FreeOrion

Forums for the FreeOrion project
It is currently Tue Dec 12, 2017 8:16 am

All times are UTC




Post new topic Reply to topic  [ 18 posts ]  Go to page 1, 2  Next
Author Message
 Post subject: Stealth Fundamentals
PostPosted: Fri May 05, 2017 4:52 pm 
Offline
Release Manager, Design
User avatar

Joined: Wed Nov 16, 2011 12:56 pm
Posts: 4288
Location: Sol III
Ok, this is going to be one excessively lengthy post, but I beg your patience to read the entire ordeal. I think we need to start a discussion what we want "Stealth" in FO to be, on a very fundamental level (hence the topic title).

The reason why I think this discussion is necessary is because there has been a lot of ideas thrown around, particularly how to make stealth a viable strategy ("Stealth as a Strategy"), stealthy empires, etc. To the point where PRs have been proposed which try to implement bits and pieces of stealth game mechanics that go beyond basic detection/stealth stats of ships/planets.

The empire stockpile mechanic which this release cycle is going to be primarily about is the most obvious example (as the entire thing is supposed to make stealthy empires possible), the proposal of stealthy supply lines the other one.

However, I think we got a bit carried away and somewhat ahead of ourselves, because IMO we don't have sufficiently thought through the fundamental concepts of what we want "Stealth" to be, in terms of actual game mechanics. But before we start to even get to some refined design of these game mechanics (not to speak of proposing actual implementations!), we need to decide what we basically want.

And before some of you object, no, I don't consider terms like "stealth as a strategy" or "stealthy empires" by themselves sufficiently precise. These are just vague ideas - certainly interesting, but not really thought through. At least that's my impression so far.

So, what I want us to do, before we proceed with designing/implementing stealth related stuff, is to step back and take the time to carefully, thoroughly and extensively consider, discuss and finally decide on where we want to go. Once we've made the necessary decisions on the basics, we can begin to focus on the details.

I'm opening this topic on the "Top Priority Game Design" subforum because, well, we should give this discussion and decision priority (considering this release cycle is about stealth related stuff and all).

After this somewhat lengthy introduction (here I'm excessively verbose again, I guess the day I manage to be brief will be when hell has frozen over ;)), lets get to the actual topic at hand (so you guys get a better idea what I mean):

When talking about "Stealth", there are several possibilities what this concept can encompass when considered as a game mechanic:

a) In it's most basic form it would just be about making ships harder or even impossible to detect, so you can hide your movements from your opponents as much as possible, or avoid detection completely. Most space 4X games I know (which have "Stealth" as a game element) implement this in some way (details vary of course). Often referred to as "cloaking".

b) A variation of the above would be to extend the idea of "Stealth" to other game objects/elements/assets: installations/buildings on planets, populations on planets, maybe large space objects like starbases if the game has such, etc. This introduces the concept of not only hiding your movements, but also other assets (or at least some of them) from your opponents. Typically you need quite advanced tech to do so.

c) The next step would further extend that to entire planets or even stars/star systems. Usually requiring very advanced late game tech, or only achievable by discovering/getting access to some sort of ancient alien/precurser assets (typical example "Ancient Ruins" ;)), or reserved to some ancient precursor aliens lurking in some reclusive/hidden spot on the map (Experimentors anyone? ;)). The idea is that it becomes possible to hide entire "bases of operation" (be it some sinister precursors or player forces), which can support large scale operations from these hidden spots.

d) Finally, extend the idea to encompass entire empires, which is what in our discussion has been referred to as "stealthy empires". For that to work, it has to be possible to hide/stealth/cloak everything: ships, planets, supply, absolutely everything an empire needs to function. If this should be possible for player empires (not only for e.g. an entire cloaked Experimentor empire), another important thing is that the means to achieve this must be available right from game start, not just as extremely advanced end game tech.

This is only a very basic summary of what one can come up with. Of course there are all kind of variations and combinations of the above concepts, but I think you get the idea. What FO currently has is probably something between options b and c.

Now, whatever option/variation or combination thereof you want to have in your game, has profound implications and consequences for the design and implementation of the actual game mechanics and all the details, especially when it comes to balancing all the related elements so you end up with something that is actually playable.

If we just go for option a, things will be comparatively simple. We need to decide if we just want to make detection harder, so the advance notice an enemy gets of your movements gets cut short (so they learn of our big invasion only when our forces arrive at their systems). Or if we want to actually make it possible for cloaked ships to slip past enemy lines and penetrate into enemy territory undetected.

In the first case we only need to come up with a mechanic that allows for a detection/stealth "race", so to speak. More advanced stealth tech means you can get closer to the enemy before you get detected, more advanced detection tech counters that. The second case is already more complex, as allowing sufficiently stealthed ships to avoid detection completely means we need to introduce some elements that counter the massive advantage you get if you can slip your forces behind enemy lines and wreak havoc in unprotected systems (e.g. by making sure that stealthed ships are significantly weaker in combat etc.). Balancing all related elements will also become more complicated/difficult in the second case.

The consequences when going for option d are vastly different, especially if we want to make stealthed player empires possible. The most obvious and challenging implication already mentioned: we'd need to make the means to achieve this available at game start. If a player needs late game tech to hide his empire, the whole thing is probably pointless (for obvious reasons).

