Spy rework suggestions here -->

tu_79

Deity
Joined
Feb 11, 2016
Messages
7,376
Location
Malaga (Spain)
Hello, everyone,

since big G is looking for a rework, let's brainstorm it a little bit.

A spying system that I liked very much, was found in a very old game, "Star Wars: Rebellion". In this game, you picked a location, a spy and a type of action. Then the spy had to travel to the selected location, taking his time of course, and began to do his job. There were three different difficulty tasks, IIRC they were Weave nets, Gather info, Sabbotage/Break defenses, capture/rescue and Incite revolt, in order of difficulty.

Weave nets was not risky. It increased the chance of success of your local spies, the longer you spent weaving nets, the more secure it was to act.
Gather info was mostly for training your RPG units, although you could use it for identifying juicy targets for sabbotage. In Rebellion there were basic units and character units, the latter could level up. There was a small chance of being detected, but high level units did this with no risk.
Sabbotage ruined whatever they were building in the planet. Break defenses destroyed a defensive building. Basic units could not do this kind of work with a reasonable chance of success.
Capture/rescue was to capture or free a character.
Incite revolt would just do that, with a chance of the planet wanting to turn sides and forcing the current holder to use force (and lose some population) or lose control.

Several basic units could be added to a character unit to increase their chances. Depending on the action, there was a chance of being detected, and a small one of being killed (basic units) or captured (character units). If they were detected but not killed, they had to leave the planet. Also, it was harder to conduct covert actions if there was any enemy spy in the planet.

Why was it a good system? It allowed the players to choose the amount of risk. They could reasonably guess how many turns it would take to succeed with some actions. They could focus their forces to increase chances, but on fewer targets. Even if they failed, there was still a good chance for not losing the units, specially units that were leveled up.


OK, now, how could we have spies in VP without changing our current interface too much?
- Sending an embassor, it will improve tourism and relationship with the other civ, provide info on the city and its surroundings, and will help to protect against third country spies.
- Sending a spy. Upon reaching the city, it will stay dormant. A dormant spy will steadily increase the chance of success for any covert op, depending on its level. A dormant spy will also reveal what's in the city.
- Select a covert operation for a dormant spy. Each operation has a set number of turns, a chance of detection, a chance of being killed and a chance of success. Higher level spies have a lower chance of detection and of being killed, while the chance of success depends on how many turns it spent dormant. Having enemy spies or police in the city just increases the chance of detection.
- Once the operation is successful, the spy levels up, the target is stolen/sabbotaged and the chance of success for covert operations is reverted to original values. If the operation fails, leveling up not happens and the spy can try again with the same built up chance of success (remember it takes some turns to perform), or the spy can stay dormant for a while to further increase chance of success.
- Regardless if the operation is a success, the spy can be detected or killed. If killed, another one takes its place in capitol, at level 1. If detected, the spy must leave the city. If undetected, even after a successful action, the spy can stay.
- A notification will show in the city when a spy reaches 100% chance of success at any covert operation, so the player does not miss the opportunity.
- In short, the mechanic will be to send low level spies to a target, let them stay and build up, then select a covert operation according to their level, so they can level up safely. Then, higher level spies will be used for more juicy and difficult tasks. The player could choose to play safely, waiting until success is guaranteed, or risks can be taken.

Covert operations:
- Gather info. It's automatic. A spy in the city will show any relevant information. It also increases chance of success by 3% per spy level per turn.
- Steal gold/tech. 5 turns. Easy difficulty. The amount of stolen gold/tech will depend on the target base yields.
- Steal great work. 5 turns. Normal difficulty. There must be a great work in the city at the beginning of the operation. If successful, it will steal the great work, regardless of where it is located when the spy has finished.
- Kill Great Person. 2 turns. Normal difficulty. The city loses its biggest accumulated GPP.
- Sabbotage. 3 turns. Normal difficulty. Destroys a random building, except wonders and defensive buildings.
- Plague. 2 turns. Hard difficulty. Destroys a random unit inside the city, might include civil units.
- Break defenses. 5 turns. Hard difficulty. Destroys a defensive building of the higher level (armory is destroyed before walls).
- Incite revolt. 10 turns. Difficulty depends on city local happiness. The city immediately revolts.

