Distance Units

[jmercer]

Star lane length could be a function of the real-space distance, but it doesn't have to be. It all depends on the algorithm used to create the galaxy. For the AI, it would only understand the total "length" and the connection with the visual view of the galaxy doesn't really matter.

What if you allow 3 tuning parameters, so that starlane distance is calculated as

d = B + I * "real distance" + A N(0,1)

where B is some base distance, I is impedance and A scales some normal (gaussian) distribution with mean 0 and stddev 1.

This allows eleazar's view of the galaxy, (B=1, I=0, A=0) and the more traditional views (B=0, I=1, A=0) as well as more noisy starlanes (B=2, I=0.2, A=4).

These parameters could either be explicitly presented to the player or have a group of preset values hidden behind syntactic fluff in the galaxy creation area.

Yes, I know this formula can give negative distances. Minor problem, easily fixed, no need to discuss.

A note on the visual representation of impedance, I think impedance should be shown as a pinching of the center of the lane rather than a narrow overall lane.

[eleazar]

Star lane length could be a function of the real-space distance, but it doesn't have to be. It all depends on the algorithm used to create the galaxy. For the AI, it would only understand the total "length" and the connection with the visual view of the galaxy doesn't really matter.

True, but a good game will have rules comprehensible to human and AI, and a visual display readily comprehensible to humans.

A note on the visual representation of impedance, I think impedance should be shown as a pinching of the center of the lane rather than a narrow overall lane.

Impedance adds no value to the game in any galactic scenario, it only adds complexity. Length on it's own provides all the same gameplay value of varying travel times, without the need of a second, less tangible, and harder to display variable.

[jmercer]

I'm not sold that impedance would offer no gameplay value.

If it were modifiable in-game, you could use it in a similar manner as building walls. You could increase the impedance of all starlanes connecting your empire with the rest of the galaxy. Now you don't need as large a fleet to guard your territory since it will take much longer for an enemy to arrive, allowing you to mass your fleet to meet them. Or perhaps you have increased the effective distance that enemy supply lines cannot support attacking troops. What if I just developed a new enemy and I have no way to counter him for a hundred turns, "walls" might be my salvation.

You might also be able to increase the impedance inside another player's empire which would increase the overhead of internal trade since colonies are now effectively further away.

Perhaps you can strengthen starlanes, reduce impedance. Why not let people research that tech rather than researching faster (potentially more expensive) engine techs.

What if starlane engines effect impedance? Perhaps the bargain basement MKII engine tears up the starlane (to some lower bound, 1/2 the original?), standard MKII do no damage and Enhanced improves a lane up to its original value, or 2x.

I think this has a pretty strong analogy with road building. In civ type games, I'm happy to get the horse back riding tech and the improved road/highway/rail tech.

We could dismiss this idea right now, but I think we'd be ignoring the strategic and tactical opportunities that this would present.

[marhawkman]

Okay..... It's not that Impedance is BAD. It's that having it as the primary reason for travel time is stupid. K'?

[jmercer]

Alright, this will be my final post on the subject. I was using the word impedance as a catch-all term to describe the quality of the starlane.

If starlanes exist in realspace the travel time would be a function of some base distance modified by the quality of the "road." The game currently supports 2 qualities of roads: connected and unconnected systems.

If the starlanes exist outside of realspace then I don't see why there has to be a connection with distance. I believe that from a gameplay stance that a strong connection should exist but under this type of starlane I see no reason why a connection can't be made across the galaxy with very short travel time, as long as this is a very rare situation.

In either of these interpretations of a starlane I think that its quality can be considered. If quality exists, it should have an associated GUI element. If users can change quality, this gives them the smallest amount of control over the terrain, even if it's just changing the weights in a weighted graph.

Many other strategy games allow the user to alter the terrain. As far a I know, the MoO series did not. Do we cherrypick only features of the MoO series or do we consider a larger set? The primary argument against quality of starlanes will be that it will "increase complexity without adding value to the game." I look at civ and say, "roads and improved roads were a good thing. To me, they added value." Is there a one to one between civ and moo? Definitely not. But, there are many transferable ideas, this may be one of them.

[Muside]

Many other strategy games allow the user to alter the terrain. As far a I know, the MoO series did not. Do we cherrypick only features of the MoO series or do we consider a larger set? The primary argument against quality of starlanes will be that it will "increase complexity without adding value to the game." I look at civ and say, "roads and improved roads were a good thing. To me, they added value." Is there a one to one between civ and moo? Definitely not. But, there are many transferable ideas, this may be one of them.

Perhaps it'd be interesting for you that our german open source project uses exactly this approach. We plan to use starlanes that can be built and upgraded by the player, and recent tests show that it improves the playing experience.

[marhawkman]

Ah... I see. That DOES make sense to me. The SE games have it that way. Although they don't have time to traverse. They just have you enter/exit instantly. The thing here is that what would be the correlation between techs for improving starlanes and techs for improving ship engines? IF you have both you need to take both into account when balancing.

