Supply crawlers, need some opinions

[Yitzi]

Yea for SE switching I was more concerned with hovering between a choice and Frontier-types, or SEs and those with no downside (CV with Power/TC, Yang with PS/Planned).

Yeah, in that case switching becomes less of a downside...but on the other hand there's also usually less reason to switch.

However, it occurred to me that the exact schedule for a double-switch is a poor approach; if you haven't fully switched to Market, and aren't heading there now, then you shouldn't have the full downsides either.  So instead, what would happen in our scenario (Green to Market, then halfway through change to Planned) would be:

Turn 0 (the turn you switch): -3 PLANET, -5 POLICE, -2 GROWTH
Turn 1: -3 PLANET, -5 POLICE, -2 GROWTH (because nothing increases by 4 this switch, there are no benefits until turn 2)
Turn 2 before reswitching: -3 PLANET, -5 POLICE, -1 GROWTH, +1 ECONOMY
Turn 2 after reswitching: -2 PLANET, -3 POLICE, -1 GROWTH, -2 EFFIC
Turn 3: -1 PLANET, -2 POLICE, -2 EFFIC
Turn 4: +1 GROWTH, -2 EFFIC
Turn 5: +1 GROWTH, -2 EFFIC
Turn 6: +2 GROWTH, +1 INDUSTRY, -2 EFFIC

Basically, getting rid of the downsides of a half-switched-to SE would only take as long as you'd already spent switching to it.  Likewise, if you switched from something and then back before the switch finished, it would take less time to get back where you were.  So if you switched from Green to Free Market, and then back after 2 turns, it would be:

Turn 0 (the turn you switch): -3 PLANET, -5 POLICE, -2 GROWTH
Turn 1: -3 PLANET, -5 POLICE, -2 GROWTH (because nothing increases by 4 this switch, there are no benefits until turn 2)
Turn 2 before reswitching: -3 PLANET, -5 POLICE, -1 GROWTH, +1 ECONOMY
Turn 2 after reswitching: -1 PLANET, -3 POLICE, -2 GROWTH, +1 EFFIC
Turn 3: -2 POLICE, -2 GROWTH, +1 EFFIC
Turn 4: -2 GROWTH, +2 PLANET, +2 EFFIC

I think as long as the positives fading in are relatively backloaded it'll be fine though.

They generally will be simply because of how social engineering works (with positives being fairly minor at low bonuses, and more important as you get more in an area.)

Bulk Matter is the +2 minerals SP.

The datalinks say +2, but it's really +50%.

One thing about late-game facilities is that they have a lot less time before the game is over to pay off.

Of course, whatever else you might have spent the minerals on also would have had a lot less time to pay off...

Although cities are much bigger late, facilities all only give a % modifier on the base E/M produced.  I'd have to run some numbers.  Something like ~15-20 turns to pay-off would be about right, I feel.

Actually, IIRC other options such as formers suggest a bit less than that.

but moreso than pay-off I think you also have to weight them against units.  Units keep you from losing cities.

Of course, there are other ways to keep from losing cities (e.g. diplomacy of various forms), and of course once you have enough units to not lose cities their value drops off fairly rapidly.

It might be more that 2:1 rushing gets rather strong, late, when considering that E doesn't cause any ecodamage, than that M costs need to stay very high.

I think it's both.  The thing is that between base growth and resources per square increasing and mineral-boosting facilities, M seems to grow a lot faster than facility costs.  An early-game facility usually takes 10 turns or more to build without rushing, whereas for a late-game facility it's usually 1 or 2.
 

I'll see about native life pops.  I wondered if perhaps the strength of pops should scale up with ecodamage #, rather than pop sequence #.

I think that makes a lot of sense.  Although rather than scaling with ecodamage #, I think it should scale with the extra beyond which the ecodamage # surpasses the # "rolled" to determine if a pop actually happens.  That way there'd be an element of randomness in the amount of worms spawned as well as in whether anything happens at all.

I can't prove this but I feel that ecodamage # is not linearly related to pop chance, that it might not go up as fast.  So many bases with low ecodamage actually are a lot more dangerous than a few bigger bases with high ecodamage.

I am fairly certain (from what I remember last time I looked at the code) this is not true, that the ecodamage # is the actual percentage chance of a pop.

The game I'm remembering might have been more of a race to VoP to get Ascent than VoP>Ascent.  Might have been something like huge boils going around and new ones stopped coming at VoP, but I had to rush Ascent in the next 2 turns and protect the Ascent city.

Could be.

[Nexii]

Yea some randomness to pops would be good if doable.

