Stealth mechanics proposal

For what's not in 'Top Priority Game Design'. Post your ideas, visions, suggestions for the game, rules, modifications, etc.

Moderator: Oberlus

Message
Author
User avatar
The Silent One
Graphics
Posts: 1129
Joined: Tue Jul 01, 2003 8:27 pm

Re: Stealth mechanics proposal

#31 Post by The Silent One »

Oberlus wrote:Another way could be to reduce stealth of fleets of ships depending on their size: a single ship would work right as it does now (if stealth>detection, ship can go undetected), but the more ships you try to sneak past the enemies or to stay undetected in a given system would require more stealth advantage. This would still be all-or-nothing in the sense that the fleets would be either detected or undetected but nothing in between. But would be less all-or-nothing than currently.
If I understand your proposal correctly, a problem with this may be that players will micro-arrange their ships into small fleets to avoid detection.
Oberlus wrote:A third way could be to only reveal part of the fleet
Or maybe that a fleet is there, but of unknown empire and/or composition.
If I provided any images, code, scripts or other content here, it's released under GPL 2.0 and CC-BY-SA 3.0.

User avatar
Oberlus
Cosmic Dragon
Posts: 5713
Joined: Mon Apr 10, 2017 4:25 pm

Re: Stealth mechanics proposal

#32 Post by Oberlus »

The Silent One wrote: Mon Jul 29, 2019 11:41 amIf I understand your proposal correctly, a problem with this may be that players will micro-arrange their ships into small fleets to avoid detection.
Yes. More than "fleet", as the group of fleets made by the player, it should be forces on range, so that many small fleets close to each other in a starlane count as one fleet for stealth effects.
The Silent One wrote: Mon Jul 29, 2019 11:41 am
Oberlus wrote:A third way could be to only reveal part of the fleet
Or maybe that a fleet is there, but of unknown empire and/or composition.
Ah, yes, yes, yes. Degrees could be something like: undetected, "something is there" (some "unknown fleet" symbol in map), "a fleet of X structure (mass) is there" (fleet symbol only reports total number of structure in the system), "a fleet of N ships with X structure is there", and fully detected.
I think I like that more than the other options.

Ophiuchus
Programmer
Posts: 3433
Joined: Tue Sep 30, 2014 10:01 am
Location: Wall IV

Re: Stealth mechanics proposal

#33 Post by Ophiuchus »

Oberlus wrote: Mon Jul 29, 2019 12:04 pm
The Silent One wrote: Mon Jul 29, 2019 11:41 amIf I understand your proposal correctly, a problem with this may be that players will micro-arrange their ships into small fleets to avoid detection.
Yes. More than "fleet", as the group of fleets made by the player, it should be forces on range, so that many small fleets close to each other in a starlane count as one fleet for stealth effects.
one question is: in which way could this scale; e.g. would having many detectors help? and does that mean detection parts or ships and planets. and how is this kind of detection applied? on directly connected starlanes? on starlane range? on universe range? only if a hidden object arrives/passes?

if we consider bulks of ships how would bleep markers work? UI wise I would imagine a single marker for the bulk in the "center" on the starlane. i think currently you need an object ID for a marker which has the advantage that ghosts can be removed automatically if you spot the object somewhere else.

as we do not have backend support for bulks - I think the simplest implementation is probably probalistic, i.e. there is a low chance of detection for every hidden object (e.g. 1%). having 10 of those hidden objects will increase the chance to detect at least one by almost 10 (9,6% and for 100 objects to 63%).
Or rather use aggregation, but for clearly defined sets of objects (e.g. all hidden ships of a single enemy empire per starlane/system).

if we wanted to make sure that bulks get detected easier, we could maybe add a temporary stealth malus in the first success, e.g. "Scrutinized" -10 stealth (or setting a marker) which will make detection more likely in the next turn

Stealth meter manipulation has two downsides though as it is independent of the scanning empire: first the detected empire is able to determine that it got detected (this could also be a good thing), second this means also non-allied empires and neutrals will have detection easier.

