Brad Wardell's views about technology, politics, religion, world affairs, and all sorts of politically incorrect topics.
The limits and the possibilities
Published on March 11, 2006 By Draginol In GalCiv Journals

I've been reading a lot of discussion both here and on other sites about ideas and suggestions players have for Galactic Civilizations II.  Stardock, I like to think, is reasonably well known for implementing pretty significant changes into its software well after release.  If we think something is a really good idea and it won't dramatically change the product, we'll seriously consider implementing it.  We've been doing that since the first software products we released over 10 years ago.

Of course, the question is, what constitutes a change that is too dramatic? And how can we determine whether a given idea is something that's good for the game or not?

Before we start out there, I should make a clarification on something.  I've seen threads where people will say "Yea, but Brad says it's working as designed so he obviously thinks it's great."  That's not what I mean when I say that.  The context is important. 

As an engineer, I try to be precise as I can be with my words. That's one of the reasons I'm so wordy. What I write tends to be full of qualifications. One of the things I tend to object to strongly is when someone will take a design choice they disagree with and simply label it as a "bug".  A bug, to me, is something that is not working as designed.  Someone may not like a given feature, but if it's working as it's supposed to, it's not a bug. But that doesn't mean that we think it's the end-all be all feature.

One of the areas I want to tackle in the post release is the economic system of Galactic Civilizations II.  But I don't want to do it alone. I want to hear what other people think too.  But such discussions can be problematic.  As with any on-line discussion, disagreements will break out.  With people all around the world, many with strong opinions, you inevitably end up with some people who will state their opinions as facts. "This is how game X did it. Do it like that."  There is no single "best" solution. We all have our own ideas. What we can do, however, is build a consensus to some degree.

Economic Systems & Strategy Games

Some parts of the game I feel strongly about, other areas are open to significant change.

For example, in Galactic Civilizations II as leader of your civilization you can set your tax rate -- the money coming in from your people.  And you can set your "spend rate" which determines what % of your industrial/research capacity to make use of.  I believe that governments should be able to intentionally have deficit spending.  I believe that your financial income should not be tied to your industrial capacity. If you have the factories and labs to do it, you should be able to make full use of them regardless of your income. It's called deficit spending and it's practiced by many governments.  Your population will get angry if you go too far into debt. And right now, we have a -$500 debt ceiling. 

Having taxation and spending separate is something I'm married to. I like it.  I realize it's more complex than in some other games but of the other systems we've contemplated over the years, I think it provides the best balance between realism and simplicity.  People are free to disagree of course.  That's natural. Some % of people will not like it.  But I think most people understand it.

So let's talk about the part of the system I'm not married to -- the UI representation of it.

So you have your spend rate -- the % of your industrial capacity that you want to make use of.  Then the question is, where do you want that industrial capacity to go?  There are 3 sliders that control how that spending is funneled.  The three sliders allow you to decide how much to fund your factories and research labs.  Military and social spending goes into your factories which produce more planetary improvements and build your ships.  Research spending goes to your labs and is converted to research points that go to getting your next tech.

I don't think it's that complicated and judging from the various forums I read, most people understand how it works. But not everyone. Some people don't understand it and others just don't like it.  The group that doesn't understand it tend to be the same people who don't understand why taxation and spending aren't linked because "game X does that".  Part of the reason I put in having taxes and spending be separate was out of frustration with other strategy games that tried to act like ones money income was somehow tied to their industrial production. As if the Germans in World War II could simply have bought more armies with money (yea, I know you can quick build but it comes as a very steep price -- on purpose). Industrial capacity has nothing to do with wealth. Hence the division.

Rhetoric

I confess, I get defensive in response to rhetoric.  I tend to have an aversion to absolutes or people giving their opinions as facts. Every game that has an economic system is going to have people who think they have a better idea on how to do it.  We obviously like our system. We think it works pretty well and we think most players think it's fine too.  But that doesn't stop us from trying to listen and make improvements to make it even better.  But when some player asserts something is "broken" that makes it sound like it's a bug and then puts us in the position of having to defend our design decision. 