That however comes close to opening Pandora's box, because that creates a lot of issues which need to be addressed, OTOH to counter the very obvious, extrememly powerful advantages a hidden empire has, OTOH to give the player of such an empire a reasonable chance to hold onto their advantage in stealth tech/ability, otherwise they could be screwed too easily, and then playing a stealthy empire wouldn't be much fun. Which in turn would make the feature pointless again.

I hope you can see where this is going at this point.

Now, don't get me wrong, it's not that I don't like the idea of stealthy empires and now try to make a case against that idea. Actually, what I'd like would be the whole thing, everything options a to d have to offer. That would be really cool!

However, please consider our current implementation of detection and stealth: it is kind of broken and apparently far from easy to come up with something that would make even option a really playable. The challenge being to come up with something that won't create excessive micromanagement (the original simple "range at which you can detect an object" = "your detection stat" - "the object's stealth stat" approach we had a long time ago), but also won't be a blunt all-or-nothing thing (our current approach, where you either totally defeat an enemies stealth or not at all, the first making stealth pointless, the second making it extremely overpowered).

And now compound that by the complexities and challenges introduced when allowing for entire empires to be hidden, and this right from turn 1. Frankly, that's the point where I get up and start looking for some Aspirin.

What I want all of you (that is, all of you who want to participate in this decision of course ;)) is to give your opinion and ideas what you think we should aim for, and to argue your case. Meaning, if you e.g. want to make a case for option d, please also outline how you want to address the particular challenges/issues which come with that approach.

At the end of this discussion we need to come to a decision (ideally by reaching a general consensus), which will be the basis for all further design discussions and implementations. Which means all work on stealth related stuff should be put on hold until we make that decision (to avoid wasting time and effort on things that might have to be discarded in the end).

And once we have made that decision, we should really stick to it. Of course, if somewhere down the road we realize that what we want to do doesn't work, or someone comes up with a really ingenious, awesome idea how something we didn't think possible might be doable after all, we'll have to/can revise our decision, but only in such cases.

Ok, this is my proposal. Personally I lean toward option c (or something like that), not because I don't like option d, but because I don't have any idea how to make that work (without creating something that is either pointless, or extremely overpowered, or so insanely complicated that it blows our design philosophy to pieces). Option c however might be possible. But I don't want to go into detail right now, as I've written more than enough at this point, and let others present their thoughts first.

So, my friends, it's your turn. Make your cases! 8)


Top
 Profile  
 
 Post subject: Re: Stealth Fundamentals
PostPosted: Fri May 05, 2017 6:47 pm 
Offline
AI Lead, Programmer
User avatar

Joined: Sat Sep 22, 2012 6:25 pm
Posts: 4390
Here's my take on this, in rough sketch form -- to me, it seems that the stealthy-empire choice should involve/require a real commitment to being/staying hidden, expanding in territory around opponents while avoiding combat with them, not just phasing in and out of view to strike when it pleases. So I think a key thing would be that the stealth techs/parts would not apply to ships bearing direct fire weapons, and carriers should either simply be revealed when they launch fighters, or at least have some chance of that, as we've discussed recently. I suppose there should probably be some parallel lines of stealth and corresponding detection techs aimed at various aspects like fighter launches and surreptitious supply lines, etc.

_________________
If I provided any code, scripts or other content here, it's released under GPL 2.0 and CC-BY-SA 3.0


Top
 Profile  
 
 Post subject: Re: Stealth Fundamentals
PostPosted: Fri May 05, 2017 8:30 pm 
Offline
Programming, Design, Admin
User avatar

Joined: Wed Oct 08, 2003 1:33 am
Posts: 12040
Location: Munich
Something that might make the more "severe" options for stealth (eg. Keeping all or almost all objects being kept hidden) viable and intetesting is to have more varied victory conditions. If the goal is not to conquer everyone else, and doesn't require controlling many highly developed planets, perhaps dpoing it stealthily could make more sense. Not to get too off topic, but extensive uplift, preserving nature, or engineering galactic peace could be goals that require alternate methods to accomplish.


Top
 Profile  
 
 Post subject: Re: Stealth Fundamentals
PostPosted: Sat May 06, 2017 8:06 am 
Offline
Space Squid
User avatar

Joined: Wed May 29, 2013 6:48 pm
Posts: 67
I would like to see point d) implemented as stealth race. Planets, ships and everything from that race is not visible to enemies until they(stealthy race) attack. Not whole empire from stealthy race is not visible, just planets, ships, supply lines etc... populated with that race. If they occupy planet populated with other race that planet is visible as normal, and other empires can conquer it. If their stealthy ship is engaged in combat it should be visible for x turns.

Lets say that race is from some interdimensional space and they occupy space in our dimension but cant interact or be seen by us. If they want to interact with us eg. attack they must move to our dimension and therefore become visible and vulnerable to our weapons and invasion.

This would allow many strategies from empire that have this stealthy race.
Eg. keep quiet and expand through colonisation destroying enemies colonies or expand through conquer and colonisation and fool enemies in thinking you are weaker (only few planets and ships visible) and lure them to trap...

