[Chris says: let's discuss!] Ship bays as a means of managing perplanet assets

[Tridus]

Boy, I feel like the odd on out here. I have all kinds of problems with this.

6. Being able to turn up and down the amount of power being used for turrets in general versus for the other things would be pretty cool.  Aka, it's like under-funding your fire department.  You can build a certain number of turrets, period, on a planet in general.  And you can have those fully powered if you want.  Or you can turn those to 50% power for a while to then have extra power to ship creation or whatever you want.

In more recent games, under-funding the fire department both leads to more fires, AND makes people unhappy at the inferior fire coverage. That was how they solved the issue in Cities: Skylines. You can't rank a zoned building up past a certain point if it doesn't have adequate coverage, and cutting funding leads to less adequate coverage. That discourages tedious micro to only turn it on when there's a fire (plus, early versions of Skylines had like a zillion fires).

That's the problem with this idea: micro. I'm encouraged to chop the hell out of local turret power until about 5 seconds before the AI attacks, because I get a ship production bonus from doing so. It's the same problem as with engineers, only moved to a different location (in my FS games, I'm constantly destroying and rebuilding engineers on whatever planet I need to build up, except the one that produces ships, because it doesn't make sense to leave engineers where they're not doing anything).

7. Hell, having engineers and remains rebuilders actually doesn't fit super well into the current design in general.  In the past those were useful on AI planets, but that means they should cost fuel.  But that really doesn't fit with the overall goals here, I don't feel like.  Having "battlefield engineers" that are used in hostile space and are a fuel-related ship would be pretty cool, but on your own planets why even have those?  Why not control everything from this sub-screen and save any worries about things like engineer travel times and whatnot.  Cutting down on the amount of waiting time, and just letting you express yourself through the sorts of screens that have been really well-refined over the years (I'm thinking of SimCity 4, not having played the later ones).

I'm not sure why this needs to be controlled at all, honestly. It takes X time to build something. If I can make that faster and it costs no additional resources to do so, you know I want to do it. Why wouldn't I?

Battlefield engineers are a good idea, though. Always fun when you can beachhead inside an AI world.

Yes, BUT.  We'd see those utility ships actually removed in general in terms of something you can control or whatever, potentially.  Instead it would be something like "invisible nanobots" that come from the command station and repair things after a certain amount of time, and you put a certain amount of funding into that from the power budget on that planet.  Same thing for acceleration the construction of ships.

It's easier for players to control in that fashion (as SimCity shows), and there's not then a bunch of micro with enemies killing your engineers and whatnot.  This whole thing ought to make refleeting easier by doing things like letting you lower your defenses on a planet in exchange for faster ship output for a bit, etc.

The difference between AI War and SimCity is that in SimCity, I'm trying to keep a city running and keep people happy. Funding a school at 5% has consequences, and those are ongoing. In AI War, funding my turrets at 5% means absolutely nothing until the AI shows up.

If I can control this stuff, the very first thing I'm going to ask for is a way to automate it, because I want it to do the same thing every single time: dump as much into production as possible until the AI shows up, then shift all of it to defense. There's no case where I want to build more slowly than I possibly can (outside of "tank the economy" stuff like trader goodies), and there's no case where I want defenses powered up and slower production when those defenses aren't doing anything. The only question is if I have to micro it or not.

And on another planet lower the defenses in exchange for increased metal production or whatever.  It would have to take time for those changes to take effect so that you can't spam it, but basically have it set to a target "funding level" from your energy stores, and it moves to that at a certain rate that leads to it taking longer to adjust the larger the change is you just made.  One candy tech might be to make the change rate on that faster.

So this would change the micro problem by rewarding me for figuring out earlier when to do it, which just makes the whole thing more fiddly and complicated. The basic choice is still the same. It's not really an interesting choice, because at any one moment in time there is a clearly correct choice and a clearly wrong one. The only question is if I can switch between the two at an opportune time.

Overall my goals with games these days is to make the minimum amount of player effort between "deciding to do a thing" and "having the thing done or in progress of being done."  Babysitting engineers and remains rebuilders is kind of the antithesis of that.

I agree with that, but I don't think this solves the problem. You've replaced one extra complexity with a different extra complexity.

IMO, the thing to do here is subtraction. Things take X time to build. Period. Construction speed boosts go away. Simple, easy to understand, no fiddly extra systems to build or micro to deal with.

 X has to be significantly smaller to compensate for not having boosters anymore, but that's fine, because frankly some of the times in AIWC seem to have been created entirely expecting them to be boosted anyway. Did anybody really wait an hour for a mk III Zenith Starship to build? And from a lore POV, if we're building defenses because we expect AI attack and we can scale up production speed without using additional resources... why are we not doing that already? What reason do the human survivors engaged in a war to the death have to build more slowly than they safely can?