Every element of the game is a choice. Why only 5 planets in a solar system? Why not 9? Why do we allow millions of people to come into the tax system in a given week? There's so many design choices that have to be made but at the end of the day, our goal is to make the game fun.  But one man's fun is another man's headache.  I've gotten emails from people who simply can't play the game as long as Earth and Jupiter are on the map in the wrong scale (Earth is much smaller than Jupiter in real life but we try to scale things so that they're usable on screen).  Heck, I should post some of the emails I get, you'd be shocked at some of the stuff.  I got one today from someone who claimed they were returning the game because all the alien races are humanoid. I kid you not. Hey, at least they're not all humans with different nose ridges!

Rhetoric matters.  When someone comes onto the forum and makes a post entitled something like "Map system totally broken" and it turns out it's because we use squares instead of hexes or because the moon rotates around the earth in clockwise or whatever it puts us on the defensive.  I think that's just human nature.  I realize some people find it tempting to say "Everyone with half a brain knows that the moon rotates counter-clockwise around the earth!11!1" But when you're on the receiving end and you know pretty certain that 99.9% of people don't care which way the moon is rotating because it's just a cool graphics effect, it's hard to champion changing it (incidentally, we are going to tweak that since it's in the customplanets.xml file).

Other Economic options

I have some ideas on economic tweaks that I could see us making.

For example: Social Production.  Social Production could be automatically transferred to ship building when all planetary improvements are done.  This would solve the potential issue of people's economy becoming crazy when all social projects are completed.  And if there's no ship to be built, it would just go back to your treasury.  It wouldn't be hard to do, would only require modest AI changes.  I can assure you the AI would love it.

Another area I could see tweaked is the relationship between research labs and factories.  Right now, spending is rationed between factories and labs.  But that's not the only way it could be done.  Other ways would require some UI thought though to keep it from being too complex. 

For example, rather than having a spending slider, you would simply have an industrial slider and a research slider that would be independent of each other. Then you'd have a dial that would let you decide how much of that industrial output was going to planet improvements and how much to ship building.  But doing it in an intuitive way would take some thought.

There are many other ways it could be done too.  All would require thought on how best to present it so that it's intuitive to players and doesn't radically change the game.

Conclusions

It's always tough trying to know where to draw the line on improvements. Game developers want to satisfy their gamers -- all of them. And often times, great ideas come from players. The whole starbase concept in Galactic Civilizations came from players for instance.

But you also have to take into account the people want to feel like they're playing on solid ground. That the game they're playing isn't that fluid. Because every change one makes is going to disappoint someone.  So we have to be very careful about how we do things.

That's my 2 cents on that anyway.


Comments (Page 9)
9 PagesFirst 7 8 9 
on Mar 14, 2006
If I have 10 factory workers and 5 scientists ready to roll if I can somehow get their salaries paid, what is it causing me to choose one or the other?


Apparently you don't have that. You have X workers, period. You can leave your labs completely empty and have the X workers work the factories, producing M units and costing M bc (if you have it), or you can put your X workers in the labs and leave the factories empty, producing N beakers and costing N bc, or you can put a fraction of the workers in each location and split the output.

Nevermind the "realism" of reassigning your entire workforce from factories to labs and back on a dime.

I think you have a point when you cite the limitation of the budget. That IS a hard limit. However Brad wants there to be two limits. There's the limit of the income, and the limit of the spending capacity, and your total output will be the lesser of the two, in the long run. So the gameplay centers around pulling the levers in such a way as to match the two types of limits as closely as you can manage to do by the point in the game in which you will reach a winning position. (After that it stops mattering, because you've won already and are just mopping up).

I don't see why having two limits would not be valid. There is a little more wiggle room than there is in having only the one limit. For instance, picking up cash through tributes or tech sales, or racing your spending capacity out front of your income but expecting the income to rise later. You spend extra up front on spending capacity infrastructure and then can ZOOM along full speed on military or research spending as the income levels are rising.

There is a fun game in there with the two limits. I have enjoyed the difference between that and all the one-limit systems. BUT the game could stand to clean up the need for player to do simple little repetitive things like haggling for exact prices or messing with focus buttons instead of being able to control individual planets more precisely.

Why can't I have lower taxes on overcrowded planets, for instance? Or run my 3-Lab Mars planet on research all the time? Why is Mars's Research Focus spending based on its FACTORY output, just because at the moment I'm running 100% military spending? Those focus buttons are a giant PITA. My Mars can put out 56 beakers if I run 100% research, but can only put out a pathetic THREE beakers if I run 100% military and have Mars "focus on research"? You've got to be kidding me. Only a programmer can come up with these kinds of things. It's not buggy, per Brad's description to open the thread -- it's doing what it is designed to do. ... This element could stand for some redesign, though.


