Some Ideas to Reduce the Effectiveness of ICS

[Selador]

I'd like to propose some radical changes to the game to limit, as far as possible, the effectiveness of ICS'ing. Feedback is much appreciated.

Why is ICS so effective?

Well, there are lots of reasons, but here's the biggest: faster growth. The faster you expand, the faster you can start harvesting more resources than the next guy. You can then use those extra resources to expand even faster, and so on. This is a positive feedback effect which will result in a geometric increase in resource income over time. Thus, even a small edge in growth will yield a massive resource advantage over the course of the game.

So, in order to make other expansion paradigms more attractive alternatives, we need to modify the growth mechanics of the game to nullify the inherent speed advantage enjoyed by ICS.

What factors lead to faster growth?

1. Economies of Scale. The more bases you have, the more base tiles you have, which do not require a worker to generate resources. Thus, a citizen can work 1 tile in an existing base, or 2 tiles by founding a new base (base tile + worked tile). So just by founding new bases you net 'free' tiles of resources. That's a pretty big opportunity cost to overcome to justify growing a base beyond size 2.

2. Growth Mechanics. Small bases grow faster than big bases. A size 1 base needs 2 nutrient rows to grow, a size 2 base needs 3, etc. More generally, the growth speed of a base is governed by:

T = R*(S+1)/N

where T is the number of turns remaining before growth (round up), R is the number of nutrients per row, S is the size of the base, and N is nutrient output per turn. So we can see that not only does building lots of new bases net you lots of 'free' base tiles, it also achieves faster population growth by keeping S small.

3. Distance and Movement. The shorter a Colony Pod has to travel to a new base location, the faster it can found the base and begin harvesting resources. Similarly, the less time a Former needs to spend moving to a different tile, the more time it can spend making improvements and the quicker you will have access to the bonus resources they yield. Both Colony Pods and Formers have only 1 movement, meaning they can get a maximum of 2 moves along a road (leaving 1/3 of a move to build/terraform) before they need to start wasting turns to get to a further destination.

(did I miss anything... ?)

What can be done about it?

Here are the changes I have in mind:

1. Economies of Scale

• eliminate the free mineral and energy from the base tile
• increase the cost of building new Colony Pods from 30 to 50
• increase the upkeep for 'multiplier' facilities
• remove hab limit restrictions

The idea here is to shift the economies of scale away from horizontal development and towards vertical development. Most of the benefits from free base tiles are gone (base tile only provides 2/0/0), and bases now require a bigger upfront investment in the form of more expensive CPs. These two factors work together to reduce city spamming: it's pretty tough accumulate 50 minerals for a CP at a size 1 or 2 base without that free mineral from the base tile.

Building many small bases instead of fewer large bases always had the advantage of getting around hab restrictions. Now big bases no longer have this handicap. Hab Complexes and Hab Domes are now useless; perhaps they can be transformed into a new facility?

The idea behind increasing the upkeep for multiplier facilities (Net Node, Energy Bank, Robo Assembly Plant, etc.) is this: multiplier facilities provide more benefits in a bigger base, but at the same upkeep as in a smaller base.
The more bases you have, the more upkeep you pay if you're building these facilities in every base. Thus, by increasing the cost of these facilities, the ICS player is faced with an unpleasant choice: build these facilities in every base and face crippling upkeep, or don't build as many of them and concede higher output per citizen to the big base builders.

2. Growth Mechanics

• reduce nutrient intake requirement per citizen to 1
• increase the number of nutrients per row from 10 to 20

Yes, this is a pretty radical change. I've tested it out a few times and it seems like it has potential. The idea is that it normalize the rate of growth between bases of different size. Remember, bases grow according to T = R*(S+1)/N. Since only 1 nutrient is being consumed per citizen, in general bigger bases will see bigger nutrient surpluses. Thus, N should increasse at a rate proportional to S, and T will be relatively constant.

The reason for doubling the number of nutrients per row is to keep the overall growth rates similar to what they were before these changes. Basically, since I'm halving the number of nutrients required to feed citizens, I'm (nearly) doubling the average nutrient output. By also doubling the nutrients required for growth, the overall growth rate is kept (somewhat) constant. 20 is also a good round number so that it doesn't mess up the Growth SE effects.