Maybe the AI should have capturable nanobot factories that increase building speed in their entire solar system or something, to give faster building if you can capture those.

[Pumpkin]

I strongly support this.

Here is my take on simplifying management:
1) No more engineers and better non-boosted build times.
2) Low-cap, expensive factories for more important AI-targets and meaningful production locations.
3) No more remain rebuilders and automatic rebuild when planet is clear (and if still controlled).
4) Automatic healing on controlled, AI-free planets (or enclave-like healing drones spawned by the OCStation).
5) "Battlefield" engineers for beachheads and healing-in-hostile-territory.

(Also merge Power production and Knowledge gathering within the OCStation.)

[chemical_art]

I'm fine with engineers. They are a resource meant to aid in projects. I would rather have one global cap to manage that rather then trying to micro resources of a different sort on a planet by planet basis. Unless it is planned to remove all of the functions entirely, that micro will exist. I prefer the status quo over some unknown new method.

[kasnavada]

I'm fine with engineers. They are a resource meant to aid in projects. I would rather have one global cap to manage that rather then trying to micro resources of a different sort on a planet by planet basis. Unless it is planned to remove all of the functions entirely, that micro will exist. I prefer the status quo over some unknown new method.

I'm not fine with engineers. They were micro needed per planet, always had to rebuild them because they either don't do anything or FRD into danger at the first opportunity. I don't count anymore the number of times that I have to move engineers again from planet to planet, having to redo the whole distribution of engineers because I hit the cap, or the numbers of games before I figured out that the best way to automate that mechanic was to automate the rebuilding per planet, by setting each planet individually. And / or having to manually set engineers to one shipyard to accelerate production. Ugh. Oh, the other annoying one, that on the homeworld, everytime it gets attacked, the engineers don't go back to where I put them after whatever they thought was useful is done. Then my whole refleeting gets shut down because they're out of forcefields, and die, unless I go check there everytime... 50 engineers or so always felt like a limitation to me.

What's written here shows solutions to those issues and helps allieviating them, with minor impacts on the game flow. Also, a "larger" game becomes possible.

@tridus:
For me this part (Chris post):

And on another planet lower the defenses in exchange for increased metal production or whatever.  It would have to take time for those changes to take effect so that you can't spam it, but basically have it set to a target "funding level" from your energy stores, and it moves to that at a certain rate that leads to it taking longer to adjust the larger the change is you just made.  One candy tech might be to make the change rate on that faster.

That basically means that what you write in your post won't be possible. Possibly I missed something ?
The goal seems more to choose "templates" for planets like the command stations were than to switch everything every 5 minutes. Unless you changed command station every 5 minutes ?

[Tridus]

@tridus:
For me this part (Chris post):

And on another planet lower the defenses in exchange for increased metal production or whatever.  It would have to take time for those changes to take effect so that you can't spam it, but basically have it set to a target "funding level" from your energy stores, and it moves to that at a certain rate that leads to it taking longer to adjust the larger the change is you just made.  One candy tech might be to make the change rate on that faster.

That basically means that what you write in your post won't be possible. Possibly I missed something ?

It's entirely possible, unless the delay is so long that having the feature at all is largely pointless. If I'm building something, I'm going to want it shifted to production to build it faster. If I see something coming, shift it back. That'll require scouting and babysitting, but it'll still be more efficient than waiting forever for stuff to finish building because the focus is on completely idle defenses instead.

It's not possible if you set a 30 minute timer on it or something, but if you're doing what, what is the point of having it at all? There's no interesting management happening whatsoever at that point. You'll always set it to one or the other and just leave it that way 99% of the time. The timer is really just a bandaid extra bit of complexity to fix the problem that this idea is creating.

The goal seems more to choose "templates" for planets like the command stations were than to switch everything every 5 minutes. Unless you changed command station every 5 minutes ?

Nope:

Basically I think of this kind of like the screens in SimCity where you can set the tax rate, set the funding rate for different programs, and enact social policies, etc.  I don't think there needs to be a different type of command station for planets at all, but rather you configure these things at the planet the way you want (and if your command station dies, next time you come back it's the same again since this data was at the planet).

Templates in this case as-proposed were a set of settings that you could save and then activate on command to change the values to whatever the template was set to. All I have to do to swap it is bring up a menu and switch from "I want to build stuff" to "I want to shoot stuff", then wait for some arbitrary delay.

I'm in no way clear on what problem this is actually trying to solve that can't be solved by something simpler.

[kasnavada]

Ah, missed the last part, thanks. Ummm. The way I understand that & the entire post is:
- there is no command stations,
- everything including bonus to turret construction / ship construction / whatever was engineers and rebuilders before can be parametrized there.
- those can be set and saved as templates to be put down. And switched, but a system has to be refined so it can't be switched at will.

