Quick Feature: Infrastructure

Past public reviews and discussions.
Message
Author
drek
Designer Emeritus
Posts: 935
Joined: Thu Jun 26, 2003 8:07 am

#46 Post by drek »

It would be convinient to have a link to where it as been decided. (A page in the design doc, a wiki, a design thread....)
http://www.freeorion.org/wiki/index.php ... activities

scroll down to "resources."

Tyreth
FreeOrion Lead Emeritus
Posts: 885
Joined: Thu Jun 26, 2003 6:23 am
Location: Australia

#47 Post by Tyreth »

My understanding of the reason for infrastructure is to ensure that a colonie's power is not just tied directly to population. In other words, a new colony will not be the same as an old colony with the same population value.

I think drek's initial proposal of planet classes solves this simply and elegantly. It's easy to understand - your colony has, eg, a nascant trait when it is first colonised, and that restricts what buildings can be built and provides a penalty. It is quite simple to understand, and solves the problem that infrastructure was asked to stop. Do we need a more involved system? I don't see a need for smooth transitions from one class to the next.

I am a Magic: TG player too though, and the planet classification appeals to that part of my gaming experience.

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

#48 Post by Daveybaby »

I dont have a problem with dreks proposal in principle, but i dont like the concept of a colony automatically changing state after a fixed number of turns.

IMO this should still be tied to population, i.e. if you start a new colony, and its population doesnt grow for some reason (e.g. its not a good environment for the species, or some plague event hits it) then it should take much longer to reach 'full colony' status than one which has lots of people flocking to it and breeding like rabbits.

So, instead of a fixed number of turns to full colony status, i'd like to see some value (i'm going to call it infrastructure, because that pretty much what it still is) which increments based on the number of people on the planet (use some nonlinear factor like square root to stop people just dumping tons of population on a planet to get its infrastructure up too quickly). Once a threshold is reached it becomes a full planet.

e.g.
infrastructure = infrastructure + sqrt(population)

a new colony requires 100 points of infrastructure.
With 5 pop this would take 45 turns
With 10 pop 32 turns
With 100 pop (the maximum) it would still take 10 turns.
The COW Project : You have a spy in your midst.

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

#49 Post by Geoff the Medio »

(If you only read one section of this post, read the second section.)

The "construction" meter could depend heavily on population, quality of planet type for race (gaia = good, inferno = bad), race traits and other such things. The "max" meter settings would be as described by drek, and the current meter settings would grow whenever the construction accumulator fills up. The net effect of this would be that a new colony has a low population and thus a low construction meter, and thus takes a long time to develop high resource production meters, so is fairly useless to start with...

... (This seems rather obfuscated and hard to understand to me, but it sorta follows from the current / max meters concept)

~~~~~~~

An alternative is that the construction meter not depend on popluation or world type, and is always "1" unless modified. (It might not even be necessary to show the player what the construction meter actually is.) In this case, the effect of a new colony being less productive emerges natrually from the calculation of resources produced:

(resources produced) = (population) * (meter value).

A meter value grows whenever the construction accumulator tops up, but the rate at which the construction accumulator fills up in not affected by population.

The main point of this suggestion is that there's no need to adjust or limit the resource meters for new colonies... the lower initial population already reduces the number of resources produced, and thus limits what buildings can be made (buildings require X per turn for Y turns... if you make less than X per turn, you can't make the building).

~~~~~~

My may concern, as with Davebaby, with a planet special trait for "Nascent Colony", is that it not be rigidly fixed to a certain number of turns. It's a nice and easily understood thing why a "Nascent Colony" has reduced production, but there should be some sense that this duration depends on things that it seems like it should depend on (ie. population and planet quality). This dependence has to be clear and understandable however... but I don't see how that can be done with multiple variables that "should" influence how long "Nascent Colony" lasts.

There's not much point in having a "Develop To Full Colony" building project for a new colong... since this is obviously the first thing you're going to do, and thus isn't really a choice and should be abstracted somehow.

User avatar
Krikkitone
Creative Contributor
Posts: 1559
Joined: Sat Sep 13, 2003 6:52 pm

#50 Post by Krikkitone »

OK, I think I can summarize some of the various ideas

I think the simplest way to think about this part of the econ model is

Infrastructure Rating*(Everything that is going to affect the economy and does Not affect Max Infrastructure) -> Output

Because everything that affects output does so by either affecting the max Infrastructure or the Output of current Infrastructure (referred to from here on out as output efficiency).