- Sirian
on Mar 14, 2006

Thanks guys for all the feedback.

To summarize what I've read:

There's 4 camps:

1) The board gamers (grognards). They want every number, every detail spelled out. There should be a way to click on anything and find out precisely how it came from. The cleaner and simpler the mechanics, the better.

Ex: 1 unit of populatoin should produce 1 unit of wealth. If I mouse over my income, it would display this information.

2) The intuists. These are the people who want the game to be more like a simulator. The more complex, the more realistic. They don't want the numbers spelled out because they want to use their intuition to lead the way. They considering learning how the relations work as part of the fun.

Ex: 10 billion people are made up of many different socio-economic classes of people, as the population increases, the wealth collected should increase at a sub-exponential rate.

3) The Realists.  These are the people who simply want a tweak here or there or a tooltip here or there or maybe just a clarification.

4) The armchair game developers. These are the people who really could make a much better game if only they had the money and developers to convey their genius onto the screen. To prove their genius, they'll use terms like "broken" or "unusable" to describe anything that they consider non-idea.

Group #4 won't ever be satisfied. It's not like there's a scenario where we'd put out some update that totally guts the economic system.

Group #3 will likely be largely satisfied by v1.1 since social production won't be "Wasted" and we'll likely put in additional tooltips.

Group #2 is going to be unhappy that social production is going to transferred to military production which is totally unrealistic. Shouldn't the economy have waste? After all, in the real world it's a fact of life.  A business, like Stardock, has to plan its resources accordingly. If we bring in too many employees, then we don't have enough for them to do and waste money on salaries.  If we don't bring in enough, we don't spend enough and the government gets that money in taxes.

But things like social spending drives some people nuts and so it's a matter of guessing whether changing a feature  akes more people happy than unhappy.

Group #1 would consider the change to remove social waste as "obvious" because such micro management wouldn't be in a board game. 

That group also was probably driven nuts by Warcraft 3 which had all kinds of fuzzy math in how battles were handled. 

This group could can probably be generally satisfied with simply more tooltips or some sort of analysis on how money is done.

For instance, tax income is non-linear.  It's your population taken to something like the .80 power.  There's a lot of exponential relations in the game.  What other games will often do is cap it or take out any mechanics that get near a race condition.

on Mar 14, 2006
4) The armchair game developers. These are the people who really could make a much better game if only they had the money and developers to convey their genius onto the screen. To prove their genius, they'll use terms like "broken" or "unusable" to describe anything that they consider non-idea.


Whatever.

Have fun with "your" game and the zero-sum, patronizing attitude. I've got better things to do.

on Mar 15, 2006
Whatever.


Did you even read the rest of the post? Sometimes it just doesn't pay to be nice. I can only hope most people appreciate the extra effort Stardock *do* actually put in.

There is one game that people can set up EXACTLY the way they want it. EVery single detail is alterable. It's call C++, you should check it out (bit of a steep learning curve though, and takes a while before you can actually get into the game).
on Mar 15, 2006
Why can't I have lower taxes on overcrowded planets, for instance? Or run my 3-Lab Mars planet on research all the time? Why is Mars's Research Focus spending based on its FACTORY output, just because at the moment I'm running 100% military spending? Those focus buttons are a giant PITA. My Mars can put out 56 beakers if I run 100% research, but can only put out a pathetic THREE beakers if I run 100% military and have Mars "focus on research"? You've got to be kidding me. Only a programmer can come up with these kinds of things. It's not buggy, per Brad's description to open the thread -- it's doing what it is designed to do. ... This element could stand for some redesign, though.


Actually the redesign required isn't necessarilly on Brad. It falls squarely on the player as well. That is no matter what the system once you figure it out it becomes a simple (or complex) exercise in min/maxing. The obvious answer to your Mars question is to not build 3 labs on Mars, bang problem solved under the current dynamics. We realize this isn't the solution you are looking for, but its the solution you have. Its also a step in the direction *against* your prefered method of jumping sliders to 100% every cycle, which again may be why you are finding it unlikeable.