I believe their whole planet should not be visible, not just what is on surface so other empires can colonise other planets in that system blocking their supply and forcing them to attack and reveal planet.

As stealth is great defensive advantage they have bad troops and combat pilots, but research and industry normal or increased.

PS. And there should be some late tech that can reveal them in case stealthy race decide to remain stealthy and not attack anyone...

Cheers!


Top
 Profile  
 
 Post subject: Re: Stealth Fundamentals
PostPosted: Sat May 06, 2017 3:38 pm 
Offline
Dyson Forest
User avatar

Joined: Mon Apr 10, 2017 4:25 pm
Posts: 241
a) Undetected ship movement: a must-have. It adds to the strategy game. Current implementation seems fitted to represent it in a reasonably (playable/believable/funny) way.

b) Hidden planetary installations/populations and space intallations. Another must-have IMO. It is reasonable to allow for and explain it. Camouflage can be relatively easier (technological-wise) to do on buildings and people on planetary surfaces, compared to hiding FTL engines emissions on the open space. And same applies to stationary space bases (if a ship can cloak, the station too).
Tech trees already cover planetary installation(buildings) stealth techs in a sensible way and the mechanics are as reasonable as the ship movement ones (e.g., it allows the situation in which a planet completely hides its buildings from distante observers but not once under close surveillance from an scout orbiting the planet: I like that). The fact that the current game mechanics only allows for full cloaking or complete detection is fine for fleets (if you detect something, you've detected it, otherwise you've not). But regarding planetary troops/defences and population it could be more interesting to show less the bigger the difference between (planetary) stealth and detection is, and considering the distance in the calculation, but it should be possible for a max-stealth empire to conceal most of its planetary assets to an orbiting fleet of a low-detection enemy, and few (but still some) to a max-detection enemy.

c) If I understood it well, I see (c), hiding an entire planet (the space body, with its whole gravitational and magnetic fields), is way trickier than anything else, except hiding whole stars (that would require to block light and many other energetic emissions from the star to the rest of the galaxy). So it would require the ultimate stealth tech, independent from the techs to hide assets (I mean, in the same tree but in a separate tech that does not affect fleet/assets detection. Should it be possible to hide the star systems to visiting scouts? I would say scouts (i.e. ships with detection equipment should be possible to detect stars and planets if they are in the system and have at least some detection tech).
It seems a bit useless if you consider that (many) systems will be detected by enemies (recorded on astronomic data base) before that ultimate tech is researched. I wouldn't go for this way of stealth.

d) Hiding entire empires. I like the idea and would go for it, I think it adds a very different strategy to follow and that adds up to the gameplay and fun.
(Note: I don't see this as the last step if ordered by technological difficulty or monumental character, for me (c) is the last one in such cases.)
You don't need to hide whole planets in order to hide an entire empire:
- Supply lines: (idealised) freight and escort ships and orbital docks that implement the supply lines of empires could use the same technology applied to hiding assets and fleets. In fact, I would use the same techs in game, and the same detection/hiding mechanics: supply lines could be detected depending on the detection strength of the spying ship/planet against the stealth strength and distance of the supply line. Thus, highest stealth empires should be allowed to pass through supply lines of lower detection empires and even through planets of very low detection empires, effectively sharing supply lines, and also avoiding blockades (maybe with malus for stealth on the formula for planets under attack).
If such shared supply lines are detected (because researched more detection tech or a scout ship or radar building appears, then the normal supply rules would apply and the highest supply strength retains control of the lanes (and previously stealthy supply lines get cut). The stealthy supply lines are not shown to the other human players (wheter on shared starlanes or not), but not sure how to represent it for the player that seeis it (either because the stealthy supply line is his/hers or because it has been detected), I guess extra colors next to each other as in the wasted PPs representation.
- All the rest to hide (planetary and space assets) have been covered on (a) and (b).

I would definitively go for stealth supply lines and the concept of stealthy empires (everything would be hidden, even stats in graphs and empires window). However, I guess such empires would be discovered relatively soon even by empires with with only basic/starting detection rates once they try to colonise/outpost an already colonised planet, making the strategy (sometimes, if not lucky) pointless. For the concept of complete stealthness to work as a main strategy planets should be possible to hide, coming back to point (c), or at least disguise in some way to discourage colonisation/outposting attempts.

Vezzra wrote:
The challenge being to come up with something that won't create excessive micromanagement (the original simple "range at which you can detect an object" = "your detection stat" - "the object's stealth stat" approach we had a long time ago), but also won't be a blunt all-or-nothing thing (our current approach, where you either totally defeat an enemies stealth or not at all, the first making stealth pointless, the second making it extremely overpowered).
The point in bold is of utter importance IMO.