3. Distances and Movement

• give formers and colony pods 2 movement
• increase movement rate along roads from 3 to 5

Well, this one's pretty straightforward. By increasing movement rates, we expand the distance a Former or CP can travel and still work in the same turn. This means you can space out your bases more without losing time to an ICS'er. Having movement rates along roads at 5 also means you can space bases up to 4 tiles apart and still get a 1 move unit from one base to another in a single turn, which nullifies a defense benefit that only applied to close base spacing (pre magtubes).

Additional thoughts

Obviously, these are pretty drastic changes. I'm sure there are other changes that will have to be made to compensate. For one, pop booming has to go, due to the fact that only 1 nutrient per citizen required now. That means CV needs to be disabled, and the SE table needs to be redesigned (sigh).

I may also have to shift some things around on the tech tree. I'm thinking of moving the Robo Assembly Plant down the tech tree so that it's available much earlier. Probably somewhere like Industrial Automation, but possibly even Industrial Economics. This should provide a good incentive to build bases up early, and should help compensate for the lack of base tile mineral. Thoughts?

Finally, I'd like to see the worldbuilder modified to promote less homogeneous terrain, i.e. more "patches" and "clustering" of rainy and dry tiles so that you need to think a little more about where to place your bases. Also, more worms and fungus to punish hasty expansion. I have no idea how to do all this though.

Whew, I think that's it for now. If there's any interest/feedback in these ideas, I'll try to get a full mod package out for testing.

[bdanv]

have you tried Mart's Unity Mod?
he's also trying to make the ICS strategy ineffective.

[Mart]

I'd like to propose some radical changes to the game to limit, as far as possible, the effectiveness of ICS'ing. Feedback is much appreciated.

Selador, that's great that you post your ideas for Smac. We have very similar thoughts about how to change ICS in Smac. Bdanv already pointed out Unity Mod, and what's there already is radical too (I think...)
Version 0.2, but many changes as you proposed in the post are already tested there.

Maybe first, there is one thing about changes to the game - AI is pretty much hardcoded in a sense, that we cannot modify it like in civ4. We could create a mod for humans only, so we do not need to worry about AI behaviour. And there may be a version of such mod - there are few things human can use unlike AI and gameplay would be even more enhanced.
A mod with AI requires compromise. And that is not that easy. If you check some UMod (Unity mod) creation thread thoughts, many things limit what and how you can change. For example economy, one has to be careful to keep AI earning and spending right. AI does not understand "build less bases." With high upkeep it will go bankrupt. In UMod this happens to Hive in ver. 0.2 and I'm not sure how to change that.
In general approach is to tweak game rules in the way that human player chooses more to follow what hardcoded AI does.

...
What can be done about it?

Here are the changes I have in mind:

1. Economies of Scale

• eliminate the free mineral and energy from the base tile
• increase the cost of building new Colony Pods from 30 to 50
• increase the upkeep for 'multiplier' facilities
• remove hab limit restrictions

"eliminate the free mineral" - done and even more. base has -48 energy, and only recycling/pressure dome returns that. The effect is, that with high upkeep of both (10 and 9) restoring productivity to base tile costs much more - no longer free.

"increase the cost of building new Colony Pods..." - done even more than you propose. good idea, cause this slows down AI too. For human player, high investment benefits better positioned base, than just ICS 2 tiles away.

"increase the upkeep for 'multiplier' facilities" - so far I found that a better solution maybe something more like civ4. Little upkeep for most of facilities, and only some very expensive. Like recycling, economy multipliers. This gives effect that we do not have the "snowball" effect with energy production in later game, possibly making late game more interesting. AI (except Hive) seems to work well with it in early tests.

"remove hab limit restrictions" - umod has something similar, they are redesigned into power plants, giving negative upkeep - so actually giving you energy. They are cheap and early, so you both want them asap and AI builds them fast too.

...

2. Growth Mechanics

• reduce nutrient intake requirement per citizen to 1
• increase the number of nutrients per row from 10 to 20

"reduce nutrient intake requirement per citizen to 1" - I tested both 1 and 3. But 2 seems to work the best.
And 1 cannot be used cause of satellites, if you want to keep them, intake must be greater than 1, otherwise you would have no population growth stopper. It accepts size 254, iirc, anyway - variable limit. I'm not sure if it flips then to size 1.
Well, actually, you might crawl nutrients, but that's more difficult, in practice One City Challenges achieve that, but regular games rather not.