Seriously Sirian, why did you hate MoO3 so much? It has all the nitty gritty controls that you are asking for here. Especially today with user mods and patches it runs and plays so much differently than even 1.25 that if you are begging for this individual planetary micro you have a game (assuming you still kept your copy) which lets you do all of that. Oh and it has ship design and fleets and all the rest of it too.
If I have 10 factory workers and 5 scientists ready to roll if I can somehow get their salaries paid, what is it causing me to choose one or the other?

If I have 10 factory workers and 5 scientists ready to roll if I can somehow get their salaries paid, what is it causing me to choose one or the other?


Because you don't actually have 10 workers and 5 scientists ready to roll. You have *capacity*. At this point I don't think you even care to understand the way its set up, you're just being a tool since you don't like the existing set up. To follow on that thought you don't realize that it doesn't really even matter what the game 'names' production, they could call it monkeys and donkeys, its the dynamic which matters, and the dynamic which needs to be understood, not the semantics of how its set up. SD could agree to change the system to what you propose, but in effect it would change nothing in the way the mechanic works (assuming they cut down the capacities). Basically you are argueing over the way the game presents this dynamic, its a pointless arguement, especially presented in the mannor you present it.

Sirian is argueing over something different, putting down individual planetary sliders. This is more a question of how much micro to add to the game. While Sirian is free to claim it won't add any, I think it rather obvious that this is not the case. Beyond which at some point Brad should simply adress this issue of where the game is meant to be focused, either at the planetary level or at the empire level. The answer as I see it right now is that the game is focused at the empire level. That means that you get to decide basic policy, what is built where, and how to fund the overall empire. You do not have individual control on planets other than to broadly focus on one aspect or another. You can quibble with that decision, but as Brad says, it's his game and he made it the way he wanted to. I for one prefer this implementation to a more micro intensive implementation (ala MoO2).
on Mar 15, 2006
Seriously Sirian, why did you hate MoO3 so much? It has all the nitty gritty controls that you are asking for here.


The only game I've ever *hated* was MOO2, for what it did to the franchise, what it did to the genre. My hatred of that title remains visceral to this day.

MOO3 that shipped was cutting losses on a product whose vision went astray. Brad categorizes some of his own customers as "armchair designers". "These are the people who really could make a much better game if only they had the money and developers to convey their genius onto the screen." Yet one doesn't have to look far to find a passel of poorly implemented games and ideas. MOO3 is a case study. Designers are exactly that: people in whom others invest money and support to enable them to bring their vision to the screen. Ideas work well or they do not. Ideas are implemented well or they are not. Getting funding for your ideas is not automatically a conveyor of success, only of opportunity. With creative works, there are no guarantees. It is a high risk business with few winners and many losers. MOO3 has a lot of ideas that didn't work well and several more that were poorly implemented. The entire planetary economic system should have been abstracted instead of automated, and that's just for starters.

Brad's planetary tile system and planet specials draws inspiration from a game called Ascendancy. He's done a FAR better job of implementing the concept. Ascendancy would devolve into an endless pit of planetary micro. Brad limits the number of plots on a planet (Ascendancy could have up to almost a hundred, IIRC) and includes auto upgrading. It works well, I think. BUT... if you are going to draw inspiration from something that revolved entirely around micromanagement, you're leaning your game in the direction of micromanagement. Players can in fact micro their planets. They can choose for each square, and choose the build order, and will have to micro the order in which upgrades occur if their tech rate outpaces their social production rate and LOTS of things are sitting in the queue. (That usually happens to me). ... This is a case of something that works smoothly. Brad added a system that you can micro, and then provided the means to control it completely.

Compare with the focus buttons. These are my beef. Allow me to boost my "net cred" by proclaiming that they suck.



Seriously, they do suck. All the micro but only half the control.

Imagine if in choosing what to build on your planets, you could only tell the governor to "focus on production" or "focus on research" but could not actually place any of the buildings yourself. What fun would that be? THAT is the paramount lesson of MOO3: abstract it or let players have direct control. Indirect, blunt and clumsy controls are not fun.

We were better off in GC1 without the focus buttons. No chance to micro, just run the various macro leans and move quickly through the turns.

Brad ALREADY made the move to planetary micro by adding the focus buttons. *shrug* All I want is for the rest of the reward to be added to the already intensive labor. If we're going to micro, by golly let us do it. Or... don't tease us with bad micro. Bad micro is... bad.


