Spies!

[Gazebo]

Anti spy buildings with the new happy system are just fine right now. I build them constantly.


The reason I am so hesitant about the spy concept is we are focusing time on features instead of balance, and there's lots of balancing left to do.

I can definitely say that spies are useful, I have never played a game and didn't use them (whereas religion I have completely ignored)

Spies have 4 primary modes, tech steal, counterspy, diplomat, cs influence.

Is there any of these that is too strong or too weak? If so...that is where I think new features could be helpful...but only if needed.

Oh sure, there's definitely a lot of balancing left to do, but that doesn't mean I can't experiment, does it?

That said, I do feel that those four primary mode are, in fact, somewhat weak. When ahead in techs, spying loses 1/4 of its utility, leaving counterspying (which might as well be an 'off' switch for your spy), diplomats (arguably their most important role, and also their least spy-ish role), and cs influence (marginal, esp. given the number of turns you can do it. Of these, there is no way to hurt civs that you are ahead of in techs. This limits the utility of spies as an offensive force, making them largely defensive and/or for catching up. This seems contrary to how spies are, historically, intended to function. Having a fifth function – disruptive actions while spying – makes them more offensive, more dynamic, and allows players not focused on techs to use spies as, well, spies.

G

 

[Stalker0]

This seems contrary to how spies are, historically, intended to function. Having a fifth function – disruptive actions while spying – makes them more offensive, more dynamic, and allows players not focused on techs to use spies as, well, spies.

G

If we look at the current "offensive" functions of spies, we see two areas:

1) Diplomat's Bonus to tourism, as tourism is "offensive culture".
2) CS Revolution.

Both of these I would argue are clunky and weak mechanics, and probably could use some improvement.


Going back to Mitsho's point, the traditional offensive spy has a few problems:

1) It can be fiddily, in a game that already has a lot of moving parts. One thing I love about the new spy system is how simple it is to implement. I just kick them off and go...and then switch them only when I need them. To me thats a feature not a bug.

2) It can be frustrating on the receiving end. In a game about control, having my empire messed with by an invisible foe can be frustrating and just plain "not fun".

Even if you add countering mechanics....to the vanilla player, you have added extra game mechanics that I have to implement, in order to solve a problem that didn't exist in the first place.



To point number 2, if you do go with offensive ideas, then I would use concepts such as "I get X gold from you due to spying" as opposed to "I steal X gold from you due to spying". while it makes less real world sense it removes all of that frustration from the defender who feels penalized by the "invisible foe".

 

[Gazebo]

1) It can be fiddily, in a game that already has a lot of moving parts. One thing I love about the new spy system is how simple it is to implement. I just kick them off and go...and then switch them only when I need them. To me thats a feature not a bug.

2) It can be frustrating on the receiving end. In a game about control, having my empire messed with by an invisible foe can be frustrating and just plain "not fun".

Even if you add countering mechanics....to the vanilla player, you have added extra game mechanics that I have to implement, in order to solve a problem that didn't exist in the first place.

Your concerns are valid, though I think the reality of these changes may affect your view. The mechanic doesn't add any new moving parts. Spies still behave the same way, there is simply a small % chance every turn that something additional will happen for the spy. Second, the fact that spies weren't truly a threat, as they didn't directly impact the player, seems, to me, a design flaw. Spies should be a problem - I wouldn't say I've added a game mechanic, but rather that I've made a game mechanic have risks and benefits that the current one lacks.

In any case, it is a minor change (relative to other changes that have been made), and can be easily toggled off by replacing a number in the SQl, so there's no cause for alarm. This was more of a 'hmm, I wonder how easily I can do this...' scenario for me. Turns out it was easy.

G

 

[mitsho]

Sure, toggling it on/off is fine. I'm a bit wary of the options file which is already huge and confusing (it took me 3 games to notice that I have to turn CivIV-Diplomacy off in the file as well ). Single versions are often easier to debug and all.