"increase the number of nutrients per row from 10 to 20" - umod has opposite, row length 6, and the reason is to promote faster growth. 6 works well with society engineering, but growth effects were redesigned to +/- 20% increments.

also umod made easier pop booming - for AI. If both human and AI can pop boom, in practice you go only to your nutrient supply limit. Fast 'jump" to that spot by both human and AI is intended, then you may have pop booming feature in game.

...

3. Distances and Movement

• give formers and colony pods 2 movement
• increase movement rate along roads from 3 to 5

I changed chassis movement rate. infantry has now 2. and military seems to play ok with it too. rover has 3.

...

Additional thoughts

Obviously, these are pretty drastic changes....

Yes, I think also that drastic changes are necessary.

Personally, I hope that it would make SMAC "fresh" again, less micromanaging and more enjoyable in late game, more builder, etc.

[Selador]

Mart, thank you very much for your comments! It's good to see that others are thinking along the same lines as me.

AI is pretty much hardcoded in a sense, that we cannot modify it like in civ4. We could create a mod for humans only, so we do not need to worry about AI behaviour.

Yeah, it's a shame that AI coding cannot be changed. However, I think it IS possible to modify AI behaviour via alpha.txt changes, to a small degree, at least. Creating a 'human player only' mod is another way to approach things, but I would prefer to not to have to go this route. Despite the crappy AI, I think most people still enjoy playing single player (I do at least), and I feel a mod that has no functioning AI would have far less appeal.

We're just going to have to use what we have, unfortunately.

In general approach is to tweak game rules in the way that human player chooses more to follow what hardcoded AI does.

Yeah, this is a pretty good modding philosophy I think. We should focus on changes that give the AI a leg up relative to the human player. That's the idea behind curtailing ICS. The AI just doesn't know how to do it, so it's basically an instant-win for any competent human player employing the strategy.

Plus, mirco-managing hundreds of bases just isn't my idea of fun.

"eliminate the free mineral" - done and even more. base has -48 energy, and only recycling/pressure dome returns that. The effect is, that with high upkeep of both (10 and 9) restoring productivity to base tile costs much more - no longer free.

Hmm, interesting idea. I'll have to take a look at all the changes you made. How did decide on -48 energy? That seems a bit much, I'm not surprised to hear Yang was having problems, lol.

"increase the cost of building new Colony Pods..." - done even more than you propose. good idea, cause this slows down AI too. For human player, high investment benefits better positioned base, than just ICS 2 tiles away.

Yup, this one's a no-brainer, IMO.

"increase the upkeep for 'multiplier' facilities" - so far I found that a better solution maybe something more like civ4. Little upkeep for most of facilities, and only some very expensive. Like recycling, economy multipliers. This gives effect that we do not have the "snowball" effect with energy production in later game, possibly making late game more interesting. AI (except Hive) seems to work well with it in early tests.

Yeah, this is what I had in mind. Low upkeep for most facilities, high upkeep for multipliers. However, it's important that the upkeeps are high enough that a small base simply cannot cover them all and so is forced to NOT build some facilities or find an expansion strategy that allows bigger bases (hmm...).

Also, the "snowballing energy" effect should already be much less, because the root cause is the exponential increase in bases and population. Assuming the mod is successful in discouraging ICS, this 'expansion explosion' won't happen.

The AI should be ok, as it rarely comes close to building all available infrastructure in any of its bases. There's also the option of creating a set of 'super AIs' which can handle this change.

"remove hab limit restrictions" - umod has something similar, they are redesigned into power plants, giving negative upkeep - so actually giving you energy. They are cheap and early, so you both want them asap and AI builds them fast too.

Cool idea . I may consider doing something like this instead of removing hab limits altogether. I'd like keep the full array of facilities in the game, if at all possible.

"reduce nutrient intake requirement per citizen to 1" - I tested both 1 and 3. But 2 seems to work the best.
And 1 cannot be used cause of satellites, if you want to keep them, intake must be greater than 1, otherwise you would have no population growth stopper. It accepts size 254, iirc, anyway - variable limit.

Hmm, that's true about the satellites, there will be no cap on growth. I didn't realize there was an upper limit on base size. There are 2 solutions I can see: (1) disable nutrient satellites, or (2) shift the second hab limit to whatever the max base size is and disable hab domes.

