Author Topic: Brainstorming about offense vs defense, and refleeting speeds.  (Read 18069 times)

Offline zeusalmighty

  • Full Member Mark II
  • ***
  • Posts: 151
Re: Brainstorming about offense vs defense, and refleeting speeds.
« Reply #45 on: April 08, 2018, 04:53:54 pm »
So taking your concept of the "radar graph" the science categories would be like:

[Fortresses] <-- [Turrets] --> [Ark] <-- [Fleet Ships] --> [Drones]

And then there's Starships, which are all about offense, and Resource Production, which is in the middle supporting each of these.


I'm a fan of this direction with tech synergy. It's intuitive and encourages a healthy amount of experimentation of offensive & defensive tech.

Given the limited amount of tech options available I'm not in favor of combining fleet and turret tech as some have suggested. This proposal seems in theory to encourage thoughtful investment in both types of tech

Offline x4000

  • Chris McElligott Park, Arcen Founder and Lead Dev
  • Arcen Staff
  • Zenith Council Member Mark III
  • *****
  • Posts: 31,651
Re: Brainstorming about offense vs defense, and refleeting speeds.
« Reply #46 on: April 08, 2018, 05:14:30 pm »
Various notes:

1. I'm agreed that if we're doing vast changes, we're probably going to get ourselves into trouble.

2. Points taken about mobile construction of fleet ships from flagships.  I'm happy for those to go either way.

3. What IS a planet controller?  Right now those are just "mini kings," really, and that's kinda boring.  Badger has been pointing out that they need to be visually larger in order to be commensurate with their importance, and that does make sense.  But we can't really make something larger without knowing what the heck it is.  Is it a space elevator?  Something that controls gravity in the favor of the owner?  I don't really know.  My thought, at the moment, is that perhaps these should be fortresses. 

- We keep talking about fortresses, and perhaps having the controllers turn into "the local fortress" and that being a thing that you can upgrade via... some sort of means... would be cool.  These fortresses would be locked to their planets, of course, since they denote ownership.  But it would also play well into the "you made it stronger, the AI took it back, and it's still stronger but now belongs to them" concept. 

- Plus you could run into the occasional mark VI fortress that makes taking a world almost impossible (maybe 1 per galaxy, by default), but if you capture it then it's an amazing defensive stronghold for YOU, now.

4. Is there a conceptual problem with having turrets all be tied to a single fleet ship?  Or rather, each fleet ship type has a corresponding turret type that you get for "free?"  I realize that if we take the current turrets and try to cross-match them, we have problems, because there are more types of fleet ships than there are turrets.

- So with that in mind, I was thinking that we could have basically no new turret graphics, and just have the AI use the turrets that currently exist.  But for the player, we'd have variant-versions that are things like "Fighter Turret" and "Autocannon Turret" and whatever else.  One turret type per fleet ship, and it's just picking whichever turret icon and visuals are closest to that fleetship type, then giving mostly the stats from the fleetship, but on steroids, and costing power instead of fuel (as now).

- That way if someone mods in a new fleetship, for instance, then it's easy to handle adding a new turret: just make a copy of the fleetship, call it "[Name] Turret," change a few variables to make them stronger and power-users, and swap out the visuals and icons with an existing turret one.

- Basically this would put the science issue regarding turrets to bed, permanently, and with minimal fuss on our part and not hemming ourselves in in the future.  The science issue regarding turrets is something that is a beef I also had with the original game... but something about the general balance there just happened to work out that it never made me fully complain to myself or anyone else about it.  But thinking back to my playtime with Classic, that was an issue, for sure.

5. Maybe player turrets should simply be made mobile?  Slow, but mobile?

6. As far as having more techs go, I do agree that I like the idea of that.  I really really love the style of techs that a game like The Last Federation or Civ has.  An actual tree, and affecting stats, and all that jazz.  We had the model for that for both TLF and SBR, and my understanding is that that made things untenably slow in terms of calculating "how strong are my shots" and things of that nature, though.

- Trying to think outside of the box, though, I wonder if we had a more TLF-like "list tech tree with prerequisites built in" that is thematically-oriented rather than tied to specific ships, if that would be better.  A given tech might give you some sort of bonus system to your Ark that does XYZ, plus it unlocks mark II of ships A, B, and C if you have them.  With all the talk of making techs more interesting/varied, my thoughts immediately head in this direction.  This is probably Way Out Of Scope (tm), but if a simple implementation is possible in the engine, the GUI should be easy enough.
Have ideas or bug reports for one of our games?  Mantis for Suggestions and Bug Reports. Thanks for helping to make our games better!