Having spies be a problem does create an athmosphere, but it moves the game away from "game" towards "simulation". Espionage then becomes it's own kind of mini-game similar to what great works swaps and archaeology digs are. Sure it can be fun, but it also means I alsmost never finish a game since turns take way too long in the late game (# of decisions). Now that's a complaint I want adressed for civ6 and not your mod

(And yes, my suggestions above aren't worth their own thread, I just needed to write them down)

 

[Lord Tirian]

Just to note stuff down so I don't forget it: I think we should steal be inspired by some Civ:BE spy actions for the random spy event system Gazebo implemented. These things provide a tangible benefit from spying and can be done in a non-punishing way:

• Divert Taxes: Gain gold equal to the target civ's gross gold income before maintenance (this should not actually subtract gold, it just generates it for you).
• Copy Research: Gain research points equal to the target civ's research per turn.
• Recruit Defectors: You can 1-3 units the target civ can build appearing near your capital (akin to militaristic city state unit gifts).

Since all of these depend on the output/tech level of the target nation, there's still incentive to go into larger, more developed nations (where the risk might be higher), so even if it doesn't "hurt" the target it still makes sense to send them to your rivals (assuming the player is one of the top dogs).

 

[Aussie_Lurker]

Hmm, makes me think of a game called "Birth of the Federation"....are you familiar with that game? In it, you could build intelligence based structures &-if you manned them-you gained espionage points that you could allocate against rival powers. This led to espionage based events you could choose to launch that could damage the relations of your rivals with other Minor & Major Powers, or damage infrastructure & the like. Of course, your rivals could defend against it by investing in Counter-Intelligence or-failing that-ascertain who was responsible (though, with sufficient points, when choosing to engage in an intelligence operation you could choose who to implicate ).

Not suggesting we should go 100% in that direction, but I do think this model has something to offer alongside the current steal tech, coup & rig elections options we currently have.

Alternatively, I don't even mind a simple system that just lets you target infrastructure (buildings) or caches (food, production, culture or the like) in cities that you have infiltrated.

Also, off on a complete tangent.....wouldn't it be cool if spies could steal Great Works? .

Aussie.

 

[Gazebo]

Hmm, makes me think of a game called "Birth of the Federation"....are you familiar with that game? In it, you could build intelligence based structures &-if you manned them-you gained espionage points that you could allocate against rival powers. This led to espionage based events you could choose to launch that could damage the relations of your rivals with other Minor & Major Powers, or damage infrastructure & the like. Of course, your rivals could defend against it by investing in Counter-Intelligence or-failing that-ascertain who was responsible (though, with sufficient points, when choosing to engage in an intelligence operation you could choose who to implicate ).

Not suggesting we should go 100% in that direction, but I do think this model has something to offer alongside the current steal tech, coup & rig elections options we currently have.

Alternatively, I don't even mind a simple system that just lets you target infrastructure (buildings) or caches (food, production, culture or the like) in cities that you have infiltrated.

Also, off on a complete tangent.....wouldn't it be cool if spies could steal Great Works? .

Aussie.

The way I've set up the 'advanced' spy actions is simply via a 1d100 roll for the function (+/- points based on rank and the value of previous actions). Most crippling events (like wonder sabotage) are lowest, gold theft is highest. There's no 'targeting' metric, as the AI for the spies, and the functions therein, are built around a 'turns to completion' model, not a 'task/completion' model. This, unfortunately, limits what we can feasibly do without punishing the AI and/or rewriting the system from scratch.

In short, 'passive' spy actions are the best I can do at present without significant investment in time.

G

 

[Aussie_Lurker]

The way I've set up the 'advanced' spy actions is simply via a 1d100 roll for the function (+/- points based on rank and the value of previous actions). Most crippling events (like wonder sabotage) are lowest, gold theft is highest. There's no 'targeting' metric, as the AI for the spies, and the functions therein, are built around a 'turns to completion' model, not a 'task/completion' model. This, unfortunately, limits what we can feasibly do without punishing the AI and/or rewriting the system from scratch.

In short, 'passive' spy actions are the best I can do at present without significant investment in time.

G

Well, the system I described above was fairly "semi-passive" in nature. From memory, you simply chose your target, you couldn't choose what kind of mission you were going to perform. I admit, though, that I am a little hazy on the details these days!

Aussie.

 

[Aussie_Lurker]

OK, I've been looking at the old BotF rules, & it was quite a passive process. You just allocated points for each rival empire (split between espionage & sabotage), then selected a espionage specialty for spies working in that empire (military, diplomatic, economic, domestic (morale), scientific or general (i.e. no specialty). You could also assign intelligence points to Internal Security.

