Well, the main problem with a regular-unit-based system is that it ties espionage to military power: you need to be able to recruit, train, and attack people using your military abilities. What I want is a system where espionage is a valid gameplay style in its own right, independant from military power. (A style, though, that you can turn off if you don't want to deal with it).

I think it would be more fun to have my cake and eat it too.

psychoak has another thrice-damned good point when he basically asks why so many TBS games have semi-buried espionage in underwhelming unit-based systems. Faced with that good crit, I find that I'd actually rather see espionage methods/skills as traits that could be attached to both champion and regular units, and leaving what those units are called called to the story layer for the given game.

Abstracted intelligence networks are also a fun and reasonable idea, and the spymaster's desk interface is just not a main map thing. And, as Denryu notes, in a game like Elemental, there should most definitely be a Third Way that is all about spell research and casting.

As I have understood there will be no premade combat units in Elemental but instead up to the player to train them as wished. With that in mind I feel the best spy solution would be to keep at least the most advanced spying abilities to only heroes + special magically conjured beings (e.g. invisible stalker: a perfect spy and assassin). Personally I would not like to mix the spying with any military unit, because the abilities are so different if Elemental turns out to treat invading armies like usual in such a game. Normally you can't enter enemy lands. If you go with a military unit onto an opponent's (at least enemy) city you actually attack it ... (e.g. GC2). Under such conditions it makes much less sense that a whole army would also have the ability to sneak around and assassinate the town governor. For example, how would a heavy cavalry with your coulourful banner held high sneak into a city, stay hidden and then backstab a ruler inside a house with a poisonous dagger? ... It would be too complex to try and make everything fit. Scouting though is another thing that should be for everyone.

Pigeonpigeon was writing that spying easily gets on your nerves because you get backstabbed yourself, you have your cities rioting etc. I would say it is much like another dimension of the tactical combat: understandably the player doesn't like to lose a city or hero or resource, but at the same time you have the chance to steal it from your opponent which is all the nicer. And there is often an even greater pleasure from managing to really hurt your opponent by spying (say by assassination and rioting) then by traditional military feats. We have all experienced the military victory a zillion times. If the spaying allows active actions and an element of chance it can be as rewarding as the military system.   

It must also be said that the system would be optional. And if you then choose to use it, the player is taking a calculated risk if not spending any thoughts and resources on protection against such attacks and problems. So it would not be unfair and "a bolt out of the blue" if a leader got assassinated.

As regards that it would be too much micro-managament, I think it would be a problem only if the system is not well done and in particular if the results do not matter for winning the game. The abilities must be important enough that you would HAVE to take it into account or risk serious consequences should your enemy try this angle of attack. This was one of the failures of GC2 were you could not do much, just spend money passively.

The beta should provide ample opportunities to tweak such a system into something people would like (i.e. those who would consider playing with the spy option, the others could still play just like they prefer).  

I want Elemental to work well for a scary-large range of playstyles, and I'm convinved that a strong espionage component will be a big help there.

Skullduggery is a major part of many fantasy epics, so it seems to me that the game ought to have an option to include exactly the sort of paranoia that you don't like so folks who *do* like looking over their shoulder every turn can have their fun too. I'd probably find switching that sort of stuff on and off giving me much more replayability than targeting different victory conditions.

I hope I haven't come out as "anti-spying". That's not the case at all. If there is a system that really makes it feel immersive, like real spying then I think that could definitely be a good thing.

I am just against paying a certain amount of coin per turn to maintain some nebulous "spy network" and then paying the again for an "anti-spying network". Even if they didn't appear on the main map, if you can give spies specific missions with varying chances of success, costs, and possible benefits - I would love to see something like that done well.

Goodmorning all,

Firstly, I to am against the pay per turn for attacks and pay per turn for defence methods, and i hope my suggestion didn't come across as being that method.


I think one benifit to my suggested method is that it is quite compatible with Magical Servents and scrying.

Most of my system involves spy units being found in your faction, and them being given specific tasks on the global level.  There is no reason why an Air elemental couldn't be crafted with basic spy abilities, or semi-animated statues, given as gifts...   Magically created units can easily be incoperated, the better the resulting spy the more versitile the resulting spy the more the magic cost...