So there are two strategies here. The player could focus on stealing as much as he can, reducing the gap against a top civ, or he could be proactive and break the other civ prior to an armed assault.
 
Last edited:
I like your ideas, but the actions you listed here are too strong. The damage done to a city should be maximum -1 citizen (plague) or revolt for 3 turns (incite revolution). Break defenses can be done, but should be temporary (reduce CS of the city by X% for Y turns).
 
If you change the spying mechanics you also have to change the "anti-spying" mechanics. Cause just as it's annoying that you don't steal or do whatever spies does it's equally annoying when you have your counterspy, all the buildings and the policies yet the AI keeps running away will all your <snip> all the time.

Also should the capital be safe from thieving? It is now. There is only diplo and tech stealing as options.

Moderator Action: If the autocensor blocks out a word, you are required to edit your post to remove it. Please keep your language on this site family friendly. - Recursive
 
Last edited by a moderator:
A spying system that I liked very much, was found in a very old game, "Star Wars: Rebellion". In this game, you picked a location, a spy and a type of action. Then the spy had to travel to the selected location, taking his time of course, and began to do his job. There were three different difficulty tasks, IIRC they were Weave nets, Gather info, Sabbotage/Break defenses, capture/rescue and Incite revolt, in order of difficulty.
That game was rebels vs empire, right? It's a 1v1, which means there isn't much difference from actions that help yourself compared to actions that hurt your enemy.

In a free for all there is a really big difference between helping yourself and hurting another civ. If the human player takes a lead, then 5 AI all send a spy on a sabotage mission it's going to have a terrible mouth feel.

For civ, I think there should be fewer 'hurt your enemies' options, it's better if spies have 'help yourself' options, like tech stealing. The target still has the tech when all is done, so really it's copying, not theft.
 
If you change the spying mechanics you also have to change the "anti-spying" mechanics. Cause just as it's annoying that you don't steal or do whatever spies does it's equally annoying when you have your counterspy, all the buildings and the policies yet the AI keeps running away will all your **** all the time.

Also should the capital be safe from thieving? It is now. There is only diplo and tech stealing as options.

Good points. A counter spy can increase chance of detection and killing, thus deleveling any foreign captured spy.

@CrazyG, yes it was 1vs1. I see your point. However, some actions that leave the target open for an armed attack should be allowed. About stealing, I agree that it might be too annoying.
 
That game was rebels vs empire, right? It's a 1v1, which means there isn't much difference from actions that help yourself compared to actions that hurt your enemy.

In a free for all there is a really big difference between helping yourself and hurting another civ. If the human player takes a lead, then 5 AI all send a spy on a sabotage mission it's going to have a terrible mouth feel.

For civ, I think there should be fewer 'hurt your enemies' options, it's better if spies have 'help yourself' options, like tech stealing. The target still has the tech when all is done, so really it's copying, not theft.

I like the direction @tu_79 suggested. However, I agree that I am not a huge fan of 'hurt your enemies' actions but could see where others might not be bother by this. If this does use the event system could it have a similar option to the "no bad events" like "no harmful espionage"?
 
However, some actions that leave the target open for an armed attack should be allowed.
Yes this action would be really interesting and I don't see it causing many problems. Spy actions will generally compete with stealing a tech, so the reward needs to be pretty big, and conquering a city is pretty big. Could be very useful for breaking stalemates.

The weaken city is probably fine (if 5 civs all planned to attack the same city it's probably not surviving very long regardless). The first time it happens if the AI takes my city because they crippled it, I'll probably be not even mad just impressed.

I was thinking more along the lines of great person assassinations getting really annoying. A better way to implement those might be a -XX% penalty to something that does NOT stack. I can live with -10% on a yield, but losing built up :c5greatperson: (or :c5gold:/:c5science:/:c5food:/:c5production:/whatever) over and over again would feel bad.
 
That gives me an idea.
What if the ambassador is allowed to perform peaceful operations? Instead of stealing techs, he would learn techs and policies, the same as a trade route does, but stronger. Instead of killing a great person, he could "recruit", meaning that the action will grant gp points to the performer, without stealing them. The ambassador will not be killed, but he can only operate with good relationships.
Then, the spy will only perform covert operations that hurt in a short term. Killing units or sabotage only available when at war. Steal technology and great works only available at peace. Incite revolt at any time, but detection being very likely.
 