Compared to setting the numbers of engineers at most planets manually having to reset them when I unlock mkII engineers, and having to target manually some of the engineers to boost ship production so they don't wander away... and also not having to "reset" the position of engineers when I have 2 different places for them to be... I'd find it simpler to use. Also it opens possibilities.

[Tridus]

Compared to setting the numbers of engineers at most planets manually having to reset them when I unlock mkII engineers, and having to target manually some of the engineers to boost ship production so they don't wander away... and also not having to "reset" the position of engineers when I have 2 different places for them to be... I'd find it simpler to use. Also it opens possibilities.

Compared to having sane ship build times to begin with and not having construction speed boosts at all? Not really.

What possibilities? What can I do with this that I'm not going to just want automated? If the AI is attacking me, I want all the power in the defenses. If it's not, I want all the power in production. Always. Without exception. There is nothing interesting going on for me to have to decide on there.

That's why comparisons to SimCity just don't work for this. Outside of small, early cities (where you can't reach capacity on service buildings), underfunding things in SimCity carries a consequence in one area for an opportunity somewhere else. This doesn't do that. At any point in time, there is an objectively correct and incorrect setting for it. All you're making me do is switch values around to keep it on the correct one, which I'm just going to turn around and ask for automation to do because it's tedious micro (aka: the very thing the original goal was to get rid of).

[kasnavada]

Compared to having sane ship build times to begin with and not having construction speed boosts at all? Not really.
What possibilities? What can I do with this that I'm not going to just want automated? If the AI is attacking me, I want all the power in the defenses. If it's not, I want all the power in production. Always. Without exception. There is nothing interesting going on for me to have to decide on there.

>Sane build time.
I'm actually hesitating what I'll be modding first, between neinzuls or a new game mode that will limit the game to 3 hours long sessions.

I feel you there, but I don't think that engineers were the answers to "sane build time". IMO they were the cause of the insane times, if anything. As everything could be potentially be boosted by a crapload of engineers... (given metal resources). I suppose they'll have to be saner if that's in the game because engineers will be gone.

>There is nothing interesting going on for me to have to decide on there.
Frankly, there is NOTHING interesting at micro-managing a resource production system in a RTS game, full dot. I'm here to pew pew stuff with my space pea-shooters. Once I captured ressource points and set-up my defenses... I don't want to have to check that collector Y is misplaced and could be 10% more efficient there. I play city builders when I want that.

>All you're making me do is switch values around to keep it on the correct one, which I'm just going to turn around and ask for automation to do because it's tedious micro (aka: the very thing the original goal was to get rid of).
Well.
That could be an idea. You seem to want to propose that the settings switch automatically given a series of criterias. Could you define those criterias, and their consequences, and we'll see what Chris does from there ?

=)

[chemical_art]

A large part of "insane build times" was not because ships were slow to build but because ships got more expensive over time. Most production facilities produce units at the same rate but as ships got more expensive they took longer to build. The use of engineers was allow flexibility in constructing those units without having to worry about the standard rate breaking the economy. Shifting more power to construction facilities leads to more cases of this problem. So the micro is shifted to turning on/off buildings otherwise economy sputters and nothing gets built.

[Tridus]

A large part of "insane build times" was not because ships were slow to build but because ships got more expensive over time. Most production facilities produce units at the same rate but as ships got more expensive they took longer to build. The use of engineers was allow flexibility in constructing those units without having to worry about the standard rate breaking the economy. Shifting more power to construction facilities leads to more cases of this problem. So the micro is shifted to turning on/off buildings otherwise economy sputters and nothing gets built.

When a single starship takes two hours to build at the normal rate, I'm less concerned about the economy and more concerned about just how long I'm intended to wait for a fleet *without* rampant engineer boosting. Granted that's at mk V, but still. Clearly, nobody is intended to actually wait for that build time.

[Tridus]

I feel you there, but I don't think that engineers were the answers to "sane build time". IMO they were the cause of the insane times, if anything. As everything could be potentially be boosted by a crapload of engineers... (given metal resources). I suppose they'll have to be saner if that's in the game because engineers will be gone.

I'm not advocating bringing engineers back. I'm advocating setting base times such that you don't need to boost everything to make it workable. Engineers weren't a great answer to the problem either, and having that go away is something I'm perfectly fine with.

Frankly, there is NOTHING interesting at micro-managing a resource production system in a RTS game, full dot. I'm here to pew pew stuff with my space pea-shooters. Once I captured ressource points and set-up my defenses... I don't want to have to check that collector Y is misplaced and could be 10% more efficient there. I play city builders when I want that.

Yep. Hence why I don't want to do that either

Well.
That could be an idea. You seem to want to propose that the settings switch automatically given a series of criterias. Could you define those criterias, and their consequences, and we'll see what Chris does from there ?