I'll have to do one or the other, because changing the nutrient intake to '1' is absolutely critical. It is the most important aspect of this mod, and without it I don't see how you can hope to eliminate ICS. I explained why in my post under 'growth mechanics'.

Leaving nutrient intake at '2' (or worse, bumping it up to 3) fails to fix the problem of faster growth in small bases, which will always promote horizontal expansion (i.e. it's better to grow by building CPs than base facilities). This is the opposite of what we want if we're trying to discourage an ICS approach.

"increase the number of nutrients per row from 10 to 20" - umod has opposite, row length 6, and the reason is to promote faster growth.

Well, yeah, shorter nutrient rows means bases grow faster, true. However, this does little to curb ICS, it only serves to speed up the game. So if that's what you're going for, sure. The only reason I made nutrient rows of 20 was to keep the game speed relatively constant so that I don't have to mess with build costs, tech rates, etc.

My goal here is to change the relative growth rates of ICS and non-ICS based alternatives, not to speed up the game. For now, I'm only trying to make changes that will cripple ICS as a strategy while leaving everything else the same (so far as possible).

also umod made easier pop booming - for AI. If both human and AI can pop boom, in practice you go only to your nutrient supply limit. Fast 'jump" to that spot by both human and AI is intended, then you may have pop booming feature in game.

How do you go about making pop booming easy for the AI?

Regardless, I don't think pop booming can be implemented with nutrient intake set at '1'.

Personally, I hope that it would make SMAC "fresh" again, less micromanaging and more enjoyable in late game, more builder, etc.

Yup, sounds good to me!

[Mart]

UMod has multiple objectives and dealing with ICS is one of them. The combination of changes has effects in various game aspects. For example, while nutrient row is 6, mineral is made 13, but at the same time many facilities are made cheaper in how many rows they cost. Even more cheaper, which in effect offsets having less bases without ICS. When you can produce quicker facilities in existing bases that is even more incentive for building that expensive colony pod later.

Faster base growth by row of 6 nutrients offsets, that you produce colony pods much slower, they cost 12 rows! This way, you reach certain size limit depending how much food there is around. Note, that crawlers are very expensive too, so crawling nutrients is so expensive that a player would use it rarely. Thus base size depends on how much terraforming you have, technologies (condenser) other like satellites. But growth is faster.

Pop booming easier is done by shifting whole growth scheme by assigning in politics all growth factors by +2:

...
Centralized, None, ++GROWTH, -EFFIC
Police State, DocLoy, ++POLICE, ++SUPPORT, --EFFIC, ++GROWTH
Democratic, Psych, ++EFFIC, ++++GROWTH, --SUPPORT
Thought Control, WillPow, ++POLICE, ++MORALE, ++PROBE, ---SUPPORT, ++GROWTH
...

Democracy is +4 then. AI can do pop boom this way easier.
SE is also redesigned, I inserted Thought control there (anyway, looks like perfect political system... for the rulling elite that is )

Also, all faction files are redesigned, as for the bonuses. Take a look into UMod, there are many changes so far, and more planned. You can see how certain combinations of new rules work. So far, I am testing version 0.2 and still much to tweak...

Some way of dealing with food satellites is to make them much more expensive. problem that AI may stop building them above certain cost, I think it may be close to 60 rows or something.

[Mart]

I forgot to answer the -48 and +49 energy for base square.
It is about giving AI incentive to build expensive (10EC/turn) recycling. I think it may be that 10 EC cost for having +49 may trigger AI, and they do build recycling fast usually. It is meant to incur flat cost per base and quite high, so you develop new bases only when your economy allows you to do it. there are many changes to economy, so take a look into alphax.txt or maybe play a quick game.
Some problem is that AI does not understand this and spams colony pods the same, but this 12 rows cost controls this ok so far.

----------
Maybe I will add it here.
There is a catch in fighting ICS.

AI does ICS on landmarks. Jungle is most obivous, cause it has usually large area, but crater, uranium is similar. Now when you kill ICS, that would hinder AI in landmark areas. For example, making preferable that bases do not overlap, that would work bad for AI.

[Quadhelix]

Hmm, that's true about the satellites, there will be no cap on growth. I didn't realize there was an upper limit on base size. There are 2 solutions I can see: (1) disable nutrient satellites, or (2) shift the second hab limit to whatever the max base size is and disable hab domes.

