Reasonable unit cost

[Alpha Centauri Bear]

I was investigating unit cost options in Yutzi's mod and found myself quite overwhelmed with all tricky formulas. Original SMAC designers tend to introduce quite complex and non linear formulas for the purpose of balancing outcomes, I guess. However, I don't see much need in unit cost fine tuning. Let me share my thoughts.

First of all, I don't think the fine tuning is going to change the experience much. Players tend to build high end units at any point in time so the choice for mass production is pretty much limited to few land units, one sea, one air. The balance between triad can be adjusted by chassis cost and balance between different weapon and armor with same chassis can be adjusted by corresponding weapon/armor cost as well. So there is no need to build up an insanely complicated cost formula.

I believe this formula should be simple and reflect a very generic dependency on parameters (weapon, armor, chassis). The better unit should cost more. So the weapon should add to cost the same amount as equivalent strength armor as they contribute the same to odds. Speed is a good thing as it allows to attack more and farther. So the initial thought on cost formula looks like this.

 cost (minerals rows) = <weapon cost> + <armor cost> + <chassis cost> - 2

Subtracting 2 to start from 1 row of minerals. As simple as that!!!! The more good things you add to the unit the more it costs. And the chassis impact can be easily adjusted by corresponding chassis cost. For example, if one would like to make sea units as cheap as land ones they just need to set foil cost to 1 and cruiser one to 2 or similar. Same story for needlejet and copter chassis. There are no much of chassis types in the game.

Now let's touch the infamous reactor and price topic. Reactor effectively multiplies both attack and defense strength by reactor multiplier. So, with all things being equal, I believe bigger reactor should multiply cost proportionally. Otherwise, a single fusion technology discovery makes your units twice as strong and twice as cheap = 4 times more economically effective. I strongly discourage such changes as they lead to inevitable conquest only oriented strategy. Stronger units already have an enormous advantage: they win all the time and become indestructible. Why to design a game where units cost decline while your economy grows? This would lead to bigger number of units for everybody and excessive mouseclicks.

Summarizing above, unit price declining = broken strategy + excessive mouseclicks. Whereas, naturally growing unit price = smooth progression with manageable army. I believe something under 50 attacking units is pretty optimal for both variability and manageability.

Of course, all above thoughts are completely subjective and speculative. My main point is that there is no need for overly complicated unit cost formula. Feel free to comment. Thank you.

[bvanevery]

In my mod, I got rid of the "weird" formula where some things were dependent on the ratio of armor to weapon or whatever.  I also got rid of free Trance and ECM abilities.  All this combo of stuff ever did, was encourage the player to spend a lot of time minimaxing all the little differences in the Unit Design Workshop so that they could get stuff for "free".  The game wouldn't figure this out for you, and would pester you with bad unit designs.  Everything is simpler if the cost is computed the same way for everything, and if almost nothing is free.  The only thing I relented on was Deep Radar for sea and air units.

In the same vein, I got rid of the complicated formula for computing Heavy Artillery costs.  If you want to make a heavily armored artillery unit, you can, it's not going to cost more than the armor.  It doesn't happen to do you any good in artillery duels because of the way the game works, so you probably won't.  Still, if you're worried about being counterattacked by something other than artillery, you can armor up.

With the exception of the tiny number of free abilities, I changed all ability costs to 1.  I'm not convinced that those that cost 2, were inherently valuable and more important than other abilities.  They would tend to come late in the tech tree and nobody got to use them for much of anything.  Generally speaking I moved them earlier in the tree, so that they'll actually get used in a game.  So if you can prove that cheap Cloaking Devices break the game somehow, because you can have them in late midgame, let me know.  In my own late midgames, I've surely been playing awhile anyways, and really wouldn't mind something that lets me win the freakin' game.  But I don't think it does.

Clean Reactors are much earlier, an early midgame tech.  The AI does actually build some Clean Reactor units.  Not enough to solve its tendency to exhaust its mineral supply though.

In sum, yes, I agree with simplifying unit costs.  I don't think Yitzi's extensive inventory of cost computations has any value.  To demonstrate value, I think you'd have to use them in a mod and then do a lot of testing of the balance.  I bet you'd find out that all that wonking on cost really doesn't make any difference.  Players are going to build units.  Then they're going to get used.  The dominant issue is what kind of tech is available and how fast people get through the tech tree.

I did cheapen the foil and cruiser chassis, cutting them in half.  That puts them on parity with a speeder and a hovertank, respectively.  In the far future, I don't see a motive for implementing "shipping is helluh expensive".  This isn't WW II.  The cheaper chasses means maritime expansion is much quicker, and sea bases are more likely to change hands.

Still, the AI is not basically good at invading over water.  In that Yang AAR where I flooded the world, various factions that were teetering on the brink of death, were actually saved by the flood.  Then later in the game then nuked me!  So I quit; that game was not basically enjoyable.

Generally speaking when looking at Yitzi's stuff, although I appreciate the sheer quantity of work he did, I think one needs to keep a critical eye on whether any given patch feature is useful for something.  Some of it almost comes across as a lot of brainstorming, rather than a concrete mod design.  "You could do this, or you could do this, or this, or...."  Yes but did anyone do that, in a mod, that people actually wanted to play and liked, that they stuck with, that a community developed around?  This is an issue of curation.  Raw capabilities don't mean anything if no one turns them into a high quality game design, that people want to keep playing.  And that's a lot of work; it's not about having a pile of modding features either.

I am inclined to guess that a lot of stuff Yitzi did, he did because he could.  Meaning, he opportunistically found something in the code that he realized he could change, without it being much work.  At least, this is what I suppose, as he's not on hand to ask.  He disappeared from around here, sometime not that long after I showed up, and he didn't seem that talkative at the time.  Probably burned out, too much been there done that!

I can relate, although my own modding hasn't completely exhausted me.  Perhaps because, I knew something about scope of work, and deliberately kept a lid on energy I would expend.  It still turned into 4 person months of full time work anyways, monopolizing my psychological energy for 6 months.  For the raw labor, that's equal to a 4 month full time project I did back in the Wesnoth community awhile ago.  That was my example of don't get sucked into doing more than that thing.  I succeeded....

Um, reactors.  Almost forgot to comment on that.  I'm not opposed to taking different sized reactors completely out of the game.  Pushing Fusion, Quantum, and Singularity to the end of the tech tree doesn't work.   I tried that, but it was a complete mess when you actually got there.  Modding the reactor sizes doesn't seem to actually do anything.

I wasn't ready to remove different reactors from the game, because I didn't feel I'd gotten enough playtesting reports from people to make a decision on that.  I still feel that way.  But if I heard from a quorum of people who said yes, Fusion reactors suck, I'd get rid of them.