Ok, you do your straight up skill match with no random elements.  A simple pass/fail system going by two preset numbers.

 

How is that defense determined?  If it's predictable, you've just created a memorization game.  It becomes pointless once you figure out the defense levels of your targets, you always succeed.  If it's random, you're taking a shot in the dark, it's still luck based, just on whether the difficulty is higher.  If it's based on the actions your opponent takes, you've just created one of those annoying games where you have to dump resources into a defense system just to make an entire category of activities pointless.  After all, if it's pass/fail, no one will ever get lucky and become powerful enough to circumvent them once they're already too good.

 

I meant it when I said I couldn't think of a good way to do it.  I've despised every espionage system I've ever used, but I've never come up with even an acceptable system to replace them.  They always add chaos to an otherwise orderly system by breaking the rules and turning a game of skill into a slot machine.  Luck is nice, but only with large numbers, and no one wants to manage hundreds of spies, which means the dreaded abstraction.

This is the same problem that faces many mechanics in TBS games. It's about balancing randomness with predictability.

I think you're overly harsh on espionage systems. I'm not a huge fan of the civ-style system, but most space empire games have playable systems.

Playable, sure, but are they fun?  Does the espionage system in GC2 add anything but a nuissance?  You run around neutralizing each others saboteurs, spending a portion of your income that could have gone towards your fleet to do something vastly less amusing.

 

Bigger ships with better guns...  Or you can blow 5% of your budget on a slider.

 

Maybe I'm overly harsh, maybe I'm just not pessimistic enough to think people actually prefer the monumentally boring espionage mechanics over more warfare in space games from Master of Orion to GalCiv2.  Perhaps if they decided victory by way of a pong match?

Yes, actually you are. Even the GC2 system had a little bit of a thrill factor wshen I went on a spy blitz, and when I didn't feel  like dealing with it in a particular game, I just turned espionage off.

I actually like the espionage mechanics in Master of Orion II. Although, I am probably just be blinded by my love for the game in general.

A possibility I'm toying with is for espionage to work on a system similar to the 'hero' system. Each player can recruit 'agents' -- like heros -- who can be sent off on various missions. Again, like heroes, they have attributes which determine their success and can gain levels. Unlike heros, agents are controllable only through their portraits in the main user interface. I think, from a UI perspective, this is the best way to handle espionage. If I have some time off, I might throw together something -- I need to practice writing programs, anyway.

Of course, this doesn't solve the problems that come with balancing the mechanic, that is, the more important problems. This is commonsensical, but any system would need to be balanced: (1) externally, ie. the mechanic is balanced in relation to the rest of the game in the sense that it is not so strong that it overshadows other mechanics, but not so weak that it itself is overshadowed; and (2) internally, ie. the mechanic is balanced  in the sense that the costs are proportional to the benefits. I think your're right that most games don't achieve this. (Although I think that some are more successful than you give credit.) 

Goodmorning all
ZigZag, that system is much like the one i am proposing,   a system where most factions have a certain number of agents (~1 per opposing faction). they are given missions on the global level and handle thier own travel.  Every mission takes pleanty of time, and instead of RNG to fail or succeed,  the RNG adds or removes turns left to complete(weighted heavily by skill and level of the spy), with a very small chanse of being detected outright.  When a spy is caught it can be killed, ejected from the state, or fed fake information and used as an unwitting/unwilling double agent.

What do you think?

Robbie Price

More or less. Although, for simplicity's sake, I don't think that there should be any distance calculations. I think having the RNG affect both the duration of the mission as well as its outcome is a good idea.

Perhaps have one RNG which selects between three outcomes each turn the spy is deployed on a mission: (1) success (very low probability), (2) failure (very low probability), (3) no change (very high probability); and another RNG which selects between two additional outcomes: (4) spy killed (very low probability), (5) spy injured (very low probability), and (6) spy survives (very high probability). These two rolls would not be mutually exclusive. There could, for example, be a spy who is killed while successfully completing a mission or another who survives while failing. Obviously, all of these dice rolls are highly affected by the spy's statistics.

Again, for simplicity, I think that there should be a maximum of one spy per mission. In general, the benefits should be commensurate to the mission's duration. (ie. long missions, no killing the enemy's king, etc.)

The only problems are that:

1. RNGs do not reward skill, and encourage just save-reloading.
2. They also have a tendancy to get on people's nerves, especially when placed "ontop" of each other (one RNG leading to another leading to another ad infinitum(sp?)). You get through the first series, then the next bumps you back.
3. It makes it very difficult to coordinate a strategy, since you don't know largely what your next move will be.