I agree the Paranoia funcitonality could prove a turnoff if allowed to be too powerful, But consider the costs for excicuting a paranoia worthy attack,  Get a very good spy, pay extra to train them to take a place in the opposing faction,  Risk loosing the spy for nothing, risk the spy getting caught every time you call on it for information, (including location and recennt travles on the overland map, you wouldn't see that unless you requested it.) you would Get very little from the unit for a Very long time as the unit rose up the ranks (or get some, but risk of being caught.)  You would also risk the Spy turning double agent on you, because they learn to truely respect the faction you sent them to.   And you risk being lied to when you do call for information (because the enemy may know about your spy and feed you false information).  Still if all those deterents still leve you with a looking over your shoulder fealing you don't like... an off button can always be included. 

I think having a 'spy unit' on the map would be a mistake,  Having an hidden assiasin gorrilia unit that stelthed around, SURE!!  But have that be prat of the WAR and Millitary tree,  not the spy/war/diplomatic tree.


In short, I see spy units as individuals you give missions to, you pay them you train them, and they go away and come back reporting on thier successes/failours, controled on the global scale.  The more interesting/powerful missions would take many turns to complete. But supplimenting this is a wide array of minor spying missions to learn basic informations.  Subtifuge and misdirection, Politics and spygames.  This is what i hope for.

One thing i really want to see is diplomacy taking several turns, and somebody has a spy they can risk the spy to -tweak- or sabatage the meeting.

Take care all,

Robbie Price

I'm all for an awesome spy mechanic. Being able to send someone in to another kingdom and gather information about it would be wonderful. What I am opposed to, unless Stardock comes up with something truly revolutionary that makes all the things I hate about subterfuge in TBS games go away, is actual subterfuge, sabotage, assassination. 

Sure, doing those things against the enemy can be fun... Except it's usually like hitting the slot machines. You succeed or you don't, your skill has nothing to do with it. The dice roll and the barracks burn down or your spy gets caught and executed... Or sometimes manages to escape. Even worse, though, is when it happens to you. An assassin sneaks through your intelligence net and manages to kill your hero - did you have an opportunity to step in and prevent the assassin? No, not really. If the assassin gets close enough, and gets a lucky roll, boom. Dead hero. No skill involved, other than in sneaking the assassin in. But even that is often a matter of luck - in every system I've seen that isn't abstracted away to oblivion, sneaking in is a matter of avoiding enemy scouts/spies/whatever - but those are often invisible to you and you don't know where they are until you can both see each other... It's just so.. ugh. High stakes, Luck O' The Draw mechanics don't interest me even one bit.

If Stardock can think up a system that doesn't suffer from those flaws (which I think are fundamental to the concept, really), I'll be happy to support it, though.

If the spy system is well done, based on training your spies in various abilities/skills (just like you train a warrior), there is not much difference when it comes to "luck" and "bad luck" compared with the usual combat system. For a fighting unit you train it and it goes out in the fog on the map, meets another unit and dies are rolled ... One is killed and one survives (much depending on your earlier military investments). The only difference with spies is that per definition they try to stay hidden, close to you. But there is the counter spy defense. The best game solution is that when you have such a defense (various levels and types could be built up) you have a chance every turn to get to know that there is a spy in your realm, or the precise city. Or if you have a really good network and luck you would actually be able to catch the spy directly. In this way, you could learn in say turn no. 14 that there is a spy in your capital. You would then be under pressure to decide what to do: try to catch him by developing better protection in the capital, or take a calculated risk and wait ... He might try several missions, but not succeed, at the same time as you reinforce your defenses. And in turn no. 20 you might finally catch him (and feel more on the safe side) or have say a local leader killed by him, depending on not just chance but chance in combination with your abilities and your opponent's. Then it doesn't feel like you are powerless to stop e.g. an assination or sabotage. Instead, it becomes a real story.

After all, often in military combat you have no way of preparing for or knowing everything about the opponents' strength, especially when magic is involved and conjured creatures can strike unexpectedly, etc. For me, that's part of the fun. A certain level of paranoia is good for a strategy game, if the system is just well thought out and handled. So I would strongly condone a well-crafted spy system with real tactical and strategic choices for both sides. Skill should definately be part of the spy system as you suggest. And I would agree with above, against a "pay twice" system, where you just pay once for "spying" and again for "counterspying" without active choices or skills.

No, that's completely false. Yes, ultimately when two soldiers fight in combat their fates are determined by dice roles. But how they got there is totally different. They function as a tiny part of the whole. When you engage in combat, you tell your troops where to go, how to go there, and how to act. Will they be used as meat shields? The backbone of your army? Will you send troops around to flank your opponent? The is strategy and player input in how combat plays out and how troops die. The enemy can do something unexpected, but unexpected is a very different thing from random dice rolls. The same is not true for spies. The only relevant factor tends to be the spy's level, and your luck with the virtual dice.

