Unit cost formula: Alternate approaches (taking requests)

[Earthmichael]

I still like the idea of clean reactor costing a constant 1 extra row, regardless of reactor.  As you get later in the game, where the higher reactors show up, you have less time to get your payback.  And I like eliminating the need for micromanaging what city is supporting what unit, to get things balanced, so I prefer to make most everything clean.  I don't think increasing the cost of clean by reactor size makes sense.  In fact, weapons and defense costs go down with higher reactors!  So I think a constant 1 row for clean is a good idea.

Otherwise, I will just keep making my reaction 1 clean formers, which seems stupid when I have fusion technology ( or higher) but while my massive attack unit have significantly lower cost with higher reactors.  Makes no sense to me.

As for formers and crawlers, I tend to place my crawlers and then forget about them.  No micromanagement needed.  Formers I tend to reduce the micromanagement by building road networks, and then sending them out in packs of say 4 formers.  I find it almost as fast to select a target and send a pack of 4 formers, than to do this with a single former, so the micromanagement time goes down by a factor of 4.

[Nexii]

Yes, that's the issue with reactors and non-combat units.  There's little benefit to non-combats otherwise and that's why in the unmodded game, they made reactors also reduce unit costs.  The problem with that was that reactors became much too crucial for military units.  For example Fusion reactor is like going from 5 attack to 10 against someone without it, *and* 3 to 6 defense.  There's no sense going down those sides of the tree early.

I think that the best available reactor and design should be encouraged on non-combat units - and this is why when I was making my formula for mode 2/4 that I said really reactor cost reduction should just be stripped out for military units.  It should be a survivability increase only.  However you're right in that non-combats would need some benefit from better reactors.  Hitting them twice then with higher maintenance on top of that wouldn't make sense.  So they would have to be exempt.  I still say formers need a maintenance structure all to their own, and they're most of the reason for the super-fast mid into late game.

[Yitzi]

Good points.  Maintenance on a per-unit basis is probably best. 

Keep the #2/#4 modes as-is then.  I can balance things around ECM/AAA cost and power from there.

And yea, I think that's my only real remaining gripe is the former/crawler cost.  They're fine to start but should increase the more of them you have.

So an extra cost per X formers/crawlers?  It could be done, but it's awkward enough that I'll delay it until I'm ready to take general requests (currently I'm only taking stuff that I want, or that I like once I hear it.)

I still like the idea of clean reactor costing a constant 1 extra row, regardless of reactor.  As you get later in the game, where the higher reactors show up, you have less time to get your payback.

But whatever else you'd spend that production on would likewise have less time to be effective.

And I like eliminating the need for micromanaging what city is supporting what unit, to get things balanced, so I prefer to make most everything clean.

Why would they need to be balanced perfectly?

I don't think increasing the cost of clean by reactor size makes sense.  In fact, weapons and defense costs go down with higher reactors!  So I think a constant 1 row for clean is a good idea.

It isn't increased with reactor size; it's increased with unit cost, which I think could work well if support costs increased similarly; improving reactors then decreases the cost of clean, but improving weapons and armor increases it.  Increasing support costs with reactor was an interesting idea, but I think making it based on production cost is a better approach.

Yes, that's the issue with reactors and non-combat units.  There's little benefit to non-combats otherwise and that's why in the unmodded game, they made reactors also reduce unit costs.  The problem with that was that reactors became much too crucial for military units.  For example Fusion reactor is like going from 5 attack to 10 against someone without it, *and* 3 to 6 defense.  There's no sense going down those sides of the tree early.

Keep in mind, though, what else is at the same level as fusion: Copters, satellites, maglevs, genejack factory...there's a lot of strong stuff at that stage.  Of course, with the new modding options, you can weaken the overpowered stuff as needed.

I still say formers need a maintenance structure all to their own, and they're most of the reason for the super-fast mid into late game.

Any ideas?

[Nexii]

And I like eliminating the need for micromanaging what city is supporting what unit, to get things balanced, so I prefer to make most everything clean.

I have to agree that this whole mechanic really is tedious and very pro-ICS.  Ideally I think the game would consider support based on total citizens (or even just bases like it does now), and then allocate out support costs automatically across your cities.   Kind of like how waste is auto-calculated for energy based on EFFIC SE. 