Alright, this will be my final post on the subject. I was using the word impedance as a catch-all term to describe the quality of the starlane.

you wish. Please, do discuss. Just don't get upset if we don't understand.

I like the idea of making it so that the map is less uniform. I get the feeling the coders don't though.

[jmercer]

I was intending that to be my last post on the subject if it didn't sway any views. I don't want to beat an unpopular idea to death, which unfortunately, can sometimes be my style. Apologies to anyone who was upset.

marhawkman, you do have a point that there would be balancing issues. There will be balancing issues anyway for the amount of research required to get Engine Tech II. I suppose the developers will make some educated guess, we playtest it a bit and give feedback for tuning. If starlanes can be improved, I'd say balancing would happen in much the same way.

I'm still not sure if spacelanes are real lanes in space or "worm holes." If they are the latter, then one way to cleanly visualize a spacelane from one side of the map to the other is to draw it as normal but have it fade to complete transparency a few centimetres (3 game distance units or something) from the source and destination unless you mouse-over those visible areas in which case the rest of the lane would fade in.

Also, if they are wormholes, have people thought about one-way travel or periodically appearing/disappearing. The quality could fluctuate in some sinusoidal manner as well. I think these ideas actually do complicate things without enough added gameplay but I'll throw them out there for others to make their arguments.

Muside, sounds interesting. Does your project have a website? I haven't used my German for years, but I could probably understand a website.

[marhawkman]

I was intending that to be my last post on the subject if it didn't sway any views. I don't want to beat an unpopular idea to death, which unfortunately, can sometimes be my style. Apologies to anyone who was upset.

marhawkman, you do have a point that there would be balancing issues. There will be balancing issues anyway for the amount of research required to get Engine Tech II. I suppose the developers will make some educated guess, we playtest it a bit and give feedback for tuning. If starlanes can be improved, I'd say balancing would happen in much the same way.

I'm still not sure if spacelanes are real lanes in space or "worm holes." If they are the latter, then one way to cleanly visualize a spacelane from one side of the map to the other is to draw it as normal but have it fade to complete transparency a few centimetres (3 game distance units or something) from the source and destination unless you mouse-over those visible areas in which case the rest of the lane would fade in.

Also, if they are wormholes, have people thought about one-way travel or periodically appearing/disappearing. The quality could fluctuate in some sinusoidal manner as well. I think these ideas actually do complicate things without enough added gameplay but I'll throw them out there for others to make their arguments.

It WOULD add gameplay value. A starlane that has a very short travel time would be sought after for it's strategic importance. Thosee with longer times would be largely avoided.

I do think it's a good Idea. It's gonna need a bit of brainwork though.

[jmercer]

I apologize, I meant that a starlane with changing quality or availability over time might not add to gameplay. In this case a starlane would have some effective "distance" of 10 on one turn then 12, then 15, 17, 20, 17, 15, 12, 10 and so on. Kind of like it gets stronger and weaker in some cycle.

I wouldn't want to see all starlanes have this feature, but perhaps it would be ok for very long (space distance), high quality (fast) starlane which should be rare. Special rules for these entities might not be bad. We could call them ancient starlanes, perhaps they are connected to specials on planets, throw in some "mysterious race of elders" type of back story and have their quality linked to the position of the planet around its sun, or something of even longer cycle length.

If planetary efficiency is linked to its travel time from the homeworld then the starlanes wouldn't add any complication. Even starlanes with bizare rules. Each turn you calculate the time it takes your fastest ship (since diplomats will travel 1st class) to move from the homeworld to a colony. Special rules are implicitly taken care of.

[marhawkman]

thatactually DOES sound like agood idea. but if used as a rare, special type of starlane.

[eleazar]

• How is the game supposed to clearly display that information to the player?

• A principle of game design is that the game shouldn't make the player do annoying or tedious tasks to succeed. "Tedious" or "Annoying" may be defined differently for various players in the contexts of various games. But the necessity of checking a "starlane fluctuation schedule" to find the optimal time to travel between two stars, IMHO clearly has all the drama and excitement of reading a bus schedule. Making the game more complicated does not equal making it better.

• I suggest you (or anyone who wants to generate ideas) read the 0.2 and 0.3 specifications to get an idea of what decisions have already been made, and more importantly an impression of which new ideas are congruent with what already exists.

• It's quite likely that latter-game tech will allow the creation and destruction of star-lanes of the same sort that the galaxy is generated with.

In reguards to:

I'm still not sure if spacelanes are real lanes in space or "worm holes." If they are the latter, then one way to cleanly visualize a spacelane from one side of the map to the other...