For How the Infrastructure is generated

PART I
What should effect Max Infrastructure?
Note: Effect of choice This is the thing that will NOT give a quick immediate benefit to Output, but will require investment of (whatever is chosen for Part II) before an economic benefit is chosen

Any combination works (except of Nothing)

Nothing (Infrastructure is a 0-100% rating the same for all conditions)
Population
Technology*
Race Picks*
Planetary Environment*
Wonders*
Government*

Because these all come in multiple types, it is possible that some of these may affect max Infrastructure and others may affect Output Efficiency, so only pick those that dominantly affect max infrastructure



PART II
What should Infrastructure Growth Require/ie consume?
again in any combination, but None is no longer an option

Time**
Production**

** The Amount of these two that is required for Improvement may be affected by other factors like population, planet size, Leaders, Wonders, etc.


My Infrastructure Model would be

I. Population+Technology+Focus (maybe others) Not Planetary Environment

II. Time + Production



Concerns with production required for development

Hrm. That would involve a slider or a slider diguised as a queue item. It opens another can of worms as well: what if I want Planet Y to recieve resources, but not Planet X?

Infrastruture was never meant to be the result of economic investment. It's just a gauge of how long a planet has been in the empire and a number to guage the effects of tradegies.

Economic investment comes from researching economic techs, Build Projects, building colony ships.


If you want Planet X to receive resources, but not planet Y, then they need to be working on different things, if they are both working on their 'sliders/infrastructure' then they both have equal priority. But if X is working on a Building/Ship and Y is working on its 'sliders'... then the priority is set on the Imperial queue by saying what gets priority 'Sliders' or particular Buildings/ships. (If Both are working on Ships/Buildings then the priority is set by setting the priority of those Ships/Buildings...of which there won't be too many.)

If you want your Economy to be growing at full speed no matter what, then you put the 'Economic Development' at top of the queue. If you want some ships/Buildings Now because they are more effective at boosting your win/survival chances than some economy, they go ahead of 'Economic Development'.

The problem with your forms of economic investment are two fold.
1. They are all too individual, If I want economic development to be a certain priority, I should just set it, instead of going searching around for new techs, barren worlds, and buildings.

2. Two of those, Techs and Colonies, shouldn't be economic Investment but economic Opportunities for investment


I don't see the problem with an Empire wide setting (slider or "slider disguised as a queue item") Admittedly Sliders per se are ineffective (given the 100/0 being the most efficient settings) but queues solve that problem. The point is they allow a simple setting of Strategic game choices.

We could have shipyards pump out ships automatically with the type+quantity of the ships being based on the Wonders you have built and the Leaders you have assigned to that world. That might make for an interesting card game, but it would not feel like managing an empire, but a card game. Techs could be gathered like spells in HOMM, each world has some 'techs' you can find there as the planet develops, and then you must send your fleets abroad to collect all the high level techs from developed worlds.

No instead we say how many ships we want of what type and what priority each group of ships gets. We determine how many planets are Research Focused and we add techs to a queue determining which ones we get first. These are simplified decision making mechanism for simplfied decisions. They allow strategy to be maintained, and allow the adding of the occasional card game elements, but they don't make a game within a game... The Economy should not take That much time trouble what ifs.. The question should be can I bypass getting XYZ sooner to get a generically bigger economy or do I need XYZ now. Yes/No That says where it goes on the queue.



There's not much point in having a "Develop To Full Colony" building project for a new colony... since this is obviously the first thing you're going to do, and thus isn't really a choice and should be abstracted somehow.


That is the problem I am trying to solve.. the effective elimination of the choice... You are obviously going to make every colony a 'Full Colony' The issue is if that is the First thing you are going to do. Perhaps it is more important to get ships to defend that New Colony, and if the cost of development is enough and production shared, then Developing New Colonies (rather than just making them) can actually limit your ship production. Of course New Colonies are likely to have a limited amount of investment that you can put into them, but developing many half developed colonies could be a serious drain on your military capability.

Tyreth
FreeOrion Lead Emeritus
Posts: 885
Joined: Thu Jun 26, 2003 6:23 am
Location: Australia

#51 Post by Tyreth »

Daveybaby wrote:I dont have a problem with dreks proposal in principle, but i dont like the concept of a colony automatically changing state after a fixed number of turns.
Personally, I quite like the idea of a fixed number of turns. It also opens up potential for empire-wide wonders. For example, a wonder that reduces the number of turns a nascent colony takes to become a full fledged one. It's simple and easy to understand.
IMO this should still be tied to population, i.e. if you start a new colony, and its population doesnt grow for some reason (e.g. its not a good environment for the species, or some plague event hits it) then it should take much longer to reach 'full colony' status than one which has lots of people flocking to it and breeding like rabbits.
Plagues and environments provide their own disadvantages already. For a nascent colony to reach the end of its tenure, it is simply saying "This colony has now finished construction of the infrastructure relative to its needs". Obviously, a race that prefers terran colonising an inferno planet will already have disadvantages on that planet - both for pre-nascent and post. I don't think we need to slow or speed it up, since the infrastructure describes the readyness of that colony for the environment it's in.