Keep in mind, though, what else is at the same level as fusion: Copters, satellites, maglevs, genejack factory...there's a lot of strong stuff at that stage.  Of course, with the new modding options, you can weaken the overpowered stuff as needed.

Yea I compensated the reactor power vs weapon/armor a bit more by increasing the weapon/armor scaling rate.  And that might be true...I'll have to try a few games going down the 'defensive' tree over the 'satellite' tree.  I think reducing satellites and CBA would help a bit here too.  There are a lot of mechanics I feel are sort of 'broken' in the sense that when a new tech becomes available, you spam it solely (e.g. Copters).

[I still say formers need a maintenance structure all to their own, and they're most of the reason for the super-fast mid into late game.]

The ideal fix would to tie non-combat unit effectiveness (esp formers'crawlers) more closely to their actual unit costs.  a 20 mineral reactor former should terraform a lot slower than one that has 60 minerals invested.  Right now there's only one real upgrade and that's Super Formers.  The ideal way of them doing this would have been to put in a few terraforming 'weapon' types up the Centauri tech chain.  And similarly for colony pods, probe weaponry, transports, etc.  Alternatively non-combats could just scale with their reactor since the HP doesn't help them otherwise.

A lot of things in the game can be evaluated in terms of mineral payback period (at least in the early game).  Clean at 10 cost is a *really* good payback (10 turns, and 5 at low SUPPORT SE).  Crawlers can be even better (30/6=5 turns if you crawl a borehole).  Or even in the early-mid, a 4N crawler feeds four forest tiles for 8M/4E, for an even faster payback (granted there are maintenance costs to drones).  Formers have the best payback of anything in the game, hands down.  They improve terrain, generating more and more resources for bases on an initial low investment of 20 minerals.  This is ok to some extent...without Formers the growth rate of the game would be too slow.  Also being on bad terrain would be even more punishing than it is without fixed improvements like Forest, Borehole.  There is a *lot* to consider when approaching Former unit costs.  Efficient terraforming is a skillful and still debatable part of the game that I don't think should go away.

Another idea I was toying with was formers and/or crawlers that expire after X turns (if this is even possible).  X could be increased by reactor and possibly support SE.  But I'm not sure how well that would play out.  They'd probably have to also give 0 minerals on disbandment to avoid high micromanagement.  A hybrid of these two ideas might work as well.  X could be fixed and the terraforming speed could depend on reactor and/or Centauri-like techs.

[Yitzi]

I have to agree that this whole mechanic really is tedious and very pro-ICS.

There are enough other anti-ICS features (energy-boosting buildings), though, that I think it will only end up making ICS a good idea for a heavily aggressive (and thus high on supportable units) strategy, which probably should be ICS-supporting.

Ideally I think the game would consider support based on total citizens (or even just bases like it does now), and then allocate out support costs automatically across your cities.   Kind of like how waste is auto-calculated for energy based on EFFIC SE.

Proportionally to production, I assume?  That'd be a huge hassle to code, though, as it means calculating something (total support cost) on a faction-wide level rather than base-wide, which nothing else is.

Yea I compensated the reactor power vs weapon/armor a bit more by increasing the weapon/armor scaling rate.  And that might be true...I'll have to try a few games going down the 'defensive' tree over the 'satellite' tree.

If you do and want to make it better with modding, consider removing the armor boost at probability mechanics (which gives tachyon fields anyway) and move it to advanced subatomic theory, so that every step in that tree boosts  armor.

I think reducing satellites and CBA would help a bit here too.

For satellites, try switching the techs of orbital defense pod and sky hydroponics labs, cutting defense pod cost to 80, and increasing hydroponics and nessus mining station cost to 180.  CBA would be weaker if air power were weaker, so maybe remove copters' speed bonus from reactor, make air power require AMA, move air superiority to synthetic fossil fuels, and have Probability Mechanics double the effectiveness of ECM and AAA.

The ideal fix would to tie non-combat unit effectiveness (esp formers'crawlers) more closely to their actual unit costs.  a 20 mineral reactor former should terraform a lot slower than one that has 60 minerals invested.  Right now there's only one real upgrade and that's Super Formers.  The ideal way of them doing this would have been to put in a few terraforming 'weapon' types up the Centauri tech chain.

Adding new "weapon" types, except as replacements for old ones, is not feasible.

Alternatively non-combats could just scale with their reactor since the HP doesn't help them otherwise.

It does help if they do get attacked...and then what about if you make armored formers?