I like the Civ IV system. Invest gold per turn into a pool of espionage points for each individual civ. The higher the accrued points the more "defense" you have vs that civ, as well as passive things like viewing their cities. When the points get high enough you can spend them on certain actions with a percentage chance of success. Certain buildings would accrue espionage points as well.
 
As far as fundamentals go, I like the core system that G spitballed.
  • Spy Events occur whenever a certain threshold of "Spy Points" is collected in a city.
  • Spy Points are gained when a spy is stationed in a city. Higher leveled spies gains more spy points, perhaps further augmented by buildings, policies, etc. Counterspies and spy defense buildings either decrease the spy points per turn, or increase the threshold needed to reach (whichever is ultimately easier to tune and balance).
  • Spy Events work like the event system today. The person chooses from a list and triggers that event. Some events may have costs required to trigger them, and perhaps some events require a certain level spy, or are disabled by certain spy defense buildings.
  • Nothing is random at base, no chance of failure, you get the numbers you need, and the thing works.
That is simple and clear. I get X, opponent removes Y. When X fills up a bucket, I get a thing....pretty much how everything else in civ works, so dirt simply to understand. Great!

Spy Death?: Should spies actually die in this vision? I think my answer is yes, but not in a truely random way. For example, I am fine with a spy death being the payment for my action. Perhaps a Tech Steal in a city with a Constabulary succeeds but the spy automatically dies. That is the event "price" for my action, and now its a strategic choice on whether I wish to give up the spy or not. I could also accept spy death as a counterspy event concept, see below.

Spy Levels?: So personally I prefer to think of spies as representing "organized espionage" rather than "a single super spy". This is similar to a Trade Route being organized and continuous trade, not just one dude on his caravan going off and making money. So I personally like leveling spies as a matter of your spy infrastructure rather than gaining it on missions, its something done to spies as a whole not just one spy. You have built 10 constabularies, and that is why all of your spies gain +X spy points. You took statecraft, so your spies all gain +Y spy points, etc.

Events
Note: I agree with CrazyG's discussion of positive and negative events. Its better to "copy" than perform a true "steal" in most cases, especially as this system in theory allows more pile on.

Tech Stealing: This is the principal event by which everything else is based around. Since a tech steal is one of the "strongest" events, it probably will need a counterbalance, perhaps a gold cost for bribes or something of that nature. Or do we just make this a straight up gain X science, and not make it a true straight up steal?

Art Forgery: So a change up from the thief play today. Instead of "stealing" a great work (which is a real bad mouthfeel for the opponent and really encourages everyone to hoard their GWs in their capital like misers), what if we used Forgeries to gain a new great work instead. I gain a Great Work which is super cool, the opponent doesn't lose anything, all good. In this model, we remove Thieves, forgery is just one of the event option the Spy has when he gains points.

Defector: Capital gains a free GP (chosen from a list like the normal GP selections are). I dislike losing all of my GPP in a city, again bad mouthfeel, but if the opponent just gains them than that's fine. Again this one is quite powerful, potentially on par with a tech steal. If we go with this one, than the art forgery concept is probably not necessary, as I could just generate a new great work if I so desired. Another option here is I gain a global +X% to GPP, and the opponent gains -Y%. That is still a true "loss", but its better than losing all of your current progress.

Propaganda: So I see this as similar to Ideological pressure, generating a global unhappiness modifier that can depress my overall happiness, or local unhappiness within the city. We might put a cap on it to prevent pile on.

Incite Revolt: Revolts can be insanely strong depending on how they are timed, so I think this is an event that requires very specific conditions, such as:
  • The Global unhappiness of the civ is below 35% (aka the civ is already showing revolt signs)
  • The city in question has local unhappiness (this prevents capital hits unless you are in real dire straits).
So if the conditions are right, yes you can pounce and hurt a civ badly, but the person has to be "doing something wrong" before you can, its not just something that can happen.

Delay 1 Turn / Delay 5 Turns: I think there will be circumstances where the spy is ready, but the player is not. They want to wait a few turns for the right conditions to do something, so I think its worth having two delays, a 1 turn delay if you are just not quite ready, and a 'come back to me later' delay that won't be too onerous. I like 5, some people might like 10 or so, whatever people like.