Offline chemical_art

  • Core Member Mark IV
  • *****
  • Posts: 3,952
  • Fabulous
Re: Brainstorming about offense vs defense, and refleeting speeds.
« Reply #47 on: April 08, 2018, 05:31:21 pm »
There is plenty to digest over what you said Chris but my knee jerk reaction to having turrets and fleetships being so directly tied to each other is it will force all fleetships going forward to having a turret tied to it. Which is its own can of worms going forward. Before I go on let me lay down a few assumptions:

-Any fleetship that has a turret is superior to one which does not, which means all fleet ships need to have a turret.
-A turret needs to model its fleetship closely for the sake of sanity for both those designing it and for those using it.
-Turrets need to be balanced fairly equally to each other, to prevent situations were a player is stuck with too many underpowered turrets and so cannot be stuck in an awful defensive position.

The result is that all fleetships need to be designed that they are good turrets. I'll go with my first thought: How useful would an eyebot turret be, or a raider? Or any of the other light ships whose primary advantage is being fast and/or infiltrating? They wouldn't. They are terrible defensive designs from the start, locking them to defense only even if they had full engines just isn't going to work. By forcing a tie in of turrets to fleetships designed, you are forcing all fleetship designs to be good on defense as well as offense. Perhaps from a 10000 foot view having that done to fleetships anyway might be good but it does hinder growth and variety of fleetship design. I never had a use of those kinds of ships in AIW 1 but it wasn't really crippling. Having them take a fleetship slot AND a turret slot though? Rage. Which also brings up that a player is much more at the mercy of rng for ship design: In AIW 1 your turret bench was very stable so you could release stable defenses across games. In AIW 2 if you get bad fleet rolls you also get bad turret rolls.

That's my first thought.
Life is short. Have fun.

Offline x4000

  • Chris McElligott Park, Arcen Founder and Lead Dev
  • Arcen Staff
  • Zenith Council Member Mark III
  • *****
  • Posts: 31,651
Re: Brainstorming about offense vs defense, and refleeting speeds.
« Reply #48 on: April 08, 2018, 05:39:07 pm »
Further thoughts over here: https://forums.arcengames.com/ai-war-ii/my-comprehensive-feedback/msg218581/#msg218581

In terms of the turrets and fleetships issue that you bring up, chemical_art, I see that as just a purely-xml issue.  It's not a problem of art, or coding, or anything else.  Not even a major design challenge, although it's a minor one.  It's mostly just making a copy of the fleetship in xml and then making tweaks to make it a good turret.  Tune to taste, essentially.

But actually, the idea of some fleet ships that are amazing in their mobile variant, but underwhelming as turrets, or vice-versa, has some interest, too.  Yay variety, and all that.  I'm not saying I'm completely sold on any of these ideas, but these are just some things popping out of my head at the moment.
Have ideas or bug reports for one of our games?  Mantis for Suggestions and Bug Reports. Thanks for helping to make our games better!

Offline keith.lamothe

  • Arcen Games Staff
  • Arcen Staff
  • Zenith Council Member Mark III
  • *****
  • Posts: 19,505
Re: Brainstorming about offense vs defense, and refleeting speeds.
« Reply #49 on: April 09, 2018, 08:13:51 am »
3. What IS a planet controller?  Right now those are just "mini kings," really, and that's kinda boring.  Badger has been pointing out that they need to be visually larger in order to be commensurate with their importance, and that does make sense.  But we can't really make something larger without knowing what the heck it is.  Is it a space elevator?  Something that controls gravity in the favor of the owner?  I don't really know.  My thought, at the moment, is that perhaps these should be fortresses.
I think we're going to need fortresses to be able to travel between planets, as that's the most viable idea I've seen so far for "something to avoid the need to bring the fleet back to deal with a wave that was unexpectedly rough".

Another unit could be used, of course, but I think we'd be better off with one of these approaches:

1) Stick with controllers as the equivalent of AIWC command stations. Weak objectives that mostly just serve as the "flag" you plant on the planet.
- They could be renamed to Command Node or even Command Station, if that's helpful; in general we want a name that fits for the "terrain" objects that cannot be destroyed (only killed and claimed).
- Perhaps we make them upgradeable like AIWC's, which would make them more interesting.

or

2) Get rid of controllers, and make it so that planet ownership is taken away by killing all the Guard Posts (currently Power Distribution Nodes; maybe a better name would be "Guard Nodes" or "Command Nodes" or... heh, this just came to mind... "Control Nodes") and gained by claiming them.
- Due to the way the engine is written the controller entities will still need to exist, but they can be invisible, untargetable, off in the middle of nowhere, etc.
- Taking planets would take longer, which may be an issue for cases where the player can really just steamroll the planet, but possibly FRD would suffice for that.
- We'd also need to decide if the controller's current functions of being able to repair/claim anything else on the planet need to be retained in some way.