For all or any of these to work as a late-game strategy, stealth empires should get some benefit from their tech even after other empires have researched ultimate detection tech. How?
* For fleets, population, planetary troops/defences: a formula that considers detection and stealth stengths along with distance so that
- zero stealth empires are always detectable at maximum detection range regardless of detection tech,
- non-zero stealth empires always get some benefit from it against any detection level (except on direct combat if stealth isn't higher than detection),
- when detection and stealth are of the same tech level the objects remain hidden (undetected) at mid detection range but gets revealed when scouted (i.e., there must be radar equipment/buildings) from the same system,
- when stealth is at least 1 (or 2?) tech level(s) higher than detection, the objects remain hidden even in the same system.
- when stealth is at least 1 (or 2?) tech level(s) lower than detection, the objects only get partially revealed (i.e., only a percentage of the total figures is shown), so that the higher the difference the more is revealed, up to 100% when lowest stealth is faced to ultimate detection.
* Planets, star systems: showing a percentage of a planet might make no sense, so I would go for completely un/detected, using the same mechanics than rest of objects except for the showing only a percentage.

With the above, fleet detection mechanics do account for benefits from stealth even against high-end detection empires, but fleet combat may still need a fix to its binary behaviour (all hidden or all detected in the first round) to make stealth interesting even in late game. For this I would go one of these (sure there are more ways):
- Hide a percentage of the stealth ships (same mechanics sketched above). Thus high stealth vs low detection would get the whole benefit of the 1st-round first-strike that we currently have while low-mid stealth vs mid-high detection would get some benefit instead of nothing.
- Change whole combat system to assign probabilities of dodging direct shots: the bigger and/or slower the ship, the easier to hit it, pilot skills considered also. This brings up more possibilities, like assigning bonus or malus to hit to given weapons depending on kind of target (planet, huge ship, small ship, fighter, missile), thus giving extra importance to scouting of your enemies and carefule ship design (that can be seen as extra, unwanted micromanagement) but also allowing for early-on techs like mass driver and laser to have chances to be useful for particular purposes on the late game (and, IMO, this would bring more varieity, strategy and fun to the game)


Extra suggestion:
Two forms of "stealth" (not sure this is good English):
One could desire to go unseen (i.e., appearing as less/smaller than the actual number/size) to its enemies.
But also one could want to appear as more numerous or bigger (i.e., showing larger number of ships, or higher than real damage/structure, and/or larger number of defending troops, planetary defences and shields than the actual figures). So this would be a way of deterring enemies from attacking you (risky because if they get better detection or just suspect it and go on reckless attack, you will be materially unprepared for the onslaught).
The fundamental for the mechanics of the pretending bigger way:
- The larger (while positive) the difference between pretending tech and (enemy) detection tech the higher-than-real figures you can/will show to your enemy.
Unrealistic figures would always be suspicious so it can be reasonable to set a not-too-high upper limit for the illusionary increase of strength. Therefore, max pretending strength should be set considering max detection strength and the formula chosen for the mechanics, so that max pretending against minimum detection produces something like defences/troops/ship's strength*3 while against max detection won't work. Once a combat is triggered, the illusionary (planetary/fleet) forces would be discovered and wouldn't affect the combat results. But it would force the use extra, unnecessary troops for invasions, effectively forcing the enemy to waste extra PPs on your invasion unless improving detection.
I like the idea of having too separate tech trees for those two purposes: for hiding, the one(s) already in the game; an a new tech lane for pretending to be stronger than real (maybe with all techs shared between planetary installation/defenses and fleets or with dedicated,separated techs for planetary and fleet pretending in a highly tied tree such as the current one for stealth).

Both tech lines would be indirectly imposing in the enemy a production penalty regarding planet invasion: if you see half the actual troops you (are forced to) attempt an invasion that will fall short, having to wait for the next turn to retry the invasion (under planetary defenses fire) or even have to wait for extra troops to come; similarly, if you see twice the actual troops in a planet, you my choose to play save and gather an stronger than necessary military force (not that bad, indeed, but buys time for the stealth empire). Regarding fleets, the same applies: stealth fleets may lure enemies to attack overpowered doomstacks while pretending fleets may force the enemy to amass excessively before attacking or to keep overpowered defending forces in different places hindering his attack tactics, gaining time for the pretenging empire.

In any case, I do think the stealthy way seems better than the pretending one so it may be hard or impossible to make tha latter desirable or effective, unless it is meant only for early on game with techs cheaper than the real stealth way.


Top
 Profile  
 
 Post subject: Re: Stealth Fundamentals
PostPosted: Sun May 07, 2017 4:26 am 
Offline
Programmer

Joined: Sun Feb 14, 2016 12:08 am
Posts: 359
As I understand them, the options are:
  • A: Only ships(includes anything currently attached or contained)
  • B: Anything containable by a system or containable by such objects (troops). Each with separate meters(to view one, the container must first be visible).
  • C: B + Systems(includes anything attributable to a system being visible only if the system is).
  • D: C + Anything displayable on the map (starlanes, fields), with attributes having a separate meter. (supply lines)
If that is correct, it seems we are already at C. But please clarify if I misinterpreted the options.

Quote:
please consider our current implementation of detection and stealth

Would prefer object distance and detection range to play a factor in detection strength.
I was hoping to bring up Geoff's proposal for a "noisiness" stat, but think that may derail the topic.
Seems obvious enough to have been asked before, but I'll risk asking again: why not use a probability for detection?


Top
 Profile  
 
 Post subject: Re: Stealth Fundamentals
PostPosted: Mon May 08, 2017 9:56 pm 
Offline
Dyson Forest

Joined: Tue Sep 30, 2014 10:01 am
Posts: 213
For me it is all about opening up interesting options in playing style.