Spy Hunt (Counterspy): So back to the question of spy death. One idea is to give Counterspies a mission that can kill the spy in the city. Not sure if I like that, it feels too strong. Or a % chance to kill....but too random. I personally am fine with teh notion that a counterspy just subtracts Y spy points from the pool every round, slowing down your events. I would also be fine if the counterspy adds a "kill spy" cost to the mission event. So example, normally the defector mission is "free", but if there is a counterspy in the city, then my defector succeeds but the spy dies.

I am just really adverse to the random element of spy killings...again going back to my notion that spies aren't a single person that got killed, but a true removal of the intelligence infrastructure in an area. So I like the idea that completing mission means you have to blow covers (aka pull back agents home) and trade favors that you cannot get back....that is what spy "death" actually means. I also like that this puts a strategic choice back in the spy players court instead of just randomly hitting them with something.

Weaken Defenses: City becomes blockaded for 1 turn. Pretty powerful if used at the right time.


Gain Support (Diplomat): I do like Tu's idea of letting diplomats get in on the fun. You gain +X votes in the next election (we could go with +X votes, -X votes to the opponent....but that is a BIG swing in votes and could be crippling in the right circumstances. I think more votes is fine.

Secondary Alliance (Diplomat): Gain +X influence with all CS that the opponent is allied with.
 
I'm fine with removing chances, just paying for the outcomes.
I personally dislike the events system for this. I'd rather change the spy screen.

In the redesigned spy screen we would have:
City Target - A foreign city or a city state. Own cities are not permited.
We can target as many cities as we have spies. If there are spies unassigned, we consider they are defending homeland.
Each city has a range of options to perform. Some won't be allowed (being at peace, at war, bad relationship, nothing to learn from), and it will show what requisite we are missing to be able to perform that action. The actions we can actually perform will be shown the same way as if we were building a building in a city, showing their cost in turns. Once the action is finished, we will get the yields.
It will show also the relevant bonuses in the city that increase the cost of the actions.
At any time we can dissengage from the current action and choose a different one, losing the turns invested in it. Moving a spy to another city will take just 1 turn.

Every spy on duty will consume 1 paper, so this resource will be shared with diplomats. We may increase the availability of paper if needed.

Spy actions:
Survey. Low cost. Shows the city and its surroundings. Revealing a city unlocks covert operations.
Improve relationship. Normal cost, peace required. Increases tourism and good opinion modifier.
Corruption. Big cost, peace required, WC required. If completed before the next WC session comes, owner gets 1 positive vote from target, except for world leader. Only 1 vote per civ per session.
Study technology. Big cost, peace required. Produces a percentage of the city science production for the owner.
Cultural appropiation. Big cost, peace required, great work in city required. Creates an artifact of the same characteristics as a great work present in the city.
Defector. Huge cost, peace required. A random Great Person will be created for the owner in the target city. Cost will be reduced by the GPP produced in the city.
Incite revolt. Low happiness required. ---
Propaganda. Requires ideology, increases ideological pressure in the city.
Plague. Requires war, high cost. Garrison dies.
Break defenses. Requires war, normal cost. Forces blockade on the city.
 
Propaganda is covered by the existing Diplomat, which gives a % tourism modifier towards that civilization.

As I understand the limitations we are working with w.r.t. the existing espionage UI and the Spies/Diplomats, I think we are limited to passive bonuses for Diplomats and punctuated actions from Spies

Spoiler Diplomats :