A lot of things in the game can be evaluated in terms of mineral payback period (at least in the early game).  Clean at 10 cost is a *really* good payback (10 turns, and 5 at low SUPPORT SE).

I think formers also tend to be in that area.

Crawlers can be even better (30/6=5 turns if you crawl a borehole).

Though once you have a borehole, wouldn't you be better off working it?

Or even in the early-mid, a 4N crawler feeds four forest tiles for 8M/4E, for an even faster payback (granted there are maintenance costs to drones).

Again, 4N is a lot of terraforming work that could usually be better used than placing condensers everywhere (the only way to get 4N before advanced ecological engineering.)

Formers have the best payback of anything in the game, hands down.  They improve terrain, generating more and more resources for bases on an initial low investment of 20 minerals.  This is ok to some extent...without Formers the growth rate of the game would be too slow.  Also being on bad terrain would be even more punishing than it is without fixed improvements like Forest, Borehole.  There is a *lot* to consider when approaching Former unit costs.  Efficient terraforming is a skillful and still debatable part of the game that I don't think should go away.

Another idea I was toying with was formers and/or crawlers that expire after X turns (if this is even possible).

This would also be highly awkward, as well as difficult to code.

[Nexii]

There are enough other anti-ICS features (energy-boosting buildings), though, that I think it will only end up making ICS a good idea for a heavily aggressive (and thus high on supportable units) strategy, which probably should be ICS-supporting.

Proportionally to production, I assume?  That'd be a huge hassle to code, though, as it means calculating something (total support cost) on a faction-wide level rather than base-wide, which nothing else is.

That's true.  Instead of looking at this issue of assigning units to bases, from the top-down, how about bottom-up? Would there be any way of auto-assigning a unit to another base with excess support capacity when it's built, if the current base is at its support limit?

Adding new "weapon" types, except as replacements for old ones, is not feasible.

I figured not.  In fact when I increase weapon values the game seems to use the wrong icons.  I'm thinking weapons were hard coded in differently from armor for some reason (perhaps tied to military/non-military unit code).

I think formers also tend to be in that area.

For their first payback perhaps.  But no, formers are much much stronger than crawlers or anything else *in the long run*.  In the short run sometimes crawlers can be stronger and I'd have to really sit down and try to analyse optimal early to mid game terraforming mathematically.  Has anyone done this to date?  I'm sure most veterans already have a better 'feel' for what gives the best growth than I do. 

Anyways, the reason is formers continually improve terrain, not just once.  Think of it like a base improvement that builds base improvements.  So I pay 20 minerals for a former.  The first forest I make might be +2 minerals on bad terrain.  So that's 10 turns to payback.  The thing is they don't have a one time benefit like crawlers and structures.  The former then goes off and ~5-6 turns later has made another forest.  This forest then does the same and I'm gaining 4 minerals - already exceeding a crawler on forest or really anything else in the game at that cost.  And so on.  Eventually a former has improved 10 forests and *every* turn it's giving 20 minerals payback.  The limiting resources are workable squares and former time, which make this a bit complex to work out.  Often times workable squares is the bottleneck - more formers are rarely difficult to make as each only requires 1 mineral upkeep.  Obviously a base can only hit size 3 on Forests alone and so then you need to build another base or give it more nutrients.  The more I play I find nutrients are always the best thing to crawl and never minerals or energy.  I think the optimal setup might actually be all boreholes and crawling all nutrients on condensors once they're available.

But to get back on topic about unit costs.  It's quite apparent that formers (and to a lesser extent crawlers) are what drive the non-linear growth rate in the game.  Is there any way to make it so later game formers and crawlers are significantly better for their unit cost? (i.e. improvements with reactor upgrades) Perhaps making formers terraform faster - I imagine any sort of more exotic new terraforming abilities are not possible.  With crawlers perhaps a capping system that eventually allows uncapped crawling of multiple resources (or a bonus to energy/mineral crawling).  My thought was that losing all your early game formers/crawlers shouldn't necessarily be game over.  The way the game is they're as vital as cities and much harder to protect.  And you can only replace those at the rate of 1 per turn, no matter how big your cities are.

It does help if they do get attacked...and then what about if you make armored formers?

Yea this is true.  And if you go with a costing model that makes reactors only give HP (no cost reduction), it's probably 'good enough' of a benefit.  But it also requires a unit cost model that allows for free armor on non-combats.  Else you're paying way too much in unit costs.  And that's a bit of a balancing act - formers should have some defense against obsolete designs but not be a replacement for military sentinels.