There are certain kinds of playing style which I imagine would be nicely achievable by stealth:
* deep exploration - you are curious about the universe and send your ships out to explore
** can be achieved by stealthy ships (your option A) and some means of refueling
*** refueling would benefit from ship parts, stealthy refueling ships (A) or stealthy outposts (B)

* grab the bounty - you dont care so much about steadily growing your empire but hunt for bonus from building outposts/colonies on some planet
** this is first like deep exploration, but depending on the bonus you probably need to bring some ships/your troops to grab and maybe defend and maybe connect to your empire
** if the planet is "inside" of neighbors territory you need either to battle your way there or you could use stealth
*** with stealthy supply you could supply-connect the place (B+)

* peaceful politics - instead of being able to fight you evade others, forge alliances
** this is like deep exploration but needs to be complemented by a way to defend or hide your planets so they cant be invaded

* hidden expansion - you settle inbetween other empires
** this probably would be best served by complete hiding (D)
** else stealthy supply and being un-invadable also suffices (B+)
** being un-invadable also suffices for research-only colonies

One mechanic for stealth which is not there until now but makes sense at many levels:
Stealth should grow, not jump to its maximal value, at least if there is an enemy observing.

The nice thing of higher levels of stealth like your option D is:
As long as one empire doesnt know you are there they wont probably invest in detection so much.
If one empire knows you are there and they know you are a threat they will engage in the detection/stealth race.
If one empire knows you are there but they are not sure you are threat it may depend on the personality of the empire. (So a paranoid empire will certainly research detection).

Also note there are (and there could be more) ways to fight certain kinds of stealth (e.g. using mines against stealthy ships, sending out a graviton field from a system, reducing stealth).

One thing which needs to be considered is multi-empire detection and stealth.

One thing which won't be able to solve is that you detect that there are undetected outposts/colonies. (Well, you could allow multiple colonies on one planet, but that would introduce the most complexity)
Maybe we could add some nasties which would be a disincentive for researching detection. "Oh no, that wasnt the outpost of an enemy, we woke the sleeping shipeating planet monster. Now our supply line is disrupted."

In the current state of discussion I'd argue for the current system plus at least stealthy supply lines, as I think we are able to solve that level of complexity and it would open up playing styles. I also think that it doesnt hurt the general idea of supply lines as tactical assets if balanced well.

_________________
Any code or patches in anything posted here is released under the CC and GPL licences in use for the FO project.


Top
 Profile  
 
 Post subject: Re: Stealth Fundamentals
PostPosted: Tue May 09, 2017 12:20 pm 
Offline
Creative Contributor
User avatar

Joined: Fri Jun 28, 2013 11:45 pm
Posts: 3289
I've had this thread open for days trying to work out what I want to say and failing, so rather than lots of detail, some thoughts I can get back to/expand on.

1) I really like, and want, the option of stealth for combat, I like the current mechanics and am sure we can expand/improve on them over time. The Organic line is great fun to play in some respects because it relies on submarine stealth strikes for its real power, but we need to do more to expand the numbers within the range of the parts, with refinements and short range detectors, etc. I've mentioned ideas before and I'll expand later, but for now just I want this and it's what I want to work most on.

2) I intensely dislike hidden planets, except at a very high tech level: I'm fine that the Experimentor outpost can be invisible and restrict its gravitic impact on the system its in, but how on earth is a Laenfa world covered in clouds somehow invisible unless you get a ship into the system?

3) I am ambivalent about hidden colonies, but can see an interesting strategic element, especially if it combines with stealthed supply (which would be essential to a stealthed empire). I can also see approaches for species that could be completely stealthy and have different gameplay styles: gas giant or asteroid dwellers, for example, but they'd be a real pain to balance so we'd need to do some real thought into speccing that out. We need to rethink the way we stealth colonies if we're going to have such a mechanic.

An idea: would you want to go off and establish a colony on a world on which no human had ever before set foot?

When I read space opera/new colony style exploration fiction, there's virtually always a survey report from the explorers. How about we have a mechanic where all worlds are "stealthed" until we've explored them, and have survey vessels with science teams 'survey' a world using something similar to the Bombard mechanic? You can't land there until its been surveyed, and stealthed species (but not normal species) would be invisible until you'd got a survey team down, and worlds that were inhabited could reject the survey in some way or another depending on native/empire preferences.

This would become easier over time, with techs allowing it to be automated in the mid to late game, but it would give something for players to do in the early game and make expansion something you need to think about, while also allowing for planetary stealth to be less obvious—it could even allow for various options to do with who's doing the scouting, a scouting team from a Terran world would find it harder to survey an Ocean world and impossible to survey an Inferno until the right tech was discovered.

There's a horrible opportunity for much micromanagement if we do it badly, but if we do it well it might make the early game more interesting while also allowing for stealth to be there but not as obvious: until you try to land a scouting party the Laenfa swimming around their oceans just aren't noticeable until you've got the detection tech to see them, and "scientists killed by dangerous plants" would be an interesting sitrep that, hopefully, wouldn't always be too obvious.