Then you'd get event pop-ups whenever an espionage or sabotage attempt was a success or a failure.....and/or if your agents got caught in the act.

Anyway, hope that helps .

Aussie.

 

[Delekhan]

I agree that some UI feedback to the havok spies are wreaking in your cities would be beneficial.

In BE you can clearly see the intrigue levels rising in your cities as enemy agents do their work, allowing you to prepare or even counter their effects before they take hold.

 

[Gazebo]

I agree that some UI feedback to the havok spies are wreaking in your cities would be beneficial.

In BE you can clearly see the intrigue levels rising in your cities as enemy agents do their work, allowing you to prepare or even counter their effects before they take hold.

The new system gives feedback if a spy is caught, which is marginally more common now that spies are actively doing stuff every turn. With the tools available in civ, this is about as good as we can do.
G

 

[Estebanium]

Hi guys,

I play Civ 5 for more than a year now and most time I played modded Hotseat with friends. I really would like to see a system in which the player can more interact with the spies or influence them what they should do. So in my opinion the vanilla spy system is a bit boring, because once you get one you will move him to another civ and thats it. From then on you will wait, nothing more. The spy system doesn't feel like a game feature, it feels like something incomplete. I know that many of you have a different opinion, but only choosing a stolen tech from another civ each 20 or 40 turns isn't much a feature. You haven't a interconnection between a game of making so many decisions and the system of how spies work.

So I would like to pick up Gazebo's idea, but to implement a system of decisions instead of probabilities:

When you send a spy to a counterparty he will do his work as in the normal game. After X turns as normal an event will happen.
A menu will pop up in which you can choose from the following actions:

• Disrupt wonder/building production
• Try to steal an enemy's tech
• Delay a unit's production
• Remove a chunk of GPP from the city's furthest-along GP track
• Deal a small amount of damage to a city and cause it to enter temporary resistance

Each action has a chance to success by a specific formula, based on the level of the spy and a special event-modifier. So each action from which you can choose has its own modifier multiplied with the spy level for instance.

Optional: Behind the event description you can read the exact probability or another idea: you can read how your spy rates the specific action. So you could divide 100% probability by 20 for example to have five different estimations: very likely (100-80%), likely (80-60%), possible (60-40%), not likely (40-20%), very unlikely (20-0%).

Currently I only have an idea for the "Try to steal an enemy's tech" event:
[(Spy level * ("Your science production" / "Counterparty science production")) * (1 - counterintelligence)]

In real numbers: Your Spies level is 12, you produce 122 science. Your enemy 144 plus he has a building to reduce enemy spy attacks by 25%. So you would get this probability to succeed: [(12 * ("122" / "144")) * (1 - 0,25)] = 7,625%

So as you can see this is only an idea how one could calculate probabilities, based on hard figures which are already in the game and which can be influenced by both parties. One who sends a spy to an enemy and one who has an enemy spy in his city. The last one can influence the probability by specific buildings or send his spy in his own city and first one could profit from a difference in science plus from the level of his spy. In this case I imagined you would be a spy and travel to a city of a counterparty. You live there, maybe you work there in a laboratory or a factory where you will have real inside information of what they know and which technologies are used in the daily life. So my basic formula illustrate exactly this scenario.

Actually, I don't know how the probability is calculated at the moment - in the vanilla game, but it should be possible to put in some modifiers into this formula like written above. I would say that we will find a reasonable formula for each event, based on realistic and reasonable scenarios like my example of the spy in the foreign city. I think that it is not necessary to introduce new buildings for developing a formula for some event (but perhaps), but to use figures who are already there.

So what do you think?

 

[heinous_hat]

I'm enjoying the spy implementation in my first CP game so far.

Managed to get an 'arm rebels' event against my hostile neighbor's capital. The unit spawn was enough that he might have lost it had it not been defended. It was fun to watch... couldn't have happened to a nicer AI person.

To address concerns over espionage becoming needlessly punishing for the player, I would suggest tying the odds of random spy outcomes to the local (un)happiness levels in a city (literacy, boredom, security etc). That might provide some incentive to not let those sit too low while overall happiness is otherwise sufficient... at present I have little concern whether my people are illiterate criminals while I'm pulling enough luxuries. It would also give the player a more interesting and strategic way to defend (using an existing system) rather than just dumping espionage buildings everywhere or shuffling agents around.