[Yitzi]

That's true.  Instead of looking at this issue of assigning units to bases, from the top-down, how about bottom-up? Would there be any way of auto-assigning a unit to another base with excess support capacity when it's built, if the current base is at its support limit?

You mean automatically rehome it?  It'd be fairly tricky, but creating that option might work.  I still don't like it much, though (as it actually changes gameplay since you don't have to move it to the new base), and it is hard, so it'll have to wait for when I start taking general requests.

You know what would be a lot easier, though?  Adding an (s) after the unit name for any unit that requires support (actually, for any unit that required support the last time the program checked its base).  That would make the micromanagement easier without actually changing the mechanics.

I figured not.  In fact when I increase weapon values the game seems to use the wrong icons.  I'm thinking weapons were hard coded in differently from armor for some reason (perhaps tied to military/non-military unit code).

Weapons and armor are hard coded in the same way.

For their first payback perhaps.

Once you're already considering payback, you can consider two-level payback.

But no, formers are much much stronger than crawlers or anything else *in the long run*.

I don't think they get much past "10 turns payback" once you have enough for forests.

and I'd have to really sit down and try to analyse optimal early to mid game terraforming mathematically.  Has anyone done this to date?

I have to some extent, which is where I got the "around 10 turns" value from.  It depends on tech, of course; getting the extra formers to go with advanced terraforming is quite strong between ecological engineering (for condensers and boreholes) and environmental economics (for tree farms), but that's a fairly narrow window.

Anyways, the reason is formers continually improve terrain, not just once.  Think of it like a base improvement that builds base improvements.

That's a very good analogy.  Of course, a base improvement that builds base improvements is fairly indistinguishable from a base improvement that gives you extra minerals with which to build base improvements...basically what happens is that when figuring out former payback you get a quadratic equation rather than a linear one, but it can still be solved, and that's where I got the "slightly better than 10 turns" value from.  (It's better if you have free support, of course.)

So I pay 20 minerals for a former.  The first forest I make might be +2 minerals on bad terrain.

Forests as compared to unimproved are indeed extremely powerful; I'm not counting those because if you're not pop booming one former per base will keep up with that easily (and if you are pop booming formers aren't really the culprit in your fast growth.)

Instead, let's (and also to show how this calculation would be done) assume your choice is between getting fewer formers and planting forests, or getting more formers and getting advanced terraforming.  Let's further assume early-ish midgame, so you have tree farms and can raise land, but no hybrid formers, super formers, or soil enrichers.  We'll assume no free market and +0 INDUSTRY.  Finally, we'll treat all resources as the same value; it's highly situational, of course, but should be a decent average over the situations before thinkers/engineers.

So a forest is then worth 5 resources, and takes 4 former-turns.  However, 2 of those resources feed the worker, and you need to devote 1 energy to psych for drone control (with a hologram theater and tree farm, that'll be enough), so it's really only worth 2.

For advanced terraforming, let's assume you're building solar+farm with over 3000 elevation, but 1/4 of the solars are replaced with mirrors (giving around 2 mirrors next to each solar), and 1/4 are replaced by condensers to keep almost everything rainy (say 7/8 rainy 1/8 moist) despite being above cloud level.  This is not a perfect calculation, but should be a decent first approximation.  We'll assume the squares are 50% flat, 50% rolling, and ignore the energy cost to raise terrain.

Thus, for every 4 squares, we get:
-3 nutrients (1 after feeding the worker) each on the non-condenser squares, and 4 with the condenser.  Half the time one of those will be one lower.  So that's 3X1+4-0.5=6.5.
-1.5 minerals; we have to place the condensers for efficient raininess cover, so we can't place them only on the flat squares.
-4 energy from the crawler (3 after keeping the worker happy), and 6 (5 after psych) per solar, so that's 16.
So that's a total of 24 for 4 squares, or 6 per square.

But how much do those 4 squares cost?  Well, you need 30 minerals for the crawler (that's 7.5 per square), but more importantly each square needs to be raised once (12 former-turns; it's only once because once the square next to it is above 3000 it is above 2000) and needs a farm (4 former-turns); half of them also need solar (4 former-turns each), while the other half need a mirror or condenser (12 former-turns each), so that's an average of 12+4+8=24 former-turns each.  So for that extra 4 resources per square, you're spending an extra 20 former-turns and 7.5 minerals.

