DESIGN: Buildings / Build Queues / Infrastructure

Past public reviews and discussions.
Message
Author
emrys
Creative Contributor
Posts: 226
Joined: Fri Oct 24, 2003 3:44 pm

#376 Post by emrys »

Question:

Now it seems that most people have accepted that one-level, infinite capacity shipyards are best avoided like the plague, what about building on planets?

As I see it there are three possibilities:

Either a) people want to do the 'things on planets are built by sacrificing infrastructure growth, not by PP' idea, so PP spend limits are not an issue anymore.

or b) people want spending limits on planets, if in this group, how should those limites be determined.

c) people don't think there is a problem with infinite spending on non-ship things, despite the selection of similar arguments to those related to infinite rate ship building that should be considered. Is anyone still in this camp?

User avatar
Geoff the Medio
Programming, Design, Admin
Posts: 13587
Joined: Wed Oct 08, 2003 1:33 am
Location: Munich

#377 Post by Geoff the Medio »

emrys wrote: b) people want spending limits on planets, if in this group, how should those limites be determined.
There are some who advocate pooled production for ships at a system shipyard, and local-only for buildings on planets. Production limits would be the amount the planet produces.
c) people don't think there is a problem with infinite spending on non-ship things, despite the selection of similar arguments to those related to infinite rate ship building that should be considered. Is anyone still in this camp?
Most of the arguments against infinite pooling for ships don't apply to non-military buildings. Planets can't move... and if they can't attack or adversely affect nearby enemy operations with area effects, then infinite pooling for buildings wouldn't be as big of a problem. As long as there are no defensive structures that can be built on a planet and unbalancing rates, the 'rarity' of buildings and the X for Y turns limit might be sufficient...

These are rather harsh limits on the functions of buildings, however... so I still tend to prefer some sort of limit for local spending on buildings.

User avatar
Geoff the Medio
Programming, Design, Admin
Posts: 13587
Joined: Wed Oct 08, 2003 1:33 am
Location: Munich

#378 Post by Geoff the Medio »

In the discussions on buildings, we've discussed shipyards, and the idea of limiting them to a certain spending rate. Also suggested was the ability/requirement to "upgrade" shipyards to increase both a) their spending cap and b) the tech level(s) of ship which they are able to build / upgrade / repair.

It seems to me that, assuming shipyards are nice and rare like we want, the shipyard could be the nexus an "upgrade tree", much like the hierarcy of buildings and units they allow in an RTS (eg. Warcraft III). Rather than just having "Tech level 5" or "Weapons Tech Level 8, Engine Tech Level 2, Hull Tech Level 6", we could have a branching tree structure of shipyard or starbase addons that allow various ship construction options.

A shipyard itself could even be an addon to a starbase.

A starbase could also have various addons not related to shipyard construction... though I worry that if "starbases" are more common than the very-rare shipyards, then the whole excess micro problem might arise. Perhaps instead, starbases could be specialized from the get-go.
-You could make a shipyards starbase, which can have lots of interesting branching tree addons (and make ships).
-You could also make an "asteroid mining station" starbase, that doesn't have any addons... (but might have a "Mining Meter" just like a planet does... affected by new tech developments, but without any focus settings to deal with).
-Other types of starbase could be thought up... surveilance post, etc.

As long as a starbase is a big expensive (to build and maintain) and time-consuming wonder-like project, this wouldn't be a big deal for micro.

... I guess I got a bit off topic there. The real point was that shipyards could have "building-like" branching tree upgrades. Hopefully these upgrades are like a Starcraft or Warcraft branching tree, and not just "pay 20000 credits to get to tech level 5" or somesuch. In fact, we could make all tech-level upgrades free, and the facitilies act just like RTS buildings, and just allow you to build (and upgrade) various new types of weapon/hull/etc.

If the shipyard upgrades are well desgined, there should be too much of a sense of "I already researched this, why do I need to build a bunch of shipyard addons now...". Rise of Nations had something like this, for each city, I believe, and didn't suffer from that problem.

What I'm talking about aren't "buildings" in the traditional sense, but are effectively the same thing, in gameplay terms (thus I figured would be worth discussing in this thread, rather than brainstorming)

Edit: It occurs to me that I should add some v0.3 focus to this. Thus, for v0.3:

Shipyards may be built at a planet (however that's accomplished...). Basic shipyards can build scouts and colony ships (only)

Upon discovering "Mark I" tech, the "Mark I Module" can be built at a shipyard, which then allows the shipyard to build Mark I ships.

Upon discovering "Mark II" tech, the "Mark II Module" can be built at a shipyard that already has a "Mark I" module, and allows the shipyard to build Mark II ships.

Upon discovering "Mark III" tech, the "Mark III Module" can be built at a shipyard that already has a "Mark I" module (Mark II module not required), and allows the shipyard to build Mark III ships.