...Some explanation of how starlanes are supposed to work: (emphasis mine)

The most useful / desirable shortcut would be between two adjacent systems that could be, but just happen to be not connected... so that instead of a one or two turn movement, you've got to go around a big loop and it takes 8 or 10 turns to get where you're going. If you want to get somewehre half way across the map, you can just make several shorter lanes. This wouldn't be a significant increase in travel time, and the whole point of the lanes is to make some meaningful geometry to the map so you can have front lines and blockades and such... if we start allowing lanes anywhere, we're sort of back to effectively having no lanes (or hidden lanes between all stars) which is contrary to this design point.

There was some extended discussion about the relative merits of having lanes that cross other lanes and whether this looked good / was confusing / hard to make sense of. Initially I thought it wouldn't be a problem, but after trying it out, I changed my mind and now agree that it's best to avoid crossed lanes. This would extend to player created lanes as well.

[marhawkman]

• How is the game supposed to clearly display that information to the player?

• A principle of game design is that the game shouldn't make the player do annoying or tedious tasks to succeed. "Tedious" or "Annoying" may be defined differently for various players in the contexts of various games. But the necessity of checking a "starlane fluctuation schedule" to find the optimal time to travel between two stars, IMHO clearly has all the drama and excitement of reading a bus schedule. Making the game more complicated does not equal making it better.

thus the reason for making it a rare, special type of starlane. It's there to annoy the player, not to really make the game exciting. Besides I'd make it so you can fix it later.

[jmercer]

Thanks eleazar, this sheds some light on the issue for me.

Though, it doesn't actually define the nature of starlanes, it seems to lean toward real lanes in real space rather than worm holes. If Delaunay Triangulation is used as Geoff mentioned, it should reduce the arbitrary nature of connected systems that I had feared.

I find starlanes an unsatisfying solution to the problem of trying to setup blockages, defending borders etc if starlanes exist in real space. If we keep the exact same routines for generating the connection graph and relabel realspace with wormholes, this makes fixed starlanes much more believable. The travel time can still be related to the real space distance just that you need to move through these structures to get there in a timely manner. Perhaps late game techs can improve engines to the point that wormholes are not needed and you can do "flt off-roading" which is effectively making new starlanes in the real-space interpretation of them.

eleazar, you had mentioned that a variable quality starlane would introduce tedium or annoyance. Did you make the same argument against the day/night cycle in Wesnoth? I think that in that game it is an interesting feature that allows for richer strategy and tactics. As marhawkman mentioned, a fix later on the game could stabilize the route. It'd say that variable quality could be viewed as a punishment that you would need to deal with for having access to this structure.

I've made 2 arguments here, one saying that variable quality can lead to strategy and one saying its a punishment which can be rectified, they seems to be at odds with each other but I think it depends on the player. Some people like bus schedules, others like being able to improve terrain. I think that even if these can't be improved (by removing variability), they provide enough benefits to be useful, in the ultralong/highspeed case.

Just to be absolutely clear, these are my views

-The time required to pass through a starlane should (almost always) be proportional to the on-screen distance.
-Starlanes can have an associated quality which can scale this distance. Quality is presented in an obvious manner to the user.

I think we may have hijacked this thread a bit and steered it off of its intended subject. Perhaps we should move this conversation to a new one.

[eleazar]

Though, it doesn't actually define the nature of starlanes, it seems to lean toward real lanes in real space rather than worm holes.

Realspace is not a very good explanation of the starlanes as they exist in this game. For instance, Ships entering opposite ends of a lane will meet each other. And it's only possible to travel between stars via starlanes.

Perhaps late game techs can improve engines to the point that wormholes are not needed and you can do "flt off-roading" which is effectively making new starlanes in the real-space interpretation of them.

NO. as i've mentioned before in this thread, no-offroading is a foundational decision of this game. And IMHO it's a good one. Search for "off-roading" and read the threads before assuming that you have something new to add to the discussion.

eleazar, you had mentioned that a variable quality starlane would introduce tedium or annoyance. Did you make the same argument against the day/night cycle in Wesnoth? I think that in that game it is an interesting feature that allows for richer strategy and tactics. As marhawkman mentioned, a fix later on the game could stabilize the route. It'd say that variable quality could be viewed as a punishment that you would need to deal with for having access to this structure.

ToD (Time of Day) was introduced before i became involved with Wesnoth. However, i have no objection to it, because it's very KISS. Everything a player will ever need to know about ToD can be learned in a few minutes 1) the intuitive schedule, and the bonus/malus applied to units of the three alignments. In most cases a player could correctly guess which units belong to which alignment. And a glance at the screen makes it obvious which ToD is currently in effect.

ToD in Wesnoth provides a high amount of tactical interest, for a very low investment of rule-learning.


However, you are proposing something with a much more complicated schedule, (not quickly memorizable) that has no obvious or intuitive way to be depicted graphically, and would require most players to frequently refer to some info panel to discover the current and future state of a star-lane.

-The time required to pass through a starlane should (almost always) be proportional to the on-screen distance.

So if it's undesirable for nearly all starlanes to have some sort of variability, it begs the question: Why is it necessary to have additional game rule, info panels, GUI conventions, documentation for something that rarely occurs, and the player won't expect?