For late-game play, I'm weighing facilities more against units and not SPs/sats.  Units don't just defend your cities, they can pressure enemies and capture their cities or SPs.  A pacifist/defend style might work vs the AI, but for higher level play you might get rolled over trying to go 100% economy. 

I think increasing the labs curve would suffice for slowing down the endgame.  Making facilities cost even more M late...I don't know.  I feel they are pretty high already costed, if you're running like 80-100M a base you'll have huge ecodamage.  I do agree with the 10 turns/fac : 1 turn/fac problem as M ramps up problem though, but it does apply to units also.  So it's relative.  Probably more an issue from late-game micromanagement perspective than anything else.  Civ2 also had this problem.  I think generally later game facilities probably should cost more, but they would need more benefit than they do right now.  Most late-game facilities only give similar benefit to early-game facilities.  Higher costs aren't such a bad thing, I'm envisioning M and E specialised bases to make the economy side more interesting.  Of course this also requires E/M specialized terraforming choices.

[Yitzi]

Yea some randomness to pops would be good if doable.

Not certain, but I'm guessing it'd be quite easy.

For late-game play, I'm weighing facilities more against units and not SPs/sats.  Units don't just defend your cities, they can pressure enemies and capture their cities or SPs.  A pacifist/defend style might work vs the AI, but for higher level play you might get rolled over trying to go 100% economy. 

And I think that should be a balance, but until the super-late-game there should be a notable defender's advantage.

I think increasing the labs curve would suffice for slowing down the endgame.  Making facilities cost even more M late...I don't know.  I feel they are pretty high already costed, if you're running like 80-100M a base you'll have huge ecodamage.

Of course the facility cost would have to be dependent on ecodamage formula; I favor a formula where ecodamage ramps up with minerals a lot slower than default, so that will of course influence things.

However, even with 30-40 minerals a base, the current lategame facilities are too fast and could use a 25% or 50% boost.  (Projects could use more, but that's because you're using resources from several bases to build them.)

I think generally later game facilities probably should cost more, but they would need more benefit than they do right now.  Most late-game facilities only give similar benefit to early-game facilities.

Of course, the more advanced tiers are always more expensive for the same benefit...but more benefit is still an interesting idea.  (I don't think it's necessary except for the paradise garden, but it's still interesting.)

Of course this also requires E/M specialized terraforming choices.

Those already exist, though whether they're balanced is another question.

[hackerjack]

OK, I haven't read up on absolutely everything in this thread so probably repeating things said already but here is my take on what crawlers should/shouldn't be and in an ideal world how their effect should be handled.

1. First of all the whole recycling for 100% production thing is and always has been bonkers, recycling in general in civ/smac is bonkers of course but crawlers doubly so.   Just put crawlers at the same level as anything else that is recycled.

2. Convoys should really be separate to Crawlers and should be able to convey 1/(x*y) excess unit from one base to another every turn with x being the distance bases and y being a modifier based on technology.  This would have the same effect as saying that one convoy can transport 1 unit between bases per x turns based on distance, which feels realistic.  The tech could be related to Speeder/Hoverjet discoveries as that seems to fit.

3. Ideally I'd like Crawler number to be limited by a new specialist type (Technician/Engineer), a bog standard base being able to run 1 crawler at full efficiency but after that a specialist would be required to each additional one in the same manner with each crawler over the limit being half as effective as the last (averaged over the fleet).

while i=1 to num_crawlers
  if i<=num_specialists+1 then
    tot_crawler_prod = tot_crawler_prod + (1/((2^(num_crawlers-1))-num_specialists))
wend
avg_crawler_prod_modifier = tot_crawler_prod / num_crawlers

with a minimum of 1 unit production for each crawler just to be fair.

4. Crawlers should also be capped in terms of efficiency by tech, I would start with a maximum of 3 units that could be collected by any crawler, raised to 5 then unlimited.  Not sure what techs would make sense for this though, possibly related to 3+ minerals then boreholes (the tech only, not if you get the WP).


All of this together would make Crawlers less of a game breaking SP building tactic and much more of a strategic choice I think.  It would still allow for throwing a crawler (or three if the loss of efficiency is worth it to you) at a new base for an early boost and makes them progressively more useful later on without allowing them to be overpowering early (capping harvesting from boreholes, condensers etc after WP). 

How much of it is feasible I have no idea.

[Yitzi]

OK, I haven't read up on absolutely everything in this thread so probably repeating things said already but here is my take on what crawlers should/shouldn't be and in an ideal world how their effect should be handled.

1. First of all the whole recycling for 100% production thing is and always has been bonkers, recycling in general in civ/smac is bonkers of course but crawlers doubly so.   Just put crawlers at the same level as anything else that is recycled.