Upon discovering "Mark IV" tech, the "Mark IV" module can be built at a shipyard that already has a "Mark III" module. A shipyard that has both a Mark IV and a Mark II module can build Mark IV ships.

If there is a spending limit at shipyards in v0.3, then each module you build should increase this by some amount. An "Enlarged Assembley Facility" module could also be built, significantly increasing the spending limit.

Pick a cost for the modules... doesn't matter. Each shipyard can have at most one of each module.

(note: in future, each tech probably wouldn't require a separate module to use... you might need a "plasma physics" module, but not "plasma physics 1", "plasma physics 2", "miniaturized plasma physics" etc.)

Daveybaby
Small Juggernaut
Posts: 724
Joined: Mon Sep 01, 2003 11:07 am
Location: Hastings, UK

#379 Post by Daveybaby »

Re: infinite capacity shipyards and pooled production

If i might blow my own trumpet for a moment (oooh! how rare!), i think the system i've developed for COW could be a solution.

Basically, instead of globally pooling PP, you locally pool it instead. What this means is : each shipyard pools PP from nearby systems (i.e. each system sends its PP to the shipyard closest to it).

How is this different from the current system? Well, you also factor in transport costs based on the distance to the shipyard. e.g. for each LY between a system and the closest shipyard, 1% PP are used up in transport costs. Thus a system 10 LY away will only contribute 90% of its PP, the other 10% will be lost as transport costs. A system 20 LY away would lose 20% of its PP, etc.

So it becomes inefficient to have one single shipyard serving the whole empire. Instead you will be encouraged to build a more efficient system with shipyards spread evenly throughout the empire.

Obviously there also needs to be a maintenance cost associated with each shipyard, otherwise the most efficient solution is to have a shipyard at every system, and we are right back where we started.

The ratio of shipyard maintenance cost to transport costs must be balanced so that the player is encouraged to have a 'reasonable' number of shipyards spread throughout their empire at each stage of the game. This can be accomplished via the tech tree, by including techs which reduce transport costs as the game progresses (or just use engine speed), so that the distances between shipyards can increase without becoming inefficient.

Linky here for the COW design documentation related to this. Dunno if this is actually suitable for FO due to some differences in the way some things work (i.e. cow is not starlane based, so distances might vary a lot more in FO) but possibly some variation on the theme might be the solution.
The COW Project : You have a spy in your midst.

User avatar
Ragnar
Space Squid
Posts: 78
Joined: Wed Feb 25, 2004 11:22 pm
Location: McKinney, Texas

#380 Post by Ragnar »

I like daveybaby's proposal for PP pooling limitation.

As another alternative to limiting PP when pooling(I had posted this a while back in brainstorming, but it looks like it is time to state it again):

All PP is pooled, but for building computations divide PP into local and imported.
100% of local PP is always used. Imported PP is used with a waste ratio based on infrastructure %.

So, given 1000 total PP for the empire, planet A produces 100 PP and has 50% inf., planet B produces 100 PP and has 75% inf., and you want to build something that costs 400 PP.

On planet A you would spend 700 PP (100PP +(300/.5))
On planet B you would spend 500 PP (100PP +(300/.75))
If a planet produced 400 PP locally or had 100% infrastructure it would only cost the 400 PP.

This system naturally forces you to build at your developed or industrial worlds, but allows you to build at enormous cost on a newly colonized planet. This would solve the frontier planet ship factory problem, because it would be wasteful to build there instead of in your core systems, but gives you that flexibility if the situation warrants it. Same for wonders; it will cost you dearly to build on a new planet, but you can do it. This will allow for more depth in strategy(flexibility) without to many balancing issues. Also, need to have some infratructure/industry damage to a newly captured world to keep that issue from being a problem.

PowerCrazy
Creative Contributor
Posts: 383
Joined: Fri Jun 27, 2003 2:35 am
Location: Texas

#381 Post by PowerCrazy »

The only problem with your proposed system is that when I have 200+ planets devoted to building a shipyard even 1% efficency from 20000 PP is enough to imbalance the game. Particularly since I don't want ship yards to take 20 turns on a fully operational late game planet.
Aquitaine is my Hero.... ;)

User avatar
Geoff the Medio
Programming, Design, Admin
Posts: 13587
Joined: Wed Oct 08, 2003 1:33 am
Location: Munich

#382 Post by Geoff the Medio »

PowerCrazy wrote:when I have 200+ planets devoted to building a shipyard even 1% efficency from 20000 PP is enough to imbalance the game.
I think, if you've got 200+ planets and are running them all through a single shipyards at 1% effic, you'll lose more in raw ship producing power than you'll gain by building a bit faster far from your empire.