About randomness in a spy system:

As long as the system is mostly skill-based I don’t see any problem with bringing in the chance-factor. It is a GAME we are talking about, and luck (when handled properly) is a perfect way to create suspense. As any real strategist knows, any good strategy must also take the unpredictable into account: you can have good or bad luck, so you will have to prepare for many possibilities in advance.

 

Also, psychoak seems to put too much emphasis on the importance of the number of spies necessary to decrease the chance-factor. Well, you shouldn’t forget the TIME factor: even if every faction has just a few spies, if these spies are active from turn to turn you don’t need to play an endless game before the chance factor is quite limited (in a skill-based system just using chance as one modifier deciding the final outcome). Say 5 spies per faction in a game with 6 factions: in every turn you have 30 “die rolls” carried out behind the scene (without any micromanagement necessary). In a game of a hundred turns or much more, what chance is left is not only fully ok, it is IMOP just a good way of adding in some unpredictability. And that is in a system without “armies of spies” or any micromanagement. So, no problem at all. People who want to play Chess can look elsewhere.

 

Victory condition:

I don’t see the need for a special spy-related victory condition. Spies should be used to achieve military or diplomatic victory etc by hurting your enemies and giving you advantages like info. However, it would be really cool if there were spy-related quests (some maybe letting you win the game, I don’t know how quests will work): like: steal the life stone of your arch enemy, or assassinate the Priest God of Istar, or sabotage the Holy temple in the heart of your enemy’s kingdom, and stuff like that.

I like the ideas of Black Dawn, who clearly shows that you can indeed implement a spy system that wouldn’t necessarily suck. Here is how it is done:

 

I would propose the following as basic info (maybe not complete but …) for a spy system that at least I myself would find adding a LOT of gaming value to Elemental (would still be optional):

 

The spy is a unit you can train (and give equipment), much like you train a worker or basic militia. Buildings, technologies and faction abilities will cut the price, time of production, add skills available, increase skill level.

 

You can endow the spy with essence much like another unit (I don’t know how heroes will work in Elemental, but this could turn the spy into a hero unit, give it magical abilities etc, making it more powerful, opening up new skills maybe).

 

You can send your spy into enemy territory: as basis, it is invisible to any enemy unit (or only enemy spies can see it). But if the enemy has a certain level of counterespionage there is a risk that the spy is revealed. The more heavily populated area it moves in, the higher the risk (since more people would have the chance of noticing something wrong: this is a game mechanic so that the spy can more easily stay hidden in “less important” areas like a forest or scarcely populated farmland, while you would need a higher level spy to sneak all the way into the capital or other major cities where it could then do much more damage). The spy is weak in standard combat.

 

When the spy moves on the map it scouts. It can also enter enemy territory without you needing to declare war, which can be a major strategic advantage and option. This would allow weak realms to actually DO something against much stronger enemies without suicidally declaring war. A major point lacking e.g. in Galciv2.

   

The spy can move into a city. Every turn it stays in the city it gives you more info than normally available: how much depending on its skills and level. Very small base risk of getting caught each turn even if there is some basic counterespionage in the city.

 

You can order the spy to initiate special actions in the city depending on its abilities. When you order this, there is a certain number of rounds of preparation (depending on skill/level of spy and the action itself) during which the spy plans and prepares for his action. Also during this time he can get caught. When the time has passed, you get a message saying whether he succeeded or failed: e.g. that your spy in city X has managed to poison the well in city Y, or has managed to start redirecting resources to your own realm. Or that he did not succeed: he is then still not revealed but would need to try again. This preparation time that goes with each action is a game mechanic so that you can’t try to e.g. assassinate each turn. It also gives suspense when you wait for the moment you will know if the spy has succeeded or failed …

 

The actions could be as proposed by Black dawn (steal resources, steal research, sabotage structure, poison the wells, assassinate unit), and I would add: sowing discontent: this decreases the morale/happiness of the city. The spy will continue doing it without micromanagement intervention. Average chance of getting caught. If content drops too low normal rules on rioting would apply.

 

Every turn the game would compare the level of the spy and the level of counterespionage, with modifications for e.g. population, magic etc and with a certain random element included. Either the spy stays invisible and continues, or one of the following happens:  

 

1. the existence of the spy is revealed, but not its origin, and it is not stopped. This would be important info, (that your enemy is using spies, that you might consider counterspy efforts …), but would not affect diplomacy or the spy itself.

2. the spy and it’s origin is revealed. This would affect diplomacy (could start a war …), the spy’s present activity is stopped, but it is not automatically killed. It would be visible for all enemy units for a certain number of turns and would have to defend itself (weak in combat). If there are no enemy units around, it could escape out into the wilderness and try elsewhere later once invisible again. This would also create action and suspense since it would often lead to panic chases through the wilderness, and to panic battles between spies and enemy units you would prefer to avoid but might just have the chance and luck to survive!