Quote
4. Is there a conceptual problem with having turrets all be tied to a single fleet ship?
It's a big move away from AIWC in that you'd only have 3 (+1 for the chosen bonus type) guaranteed turret types, and for the rest you'd just hope that you got the bonus types you wanted. The forced variety is easier to work around on offense, but turtles like knowing which tools are in their potential toolbox. This would also be more pronounced if some fleet ships have poor turret-variants.

It also wouldn't really achieve your goal of "the player should never need to spend science on non-offensive defense stuff" because the player would still have tractors, tachyons, gravity, and probably minefields and fortresses to research/upgrade.


Quote
6. As far as having more techs go, I do agree that I like the idea of that.  I really really love the style of techs that a game like The Last Federation or Civ has.  An actual tree, and affecting stats, and all that jazz.  We had the model for that for both TLF and SBR, and my understanding is that that made things untenably slow in terms of calculating "how strong are my shots" and things of that nature, though.

- Trying to think outside of the box, though, I wonder if we had a more TLF-like "list tech tree with prerequisites built in" that is thematically-oriented rather than tied to specific ships, if that would be better.  A given tech might give you some sort of bonus system to your Ark that does XYZ, plus it unlocks mark II of ships A, B, and C if you have them.  With all the talk of making techs more interesting/varied, my thoughts immediately head in this direction.  This is probably Way Out Of Scope (tm), but if a simple implementation is possible in the engine, the GUI should be easy enough.
I think we need to stick close to how AIWC did the techs; anything else is a pretty big upheaval at this point.

The only exception in what I've been thinking lately is the Ark-upgrade tree, since it would have forking paths and whatnot. That's probably too far. But what you said about needing another vector of competition, and the responses largely having to do with hacking, points in a good direction for that: have Ark-upgrades happen through hacking. So you have to expose yourself to danger to "level-up", so to speak. But details on that don't really fit in this thread.

That would be for the "change the Ark's core stats" stuff (by replacing the Ark's entity type). I'd still like to have the turret and fleet ship techs both contribute to the guns/drones of the Ark (once the Ark is past the first few core-upgrades), and for the the turret techs to contribute to the fortress and the fleet ship techs to contribute to the guard-post-drones. But none of this by way of a "bonus system" or "variants" or whatever: it just literally bolts on the same guns the turrets use, or gives drones exactly like that fleet ship (just with an icon in the drone style, for clarity).

That would be similar to how different turret techs in AIWC contributed to the Enclave Starship's drones; though that was less intuitive due to the crossover from immobile to mobile.

But even that kind of crossover may be too much at this stage. I think we may find it sufficient just to fill out the defensive toolbox with more of the stuff from AIWC (fortresses, minefields) and switch power-distribution-nodes over to guard posts.
Have ideas or bug reports for one of our games? Mantis for Suggestions and Bug Reports. Thanks for helping to make our games better!

Offline BadgerBadger

  • Arcen Volunteer
  • Hero Member Mark III
  • *****
  • Posts: 1,229
  • BadgerBadgerBadgerBadger
Re: Brainstorming about offense vs defense, and refleeting speeds.
« Reply #50 on: April 09, 2018, 12:27:47 pm »
My two cents at the moment.

Definitely add minefields and fortresses back in.

Have the Power Distribution Nodes be something you can spend Science on to turn them into defensive structures.
My favourite version of the Power-Nodes-as-Defense is as follows. You can spend science to turn the nodes into Human Special Forces constructors. HSF constructors will slowly spawn drone-style ships that you don't control. They will wander around your planets and fight any enemies that go onto your planets. If enemies show up then all the HSF ships on nearby planets will converge.

You have to spend Science to enable HSF. If we want to get fancier then we can give it a tech tree; you could invest science in A. increasing the speed at which drones spawn, B. increasing the per-HSF constructor unit cap, and C. making the drones stronger.
You could also have the Ark be able to summon all the HSF drones and have those drones follow the Ark onto an AI planet, though there would be some massive cost associated with it.

