Page 4 of 6

Posted: Mon Oct 30, 2006 10:32 am
by Moriarty
I believe the primary criticism to "realspace FTL" is that it'll make the AI harder.

Posted: Mon Oct 30, 2006 1:31 pm
by eleazar
Moriarty wrote:I believe the primary criticism to "realspace FTL" is that it'll make the AI harder.
My understanding is that this approach will probably a significant benefit in AI programming. At Battle for Wesnoth, new features must pass the "can we easily make the AI use this well?" test. I believe having an AI-friendly feature set is just as important to FO, if not more so.

AI coding is hard— lots of commercial games don't get it quite right. However, it's even more important for a FOSS game, since these kind of projects get developed over longer timespans. The quality of the AI is a large part of what provides a strategy game's replay value. Having people that are interested enough to replay many-many times is neccesary to keep a project going. A developer is willing to delay gratification, but the sooner the game becomes fun, and the longer it stays fun, the healthier the contributing community will be.

Posted: Wed Dec 20, 2006 7:03 pm
by eleazar
Actually there's another option. It's the most KISS, but is a little less intuitive.

In spite of the fact that they are currently depicted as different lengths, StarLanes per Year could be the unit to measure ship speed. Once the player gets over the mental hump of understanding that each starlane takes the same amount of time for the same ship to travel, it becomes trivial to estimate transit time for your fleets.

I suppose ship engine tech will start at around .25 SLpY, and eventually get to multiple SLpY.

For this, it would be advisable to set the min and max Starlane length values near each other, but they shouldn't have to be precisely equal.

The casualty would be the small strategic value that varied StarLane lenght provides.


I'm not saying this is the best approach, but that it's worthy of consideration.


EDIT:
And in response to the original question, "how should straight-line distances be measured?" I suggest they shouldn't be— even if SLpY is not the measurement of ship speed.

The most compelling reason to have a measurement of non-starlane distance is buildings with limited effect range. But i would propose we not use that option. (the code supports off-roading, or used to, but that doesn't mean we have to build the game around it) Is there really a significant strategic difference in a building that effects a range of starlane-nodes over one that has a linear distance effect? I can't think of a compelling (non-realism) reason to use linear distance effects instead. Let's keep the rules, effects, and UI simple.

Posted: Wed Dec 20, 2006 8:10 pm
by SowerCleaver
eleazar wrote:Once the player gets over the mental hump of understanding that each starlane takes the same amount of time for the same ship to travel, it becomes trivial to estimate transit time for your fleets.
I didn't know this was the case. This would mean that the "strategic" distance of two planets could be very different from the "real" distance between them.

Maybe we can vary this concept slightly - there are a few different degrees of "impidence" in star lanes, meaning some starlanes are slower to travel than others. This "impidence" would have no relationship with the "real" distance of planets, and therefore if two starlanes have the same "impidence" then they will take the same amount of time to traverse with the same ship.

The possibility of offroad FTL in the late game can add more strategic depth to this approach. Once the offroad FTL becomes available, suddenly the "strategic" distance of planets will change significantly. A planet deep inside enemy's empire without any convenient starlanes to their systems previously will now become a beachhead for invasion. Depending on the preference as to such sudden strategic shift, we can choose to make the offroad FTL as a very hard-to-get wonder available for one race. Barring any AI coding difficulties, I would think this is something that may benefit the game experience.

foo

Posted: Wed Dec 20, 2006 9:42 pm
by tzlaine
SowerCleaver wrote:
eleazar wrote:Once the player gets over the mental hump of understanding that each starlane takes the same amount of time for the same ship to travel, it becomes trivial to estimate transit time for your fleets.
I didn't know this was the case. This would mean that the "strategic" distance of two planets could be very different from the "real" distance between them.
This isn't the case in the current and planned design. He's proposing this, iiuc. In the current design, ship speed is in terms of units of distance (no scale is implied), and the time it takes to travel a starlane is starlane's length in units of distance / speed.

Posted: Thu Dec 21, 2006 12:05 am
by marhawkman
Just in case.... I'll try my explanation.

Basically he wants to display ship speed as a function of how long it takes to get where it's going rather than how far it actually goes.

Posted: Thu Dec 21, 2006 1:58 am
by eleazar
tzlaine understands what i'm not saying, but it's not clear from the posts that my point got across. I'll try rephrasing.

I'm suggesting this idea for consideration. (i'm not yet convinced it's the best) That each Starlane be considered as equally long— for the purposes of ship travel. Thus the starlane itself become the basis of the unit of measurment for ship speed: StarLanes per Year (SLpY or something to that effect)

Thus a ship that can go .25 SLpY can travel between any two adjacent stars in 4 turns. At 1 SLpY a ship can go between any two adjacent stars in a single turn.

The point is that any other measurement of speed (light years, parsecs per turn, Warp factors, ect.) all need to be translated before the player knows what they mean in the actual game. "StarLanes per Year" needs no interpretation to understand how quickly a ship can go where you send it.