User avatar
Ragnar
Space Squid
Posts: 78
Joined: Wed Feb 25, 2004 11:22 pm
Location: McKinney, Texas

#383 Post by Ragnar »

If you are a huge powerful empire that can afford to build ships at 1% effeciency near the border, I don't see that as imbalanced, since you could build 100 times more ships at you core worlds. If your engine tech/speed is so low that you save that much time building on the border, then the system works. Your 20000PP empire is only as good as your enemy's 200PP empire that works at 100% efficeincy next to your 1% efficient shipyard.

Edit: In fact, it creates balance during the late game, because large empires don't run away with the game. You have to decide between spending time traversing your empire with your fleets or building near the front with low efficiency (as if you were a smaller empire).

PowerCrazy
Creative Contributor
Posts: 383
Joined: Fri Jun 27, 2003 2:35 am
Location: Texas

#384 Post by PowerCrazy »

I fail to see how unrestricted global pooling would create balance.

Lets say that we go with MoO2 production model, with only the inclusion of global Industrial Pooling. What wold end up happening is the player would focus ALL his production on each world, one at a time until he got to the cusp of his production limit (one building per turn ) then he would move on to the next planet. Each turn taking geometrically(perhaps exponentially if you include tech adavnces) longer to complete until by end game you are spending a few hours to make sure that your empire is "optimized": all planets are building their respective buildings in one turn (maximum efficency). Then next turn you do it all again. The strategy is nil, the micro is hell, and the winner is decided by who has the most patience. Hardly my idea of a fun game.

What we are trying to do is introduce a concept more akin to emperor running an empire. He makes the grand/fun decsions i.e. What a planet will do, where to build wonders, and how big of a fleet to build. Rather than telling farmer "Joe the Nth" to build another farm on farming planet #2397.

All the emperor cares about is "The Plan." The nitty gritty details of it are unimportant and uninteresting particularly when they are repeatitive.

Also each planet doens't HAVE to be unique and interesting. Face it, just as the US (or your country) has 1000's of small towns that are just boring and insignificant, they are the bread and butter of your country. I guarentee you that President Bush has never heard of Bigsby, Oklahoma or Jacksboro, TX, but that doesn't mean they are unimportant or useless to the country.

Davybaby I (and others) originally played with the idea of having distance be the limiting factor, but then I got this new fangled Idea of "infrastructure" which would basically be a universal number that would put a value on the lenght of time you had controlled a system, in addition to telling you how developed a planet is.

The main problem with distance is what do you do with shipyards that are fairly close to each other? Which planet contributes to which shipyard and more importantly how do you display the relationships cleanly on the UI to the player? And what if for strategic reasons I wanted planet X to contribute to shipyard a, but planet Y was closer, etc, etc. These are issues that I would rather just avoid than have the player constantly tweaking.
Aquitaine is my Hero.... ;)

User avatar
Ragnar
Space Squid
Posts: 78
Joined: Wed Feb 25, 2004 11:22 pm
Location: McKinney, Texas

#385 Post by Ragnar »

PC: I think we may be miscommunicating here.

If you look at my posts in the review thread, you will see that I voted for NO pooling, and if pooling won it should be limited. You voted with Drek and Tzlaine, with the caviat that buildings should be local production only.

I am certainly not advocating unrestricted pooling. I'm trying to help solve that problem, since pooling won the vote. Maybe part of my problem is, I assumed (I always got in trouble for not showing my work in school :) ) that some other limitations were going to be in, because they had heavy support. I was adding addtional ideas to further refine the pooling limitations. Specifically, I assumed that Tzlaine's idea for HOI type turn limits on building was in (I would also support this). My efficiency rating was in conjunction with this, as another alternative to the distance method of Daveybaby's. I do like his idea, but mine is simpler. In the distance method, you have to decide which planets are contributing to which production centers. This could be done through an interface like the Warlords series vectoring. In my efficiency model you only have to worry about the state of the planet you are assigning a build order to. I see the UI working out of the global pool; when you decide to build some wonder or ship you pick a location and you are shown the actual cost(based on efficeincy). You could then click on some other planets to see if there is a more efficient place to build, decide, and you are done, its in the que.

Do you have an alternative plan to combat the negative effects of pooling? (remember you voted for it, yet you just gave a great argument against it)

drek
Designer Emeritus
Posts: 935
Joined: Thu Jun 26, 2003 8:07 am

#386 Post by drek »

Ragnar wrote: In my efficiency model you only have to worry about the state of the planet you are assigning a build order to. I see the UI working out of the global pool; when you decide to build some wonder or ship you pick a location and you are shown the actual cost(based on efficeincy). You could then click on some other planets to see if there is a more efficient place to build, decide, and you are done, its in the que.
Something like this might work. No big objections; it might be a good idea. The player would have to make a real decision: do I want to build on a planet where my FooWonder will be constructed faster/cheaper, or do I want to build on a planet where my FooWonder will do more good?