This is actually a potential future change.

2. Convoys should really be separate to Crawlers and should be able to convey 1/(x*y) excess unit from one base to another every turn with x being the distance bases and y being a modifier based on technology.  This would have the same effect as saying that one convoy can transport 1 unit between bases per x turns based on distance, which feels realistic.  The tech could be related to Speeder/Hoverjet discoveries as that seems to fit.

Having two separate unit types would be somewhat harder, though, and I'm not sure there'd be an advantage over having them actually physically carry the resources.

3. Ideally I'd like Crawler number to be limited by a new specialist type (Technician/Engineer), a bog standard base being able to run 1 crawler at full efficiency but after that a specialist would be required to each additional one in the same manner with each crawler over the limit being half as effective as the last (averaged over the fleet).

That would require enabling new specialist types, which is fairly difficult, but that's certainly an interesting idea.  Not sure if that's the best formula; it might be better to have each crawler over the limit being half as effective as the one {number of full-efficiency crawlers} before.

4. Crawlers should also be capped in terms of efficiency by tech, I would start with a maximum of 3 units that could be collected by any crawler, raised to 5 then unlimited.  Not sure what techs would make sense for this though, possibly related to 3+ minerals then boreholes (the tech only, not if you get the WP).

Actually, in vanilla 3+ minerals and boreholes are the same tech, but in any case I don't think that would help much since if crawlers getting a lot of resources is a problem it'll be more of a problem in the late game anyway.

[hackerjack]

Having two separate unit types would be somewhat harder, though, and I'm not sure there'd be an advantage over having them actually physically carry the resources.

I don't mean have them trundling back and forth on the map, just mimic it by transporting 1/5*y (my brackets are off earlier) per turn to a city 5 tiles away once the initial connection is made.  I was thinking a seperate unit because a Convoy would be cheaper but would be consumed on setting up the route.

That would require enabling new specialist types,

Ideally yes, but tying it to the current types could be possible.

[Yitzi]

I don't mean have them trundling back and forth on the map, just mimic it by transporting 1/5*y (my brackets are off earlier) per turn to a city 5 tiles away once the initial connection is made.  I was thinking a seperate unit because a Convoy would be cheaper but would be consumed on setting up the route.

It's an idea, but I doubt that many people would prefer it to both the current system and "trundling back and forth".  That said, feel free to nominate it next time I open nominations.

That would require enabling new specialist types,

Ideally yes, but tying it to the current types could be possible.

[/quote]

So one crawler per specialist?  Interesting idea, especially if specialists are made less effective...

[Ajantaka]

Why is using crawlers to help build Secret Projects considered an exploit?  Why is it not a legitimate use?  After all you have to expend time and resources to build crawlers?

[Yitzi]

Why is using crawlers to help build Secret Projects considered an exploit?  Why is it not a legitimate use?  After all you have to expend time and resources to build crawlers?

It really depends how you use it.  If you use it to pool all your bases' production together to build a project faster, that's a legitimate use.  But if you use it to be able to build the project the turn that you get the tech, that's fairly exploitative.

[Mart]

AI, although generally does not instabuild, it is capable of doing something close to it. When having some crawlers, it can crawl, but on acquiring a technology with SP, after it starts that SP, it can direct crawlers to add them.
Theoretically, it can instabuild this way, however, in practice:
- AI does not have usually enough crawlers
- AI will not (my guess for 99.99%) plan instabuild, by gathiring enough crawlers, 10-13 at least, near a base it wants to start that SP.

If AI is improved to build more crawlers, probably the gap between human and AI would be made smaller.
There is still possibility to change game conditions with existing AI, to get better results, by modding rules.

[Mart]

...

I think increasing the labs curve would suffice for slowing down the endgame. 

At present I am testing the idea to slow research overall by around 6 times, so it is like 15-17 % of standard research rate. However, the early game should have discovery rate similar as in standard rules, so I made each faction to get appropriately more energy to direct it to research. This would work in early turns, and in endgame having minimal impact. I would expect accordingly 6 times slower research in endgame as compared to standard rules. Or even more, as I intend to slow down base spawning.
Testing mid game at the moment, so I cannot say final result.

Making facilities cost even more M late...I don't know.  I feel they are pretty high already costed, if you're running like 80-100M a base you'll have huge ecodamage.  I do agree with the 10 turns/fac : 1 turn/fac problem as M ramps up problem though, but it does apply to units also.  So it's relative.  Probably more an issue from late-game micromanagement perspective than anything else.  Civ2 also had this problem.  I think generally later game facilities probably should cost more, but they would need more benefit than they do right now. 