Posted: Thu Dec 21, 2006 5:01 am
by utilae
eleazar wrote: "StarLanes per Year" needs no interpretation to understand how quickly a ship can go where you send it.
Hardly. Each starlane is a different length. Assuming the smallest starlane is 1 SLPY and the player only has long starlanes, then how is the player gonna know how these units, when they don't even have the base length to compare it to.

The only solution to this is to make each starlane a ruler, eg if little marks along the starlane, so a starlane that is 4 starlanes long will have 3 marks.

Posted: Thu Dec 21, 2006 1:47 pm
by eleazar
utilae wrote:
eleazar wrote: "StarLanes per Year" needs no interpretation to understand how quickly a ship can go where you send it.
Hardly. Each starlane is a different length. Assuming the smallest starlane is 1 SLPY and the player only has long starlanes, then how is the player gonna know how these units, when they don't even have the base length to compare it to.
I've stated the idea as clearly as i know how, and yet you've totally misunderstood. If you care about the question, try reading my last 2 posts again.
utilae wrote:The only solution to this is to make each starlane a ruler, eg if little marks along the starlane, so a starlane that is 4 starlanes long will have 3 marks.
Have you ever played a space game that did that? I haven't, so obviously there is another solution.

Posted: Thu Dec 21, 2006 7:18 pm
by marhawkman
I agree with Utilae. Giving all starlanes the same arbitrary amount of time to traverse seems to be a bad idea. It gives you a total flatline as far as tactical importance(of the starlanes) is concerned. And it doesn't really add anything to the game. Now doing the opposite and measuring the length of the starlanes by years to traverse using some sort of standardized drive seems to be less of an issue, but there's still the problem of how to word the improved drives.

Posted: Thu Dec 21, 2006 8:09 pm
by SowerCleaver
utilae wrote: The only solution to this is to make each starlane a ruler, eg if little marks along the starlane, so a starlane that is 4 starlanes long will have 3 marks.
I doubt that would be very helpful or aesthetically pleasing. First of all, if two stars are close by, the marks would be crowded together, making it hardly discernible. Also, as engine tech develops, a ship will move at a fractionof, or multiples of. "1 starlane distance".
marhawkman wrote: Giving all starlanes the same arbitrary amount of time to traverse seems to be a bad idea. It gives you a total flatline as far as tactical importance(of the starlanes) is concerned.
A valid but surmountable point. Ways to reduce such strategic flatline are (1) adding the "impidence" concept (as I laid out in my previous post) and/or (2) introducing FTL offroading earlier (obviously slower than the starlane travel but not too slow). I still think severing the starlane distance from the "real" distance is an interesting idea, and it could be more fun and sci-fi.

Posted: Thu Dec 21, 2006 11:32 pm
by eleazar
SowerCleaver wrote:
marhawkman wrote: Giving all starlanes the same arbitrary amount of time to traverse seems to be a bad idea. It gives you a total flatline as far as tactical importance(of the starlanes) is concerned.
A valid but surmountable point. Ways to reduce such strategic flatline are (1) adding the "impidence" concept (as I laid out in my previous post) and/or (2) introducing FTL offroading earlier (obviously slower than the starlane travel but not too slow). I still think severing the starlane distance from the "real" distance is an interesting idea, and it could be more fun and sci-fi.
(1) The whole point of equalizing the length of a starlane is to make it a viable measurement of ship speed. If you throw in the concept of "impedance", that benefit is lost, and variable star-lane travel time is achieved in a much less intuitive manner. Purposely making things less obvious, is not a KISS definition of "fun".

(2) No Offroading is a foundational design decision of this game, as has been mentioned at the top of this page.

Posted: Tue Jan 30, 2007 6:31 am
by skdiw
just add in some fluff to explain why it's 80. say light travels slowly in that universe. or say FO space is warped and got a lot of wrinkles so you have to travel 80. in terms of game function, player just want to get there in one turn :p. in terms of units, just pick two letters... GM or something so this starlane has 80 GM and your warp engine travels 10 GM / turn so you get there in 8 turns.

Posted: Sun Feb 04, 2007 11:24 pm
by jmercer
I don't mind the introduction of impedance to a starlane, as long as it's clearly presented. Currently, it seems that starlanes are straight lines between systems. Why not give them more substance, make them look like pipes, or pipe wireframes. This would allow them to look fat or thin, indicating the impedance. This also allows galaxy disasters or local weapons which alters impedance levels.

Posted: Mon Feb 05, 2007 1:14 am
by marhawkman
jmercer wrote:I don't mind the introduction of impedance to a starlane, as long as it's clearly presented. Currently, it seems that starlanes are straight lines between systems. Why not give them more substance, make them look like pipes, or pipe wireframes. This would allow them to look fat or thin, indicating the impedance. This also allows galaxy disasters or local weapons which alters impedance levels.
this isn't really a bad idea. But I still think it'd be best to have the base impedance be a function of the length of the warp point.