What I've described above is essentially "Turn the Guard posts" into "Miniature Human-Allied Dyson Spheres" for all intents and purposes, so it would be quite easy to implement.

Offline zeusalmighty

  • Full Member Mark II
  • ***
  • Posts: 151
Re: Brainstorming about offense vs defense, and refleeting speeds.
« Reply #51 on: April 09, 2018, 12:44:10 pm »
3. What IS a planet controller?  Right now those are just "mini kings," really, and that's kinda boring.  Badger has been pointing out that they need to be visually larger in order to be commensurate with their importance, and that does make sense.  But we can't really make something larger without knowing what the heck it is.  Is it a space elevator?  Something that controls gravity in the favor of the owner?  I don't really know.  My thought, at the moment, is that perhaps these should be fortresses. 

This concern has bugged me and mostly from a lore perspective. I'll preface this by saying I'm glad the controller subsume all the resource gathering responsibilities. I think that in this vein, the controller needs to be understood as something that governs the activities on the planet. Since humanity is basically living on the Ark, and since the AI controls virtually all territory in the galaxy, I imagine that the controller effectively handles the terraforming activities of the respective planets--each planet has an automated terraforming AI and the controller effectively gives access to the resources provided by the planet.

So a space elevator is part of the function, but more than that the controller literally controls the terraformers. What should this thing be called? Whatever is chosen should reference the planet. The planets still feel too removed from the action and tying them into the lore would be a way to address this

Offline x4000

  • Chris McElligott Park, Arcen Founder and Lead Dev
  • Arcen Staff
  • Zenith Council Member Mark III
  • *****
  • Posts: 31,651
Re: Brainstorming about offense vs defense, and refleeting speeds.
« Reply #52 on: April 09, 2018, 03:10:04 pm »
Okay.

1. Points all well taken.  Keith, consider me conceding all the points from your last email.

2. zeusalmighty, I really like the idea of these relating to the planets.  What if these were basically "Space Elevator Terraformer Space Station Orbital Facilities" (SETSSOF for short).  (Hopefully) obviously joking on the name. ;)

3. But, that said, relating to #2, what if these things were different based on the planets that were seeded there?  So you have a SETSSOF at lightning planets that gives some sort of power boost and is called something like the "Electro SETSSOF" or whatever, and maybe gives some other passive bonuses locally.  Then on a gas giant, you have a "Hydrogen SETSSOF," which gives extra fuel and does something else like maybe giving a speed boost, etc.  Maybe on gas giants its a random roll between the "Hydrogen SETSSOF" and the "Gravity Assist SETSSOF," the latter of which gives a major gravity advantage to whoever controls the planet.

4. If the SETSSOFs are really variant in how defended they are, that also seems good.  Maybe at the lava planets, there are "Hardened Military SETSSOF" units that are like fortresses unto themselves, and give a slight bonus to all the local attacks?  Whereas some of the other ones remain pretty weak (though stronger than now).

5. Fortresses coming back, and being able to travel between planets but using power, sounds great to me.

6. Minefields are something I can do the visuals for sooner than later if desired, but I can't figure out how big to make the field.  Advice is appreciated.

7. Badger, I like your idea about the power distribution nodes being able to be turned into defensive structures.  I'm not sure how I feel about science being required for that.  Not sure if that could be tied into the hacking stuff I'm talking about here: https://forums.arcengames.com/ai-war-ii/my-comprehensive-feedback/msg218594/#msg218594

8. In general, obviously the hacking stuff I'm talking about there on the other side is pretty huge in terms of additions at this stage... but if people are on board with the general idea, I can write up a document that lays this out as just a few days of work, in the main, I think.
Have ideas or bug reports for one of our games?  Mantis for Suggestions and Bug Reports. Thanks for helping to make our games better!

Offline BadgerBadger

  • Arcen Volunteer
  • Hero Member Mark III
  • *****
  • Posts: 1,229
  • BadgerBadgerBadgerBadger
Re: Brainstorming about offense vs defense, and refleeting speeds.
« Reply #53 on: April 09, 2018, 03:36:27 pm »
I think in principle the varied SETSOFF sound cool. But I think it doesn't really help one of the issues I'm concerned about, which is the degree to which high-power planets (or the lack thereof) can really affect gameplay.