The obvious answer to your Mars question is to not build 3 labs on Mars


Bzzt, wrong answer.

Mars is still best off with three labs. Why?
* Factories are only useful for two things: building social or building starships.
* Mars will never be able to build state of the art starships quickly.
* Mars can set up some labs, freeing up slots on other planets for other things, or adding to the research capacity without stepping on starship production capacity.
* Although it grows slowly, Mars will waste no cash on temporary social projects. Every penny spent on its social promotes its three Labs. It will sit idle, spending nothing, when running high military focus, and will go full speed when running research lean.
* Don't even build Starports on your tiny planets (< 6), you're just wasting space. Even under the current system, some specialization of planets is beneficial.

Aside from this, however, your reply proves my point. The promise to micro your planets is largely a false one, in a strategic sense. The successful approach is the one that leans every planet in to the global economic balance in the most effective way. You can't really be creative. It's an exercise in macro-economic management.

But then, why do the focus buttons exist? I would urge Brad to get off the fence with those. The micro is already there, but the control is not. It would be better in either direction: back to GC1, with no micro, or forward to embrace the notion of bona fide specialized planets, with independent controls.


While we're on the topic of MOO3, I might as well make a third point. MOO3 limited fleet sizes for technical reasons. The graphics to display the fleet combats could show only a limited number of ships. It was a gameplay disaster, as the AI was not as efficient as players in creating ship designs. Players could assemble the Invincible Fleet. GC2 now has a similar problem -- and the same technical limitations on fleet size. Logistics are a better implementation and a moving target implemented for other reasons, but the results display many of the same issues as MOO3's limited fleet sizes.

Most systems in empire-building games have at least one example somewhere at some time of a game that did things well. These limited group combats are a different beast. They've only been tried twice now, and you can call me a critic in both cases. I wonder if this concept can be done successfully. At the moment, I don't see a way, but I would be happy to find evidence to the contrary. The two systems that have worked in the past are single units and unlimited group sizes, and both also have issues. Both suffer from the "Stack of Doom" issue where bigger is always better. The limited fleet size mechanism stops that problem in its tracks, but only trades it for an even worse monster, where the strongest mini-group wipes all comers off the board without breaking a sweat.


- Sirian
on Mar 15, 2006
Mars is still best off with three labs. Why?
* Factories are only useful for two things: building social or building starships.
* Mars will never be able to build state of the art starships quickly.


You assume though that your only options are factories or labs. Not true. In any case a PQ 4 planet is not going to be able to contribute much to your empire one way or the other.


Aside from this, however, your reply proves my point. The promise to micro your planets is largely a false one, in a strategic sense. The successful approach is the one that leans every planet in to the global economic balance in the most effective way. You can't really be creative. It's an exercise in macro-economic management.


I love it how you seem to think that by making more redundant choices you are somehow adding creativity to the process. If you have full micro what have you actually gained? Sure the ability to set things up differently, but in the end there is still going to be your one formula for success. You will pick a planet to be tech capitol (likely one with reserarch boni), one for manufacturing capitol, one for econ capitol, ... essentially you are trading one mechanic for another. I understand that you have a preference towards one, but its just that, a preference. In the end your decisions are still limited, your best strategy won't change, all that changes is how you have to implement it. I honestly don't have anything against changing the econ system to untie research from production, but I see that its the same thing in the end. I do however not desire to go full micro, as to maximize your empire you will be required to spend more time dinking around.

But then, why do the focus buttons exist? I would urge Brad to get off the fence with those. The micro is already there, but the control is not. It would be better in either direction: back to GC1, with no micro, or forward to embrace the notion of bona fide specialized planets, with independent controls.


I would like it better if we could set the priorities from the colony management screen rather than having to go through each planet seperately, but I don't necessarilly see how either of your suggestions is implicitly superior to the current set up. As a game mechanism it doesn't really matter how it is set up so long as everyone is required to follow the same rules. Everything else is personal preference. My question at some point though is how do these changes affect the AI. Should there be rules (or limitations if you prefer) which serve to make the AI more competitive? I would say yes, even if that means giving up some 'control' which you might otherwise want or expect. I do not know if this is a reason for the design or not, only that its something that bugs me in many games. That is the designers have some cool (usually complex) idea for how to run something, but it proves highly difficult to get an AI which can manage it even fractionally as well as a human.