I want Stealth for ship combat, roughly as it is but with better range of numbers and detection options: Geoff's idea for different weapons reducing stealth by differenit amounts might be a great addition here. I'm ambivalent about stealthy planets, but if we do have something we need to make it both interesting and non-obvious.

Oh, I intensely dislike Space Clouds and other stealth special granting monsters, they were put in as a stopgap, please let's come up with something more interesting with them/instead of them?

_________________
Mat Bowles

Any code or patches in anything posted here is released under the CC and GPL licences in use for the FO project.


Top
 Profile  
 
 Post subject: Re: Stealth Fundamentals
PostPosted: Tue May 09, 2017 1:08 pm 
Offline
Programming, Design, Admin
User avatar

Joined: Wed Oct 08, 2003 1:33 am
Posts: 12040
Location: Munich
MatGB wrote:
How about we have a mechanic where all worlds are "stealthed" until we've explored them, and have survey vessels with science teams 'survey' a world using something similar to the Bombard mechanic?
There would have to be a cost to surveying, like having to use up a survey ship to survey a planet (analogous to using up an outpost or colony ship to outpost or colonize). Otherwise, and even perhaps then, having to click to survey every planet would be unacceptably micromanagy for no strategic reason. Which is unfortunate, as a more interesting exploration (and unexplored stuff as a consequence) mechanic could be very interesting...


Top
 Profile  
 
 Post subject: Re: Stealth Fundamentals
PostPosted: Tue May 09, 2017 2:23 pm 
Offline
Vacuum Dragon
User avatar

Joined: Sun Sep 25, 2011 2:51 pm
Posts: 500
A possible way to make stealthy empires work would be to have (stealthy) ships that are production centers.

There could be ship parts that benefit from specials present in system (for example as long as they remain undetected by the "owner" of said special).

Although this approach would be hard to balance if planet-based empires could benefit from resource-generating ships as well (spamming research generating decoys, for example).

_________________
[...] for Man has earned his right to hold this planet against all comers, by virtue of occasionally producing someone totally batshit insane. - Randall Munroe, title text to xkcd #556


Top
 Profile  
 
 Post subject: Re: Stealth Fundamentals
PostPosted: Tue May 09, 2017 5:55 pm 
Offline
Creative Contributor
User avatar

Joined: Sat Sep 13, 2003 6:52 pm
Posts: 1396
Geoff the Medio wrote:
MatGB wrote:
How about we have a mechanic where all worlds are "stealthed" until we've explored them, and have survey vessels with science teams 'survey' a world using something similar to the Bombard mechanic?
There would have to be a cost to surveying, like having to use up a survey ship to survey a planet (analogous to using up an outpost or colony ship to outpost or colonize). Otherwise, and even perhaps then, having to click to survey every planet would be unacceptably micromanagy for no strategic reason. Which is unfortunate, as a more interesting exploration (and unexplored stuff as a consequence) mechanic could be very interesting...

It could take time..something like 1-3 turns, and if a survey ship is just sitting in a system, it will survey the available planets one at a time automatically, unless you tell it to survey a particular one first.

Eventually, you could survey remotely (with a certain tech colonies would automatically survey available worlds in a certain distance)


Top
 Profile  
 
 Post subject: Re: Stealth Fundamentals
PostPosted: Tue May 09, 2017 7:29 pm 
Offline
AI Contributor

Joined: Tue Feb 17, 2015 11:54 am
Posts: 224
Vezzra wrote:
However, please consider our current implementation of detection and stealth: it is kind of broken and apparently far from easy to come up with something that would make even option a really playable. The challenge being to come up with something that won't create excessive micromanagement (the original simple "range at which you can detect an object" = "your detection stat" - "the object's stealth stat" approach we had a long time ago), but also won't be a blunt all-or-nothing thing (our current approach, where you either totally defeat an enemies stealth or not at all, the first making stealth pointless, the second making it extremely overpowered).


The current binary system is horrible from a balance perspective, that is for sure.

What I think could be a solution is to make the stealth meter have a current and maximum capacity. With this framework:

    1. Different hulls have different max stealth and different stealth "regeneration". There may be some hulls which have a very high max stealth but very low "regeneration". For example, an asteroid hull hidden in an asteroid belt is hard to find but once you know its signature, it is easy to find the ship again (so it is hard to re-stealth the ship).

    2. There are parts that increase and/or decrease both maximum stealth and stealth regeneration. For example, engines will have negative stealth regeneration if the ship is moving. Some designed camouflage armor parts increase maximum stealth but offer less structure than standard armor. Cloaking devices will increase stealth regeneration. Some core slot energy source allows quick restealth (i.e. high regeneration) but makes the ship easier to detect (i.e. reduces maximum stealth value).

    3. Techs may or may not globally affect the two stealth meters, possibly depending on hull line.

    4. Being in the scan range of an enemy empire decreases the current stealth value of the ship by some value per turn. That value would depend on the detection tech level of the empire and generally be greater than the "restealth" rate of a ship so the ship will be detected in enemy empire detection range eventually. Ships would lose more stealth per turn if in an enemy controlled system (similar to the current mine techs work).

    5. Once the stealth meter drops below some value based on the detection tech level, the ship would become visible to that empire as is now. To restealth, the ship must be moved outside of enemy detection range.