As an alternative, I was thinking the player would simply be barred from constructing any buildings on planets with crappy infrastructure. And certain special buildings (true wonders, shipyards) would require planets with excellent infrastructure. This would save the player from hunting around for the best planet to build something on, but would lose that extra bit of player decision.

Daveybaby
Small Juggernaut
Posts: 724
Joined: Mon Sep 01, 2003 11:07 am
Location: Hastings, UK

#387 Post by Daveybaby »

Okay, so the main problem with the locally pooled PP seems to be in balancing the distances to shipyards to minimise losses. How about this instead:

A shipyard (or barracks, or spy training centre etc) which uses pooled PP in order to build ships, has a maintenance cost associated with it. This maintenance cost varies based on the capacity of the shipyard, and a shipyard's capacity must be upgraded in order to be able to process more PP per turn. i.e. the more PP/turn are funnelled through the shipyard in order to build ships, the higher the maintenance costs are.

However (this is the important bit) the maintenance must be funded locally from the planet around which the shipyard is based.

The formula for maintenance cost should be something like:

maintenance = base + (k * shipyard capacity)
where k is the maintenance cost per PP capacity.
where base is a basic maintenance cost associated with any shipyard, no matter how small.

This formula means that:
(a) You cant just plonk a shipyard on an undeveloped planet, since it wont have enough local PP to pay maintenance for even a small shipyard (due to base costs).
(b) You cant just have once shipyard per empire, because at some point the maintenance required for a shipyard capable of processing your masses of pooled PP is going to exceed the amount that even a well developed planet can afford to pay.
(c) You wouldnt want one shipyard per planet, since you'd end up paying loads of maintenance in base costs (plus dont forget youve still got to build and upgrade your shipyards).

Ideally you would end up with (again) a manageable number of shipyards spread throughout the empire (although this method doesnt guarantee they will actually be spread out - this could be encouraged/enforced via other means). Tweaking of base and k via the tech tree as the game progresses would allow control over the typical ratio of planets to shipyards throughout the game.
The COW Project : You have a spy in your midst.

vishnou00
Space Kraken
Posts: 157
Joined: Tue May 25, 2004 3:15 am

#388 Post by vishnou00 »

I like this. (by this I mean the whole DB post).

The "important part" means that like the food (and mineral?), the PP are first consumed at a local level, then made available for global pooling. Do I get it right?

noelte
Juggernaut
Posts: 872
Joined: Fri Dec 26, 2003 12:42 pm
Location: Germany, Berlin

#389 Post by noelte »

Hi Daveybaby, imo it's a nice idea, but do it add something valuable to how shipyards are handled? Providing maintance by the planet at which the shipyard is placed obviously prevents shipyard to be place at underdeveloped worlds, but would you do it if they weren't? A planet is most likely not developed well because it's on your border, if so, i doubt you would place it there. If such a planet is near you core systems you might easyly find another place for the shipyard.

Another maybe unwanted effect is, the planet providing maintance might get a problem to construct buildings, because of a lack of PP's.

Daveybaby
Small Juggernaut
Posts: 724
Joined: Mon Sep 01, 2003 11:07 am
Location: Hastings, UK

#390 Post by Daveybaby »

vishnou00 wrote:The "important part" means that like the food (and mineral?), the PP are first consumed at a local level, then made available for global pooling. Do I get it right?
Yup.
noelte wrote:Providing maintance by the planet at which the shipyard is placed obviously prevents shipyard to be place at underdeveloped worlds, but would you do it if they weren't? A planet is most likely not developed well because it's on your border, if so, i doubt you would place it there. If such a planet is near you core systems you might easyly find another place for the shipyard.
IIRC some people thought globally pooled PP could cause a problem where you can plonk a shipyard on a brand new border world right next to your enemy's territory, and start churning out battlecruisers using all of your empires PP funnelled through 1 undeveloped rock. Whether thats actually a problem or not is another matter, but thats what i was trying to eliminate. Plus the maintenance is a way of encouraging the use of a 'reasonable' number of shipyards, as opposed to the extremes of just having one shipyard on your homeworld (which is strategically very boring) or having a shipyard on every planet (which is both strategically boring and micro intensive)
noelte wrote:Another maybe unwanted effect is, the planet providing maintance might get a problem to construct buildings, because of a lack of PP's.
Very true. Possibly this could be countered by having the maximum upper size of a shipyard set at 50% of the planet's available PP. Thus 50% of the planets PP goes to either the global pool or local infrastructure, and 50% goes to shipyard maintenance. This would have no effect on shipyard size ranges (since you'd also just halve the maintanance costs) and would guarantee at least half of the planet's PP was available for local development if needed.

Possibly the 50% limit on maintenance should be a general limit for all local maintenance costs (assuming there were other buildings that required maintenance).
The COW Project : You have a spy in your midst.

Locked