Now, how much is a former-turn?  Well, the former needs support, so that's 1 mineral, or 20 for 20 former-turns; it also needs to be built, so that's 20 mineral-turns per former-turn, or 400 mineral-turns total.  So the value of a turn is given by 400 resource-turns+27.5 resources=4 resources/turn.  Solving for 1/turn (which will effectively give the number of turns to pay off) gives x^2-(55/8)x-100=0.  By the quadratic formula, this gives [(55/8)+sqrt[(55/8)^2+400]]/2.  (55/8)^2=47.265625, and sqrt(447.265625)~21.148; adding 55/8 gives a bit over 28, so the total payoff time is a bit over 14.

One way to think of it is that with a payoff time of 14, each former-turn costs a bit under 2.5 minerals (1.5 minerals because you build a former instead of a 20-mineral object with a direct payoff, and 1 for support), so it's a bit over 48.5 minerals for the extra terraforming, plus 7.5 for the crawlers is a bit over 56 per square, which should give a bit over 14 for payoff time.

So yes, that first former for forests is more powerful (if it boosts by 2 resources per forest and requires no support, that's a payoff described by 2x^2-80=0, or between 6 and 7 turns), but after that it drops off fast.

The limiting resources are workable squares and former time, which make this a bit complex to work out.

And something to work that forest...that's the real limitation once you have more than one former per base, unless you don't go forests or you pop boom easily (something that I feel should not be possible until late game).

I think the optimal setup might actually be all boreholes and crawling all nutrients on condensors once they're available.

Late-game, the optimal setup is all specialists and crawling all nutrients on condensers.

But to get back on topic about unit costs.  It's quite apparent that formers (and to a lesser extent crawlers) are what drive the non-linear growth rate in the game.

Base facilities also have a substantial effect; a genejack factory has payoff time equal to 200 divided by two less than the base's production without it; for a 25-mineral base, that's fairly quick.

Is there any way to make it so later game formers and crawlers are significantly better for their unit cost? (i.e. improvements with reactor upgrades) Perhaps making formers terraform faster - I imagine any sort of more exotic new terraforming abilities are not possible.

New terraforming options would be unfeasible.  Increasing terraforming rate with reactor might work; if super former gave +50% per reactor, you could get late-game singularity formers with triple effectiveness, but they'd be expensive (90 minerals each under the normal formula, 70 under formula 1).

With crawlers perhaps a capping system that eventually allows uncapped crawling of multiple resources (or a bonus to energy/mineral crawling).

I think a flat decrease to the amount brought in might work better if it proves necessary (it might not, as you still only get one crawler).

My thought was that losing all your early game formers/crawlers shouldn't necessarily be game over.

I don't think it is; it's a big setback, but you can replace them.  And even with one former per base, you can still forest (and maybe even farm/solar) faster than a base can grow.  Crawlers also are really only great on mines and squares where you're building condensers anyway for the raininess boost; building condensers just so you can get +1 nutrients when you crawl them is probably a waste of former time.

And you can only replace those at the rate of 1 per turn, no matter how big your cities are.

Of course, with a big base, you can replace them with ones that won't die easily.

Yea this is true.  And if you go with a costing model that makes reactors only give HP (no cost reduction), it's probably 'good enough' of a benefit.

Unless you armor them or change the chassis, the increased minimum cost will mean higher-reactor formers actually cost more than lower-reactor versions.

But it also requires a unit cost model that allows for free armor on non-combats.

Well, not free; you're still going to pay for the higher armor cost.  But yes, it does lose a lot of its effect if you get +1 row cost for armoring noncombat units...good think the formulas give the option not to do that.

Else you're paying way too much in unit costs.  And that's a bit of a balancing act - formers should have some defense against obsolete designs but not be a replacement for military sentinels.

Formers will never be a replacement for military sentinels, simply because they're too expensive.  Even without the +1 row cost, 0/3/1 formers cost 4 rows, whereas 1/3/1 defenders cost only 2 rows.

[Geo]

I figured not.  In fact when I increase weapon values the game seems to use the wrong icons.  I'm thinking weapons were hard coded in differently from armor for some reason (perhaps tied to military/non-military unit code).

The weapons seem to be linked to their place in the alpha(x).txt list, in combination with the names of the caviar art.