Diplomats can be sent to any foreign major capital.
Each Diplomat on Empire provides the following bonuses:
  • 5%:c5greatperson:Great Person Rate in your :c5capital:Capital
  • 10 :c5faith:Religious Pressure of the Majority religion in your :c5capital:Capital exerted on that Target Capital (requires Religion in your capital, obviously)
  • -2% :c5science:Science Cost for any Techs researched by the Target Civilization.
  • +10% :tourism:Tourism towards the Target Civilization
  • (Requires Trade Confederacy (Statecraft) +10% yields from :trade:Trade Routes to and from any of the target Civilization’s cities
Diplomats Gain levels, but they do not get more powerful from levelling up. Diplomats lose all their XP whenever they are moved, and must start back at level 1.
Once established in the Capital, an event choice appears for the Diplomat to focus on 1 specific task. This does not erase the other bonuses, but supercharges or offers an extra ability:
  • Evangelize (Available immediately): Increases :c5faith:Religious pressure of your Majority religion in your :c5capital:Capital from +10 to +40
  • Recruit Dissidents (Available after adopting a Medieval policy tree): Increases :c5greatperson:Great Person Rate in your :c5capital:Capital from 5% to 20%
  • Smuggle State Secrets (Available after adopting an Industrial tree): Increases :c5science:Science cost of Techs researched by the Target Civilization from -2% to -8%
  • Propagandize (Available after adopting an Ideology): Increases :tourism:Tourism Pressure towards the Target Civilization from 10% to 40%
  • Rally Support (Available when a Diplomat has reached level 3): Gain +1 vote in the next :c5citystate:World Congress
  • Promote Commerce (Available after adopting Trade Confederacy Policy (Statecraft)): increases yields from :trade:Trade Routes to the target Civilization from 10% to 40%

Spoiler Spies :
Spies can be sent to any city. They generate progress towards an action each turn. When enough action points have been accrued you can select an action for the Spy to undertake.

Spoiler When in your own city: :

  • Counterintelligence (Available immediately): Adds a charge of counterintelligence protection to the city. Does not scale with spy level.
    • If an enemy spy action is performed in this city, a counterintelligence charge is consumed and nothing happens.
    • If your friendly spy is moved away from the city, all charges of counterintelligence are lost
  • Sting Operation (Available after adopting an Industrial tree): Adds a modifier to your counterintelligence, each charge consumed a chance to kill an enemy spy, depending on the level of your spy
  • Targeted Surveillance (Available after adopting and Ideology): -1:c5unhappy:Unhappiness from Urbanization and removes :c5production::c5food: yield penalties for :c5unhappy:unhappiness in this city until a new action is taken

Spoiler When in a City-State: :

  • Rig Election (Available immediately): Gain :c5influence: Influence with the target City-State and lower the :c5influence:Influence of all other major Civilizations with the target City-State. Amount of influence scales with spy level.
  • Confiscate Goods (Available after adopting a Medieval Policy Tree): Steal :c5gold:Gold equal to a portion of what you could steal by demanding tribute from the target City-State without losing any :c5influence:Influence.
    • Yields are 50/60/70% of a Tribute demand, scaling with Spy level
    • Triggers :c5culture:Culture from Tribute and Authority policies, and increases :c5influence:Influence with that City-State if you have adopted the Iron Fist Tenet
  • Raid Stockpiles (Available after adopting an Industrial Policy Tree): Steal the appropriate yield corresponding to 50/60/70% of what you could steal by demanding Heavy Tribute from the target City-State without losing any :c5influence:Influence.
    • Yields are 50/60/70% of a Heavy Tribute demand, scaling with Spy level
    • Triggers :c5culture:Culture from Tribute and Authority policies, and increases :c5influence:Influence with that City-State if you have adopted the Iron Fist Tenet
  • Install Regime (Available after Adopting Tyranny tenet (Autocracy)): Gain a Sphere of Influence over the Target City-State
  • Attempt Coup (Target City-State must be under an Open Door Policy or an Enemy Sphere of Influence): % chance that all Open Door or Sphere of Influence effects will be cleared from the City.
    • % Chance of successful coup scales with level of Spy
    • % Chance greatly increased by Covert Action tenet (Freedom)
    • If Coup fails, your spy dies

Spoiler When in a Major Civ City: :