I can't see a compelling reason to account for environment other than realism - and that's arguably already present. We will already have penalties for environment type elsewhere.

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

#52 Post by Geoff the Medio »

Tyreth wrote:Personally, I quite like the idea of a fixed number of turns. [...] It's simple and easy to understand.
It is possible for something to be too simple. While it may be easy for a player to understand that a colony will take 20 turns to develop to usefulness, s/he will undoubtedly want to know why this can't be sped up a bit.
It also opens up potential for empire-wide wonders. For example, a wonder that reduces the number of turns a nascent colony takes to become a full fledged one.
This is not dependent on early colonies having a "nascent" trait. Almost any system could incorporate this sort of effect. (the idea itself is fine... just not the implication that it's an advantage to the "nascent colony" special with fixed (but now variable?) duration)

As I said before, there's no need to adjust or limit the resource meters for new colonies. The lower initial population already reduces the number of resources produced, and thus limits what buildings can be made (need X per turn to make something requiring X per turn for Y turns). As far as I know, the nascent colony trait would just lower some meters. If so, it is redundant.

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

#53 Post by drek »

Tyreth wrote:
Daveybaby wrote:I dont have a problem with dreks proposal in principle, but i dont like the concept of a colony automatically changing state after a fixed number of turns.
Personally, I quite like the idea of a fixed number of turns. It also opens up potential for empire-wide wonders. For example, a wonder that reduces the number of turns a nascent colony takes to become a full fledged one. It's simple and easy to understand.
The turns aren't entirely fixed, as Tyreth guessed. There'd be a Construction meter. Buildings/techs/enviroment/leaders and racial picks would be able to apply bonuses/penalties to this meter.

Internally, there'd be an variable hat's chipped away each turn by the value of the Construction meter. On the UI, the player would see "(variable/Construction meter) Turns Remaining".

We'd set the default of the C meter so that it would take X turns for the average empire to remove the "Nascent Colony" status from a world. If the empire has the racial pick "Fast Builders" or builds a "Make Colonies Fast" wonder, then the C meter would be improved.

Of course, as Geoff said, the C meter could be used in either the label system or the more detailed idea he's advocating.
As far as I know, the nascent colony trait would just lower some meters. If so, it is redundant.
If the food meter is reduced to 0 or below, the colony wouldn't be able to feed itself. That be the major effect of Nascent/Ruined colonies: you'd need outside Nutrients if you plan on keeping these colonies alive. (their Health meter would also be reduced, meaning a lack of food would easily kill off the colony.)

If during early game you are focusing your homeworld on farming so you can expand, player B might be focusing his homeworld on industry so he can blow up your defenseless expansions.

edit: Even if we use the Geoff/Krik system, we might still need labels of some sort if buildings are constructed through a global queue with a global pool of PP, so that we can restrict buildings to planets with X infra.

tzlaine
Programming Lead Emeritus
Posts: 1092
Joined: Thu Jun 26, 2003 1:33 pm

#54 Post by tzlaine »

I find it ironic that this is called a "Quick Feature". Could someone break this stuff down for me, here or in another thread if need be? I've been coding with nearly every spare minute, and I'm having a hard time with the disorganization in this thread. What are the different proposals A, B, C, etc...?

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

#55 Post by drek »

Tzlaine:

There's two points of contention. First, the general behavior of infra, though ultimately all these ideas have about the same effect:

A: Powercrazy's idea
Infrastructure is a percentage 1-100%. As new techs are researched, the percentage drops to indicate increased capacity for infra. Infra multiplied with (resources + tech bonuses)= resources produced.

B: Geoff/Krik's idea:
Each resource (Money, Farming, Science, Mining, Industry) has a max amount and a current amount. The max amount is determined through focus, techs, buildings, racial picks, etc. Planets slowly build current amount up to the max amount. (When focus changes, so does the max amount, creating a penalty for focus switches.)

The visual for this idea:
Image

C: My dumb idea:
Infra is almost entirely abstracted. There are three or four different states that a world can be in, including "Nascent Colony", "Ruined Infrastructure", "Transitioning Between Foci" and "Core World". A newly colonized world would be a "Nascent Colony", for example. Each penalty inducing state lasts a number turns, after which it disappears. The states stack, so you can have both a "Nascent Colony" and "Ruined Infra" tag on a planet.