    6. The restealth rate may or may not be increased in owned systems, nebulas / ion storms or maybe some planetary specials / buildings.

    7. Engaging in combat may reduce the stealth meter or disallow regeneration for some turns. This could also be the place to introduce the "noisiness" mechanic: Each shot fired or fighter launched reduces the current stealth meter of the ship. Possibly, the ship wouldn't automatically decloak after first turn but stay in stealth until it drops to some minimum value (balance-wise, that value should probably be higher than the galaxy-map stealth; the added complexity may not be worth it).

So what are the advantages of this system?
    1. The system is deterministic while avoiding the "binary" approach we have right now. "Outdated" stealth technology is not rendered immediately and utterly useless but will simply limit the time stealthed and the effective (stealthed) range of the fleet. On the other hand, even with worse detection tech, stealthed ships are "eventually" detected.

    2. The meters would allow for different design choices (camouflage armour, restealth-rate, different hull types ...) allowing for interesting decision-making based on the current ingame situation. Do you aim for quick restealths or a slow-to-restealth ship which can stay a long time in enemy detection range? Do yo u aim for maximum stealth or do you only aim for a stealthed first-strike capability or even a good old blunt unstealthed gunship?
    This directly increases the number of (situational) viable designs and strategies which therefore increases the amount of (interesting) decisions the player can make. At the same time, this is a rather high-level decision-making avoiding excessive micro-management.

    3. Detection and stealth techs as well as species traits may become more nuanced by having them apply to the different meters.

The same system would, in principle, also be applicable to planets.


An arguable weakness is that the stealth meter would be a single value against all empires (having a value against each empire seems too complex). To come up with some random fluff-explanation: The stealth from an object comes from some sort of "cloaking field" which becomes weaker when it interacts with sensor technology. Obviously, the cloaking field is configured correctly to not interact with an empire's own detection devices...

_________________
If I provided any code, scripts or other content here, it's released under GPL 2.0 and CC-BY-SA 3.0


Top
 Profile  
 
 Post subject: Re: Stealth Fundamentals
PostPosted: Tue May 09, 2017 7:51 pm 
Offline
Programmer

Joined: Mon Feb 29, 2016 8:37 pm
Posts: 204
I will follow MatGB's lead and comment before the topic goes cold.

The reason that we should strive to make stealthy empires work, is that they are
just fun. I player two or three games with the stealthy supply mechanic that I
worked on. Exploring and expanding through the galaxy backfilling in the "Good"
worlds that other empires left behind felt exciting.


Vezzra

WRT range vs detection strength.
Why did reducing range of detection with the stealth strength not work well?


Dilvish

Stealth is so advantageous in combat, because we have balanced combat to
be short, 3 rounds, usually completely resolved in 1 game turn and biased
towards more power than armor/structure/shields. That accentuates the first round,
first strike advantage. The grindier combat is the smaller the benefit to
stealth as a first strike advantage.

That is why the Solar hull is so powerful, it confers a stealth bonus of +120 to
all allied ships in the fleet.

A simple implementation to Geoff's "noisy" idea would be to add a stealth malus
to some/all of the weapons parts. That would limit the number of weapons that a
ship could add and still be stealthy enough to avoid detection, without
preventing players from exploring strategies like large numbers of stealthy
ships with a single weapon each.


Geoff

I like the idea of changing victory conditions to achievements. We could
add achievements like "met all the empires" (would require some changes), "have
at least one colony of each species in the current game", "exterminated all
other species". And finally, "the achieved all achievements in one game
achievement", met all empires, made peace with all empires, had a colony of each
species, destroyed all other empires, conquered the Experimentors, exterminated
all other species and transcended.

This also expands the game concept from a military zero-sum game to a galactic
empire simulator.



Oberlus

Your decoy idea is brilliant.

A way to implement the mechanic might be to allow designing ships and adding a
decoy generator internal part. That part makes the ship very cheap to build,
but none of the other parts are real. The decoy generator strength plays the
same role as stealth strength in defeating detection strength up to the decoy
strength. In combat their weapons do no damage. The tell-tale sign of a decoy
ship would be that one internal part is empty, unless we allow it to mask itself
with some other part.


dbenage-cx

I agree with you that we are already at C. I think that the changes to supply
lanes stealth representation just make things more consistent, so that visible
supply lanes don't reveal an otherwise hidden colony.

I don't like a probability of detection for mechanical and narrative reasons.

It is difficult to make a probabilistic mechanic transparent. Without a UI
element that shows the percentage chance of discovery each turn, so that it is
clear on the turn after you detect the stealthy ship/colony why it disappeared.
It is easy to reason that if my detection is higher than their stealth then I
will find them. It is harder to reason if I have a 20% chance of detecting an
object with X stealth each turn then I have a 90% chance of detecting a
stationary object over 10 turns.

Narratively, we describe the stealth technologies as completely new tech at each
level. By analogy, I imagine a ship of the line trying to detect the
communications of a modern destroyer. No matter how carefully they examine the
flag signals of the destroyer, they don' have the smallest chance of detecting
radio communications.

I like discrete thresholds. It imposes choices on the players.