  • Steal Technology (Available Immediately): Steal a :c5science:Tech researched by the target Civilization. Effect does not scale.
  • Steal Gold (Available after adopting a Medieval Policy Tree): Steals a % of the stockpiled Civilization's :c5gold:Gold, scaling with the Spy's level.
  • Sabotage Defenses (Available after adopting a Medieval Policy Tree. Must be at War): Lowers the :c5strength:Strength of the Target City until a new action is taken, scaling with the Spy's level.
  • Sabotage Construction (Available after adopting a Medieval Policy Tree. Target City must be building a World Wonder): Reduces :c5production:Production Progress towards a World wonder in the Target City by a % of total production needed, scaling with the Spy's level.
  • Corporate Espionage (Must have a Corporation. Target City must be an Origin city for 1 or more :trade:Trade Routes): Immediately recalls all Trade Routes Originating in this city without triggering their Trade Route Completion bonuses.
  • Instigate Riot (Available after adopting an Industrial Policy Tree): Target city will go into :c5razing:Resistance for 1/2/3 turns, scaling with Spy level.
  • Arson Attack (Available after adopting an Industrial Policy Tree): Destroy a random building in the Target City.
    • Level 2 spies will ignore Ancient Era buildings when selecting a target
    • Level 3 spies will ignore Ancient and Classical Era buildings when selecting a target
  • Forge Documents (Available after adopting an Ideology. Must be at Peace): Forces Open Borders with the Target Civilization until a new action is taken. Effect does not scale.
  • Foment Revolution (Available after adopting an Ideology. Must be at War): Generates :c5unhappy:Unhappiness in the target city, scaling with the Spy's level, until a new action is taken.
  • Infiltrate Spy Network (Available after adopting Double Agents Tenet (Order)): Clear all Counterintelligence in Target City. While performing this action, your spy is immune enemy Sting Operations

 
Last edited:
As I understand the limitations we are working with w.r.t. the existing espionage UI and the Spies/Diplomats, I think we are limited to passive bonuses for Diplomats and punctuated actions from Spies

The trick here is what triggers can be used with the Event system. For example, it might be possible when you send a diplomat to the city that the event screen immediately pops up with your "missions" that you select. Or perhaps when a certain "spy point" value is reached. That's going to be a G question, not sure how robust the event triggers are.

Each Diplomat on Empire provides the following bonuses:
Once established in the Capital, an event choice appears for the Diplomat to focus on 1 specific task. This does not erase the other bonuses, but supercharges or offers an extra ability:

I like the concept of this, effectively choosing "specializations" and then just letting the Diplomat do its thing. The trick would be how do we set up the UI to indicate to the user what specialization a current diplomat has?

(Available after adopting a Medieval Policy Tree)

I'm fine with era requirements or specific tech requirements but I wouldn't want to use policies. Some people like the ability to Christmas tree the policies, take 2 ancient era trees as an example. No reason to penalize further the people that want to play that style. I would be okay with number of policies though, once you hit 6 policies as an example.
 
The trick would be how do we set up the UI to indicate to the user what specialization a current diplomat has?
Right now the Diplomat always just says "schmoozing". Ideally we could just change that text to whatever, but I don't know what can and can't be done.
 
For me it seems spying works quite as expected (my expcetations) until I research Banking. Then all bets are off and my spies usually get caught asap and if I don't remove them they get KIA. So if I'm behing in tech I research Banking as late as possible.
 
For me it seems spying works quite as expected (my expcetations) until I research Banking. Then all bets are off and my spies usually get caught asap and if I don't remove them they get KIA. So if I'm behing in tech I research Banking as late as possible.

The suggestion here then is that advanced spy actions (which banking triggers) dramatically increase the risk to your spies....but with only a small benefit in return.
 
The pushback I am getting on troop movement micro in another thread makes me wonder if people would really appreciate spies constantly prompting you to select missions for them? People are acting like I killed their mother over there, and here we are talking about implementing an entirely new events-driven mechanic where there is currently almost zero input from the player.
 
Last edited:
Current plan:

Advanced Actions are being replaced with an Event-based model. The event uses the CityEvent system, so actions affect the city the spy is based in (home or away). Events are currently designed to select from a pool of options (weighted), with a set 'cooldown' on that Spy based on the choice you take.

i.e.

'Sap Production in City.'
a. Quick: -50p in city, scaling (20 turn cooldown)
b. Dirty: -10%p in city (50 turn cooldown)

etc. etc.

Integrating this system is already a lot of work (activating events on other civ cities is not easy) but it has the advantage of already having a robust selection of modifiers available for event creation. This is a good thing.

We're going to do this change first, and adjust the speed of GW theft/Tech theft.

We have to work within the confines of the UI, and within the confines of the netcode. A lot of suggestions here are fun but we have limited connectivity to the UI via netcode.

G
 
Top Bottom