On the flavor side, it seems intuitive that a spy would have an easier time causing disruption amongst a poor, unhappy populace. I think it would help make the underlying, finer details of the happiness system more prominent.

 

[Gazebo]

I'm enjoying the spy implementation in my first CP game so far.

Managed to get an 'arm rebels' event against my hostile neighbor's capital. The unit spawn was enough that he might have lost it had it not been defended. It was fun to watch... couldn't have happened to a nicer AI person.

To address concerns over espionage becoming needlessly punishing for the player, I would suggest tying the odds of random spy outcomes to the local (un)happiness levels in a city (literacy, boredom, security etc). That might provide some incentive to not let those sit too low while overall happiness is otherwise sufficient... at present I have little concern whether my people are illiterate criminals while I'm pulling enough luxuries. It would also give the player a more interesting and strategic way to defend (using an existing system) rather than just dumping espionage buildings everywhere or shuffling agents around.

On the flavor side, it seems intuitive that a spy would have an easier time causing disruption amongst a poor, unhappy populace. I think it would help make the underlying, finer details of the happiness system more prominent.

Good feedback. The 'arm rebels' and the 'riots' events only fire if the target empire is unhappy, but I'm sure I could include a % buff/debuff looking at city unhappiness. Is a good idea.

G

 

[heinous_hat]

So I lost that game to the same hostile neighbor (Harold).

In the closing turns, he scored a successful sabotage act on a cannon that was near completion (and needed desperately). I, of course, had my only spy elsewhere. Good use of naval bombardment ended my game.

I played militarily passive, except for a modest border guard which ended up out-teched. The espionage acts (especially the revolt and the sabotage) were welcome events that actually made an impact (for and against). I really should have used my spy defensively at the end.

The only other thing to note is that spies do rank up pretty quickly, at least compared to glacial pace that occurs with vanilla. My own agent was never caught after gaining maximum experience, and did kill a number of enemy spies.

I may take a look at whoward's Espionage Reports mod and see if that will mix well for the next game.

 

[Aussie_Lurker]

Hi Gazebo.

I know this thread has been quiet for a while now, but I've been giving more thought to spies....now that I've been playing more games.

I came across the sabotage mission, which I thought was a cool touch....but then thought it would be great if the sabotage could extend to other areas of city functions-like gold production, science production, food production, culture production....or even constructed buildings?

Another possibility I have been thinking of is to extend stealing from just Technologies into Great Works, gold, production & culture (& possibly food).

Other thoughts I had were the possibility of convincing a garrisoned unit to turn on its city or creating religious discord within cities which have more than one religion in them (shame cities don't have mixed cultures anymore, as that could be another potential mission-incite ethnic unrest ;-) ).

Anyway, this is just some new thoughts I had. Don't know how many of them are actually feasible, but thought I'd just put them out there .

Aussie.

 

[Delekhan]

Sabotage really hurts if you're building a wonder and you're almost done. Actually quite enraging.

Does having a spy in your city prevent all of these events from triggering there completely?

 

[Valvts]

Sabotage really hurts if you're building a wonder and you're almost done. Actually quite enraging.

Does having a spy in your city prevent all of these events from triggering there completely?

From my own experience : No, but I helps

 

[Gazebo]

Sabotage really hurts if you're building a wonder and you're almost done. Actually quite enraging.

Does having a spy in your city prevent all of these events from triggering there completely?

Nope, but it reduces the possibility dramatically. There is around a 4% chance of a wonder sabotage happening, less if you have a good spy. Rare, but maddening (but you should hate spies, no?).
G

 

[Aussie_Lurker]

Sabotage really hurts if you're building a wonder and you're almost done. Actually quite enraging.

Does having a spy in your city prevent all of these events from triggering there completely?

See, I was the victim of the sabotage event I mentioned above, but far from being angry I was really glad to see this feature back in the game-in some form or other. In that same turn one of my spies was able to kill a Danish spy & gain an extra level-so swings & roundabouts I guess ;-).

BTW, just checking. Are we meant to get 3 spies from the get-go, or is that a bug?

Aussie.