meta:
  • speed and stealth and "being small" help preventing discovery
  • do not scale with number of detector objects
  • ignoring detection range completely is ok
  • discovery could be similar to blocking (in the sense that you need a stationary detector)
implementation sketch:
  • discovery operates only on directly connected starlanes and the current system - this helps fast ships and makes hiders undiscoverable on short starlanes (if the ship speed is higher than incoming+outgoing starlane length it can pass by certainly undiscovered)
  • do a single discovery attempt for every system where you have stationary detectors - this means also you can only scale discovery by having detectors on both sides of a starlane
  • optional: make this detection and distance based (taking the highest detection of all stationary detectors), higher chance of detection if closer - this will penalize badly manoeuvring (and slow) ships a bit more; downside - invites sling-shotting micro.
  • probabilistic variant: discovery attempt is (WIP) a dice roll vs probability: (lowest_stealth + number-of-ships + sum(max_structure/100))/detection/10 ; e.g. if a single stealth 50-scout is scanned by a 30-detection empire the discovery probability is (50+1+0)/30/10 so about 51 in 300 ~1/6
  • probalistic variant: for simplicity: discovery works for the whole starlane/system - all objects of a hiding empire on the starlane are either undiscovered or discovered; - this means splitting hiders between two empires makes it possible that only one half gets discovered
  • optional: different levels of discovery based (based on how much success the discovery attempt has)
  • optional: add almost-discovered stealth penalty (based on how much success the discovery attempt has); not sure if it is ok if known to the other empire
  • optional: always add stealth penalty if an object is detected this turn; but this MUST NOT be known to the other empire (this could be the same stackinggroup as the almost-discovered stealth penalty)
  • non-probabilistic variant: discovery attempt is (WIP) compare detection strength vs (stealth - number-of-ships - sum(max_structure/100)) (this could also use individual stealth). if detection strength is higher, the discovery is successful
Any code or patches in anything posted here is released under the CC and GPL licences in use for the FO project.

Look, ma... four combat bouts!

User avatar
Oberlus
Cosmic Dragon
Posts: 5713
Joined: Mon Apr 10, 2017 4:25 pm

Re: Stealth mechanics proposal

#34 Post by Oberlus »

Ophiuchus wrote: Wed Jan 19, 2022 9:08 am
Oberlus wrote: Mon Jul 29, 2019 12:04 pm
The Silent One wrote: Mon Jul 29, 2019 11:41 amIf I understand your proposal correctly, a problem with this may be that players will micro-arrange their ships into small fleets to avoid detection.
Yes. More than "fleet", as the group of fleets made by the player, it should be forces on range, so that many small fleets close to each other in a starlane count as one fleet for stealth effects.
one question is: in which way could this scale; e.g. would having many detectors help? and does that mean detection parts or ships and planets. and how is this kind of detection applied? on directly connected starlanes? on starlane range? on universe range? only if a hidden object arrives/passes?
We were talking (back then) about making bigger fleets less stealthy while on flight (not only while in-system), and TSO pointed out the issue with leaving all your ships in single-ship fleets to avoid the stealth malus while on-the-fly. And I commented that the for that issue the stealth malus should be calculated considering nearby ships in the starlane instead of only the ships in current fleet.
So I don't understand what does this have to do with stacking sensor parts in a fleet.
The idea is to modify the effect to count allied ships at less than (maybe) 10 uu. Dunno if possible or how hard.

I am a bit lost at the rest of your suggestion, I'll ponder it better later.