My impression is too, that late game may have more expensive stuff.
Some solution is to change minerals curve, by giving the player mineral multiplications earlier. Then in midgame to endgame increase in minerals would be slower and mostly due to more worked tiles.
Thanks to Yitzi new factors in #RULES we have control over ecodamage and we can fine-tune it.
So cost of all facilities can be also adjusted to the new mineral curve.

Most late-game facilities only give similar benefit to early-game facilities.  Higher costs aren't such a bad thing, I'm envisioning M and E specialised bases to make the economy side more interesting.  Of course this also requires E/M specialized terraforming choices.

However, in late-game player has more bases and more population, so the same effect of +50% of something is accordingly larger. Increasing such effects would contribute again to "snowball" effect.

[Nexii]

By making early bases have 6E rather than 1E?  This tended to encourage ICS I found unfortunately.  The problem was more with the curve.  Midgame will be relatively slower, late game will feel about right in your case.  It'll range from like 5T/10T/5T from early/mid/late which isn't too terrible

But on the topic of crawlers, if I didn't suggest this before.  What if the amount crawled into SPs was inverse to the progression amount?

For example let's say we are using a 40 cost crawler to boost a SP
0% SP = 40 minerals
If the SP was 50% complete, the crawler would only get 20 minerals (50% left*40)
75% complete, (25% left*40) = 10 minerals
ETC

A better formula might also account for supercosted crawlers (though I play with flat cost, so less of an issue)
For example a 200 mineral supercosted crawler probably shouldn't boost a 200 mineral SP instantly.
Something like a max gain of 25% per crawler (50M), or maybe 50% (100M) might do.  Or SPs could have a mineral% cap per turn, though personally I don't mind 1 turn SPs if you've stockpiled

[Yitzi]

By making early bases have 6E rather than 1E?  This tended to encourage ICS I found unfortunately.  The problem was more with the curve.  Midgame will be relatively slower, late game will feel about right in your case.  It'll range from like 5T/10T/5T from early/mid/late which isn't too terrible

But on the topic of crawlers, if I didn't suggest this before.  What if the amount crawled into SPs was inverse to the progression amount?

For example let's say we are using a 40 cost crawler to boost a SP
0% SP = 40 minerals
If the SP was 50% complete, the crawler would only get 20 minerals (50% left*40)
75% complete, (25% left*40) = 10 minerals
ETC

A better formula might also account for supercosted crawlers (though I play with flat cost, so less of an issue)
For example a 200 mineral supercosted crawler probably shouldn't boost a 200 mineral SP instantly.
Something like a max gain of 25% per crawler (50M), or maybe 50% (100M) might do.  Or SPs could have a mineral% cap per turn, though personally I don't mind 1 turn SPs if you've stockpiled

Actually, to deal with supercosted crawlers, you could just set up the formula so that it's as if minerals from the cost were added one at a time at that penalty.

The real problem with your suggestion, though, is that rather than encouraging the use of crawlers to put your whole faction to work on a project and discouraging their use to instabuild, it has the opposite effect, as the best way to use crawlers on a project is then to have them pre-built and then finish the later portion in the base itself.

[Nexii]

Yes, you'd want to prebuild crawlers in that case.  But it's not that exploitative, those crawlers could have been put to use making facilities.  They're sitting around doing nothing until you get the SP tech.  Alternatively you could go with more of a flat formula like 50% return (thus making crawler dumping similar to selling a facility), and/or limit it to 1 crawlers per turn dumped (like how one facility a turn can be sold)

That being said, you'd want rush costs to still be significant.  If you set hurry rows to the max, then that's 16E:1M on SPs and that makes it hard to rush the remainder in the base.  There's a lot of other issues here around SPs of course.  Making more meaningful SP races raises the problem of what sort of penalty is right when you lose a race (the exploit of switching to a newly discovered SP from a 'lost' one)

[Eadee]

It may be just me but I really think loosing the desired SP is penalty enough. Lets say you're Zakharov and didn't get the hunter-seeker-algorithm. Instead you have to go for another project that may be completely useless for your plans (one of those psi-bonus SPs?). I really think there is no need to get another penalty. However if you have a problem with SP-switching would it be possible to simply "cash out" minerals invested in a SP that was already built (at the same rate as stockpile energy works)? I'd find it most interesting if those minerals wouldn't turn into EC but instead turned into research-points (at stockpile energy ratio without any base/facility -modifiers for science output).

This way you might get another tech which allows you to race for a different SP, but it doesn't give you an instant-SP you never worked on before.

However I have no idea if this is even possible through modding without serious amounts of work so.... its just an idea.