MatGB

I was going to suggest adding more dynamic effect concealing planets to add
cover for stealthy empires and then I read your post and I get that there is
something not quite correct about that solution.

However I thought, if we adopt stealthy supply, then there is no absolute need
for a stealthy empire to be completely hidden. With stealthy supply, the
stealthy colony is still functional because it can't be effectively blockaded.
The non-stealthy empire, knows that there are sneaky foreigners about, but that
is just part of the game. I think that after we add stealthy supply there is no
further need to conceal the presence of the stealthy colony.


Morlic

I'll give your suggestion some thought.


Here are some additional comments:

1) global vs local (or the wrong scales of interactivity)

In the game currently detection upgrades are global, instant and free after the
research costs while stealth is local and not upgradable. This compares with
weapon, shield, armor and hull upgrades which are all local and require an
additional production expenditure.

Consequently, it does not seems satisfying to either research a new detection
level and reveal your opponents everywhere or be revealed by the converse.

Try removing ranged detection from planets. The only ranged detection would be
mounted on scout ships. Require the upgraded detectors to be added to new ships,
the same as weapons upgrades.

This makes scouts an essential part of the game, right through to the endgame.
It also makes taking out scouts a valuable pre invasion service.

2) Consider a stealth penalty for number of allied ships in a fleet, (or in a single
system). One stealthy ship is stealthy. One hundred stealthy ships are not.
This also fixes the combat issues, because a big fleet no longer gets a first
strike advantage.

3) If we add a stealth penalty per ship in a fleet, then also consider
dispensing with the attacking reveals the attacker trope. In stories, people
invented the trope to force an interaction. We use it to prevent stealth in
combat being the ultimate weapon. However, if massive fleets lose their
stealthiness it would mean that stealthy fleets have to be small, and can only
work as harassers against a larger fleet. They could overtime grind down
a fleet.


Top
 Profile  
 
 Post subject: Re: Stealth Fundamentals
PostPosted: Sun May 14, 2017 9:34 am 
Offline
Space Floater

Joined: Sat Oct 24, 2015 1:58 am
Posts: 20
I'm worried that stealth empires won't be fun to play against. If stealth doesn't neutralize the threat of conventional empires they aren't working and if they are they'll be untouchable and horribly frustrating to play against. This is bad for multiplayer, obviously, but one way or the other it's bad for single player as well. Either the AI can use the strategy, in which case it's no fun for anyone who is self aware enough to experience fun, or the AI can't and using it is the sort of AI stomp that gets old real fast.

Stealth needs to stay in a regime where hiding some of the things some of the time is useful because 0% or 100% stealth are both not fun. Well, 0% stealth is okay if you haven't invested in stealth, but having your stealth invested military drop from super-powered to completely useless because your enemy discovered one tech is not fun. Having your enemy suddenly go from challenge to chumps because you caught up with them in detection is only briefly fun. Then you realize there's no challenge left and the game becomes a slog of conquest because their entire fleet is nearly useless. I think that means it has to be entirely tactical. Even hiding colonies is a problem since it makes them immune to conquest. Stealthed logistics is probably fine if planetary conquest is still allowed, but otherwise I think it should be limited to ships only. I guess logistics is abstracted ships so that's not really an exception.


Top
 Profile  
 
 Post subject: Re: Stealth Fundamentals
PostPosted: Sun May 14, 2017 12:56 pm 
Offline
Creative Contributor
User avatar

Joined: Sat Sep 13, 2003 6:52 pm
Posts: 1396
Atarlost wrote:
I'm worried that stealth empires won't be fun to play against. If stealth doesn't neutralize the threat of conventional empires they aren't working and if they are they'll be untouchable and horribly frustrating to play against. This is bad for multiplayer, obviously, but one way or the other it's bad for single player as well. Either the AI can use the strategy, in which case it's no fun for anyone who is self aware enough to experience fun, or the AI can't and using it is the sort of AI stomp that gets old real fast.

Stealth needs to stay in a regime where hiding some of the things some of the time is useful because 0% or 100% stealth are both not fun. Well, 0% stealth is okay if you haven't invested in stealth, but having your stealth invested military drop from super-powered to completely useless because your enemy discovered one tech is not fun. Having your enemy suddenly go from challenge to chumps because you caught up with them in detection is only briefly fun. Then you realize there's no challenge left and the game becomes a slog of conquest because their entire fleet is nearly useless. I think that means it has to be entirely tactical. Even hiding colonies is a problem since it makes them immune to conquest. Stealthed logistics is probably fine if planetary conquest is still allowed, but otherwise I think it should be limited to ships only. I guess logistics is abstracted ships so that's not really an exception.


I do think the 0% v. 100% problem can be solved somewhat

100% stealth... enemy can't touch you

0% stealth... normal combat

X% stealth (some tactical numerical bonuses)

so if the enemy's detection is Just high enough to see your forces, you have to do normal combat... but you have bonuses for normal combat (until thier detection is say 2x as high as your stealth, then all the bonuses are gone)


Top
 Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 18 posts ]  Go to page 1, 2  Next

All times are UTC


Who is online

Users browsing this forum: No registered users and 1 guest


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  
Powered by phpBB® Forum Software © phpBB Group