=)

Except that if a game system works best when fully automated, you have to ask yourself if it existing at all is adding any value. I already explained how I'd want this automated if it could be automated, which is pretty much why I think it shouldn't exist at all. We should have fast standard build times, no boosts, and no "allocating power between building and defense" aside from the obvious: building defenses uses power.

[ptarth]

I'll admit I've had a hard time devoting any time to keeping track of what's going on in the design. I try to browse the design document a couple of times per week. I try to keep track of discussion but, in general that's chaotic due to the lack of summaries. I also think that many good points are brought up and then buried under other discussion. So rather than soapbox, I'll try my classic bulleted list approach.

• Macro is fun, to an extent, M003 (argueably) took it too far.
• Micro is fun, to an extent, MOBAs (pick one) took it took far.
• Some RTSs seem to get the balance okayish. E.g., Total Annihilation, Supreme Commander. However, those titles did not make the millions that Blizzard did with its RTSes.
• This discussion is basically, how much macro/micro do we want for the planetary interactions. (similar to other discussions just replace planetary with ship/starbase/fleet combat/etc/etc).
• So perhaps instead of fighting about the position, we think about what is fun about planetary management, and just do that?

What I like about planetary management.

• Seeing my minions do my bidding.
• Designing something and playing with sliders, for a little while.
• Making a template, that matters.

What I don't like about planetary management.

• Excessive micro (i.e., engineers).
• Manually activated and targeted abilities. (e.g., from engineer production boosting to bloodlust to healing/repairs
• Having to specify that I want to build up the economy in a freshly conquered area.
• Having to do building/turret/minefield placements due to lack of good AI spatial planning.
• Repeating things that could be templated designs.

Would empire wide sliders with priority focuses work?

• Each planet has a defense and economic resource allocation level.
• This is how much of your empire wide budget is allocated to the planet.
• Example: 0-5 scale (click to cycle through). 1 means 1 share, 5 means 5 shares.
• Emergency button: Allocate all resources to this planet.
• AI then builds whatever given that budget.
• Locations are set automatically, spaced to defend intelligently. Removing positioning micro is really important to the reduction of micro.
• Placement in Space Games is always weird. First, scales are completely wrong, the amount of space you'd need to cover to protect a planet against ships moving at lightspeed+ speeds is incredible. Thus everything is really an abstraction where a ship represents fleets and defense turrets represent entire arrays. Second, Space is at least 3d (possibly more ds). But combat in almost all games is 2d. Even Homeworld was really 2d with a mostly unused third plane. Wormholes/star lanes are attempts to disguise this, but even then.... Plus 3d controls and visuals are generally horrible. Also, let's not even get started about how real construction networks would be better off outside of star systems.
• You could add specific modules/shipbays/whatever you want to call them to planets.
• These would be things like: minefields, turrets, shield generators, etc or perhaps just a global defense level and the ai auto-chooses.
• The AI would then build these up with the resources it is allocated.
• For economics once a planet is maximized, the excess funding is returned to the budget to be allocated else where.
• For defenses, there is no maximum, the AI keeps building defenses. You can control what type (i.e., all turrets, minefields and turrets, freeze turrets, etc, etc)
• Everything has an upkeep cost. so your production speed is relative to your resource excess. Once your budget is 100% spent on upkeep, you need to either expand or lose/recycle  ships/defenses/buildings/etc to free up resources for other purposes.
• Ship Construction is based on orders to linked secured systems.
• Linked secured systems are systems that you control (without enemy presence or threat) that are connected by starlanes.
• Any order placed for ships included a gathering point. Ship orders are then distributed proportionally to construction speed (including travel time) to all linked secure systems, and those ships automatically path to your selected gather point.

Conclusion

• That seems reasonable, but that also sounds like M003, which isn't AI Wars, nor universally loved.
• So.. I dunno. But I do keep seeing these arguments that parallel the above scenario, just in different words.

[kasnavada]

That seems reasonable, but that also sounds like M003, which isn't AI Wars, nor universally loved.

Nitpicking here (I'm not contradicting your post), but MoO3's management of micro / macro was not the only issue it had. Like the AI not being able to manage its economy even with huge bonuses, the "fleet" idea no one cared about... basically it could have worked if it was not called MoO. Sadly it was. I'm not that worried about AI's War's AI throwing their economy into a bus =). A better example of everything being possibly automated is distant world (even if I hate that game, it seems to have quite the success).

[chemical_art]

I adore distant worlds.

I will say comparing AI Wars to Distant Worlds is like trying to compare a documentary on shrimp to a documentary on whales. They are both documentaries and both feature oceans, but the comparisons end there.

From the absolute basics on up they are two entirely different games.

[kasnavada]

Yes, but as automation there is done mostly right, we can get some inspiration from it when some (if any) automation's needed in AI War. Rather than from MoO 3 where it's mostly... not.