Ophiuchus
Programmer
Posts: 3433
Joined: Tue Sep 30, 2014 10:01 am
Location: Wall IV

Re: Stealth mechanics proposal

#35 Post by Ophiuchus »

Oberlus wrote: Wed Jan 19, 2022 10:19 am
Ophiuchus wrote: Wed Jan 19, 2022 9:08 am
Oberlus wrote: Mon Jul 29, 2019 12:04 pm Yes. More than "fleet", as the group of fleets made by the player, it should be forces on range, so that many small fleets close to each other in a starlane count as one fleet for stealth effects.
one question is: in which way could this scale; e.g. would having many detectors help? and does that mean detection parts or ships and planets. and how is this kind of detection applied? on directly connected starlanes? on starlane range? on universe range? only if a hidden object arrives/passes?
We were talking (back then) about making bigger fleets less stealthy while on flight (not only while in-system), and TSO pointed out the issue with leaving all your ships in single-ship fleets to avoid the stealth malus while on-the-fly. And I commented that the for that issue the stealth malus should be calculated considering nearby ships in the starlane instead of only the ships in current fleet.
So I don't understand what does this have to do with stacking sensor parts in a fleet.
just a general design/implementation question when thinking how often and with which strength a discovery test should be made. one side of the scaling when "bulks of ships" is to hard to implement. check defaults to "once-per-scanning-empire". we can safely drop this part of discussion.
Oberlus wrote: Wed Jan 19, 2022 10:19 am The idea is to modify the effect to count allied ships at less than (maybe) 10 uu. Dunno if possible or how hard.
those i meant with "bulk of ships" as opposed to fleets.

if you have e.g. on a starlane
fleet1 - 6uu - fleet2 - 6uu - fleet3 - 6uu - fleet4 - 6uu - fleet5

are bulks entities or are they just a set of objects?

what kind of 10uu-bulks are expected to be the outcome?
3uu - bulk(fleet1,fleet2) - 6uu - bulk(fleet2,fleet3) - 6uu - bulk(fleet3,fleet4) - 6uu - bulk(fleet4,fleet5) - 3uu

or one of those? and which one?
fleet1(bulk1) - 6uu - fleet2(bulk1) - 6uu - fleet3(bulk2) - 6uu - fleet4(bulk2) - 6uu - fleet5(bulk2)
fleet1(bulk1) - 6uu - fleet2(bulk1) - 6uu - fleet3(bulk1) - 6uu - fleet4(bulk2) - 6uu - fleet5(bulk2)

So I guess you guys meant the objects have individual tests and get modifiers on that?
fleet1(2 ships) - 6uu - fleet2(4 ships) - 6uu - fleet3(1 ship) - 6uu - fleet4(1 ship) - 6uu - fleet5(1 ship)
fleet1(2+4 ships, -5 stealth) - 6uu - fleet2(4+2+1 ships, -6 stealth) - 6uu - fleet3(1+4+1 ships, -5 stealth) - 6uu - fleet4(1+1+1 ship, -2 stealth) - 6uu - fleet5(1 ship, +0 stealth)

that should be possible (either in the backend detection check or in FOCS by setting stealth meters with the usual 1 turn delay).
Any code or patches in anything posted here is released under the CC and GPL licences in use for the FO project.

Look, ma... four combat bouts!

User avatar
Oberlus
Cosmic Dragon
Posts: 5713
Joined: Mon Apr 10, 2017 4:25 pm

Re: Stealth mechanics proposal

#36 Post by Oberlus »

Ophiuchus wrote: Wed Jan 19, 2022 10:45 am
Oberlus wrote: Wed Jan 19, 2022 10:19 am The idea is to modify the effect to count allied ships at less than (maybe) 10 uu. Dunno if possible or how hard.
those i meant with "bulk of ships" as opposed to fleets.

if you have e.g. on a starlane
fleet1 - 6uu - fleet2 - 6uu - fleet3 - 6uu - fleet4 - 6uu - fleet5

are bulks entities or are they just a set of objects?

what kind of 10uu-bulks are expected to be the outcome?
No bulks I'd say.
Each ship on a starlane gets an effect that counts allied, not-in-system ships in X range abd applies a stealth malus based on that.

wobbly
Cosmic Dragon
Posts: 1873
Joined: Thu Oct 10, 2013 6:48 pm

Re: Stealth mechanics proposal

#37 Post by wobbly »

I'd just count the whole starlane, anything on it is 1 fleet.

I'm expecting funkiness in the current set of mechanics where everything is actually the stealth bonus from 1 turn ago, and that meaning ships on the starlane last round don't have a penalty, but I haven't really tried it out yet.