[Nexii]

This illustrates what I see for weapons better...their graphic depends on the power and not their name/tech in alphax.txt

#WEAPONS
Hand Weapons,         Gun,            1, 0, 1, -1, None,
Laser,                Laser,          2, 0, 2, -1, Physic,
Particle Impactor,    Impact,         4, 0, 3, -1, Chaos,
Gatling Laser,        Gatling,        6, 1, 4, -1, Super,
Missile Launcher,     Missile,        8, 2, 5, -1, Fossil,
Chaos Gun,            Chaos,         12, 0, 6, -1, String,
Fusion Laser,         Fusion,        16, 1, 7, -1, SupLube,
Tachyon Bolt,         Tachyon,       20, 1, 8, -1, Unified,
Plasma Shard,         Shard,         24, 2, 9, -1, Space,
Quantum Laser,        Quantum,       30, 1,10, -1, QuanMac,
Graviton Gun,         Graviton,      40, 0,11, -1, AGrav,
Singularity Laser,    Singularity,   50, 1,12, -1, ConSing,
Resonance Laser,      R-Laser,        8, 1, 7, -1, Bioadap,
Resonance Bolt,       R-Bolt,        20, 1,11, -1, SentRes,
String Disruptor,     String,        60, 1,13, -1, BFG9000,
Psi Attack,           Psi,           -1, 2,10, -1, CentPsi,
Planet Buster,        Planet Buster, 99, 0,30, -1, Orbital,
Colony Module,        Colony Pod,     0, 8, 3, -1, None,     ; Noncombat packages
Terraforming Unit,    Formers,        0, 9, 2, -1, Ecology,
Troop Transport,      Transport,      0, 7, 2, -1, DocFlex,
Supply Transport,     Supply,         0,10, 3, -1, IndAuto,
Probe Team,           Probe Team,     0,11, 3, -1, PlaNets,
Alien Artifact,       Artifact,       0,12,36, -1, Disable,
Conventional Payload, Conventional,  12, 0, 8, -1, Orbital,
Tectonic Payload,     Tectonic,       0,13,12, -1, NewMiss
Fungal Payload,       Fungal,         0,14,12, -1, NewMiss

[Earthmichael]

My issue about micromanagement verses clean is that if I support 2 free units per city, I don't want the micromanagement of trying to figure out which cities have some unit support left.  I would just rather make everything clean.

I still have no seen a good argument why clean should be different costs on different units.  I still like the idea of 1 row, period.  This way it is a fairly no-brainer choice for most units whether to build clean or not, or to worry about city resource micromanagement.

I also don't like the idea of reactors increasing the costs of non-combat units like formers.  I usually still build fission reactor formers for this reason.  It would be nice to get the extra defense from a better reactor, but I am not going to pay MORE for the unit (when in theory better reactors drive unit costs down).  I am not asking for a reduction in cost for formers, just a breakeven.

[Earthmichael]

On the topic of non-linear growth, all games of this kind have exponential growth until they reach some sort of game cap.  It is the nature of the game: 1 base builds 2 bases, which builds 4 bases, which build 8 bases, etc.

Formers can make the bases more productive, and so can city structures.  But formers are not the reason for exponential growth, cities are.  Limiting formers just push toward the small cities/ICS approach.  I personally think large cities are stronger, but only because of facilities, and to a lesser extent, the availability of formers to make enough productive land available for a large city.

Formers and crawlers are powerful, and are the key to logistics side of SMAC, and in my opinion, what makes SMAC better than any Civ game before or afterwards (which did not include crawlers, and where formers where much more limited in abilities).  Otherwise, it is just military verses military.  Part of your military goal now must be to protect your logistics side, crawlers and formers.  I think this makes the strategy much more interesting if it were only military verses military.

I do not like the idea of improving crawlers to crawl more than one resource; I the crawlers are powerful enough as it is.  As for the borehole, I almost always make them in range of a city so the city can work the borehole for full value.  I think it is fine that a crawler can get only half of the value of a borehole.

Once I get enough formers than control become tedious, I automate a bunch of road building formers, and I also automate a bunch of fungicidal fungus removing formers.  The rest of the formers I tend to bunch up in groups of 4, and move an terraform them as a pack.  This makes management of formers much less tedious when you have a lot.

As for crawlers, I usually place them and forget about them; no real micromanagement required.