Second point of contention is how infra improves:

1: Automatically. Infra improves on the planet without any PP cost or direct player intervention. Wonders, techs, racial picks, leaders, etc. can improve the rate at which infra improves. (through, for example, the Construction meter)

2: Infra costs PP. The player uses some method of assigning PP to Infra, either Empire wide or for each planet. A queue item has been mentioned, though I confess I don't understand how that would work exactly.

------
I'm voting for either B or C with infra improvement method 1. I'm strongly against infra improvement method 2.

If I got anything wrong, write a brief correcting paragraph and maybe I'll edit this post.
Last edited by drek on Fri Jul 02, 2004 4:03 am, edited 1 time in total.

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

#56 Post by vishnou00 »

As the opening post stated, it's "both a debate thread and a public review rolled into one ugly mess".

The keyword here is "ugly mess", as in "not ready for consumption". So far, it's been quicker than building imho.

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

#57 Post by Geoff the Medio »

drek: Are "My dumb idea" and "Geoff/Krik's idea" mutually exclusive?

I see your point about "nascent" colonies needing outside food and the benefits of having the specials. This, and the other speicals, excepting "focus transition" would work in the "Geoff/Krik's idea" model as well, wouldn't they? A nascent colony would have maximum meter ratings of 0 in all categories. Pre-existing infrastructure could give a "starting bonus" to infrastructure in some levels... example: When you colonize a new world, it starts as a nascent colony, and produces nothing for some turns due to having 0/0/0/0/0 meters. After beomcing a full colony, the ruined infrastructure means that it now has 1/2/3/1/0 meters, and can start building up from this state. The specials wouldn't "stack" in any meaningful sense, but they could be had by the same planet. ("core world" would give +1 to all meter max's... other specials could give others)

I guess the real gist of your system is that the specials are all there is... and there's no other system that they work with or interact with. Simplicity to the extreme.

vishnou00: I think / hope that the scope of this thread is sufficiently limited, and any possible brach-off discussions have been sufficiently already decided that we've more or less finished the design part of the discussion... (unless I just jinxed it...), so all that's left is seeing what systems people like...

tzlaine
Programming Lead Emeritus
Posts: 1092
Joined: Thu Jun 26, 2003 1:33 pm

#58 Post by tzlaine »

Thanks, Drek. I like B and 1. I also like the nascent colony part of C, but I can see that B subsumes this, as Geoff points out. And I'm completely against 2.

Now, back to fixing the pathfinding code.

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

#59 Post by drek »

drek: Are "My dumb idea" and "Geoff/Krik's idea" mutually exclusive?
Obviously not. Which is why I support either/or and took the time to spell out how I'd like to see system B written up in the design doc.

I'd like to be able to restict buildings, so that they can't be built on new/messed colonies via the global PP pool. If we use system B, there'd should be a metric for determining when a colony is "Nascent", "Ruined" or whatever.

Perhaps, if total infra < X, then the world is flagged as not being a target for build orders. If infra is reduced below X, then any current build orders are canceled (the project has been halted, due to earthquakes, enemy action, chaos due to focus switch or whatever caused the drop.)
I guess the real gist of your system is that the specials are all there is... and there's no other system that they work with or interact with. Simplicity to the extreme.
Yes. That's why I like it.

But I can deal with the moderate complexity of the Geoff/Krik thing, so long as all the t's are crossed and the i's dotted.

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

#60 Post by Geoff the Medio »

drek wrote:I'd like to be able to restict buildings, so that they can't be built on new/messed colonies via the global PP pool. If we use system B, there'd should be a metric for determining when a colony is "Nascent", "Ruined" or whatever.
Hmm... for some reason I'd been thinking in terms of pooled PP for ships, but local only for buildings... there was some discussion of that elsewhere, and the public review was mostly about ships. I guess this suggestion won't work: new colony has low population so produces only a few PP / turn. If this is less than X in "X / turn for Y turns for building Z", you can't build Z. nascent colony special would do it, I guess. (I'll avoid any comments concerning problems with pooled production :lol: )
Perhaps, if total infra < X, then the world is flagged as not being a target for build orders. If infra is reduced below X, then any current build orders are canceled
Could also just the "production meter" for this. It's just an on-off switch, but like smac, if it falls below a certain level (say if it goes to 0) then you can't build anything at the planet. The bonuses from meters don't have to be just increasing the resource production...

Locked