Ophiuchus
Programmer
Posts: 3433
Joined: Tue Sep 30, 2014 10:01 am
Location: Wall IV

Re: Stealth mechanics proposal

#38 Post by Ophiuchus »

wobbly wrote: Wed Jan 19, 2022 12:28 pm I'd just count the whole starlane, anything on it is 1 fleet.
+1 lets start with that
wobbly wrote: Wed Jan 19, 2022 12:28 pm I'm expecting funkiness in the current set of mechanics where everything is actually the stealth bonus from 1 turn ago, and that meaning ships on the starlane last round don't have a penalty, but I haven't really tried it out yet.
hm if you are on a starlane at the end of movement, i guess you get the malus probably for this and next turn as detected objects get updated multiple times in-between-turns. For combat it will apply the turn.

did you guys also imagine a stealth malus for in-system ships?

i imagine having a stealth malus for being bulky (so e.g. -1 stealth per ship in the same system / on the same starlane)

i would maybe simply add some generic stealth bonus/malus like +5 if stationary, -5 if detected by enemy empire (if you participate in battle you also get detected), maybe some extra +5 for ending a movement on a starlane (that would make speed more important, so there is a tradeoff between having more fast ships and fewer slower ships).

Dynamics which could happen:
So maybe you are 2-jumping a undetected stationary ship over a starlane and the enemy has 10 detection less than ship stealth. You get 10 stealth less the first turn. Enemy detects you because of that. Second turn it gets +5 for not being on a starlane, but -5 for being detected. Third turn you are stationary +5 and the enemy looses sight of the ship. Fourth turn you get +5 for not being detected the turn before.
Any code or patches in anything posted here is released under the CC and GPL licences in use for the FO project.

Look, ma... four combat bouts!

User avatar
Oberlus
Cosmic Dragon
Posts: 5713
Joined: Mon Apr 10, 2017 4:25 pm

Re: Stealth mechanics proposal

#39 Post by Oberlus »

Ophiuchus wrote: Thu Feb 03, 2022 10:41 am did you guys also imagine a stealth malus for in-system ships?
I am sure I once, recently, saw a commit by Geoff implementing a stealth malus depending on number of allied ships InSystem. Can't seem to find it now in the code base.

wobbly
Cosmic Dragon
Posts: 1873
Joined: Thu Oct 10, 2013 6:48 pm

Re: Stealth mechanics proposal

#40 Post by wobbly »

Oberlus wrote: Thu Feb 03, 2022 10:56 am
Ophiuchus wrote: Thu Feb 03, 2022 10:41 am did you guys also imagine a stealth malus for in-system ships?
I am sure I once, recently, saw a commit by Geoff implementing a stealth malus depending on number of allied ships InSystem. Can't seem to find it now in the code base.
This one?

https://github.com/freeorion/freeorion/ ... 6dd220a728

User avatar
Oberlus
Cosmic Dragon
Posts: 5713
Joined: Mon Apr 10, 2017 4:25 pm

Re: Stealth mechanics proposal

#41 Post by Oberlus »

wobbly wrote: Thu Feb 03, 2022 11:06 am This one?

https://github.com/freeorion/freeorion/ ... 6dd220a728
Right that. I didn't thought of looking into a tech named "PLANET_STEALTH_MOD" when searching for ship modifiers.

Ophiuchus
Programmer
Posts: 3433
Joined: Tue Sep 30, 2014 10:01 am
Location: Wall IV

Re: Stealth mechanics proposal

#42 Post by Ophiuchus »

Oberlus wrote: Thu Feb 03, 2022 11:14 am
wobbly wrote: Thu Feb 03, 2022 11:06 am This one?

https://github.com/freeorion/freeorion/ ... 6dd220a728
Right that. I didn't thought of looking into a tech named "PLANET_STEALTH_MOD" when searching for ship modifiers.
thanks, i forgot about it somehow
Any code or patches in anything posted here is released under the CC and GPL licences in use for the FO project.

Look, ma... four combat bouts!

Post Reply