I, personally, would go with option 2, but that is purely subjective.

I'll have to do one or the other, because changing the nutrient intake to '1' is absolutely critical. It is the most important aspect of this mod, and without it I don't see how you can hope to eliminate ICS. I explained why in my post under 'growth mechanics'.

Leaving nutrient intake at '2' (or worse, bumping it up to 3) fails to fix the problem of faster growth in small bases, which will always promote horizontal expansion (i.e. it's better to grow by building CPs than base facilities). This is the opposite of what we want if we're trying to discourage an ICS approach.

One thing that I discovered is that, if a faction has an Efficiency of -4, they will get one Bureaucracy drone for each base that they build. This means that lots of small bases would lead to...significant drone problems.

Implementing this would mean that all Faction and Social Engineering effects on Efficiency would have to be removed save either a) giving each faction a -4 Efficiency penalty or b) giving each SE choice in a certain branch a -4 efficiency penalty.

This would, of course, necessitate a new penalties for Lal, Svensgaard, Police State, and Planned, as well as new bonuses for Deidre, Aki, Democracy, Knowledge, and Cybernetic.

SE is also redesigned, I inserted Thought control there (anyway, looks like perfect political system... for the rulling elite that is )

A) The button for Thought Control will be cluttered.

B) Thought Control definitely belongs in the "Future Society" category - in part because it is highly similar to Police State in certain regards, i.e., there is enough overlap between the two that it should, theoretically, be possible to run both at the same time.

[Mart]

Future society was renamed into just Society

And it is somehow hidden now. Thought Control is future, just that it appears in politics.
Society has in early game 2 choices: anthropocentric and fundamentalist. later future choices, eudaimonia and cybernetic are available.

This new system makes the default 4 choices somewhat more functional, no just plain 0 effect. AI can choose them for some reasons and also human player, that's true, they probably cause no AI feelings (not sure, is default choice hardcoded to not cause AI feelings?) but overall having a bit more peacefull game does not hurt. AI eventually attacks anyway. Political/diplomacy with AI gameplay seems ok to me.

[Quadhelix]

Future society was renamed into just Society

And it is somehow hidden now. Thought Control is future, just that it appears in politics.
Society has in early game 2 choices: anthropocentric and fundamentalist. later future choices, eudaimonia and cybernetic are available.

I still say that Thought Control and Police State are too similar in concept. If anything, I would make Cybernetic a Politics system (it is, after all, a system in which people have "turned over the tasks of governing to sentient computers") and make Thought Control a "Society" choice...of course, it would have to be a choice that is also different from Fundamentalist.

For example, in a concept that I was working on for a different mod, Fundamentalist politics' Probe bonus was removed so that I could move the +2 Growth from Democracy to Fundamentalist (Democracy got a research bonus to compensate). You could do something similar: replace Fundamentalism's Probe bonus and Thought Control's Morale bonus with something else (e.g., Growth for Fundamentalist and perhaps Industry for Thought Control).

Of course, this would require a few changes to other SE choices, depending on which bonuses you chose to change (for example, if you can run Fundamentalist, Democracy, and Planned at the same time, then only two of them can have the +2 Growth bonus).

[Mart]

Quadhelix, thank you for remarks. Good points and ideas.
SE table is not very clear anyway. For Umod, I think it will be redesigned, for now it's one of the testing versions. Things like politics, values and society models are overlapping and maybe that is why it's so difficult to find some very good model.

Some problem is that certain secret projects have hardcoded SE choices changes.

How about the ICS effects? That is main topic in this thread. Playing with Umod, one can try ICS and see how it works with changes, that many are similar that Selador proposes.

Selador, do you have an alphax.txt modified and playable with the changes you think to put in?

[Selador]

Selador, do you have an alphax.txt modified and playable with the changes you think to put in?

Not at the moment. Unfortunately it's not just a matter of going into alpha.txt and making the changes I listed; it needs to be "playable" as you say. There's going to be a bunch of other changes I need to make to compensate (like social engineering for example). Part of the reason I made this thread is for people to point out what some of these changes might be.

Basically, I'm just brainstorming ideas for now and wanted some feedback about the effective of the changes I'm proposing and possible consequences I didn't foresee. However, I am planning on getting one up for testing once I have a more concrete outline on exactly what changes need to be made.