While we're on the topic of MOO3, I might as well make a third point. MOO3 limited fleet sizes for technical reasons. The graphics to display the fleet combats could show only a limited number of ships. It was a gameplay disaster, as the AI was not as efficient as players in creating ship designs. Players could assemble the Invincible Fleet. GC2 now has a similar problem -- and the same technical limitations on fleet size. Logistics are a better implementation and a moving target implemented for other reasons, but the results display many of the same issues as MOO3's limited fleet sizes.


MoO3 as shipped, and even with 1.25 had more than its share of problems, you will get no arguement from anyone on that. MoO3 as it exists now is a completely different beast. You can now change TF sizes, you can change number of allowed TFs in a battle, you can do a number of things to improve the AIs ability to compete. There is even a way to autodesign AI ships so you get something more logical. The SoD problem exisists in almost every game out there, it is a really tough nut to crack, hell it still exists in Civ4 to some extent, though the splash damage tries to counter the effectiveness of it. But short of MP (which we don't have to deal with in GC) there will nominally always be a SoD problem facing the AI, once the player figures out how to exploit it. The biggest issue with fleet combat in GC seems to be the value of being the attacker. Ideally this will be tweeked to limit the value of always being the attacker.

Personally I feel that having an open galaxy in GC is a bad design choice, though one which seems unlikely to disappear. The system used in SEIV (and MoO3) with warp lanes meant that you could build up defenive points and force the attacker to come to you. Seemingly this is an easier situation for the AI to deal with, though it limits the players freedom. There are countless ways to tweek combat though, the issue for SP is how to tweek it so that it is still entertainting to the player, while not completely impossible to design a competant AI to deal with.

I understand that moving away from the open galaxy takes out anomolies, starbases and trade as they are currently implemented. That doesn't really bother me personally, but I think Brad is more or less married to this design.
on Mar 23, 2006
I strongly support post #16. It doesn't make sense that, in order to use a higher percentage of your labs, you have to use fewer of your factories, and vice versa. Especially, as someone said, if you have a lot more factories than labs, or vice versa. If you have 300 mp and 10 tp on a planet, why should you have to give up all 300 mp if you need to use the 10 tp?

I also generally agree with Sirian's point that the predominant "best way to play" is to always run either 100% military or 100% social (going back and forth between the two as strategy dictates). But I don't really agree with his conclusion from that (maybe I don't understand it). It's ok, to me, for the player running at 100% social to have waste because some of his planets don't have any social projects he wants. Or the player running at 100% military to have waste because he doesn't really want to build ships at some planets. This is just a cost of running a big empire en masse, and can be taken into account when planning overall strategy.

What I don't like is that if I build a factory that costs 50 mp, and my planet generates 24 mp/turn, then on the third turn, of my 24 mp, 22 of them are wasted. This creates a lot of micromanagement. Sirian endorses (I think) crediting the 22 mp back to the treasury. That would certainly be a major step forward, but I don't think it's enough. I'd like to see the excess 22 mp applied to the next social project in the queue (if there's none, then back to the treasury is fine).

There's the same problem with ships, and it's even a bit worse because rush-buying ships works differently than rush-buying social projects. When I rush-buy a social project, it appears immediately, and I can then use my social production on the planet to start building something else. But when I rush-buy a ship, it doesn't appear until next turn, and any military production on that planet is totally wasted. As above, I'd like the excess military production to be used to start building a new ship (it would be ok to default to the same type of ship I just finished, although a queue would be better).
on Mar 23, 2006
P.S. I also strongly suggest that the spending slider should begin a new game, by default, at 100%. Everyone wants it at 100% at the beginning of the game. To put it lower is just confusing newbies.
on Mar 23, 2006
4) The armchair game developers. These are the people who really could make a much better game if only they had the money and developers to convey their genius onto the screen. To prove their genius, they'll use terms like "broken" or "unusable" to describe anything that they consider non-idea.

Group #4 won't ever be satisfied. It's not like there's a scenario where we'd put out some update that totally guts the economic system.


Hahahaha thats a really nice way to say I'm not listening to you %&*$&^#^ who want me to change everything for their benefit only.
9 PagesFirst 7 8 9