Since I think formers and crawlers are the most interesting aspect of SMAC, I would take care before I made any serious modifications (like having high reactors result in faster terraforming).  I would not go that route.  The only change that made sense to me is allow formers to use higher reactors with no great cost than fission reactor for the same capabilities.  This gives basic formers a nominal bit of extra defense from the reactor.

[Nexii]

Since I think formers and crawlers are the most interesting aspect of SMAC, I would take care before I made any serious modifications (like having high reactors result in faster terraforming).  I would not go that route.  The only change that made sense to me is allow formers to use higher reactors with no great cost than fission reactor for the same capabilities.  This gives basic formers a nominal bit of extra defense from the reactor.

Yea - as I said I tend to favor all unit costs not being tied to reactor - it over complicates the cost formula.
1 - Reactors already give a survivability increase to military.  No need to double dip with a cost reduction as well.
2 - For non-military they drive the cost up more than they're worth.
3 - Native life has a constant cost and ignores reactor.  Therefore military units should have a constant cost ignoring reactor as well.

Try cost mode #2/4 that Yitzi put in.  Of course this requires some re-tweaking of all the base armor and both military and non-military weapon costs.  I've been playing around with mode 2 and some customized armor/weapon/chassis balances.  Still testing...I noted a few things but haven't had time to analyse..

And yea...I need more understanding of growth in smac.  It's actually time that drives growth.  Cities are just an abstract much like formers, crawlers.  Non-linear growth is fine and probably outside the scope of unit costing...at least for now.

[Dio]

This illustrates what I see for weapons better...their graphic depends on the power and not their name/tech in alphax.txt

#WEAPONS
Hand Weapons,         Gun,            1, 0, 1, -1, None,
Laser,                Laser,          2, 0, 2, -1, Physic,
Particle Impactor,    Impact,         4, 0, 3, -1, Chaos,
Gatling Laser,        Gatling,        6, 1, 4, -1, Super,
Missile Launcher,     Missile,        8, 2, 5, -1, Fossil,
Chaos Gun,            Chaos,         12, 0, 6, -1, String,
Fusion Laser,         Fusion,        16, 1, 7, -1, SupLube,
Tachyon Bolt,         Tachyon,       20, 1, 8, -1, Unified,
Plasma Shard,         Shard,         24, 2, 9, -1, Space,
Quantum Laser,        Quantum,       30, 1,10, -1, QuanMac,
Graviton Gun,         Graviton,      40, 0,11, -1, AGrav,
Singularity Laser,    Singularity,   50, 1,12, -1, ConSing,
Resonance Laser,      R-Laser,        8, 1, 7, -1, Bioadap,
Resonance Bolt,       R-Bolt,        20, 1,11, -1, SentRes,
String Disruptor,     String,        60, 1,13, -1, BFG9000,
Psi Attack,           Psi,           -1, 2,10, -1, CentPsi,
Planet Buster,        Planet Buster, 99, 0,30, -1, Orbital,
Colony Module,        Colony Pod,     0, 8, 3, -1, None,     ; Noncombat packages
Terraforming Unit,    Formers,        0, 9, 2, -1, Ecology,
Troop Transport,      Transport,      0, 7, 2, -1, DocFlex,
Supply Transport,     Supply,         0,10, 3, -1, IndAuto,
Probe Team,           Probe Team,     0,11, 3, -1, PlaNets,
Alien Artifact,       Artifact,       0,12,36, -1, Disable,
Conventional Payload, Conventional,  12, 0, 8, -1, Orbital,
Tectonic Payload,     Tectonic,       0,13,12, -1, NewMiss
Fungal Payload,       Fungal,         0,14,12, -1, NewMiss

You could also simply swap the caviar file names in the game folder. In this case, VW03.cvr becomes VW04.cvr and VW04.cvr becomes VW03.cvr. I had that same issue and found an acceptable workaround so long as you do not exceed the pre-determined range of weapon values for each weapon slot.

This also ties into my desire to figure out how to make the weapon graphics work in the design shop.

[Geo]

This illustrates what I see for weapons better...their graphic depends on the power and not their name/tech in alphax.txt

I see. Thanks for clarifying. 

[Geo]

On the topic of unit maintenance, I wonder if it is code-wise possible to let a popped independent crawler be re-homed to the closest base of the faction popping it.
I know a player can do it manually, but I've seen AI independent crawlers happily crawling tiles to the benefit of no-one.