3. the spy is caught and it is executed by the populace / military. Affects diplomacy as above and the spy is lost.

 

You can order one of your own spies in one of your own cities to search for enemy spies. This would increase you level of counterespionage in that city (quite a lot in the city, and a little bit in the realm as a whole).

    

There would be spells against spies: like True seeing revealing all of them at least at that moment, Scrying for them in a particular city or region, Aerial/Spirit protectors drastically increasing your counterespionage in a city, Talking trees: any spy moving through this forest region is automatically revealed, … There would be spells for spying and aiding spies, like: invisible stalker: much harder to catch and with assassination as special ability, aerial servants able to fly quickly to any city to spy there, invisibility potion letting one of your spies avoid getting caught once …  Hero spies could have drastically improved versions of the basic abilities, like sowing discontent with much higher chance of rioting, or stealing spells or possibly even essence.  

 

There would be technologies for and against spying: these would be needed to perform any “active actions” like poisoning wells or sowing discontent. By investing in counterespionage techs like local anti-spy militias, citizen networks etc you could create a rather spy-safe realm except maybe from the most fearsome master spies aided by magic.

 

Some factions would have natural spying or counterspying bonuses or abilities (+ or -).

 

There would be no double agents, since this would be too complicated to handle in my view.

 

I really hope Stardock would consider including such as system, since the game would be so much poorer without it.

As long as the options are there more or less as described I wouldn't care if it is a fully "separate" system or if these were standard abilities for heroes. But I don't really see how you would achieve the same thing with just individual abilities you give to any hero. For example, would it be an ability to move invisibly through enemy territory? And if you don't have it, could you then use any of the other actions ...? If there are just a lot of abilities, what "spy-hero" could the enemy detect? And you would still need the spying and antispying techs etc. In the end it might work the same, just broken down to each ability, but to me that actually seems much more coplex to handle while not giving much added value (more flexibility yes, but I think the system a described could already be made rather flexible). 

Sure. You could certainly create spy-only abilities which would make them unique and interesting units.

That isn't really my main gripe, anyway. I would rather have a system where spies aren't represented by units on the gameboard.

zig, I'm not saying that it's a completely deterministic system: just as long as the RNG creates exceptions as opposed to norms, it's OK.

At this point I'm all for abstraction, at least at first. I'd much rather set a slider and forget it than spend a part of every turn screwing with spies, counter spies, training new spies, retasking spies, ad nauseum. Especially for the first parts of the beta just screw the espionage and come back to it later. Any espionage system is going to please only a fraction of the users, piss off a similar size fraction, and leave the rest ambivalent at best, resenting the time sink at worst.

If a player is spending more that 1% of their turn time on espionage, scrap your system and rethink from the beginning.

Scoutdog, okay, then we agree with each other on that point.    

Goodmorning all

Regarding distance calculations, a mission in a faction at the other end of the globe aught take longer, if your spy isn't already inplace. Actually moving your spy from town to town, or while in the faction tracking WHERE the spy is would be unnessisary, but having tw factions, one right beside you and one at the far side both take 5 turns to do the same (very simple) mission... seams odd.

I would change the two slightly.

When you select a mission you get a # of turns, shown to you, it's an estimate.   The game has a real # of turns, based on the spy and the faction and the mission, and pressents of counter spys. 

Each turn the spy is actively doing it's mission you add or remove 1 or more real turns remaining,  A very bad roll adds many days, and good roll removes 2 or 3 days, most rolls remove 1.  You then have a second roll, very high probabilty nothing, small chance to make a report  (gives you some information collected, and reports how many days left till compleation[real # +/- some]), a small chance to increase susption (but the spy doesn't know that the've been detected). a slightly higher chance of increasing susption and knowing it.   Most spywork as i see it has little or no risk to injury or death, except being detected and punished which only happens when your suspition rate increases beyond a theashold no 1 bad roll can end a mission. many bad rolls, and or poor skills are needed to end a mission.  Thus direct rolls to die/live are in my eyes too abrupt. Also at any given turn you can request a report, and this would roll twice (because the unit forced to make a report rather then waiting for a perfect window to make an undetected report).

This means that the RNG doesn't just occur once at the end, and that's it, yes or no. It happens many many times, (therefore results are much more like those of battles, so many RNG calls that it all averages out in an acceptable way.)