Lets say we do the SETSSOF approach where you get different buffs based on the planet type. Lets say I start a game where all the obvious choke points are low power planets and the sort that don't have tanky SETSOFF. I'm going to have a much harder time winning than if those were high power planets with tanky SETSOFF. But I won't know that till I start the game. I would actually kinda consider starting a game repeatedly until I find a high-power planet for my choke point, which seems dumb.

Edit: If the problem is "We need more avenues for skill expression" then having SETSOFF be chosen at random based on the planets you've captured, that's not great. Lets say though that capturing a planet of SETSOFF type lets you change another planet's SETSOFF to that type. So say when you capture a new planet you get a vanilla SETSOFF and you get 1 upgrade point in the SETSOFF type that the planet originally has. Then you can choose to upgrade any of your SETSOFF to the type you just captured (or upgrade an existing SETSOFF of that type).

So lets say I'm defending Murdoch and I have a Gravity SETSOFF. It's not strong enough though, so I go capture a few more planets with Gravity SETSOFF, then upgrade Murdoch's SETSOFF a bunch.
« Last Edit: April 09, 2018, 03:46:26 pm by BadgerBadger »

Offline x4000

  • Chris McElligott Park, Arcen Founder and Lead Dev
  • Arcen Staff
  • Zenith Council Member Mark III
  • *****
  • Posts: 31,651
Re: Brainstorming about offense vs defense, and refleeting speeds.
« Reply #54 on: April 09, 2018, 03:51:20 pm »
That actually sounds fantastic to me.  I hadn't considered the re-rolling issue, but you're absolutely right on that.

And that then gives us one more avenue for defensive expression as WELL as a new capturable type.
Have ideas or bug reports for one of our games?  Mantis for Suggestions and Bug Reports. Thanks for helping to make our games better!

Offline keith.lamothe

  • Arcen Games Staff
  • Arcen Staff
  • Zenith Council Member Mark III
  • *****
  • Posts: 19,505
Re: Brainstorming about offense vs defense, and refleeting speeds.
« Reply #55 on: April 09, 2018, 03:57:58 pm »
If we need to make planets all have the same power output, that's fine. There's no corresponding cap in AIWC. I had thought that "having to work with the terrain of the map" was a good thing, but I understand that chokepoints are one of those things where players don't want to be hampered by any factor outside their control (random availability of resources, etc).
Have ideas or bug reports for one of our games? Mantis for Suggestions and Bug Reports. Thanks for helping to make our games better!

Offline BadgerBadger

  • Arcen Volunteer
  • Hero Member Mark III
  • *****
  • Posts: 1,229
  • BadgerBadgerBadgerBadger
Re: Brainstorming about offense vs defense, and refleeting speeds.
« Reply #56 on: April 09, 2018, 04:04:08 pm »
Might it be more fun then to just make sure that everything that looks like a real choke point has high power?

Offline x4000

  • Chris McElligott Park, Arcen Founder and Lead Dev
  • Arcen Staff
  • Zenith Council Member Mark III
  • *****
  • Posts: 31,651
Re: Brainstorming about offense vs defense, and refleeting speeds.
« Reply #57 on: April 09, 2018, 04:09:11 pm »
That works for me, yes.  How to calculate what useful choke points are is going to be tricky, though.  The thing that comes to mind for me is a flood fill algorithm, but that wouldn't work when there aren't any "true" choke points.  Keith might have better ideas on how to detect these.
Have ideas or bug reports for one of our games?  Mantis for Suggestions and Bug Reports. Thanks for helping to make our games better!

Offline keith.lamothe

  • Arcen Games Staff
  • Arcen Staff
  • Zenith Council Member Mark III
  • *****
  • Posts: 19,505
Re: Brainstorming about offense vs defense, and refleeting speeds.
« Reply #58 on: April 09, 2018, 04:33:30 pm »
It's fairly straightforward to compute "if I block node X, how much distance does it add to the paths between each of its neighbors", which is a reasonable definition of chokepoint-potential. The top 1/4 or 1/5 or whatever of those could be flagged for this purpose.

So those planets could be assigned max power, and they wouldn't have max of any other resource, so taking one has a slight tradeoff in that sense.

Would that help?
Have ideas or bug reports for one of our games? Mantis for Suggestions and Bug Reports. Thanks for helping to make our games better!

Offline Magnus

  • Newbie Mark III
  • *
  • Posts: 34
Re: Brainstorming about offense vs defense, and refleeting speeds.
« Reply #59 on: April 09, 2018, 04:35:44 pm »
Let the players build their own chokepoints.

"Warp gate nullifier warhead" (not usable if 2 or less warp gates present on planet)