[Mart]

Modding takes time. Playable - I mean also like "not crashing," but I agree, that making just a few changes at a time will prolong mod building and one can spend time for effects that may be very different when bigger set of changes is in place.

Smac comunity is now so slow, that there will be probably only few of us sharing ideas. Good place to see what others tried is to also search this forum and its smac archives. There are many modified alpha/x files, and people were discussing it in threads, and there are several mods with larger modifications that come to my mind:

SNAC - one of the earliests
Aldebaran, Aldebaran 2 (smax)
SMAniaC

[Quadhelix]

Well, one possibility to discourage (though not invalidate) ICS might be to increase the effectiveness of specialists (Technicians, Librarians, etc.) such that a base with 2 Technician and 2 Librarian can match/exceed 4-5 pop-1 bases. This may reduce the need to change nutrient requirements.


So, let's analyze this for a second: firstly, you've already changed the standard base square to 2,0,0. This means no free minerals/energy from multiple bases. Now, it's just down to the citizens.

With decent terrain, a Worker can net, on a farm/panel tile, say, 3-1-3. Thus, a size 1 base with no Recycling Tanks will have an output of 5-1-3. Five such bases would produce 15 energy/turn-let us split this into 8 Energy Credits (EC) and 8 Research Points (RP); I know that 8+8 is 16, not 15, but that doesn't really matter here. Note, also, that I am ignoring the effects of inefficiency - this is an extra boon for the Specialists because, IIRC, their addition to a base's output is factored in after the base's energy loss to inefficiency.

Now, let us concentrate these 5 citizens into a single base. Firstly, in order to get the 10 Nutrients needed to feed the base you have to run 3 Workers, leaving room for 2 Specialists. Now, these 3 workers return 9 energy, which splits into 4 EC and 4 RP. If, for example, Technicians produced 4 EC and Librarians 5 RP, then this would outproduce the 5 independent bases. Higher populations are just gravy


Now, there is the problem of how to keep the size-1 bases from growing to size-2 bases? I was thinking to make the maximum base size without a Hab Complex 1, give Hab Complexes have no technological prerequisites (or perhaps require industrial base, or some other low-level technology) and are fairly cheap to build (3 mineral rows) - but not to maintain (3 EC/turn). This makes any base larger than 1 an important investment. It does somewhat negate the strength of Specialist bases, but it also negates the "faster growth" advantage of ICS: there is no growth unless you build a Hab Complex, which is a long-term investment.


Another possible alteration is to give Recycling Tanks an output of 1/3/2 (to make up for Workers that have been converted to Specialists) but an upkeep of ~3 EC/turn. This way, if you build Recycling Tanks at your 5 ICS bases, you get an upkeep cost of 15 in exchange for 10 energy...much of which will be lost to inefficiency. For the Specialist base, however, the upkeep is more manageable (by allowing the base, in theory, to grow another size, it allows the application of an extra Technician/Engineer, which more than offsets the extra cost - indeed, the extra Technician almost completely pays for both the Recycling Tanks and the Hab Complex).


Of course, there is the possibility that my analysis if fundamentally flawed, but I figure you guys would spot that better than I could.

[Mart]

There is somewhere a good analysis of ICS, on this forum, or maybe another forum. I believe Petek, or someone else, not long ago posted a link.

Anyway, one of the ICS features is that you place bases close to each other. In a dense scheme you can place 1 base on 4 tiles - they cannot be on neighboring tiles, so this is as dense as possible. In between you put condenser, borehole. Base then has quite a number of nutrients and minerals from just 4 tiles. Benefits are from specialists. Crawlers and satellites make it even more possible to grow to large size with majority of population being specialists.

This is one of ICS things, but to force a player to spread bases more would be to make a worker more preferable than a specialist. Then you would rather have 20 workers in a base and specialists from population above 20. However! this works well with deteriorating base square profitability and high base upkeep. Making a base expensive helps, like recycling tank costs 10 EC/turn, for example.

[Mart]

Something to consider is that when bases do not overlap, we have more or less 1 base for 20 tiles, in dense ICS it is 1 base for 4 tiles. So roughly in ICS we have 5 times more bases per given terrain area.

Now, the balance can be estimated how to do it, that it is better to have 1 base covering 20 tiles than 5 bases doing the same.