Most of your post would be better applied to general stealth, spys as board and combat units just doesn't add anything in the way of game play. I think if you want stealth units, Garrilla units are the way to go, but they are not spys.  Spys are diplomatic units with alternate modivations, or implanted units with outside loyalties.  Most of the system your suggesting lacks subtefuge. I want a spy system that is dark and nueanced, rather then just another military unit. 

Also how does your system mittigate the RNG abuse, or frustration. All your spy actions appear punctual, one roll to succeed or fail, one role to be detected. it just sounds like a painful system to me.  althought in gerneal i think spells to aid spy rolls are a good idea, and would have to be thought out. But a mixed system is almost nessisary.

I like this idea...

A suggestion I would like to add, if your spy thinks he might be detected, you should have the option to abort the mission and recall the spy. That way if you do have a serie of bad rolls, you still have a chance to save the spy. Off course, recalling the spy might put more attention to him, raising the chance he get's picked up for a friendly chat.

If you recall the spy, all progress towards the mission is lost, so another spy should start over from scratch if you send someone else on the same mission.

Goodmorning all.

Ya, i was thinking something like that,  When a spy sends back a report it has 

turns to compleation ( say 10, real number would be between 8 and 12)
precived Level of Suspition (none,low,mid,watchful,Coverblown(trying to get out).)  The real suspition level would almost always be higher then the precived level,

But i think you get the idea

 Take care

Congrats on returning to boringville.

Goodmorning

Psy... which post was your comment dirrected at?   can you expand on what part you find boring?

thank you for the contribution.

RP's idea is (I think) less boring than just a simple slider like the early GalCiv2 spy system, a system that isn't purely RNG based, has player interaction but not too much... It doesn't sound THAT bad to me...

Let me begin by stating that I think that espionage should be part of the game,but I am not sure how/if it could be implemented as a victory condition.

Having spend the better part of an hour reading through this discussion here is my $0.02.


Perhaps spying should be broken into 2 functioning catagories.

1) The Spy network (no micro-managing, just spend money to create a spynetwork in enemy kingdoms) A spynetwork will give your kingdom access to general information about the enemy such as income, unit types available, technologies learned, spells known, resourses controled, and an updated map of the enemies citys.

2)"SuperSpys" (Micro-Managed, expensive, specialized, tideturning inividual units) Can be equiped with items and trained with single or multiple specializations such as stealth, general information gathing, battle plan gathering, money theft, tech theft, spell theft, discord sowing, well posioning, unit capture/side change, unit kill/assisination.   

Skills:

Stealth - This gives the SuperSpy unit the ability to move undetected to target city/building/army/unit etc.

Information Gathering - Basically this skill would act as a boost the the spynetwork in a particular city. Allowing your kingdom to know what resources were housed there, what units were there etc.

Battle Plan gathering - This would give you an upto date map of the enemies units and their paths (maybe based on the Spys "sphere of influence" or if the spy was housed in the unit's home city)

Money Theft - Spy can either steal lumps sums or start too funnel money.

Tech Theft - As it sounds
Spell Theft - As it sounds
Discord Sowing - As sounds
Well Posion - As sounds

Unit Capture/Side change - Superspy moves to enemy camp (unit) and causes enemy forces to change sides (and maybe not to your own if you want to be tricky and if you know enough about another kingdom)

Unit Kill - Moves to enemy camp/unit and kills selected enemy. (While this would rely on RNG it could also cause the spy to have to fight with one or more enemy units in a tactical battle making a better equiped spy more likely to survive).


On the defense/magic defense side you could have unit perceptivness abilities, city antispy buildings, and  spells such as a Ward which detects spys around citys and/or personal wards which will allow spy detection around individual units or armys.

 

You're welcome.

 

Sliders.  Boring.  If espionage is a matter of adjusting a slider, why bother?  Skip it.  If you reduce it to something as interesting as changing the tax rate in civ, you're not adding anything even resembling an entertaining feature that adds to the game.  You're including espionage just for the sake of including espionage.  You can "spy" on people with a simple stealth function, a boring budget slider isn't necessary to do surveillance.

 

Abstracted units that don't exist and go on missions with a random result.  How about we just put a roullette wheel in the game?  Abstracted units that don't exist and go on missions without a random result, um...  Why?  What is the point of adding espionage if it's not going to be much of a strategic implementation?  Getting to gamble on whether it takes more or less time is not strategy, and it's only fun if you find slot machines and other mindless games of chance to be fun.

 

Between tactical combat using custom designed armies, and some lame espionage menu or even just a slider, I don't see the draw.  All I see is a nuisance I have to waste time on.

Indeed. The only benifit I can see to a slider system is that it miiiiiight make it a TINY bit easier to mod in a good workable espionage system..... but since it would also probably set the loud people off, why bother?