And like I said, I want there to be spies. I want to be able to gather information and all that and all those mechanics you mentioned above would work wonderfully or spies. I'm only opposed to saboteurs and assassins. If I am forced to constantly comb my kingdom (aka micromanagement) with scouts and my own spies every turn to make sure no one dangerous is sneaking in, I will be angry and bored by it. Alternatively if they abstract that away so that I don't, then it turns into a "pour more money into espionage to prevent enemy scouts from entering." The reason why this doesn't bother me for spying (gathering information) is because just spying isn't quite as high stakes as the other elements of subterfuge, nor does it cause the same annoying administrative requirements.

Well, if we were to make spying optional, which it would pretty much have to be), there's no reason not to make sabotage optional (seperately) as well....... I'm currently trying to think of a non-dice-roll system...... Perhaps something could be done with the people-as-a-resource idea..... all depends on how you control your population, and how much....

One idea on how spying could be a bit more fun.

1. I agree they should not be able to assasinate or sabotage.

2. The only defense against spies should be magical in nature, or possibly a default amount of anti-espionage that costs nothing. Needing to sink money into anti espionage is a not fun option.

3. Spies can steal tech (spy owner gets a tech), report on spells known by the person being spied on (each spell could have a saving throw against being reported), join an army (spy owner sees army composition and "sees" from the army stack as if it were his own unit (he has no control of course, but just clears fog of war, etc as if it were his) - maybe some other abilities along those lines. Spies should be fairly expensive, but also highly successful at what they do. Once a spy is assigned to a task, that is the only task he can do until he is detected. i.e. if he steals a tech, that spy is effectively used. Same with the spell report - it is a one time static snapshot. The join army option is an ongoing ability that continues until the spy is detected, at which point he is gone. Maybe there could be other things he could do, but I think sabotage or assassination is too powerful.

Now, this join army thing is interesting... what if the spy went into battle against you or your allies? Sooooooo many possibilities....... One thing I thought of is since people are a resource acquired largely through immigration, you could have spies be special population units that the victem cannot easily see as spies.... you could have an enemy city where all the people are spies for you! (I think I saw an episode of "Mission Impossible" like that once....)

If they used the ability to join an army, I don't think you would want to complicate it with "what if the spy fought in a battle against me? Just treat the army as a normal stack as far as the victim goes, and the spy is giving you the info of what units and how many are in the army, as well as giving you visibility from that enemy stack. (during your turn you could inspect the stack, and it would clear fog of war for you. At least that was the concept I was thinking of.

Well, in realtiy the spy would probably desert, and considering their stealth skills they could probably get back with little trouble..... probably wouldn't affect the unit much, so yeah, ignoring the spy would probably be the best option...... although a few feautres like "backstab", "begin rout", and so on might be fun.

No, that is not correct. Like stated earlier, you can first train your spy much like a soldier, then decide what to do with him. You could send him to the most heavily protected place against spies (e.g. enemy capital) which would be very dangerous. Much like the risk in attacking a powerful army. Or you could stick to easy targets, like small towns on the outskirts of your opponent's realm, which would mean you have a much higher chance of succeding - much like engaging a small army you know you can defend. Of course the combat and the spy system would never be identical in level of detail, and the combat would always be more developed (after all it is not a game about spying), but there is no reason to critize a spy system including e.g. assasinations for automatically being only about chance and without tactics and strategy. We don't even know anything about the system, or even if there will be one at all!... Another thing is that the combat viewer will be optional as far as I understand, so you might well just let chance play out also all military battles. I understand your reluctance to include other actions than info gathering but that's a personal preference, not a problem with all spy systems. After all, it will depend on how the system is actually made - in principle it could be extremely detailed, in fact as detailed as the comabt system itself altough I wouldn't recommend that (just to show that it's not an inherent system problem). 

In sum you can't say it is only about "the spy's level" and "the dice", since it would be easy to include 1. other abilities, 2. magic and 3. tactical choices like movement, evasive actions etc to go in the opposite direction.  

I will say I do not like the idea of an espionage victory condition. I would need to see the final implementation but as a general idea I do not like that (even if there is an espionage system in the game, I can't see how you could have a good victory condtion based on it.)

I suspect that part of the 'problem' here is the term victory condition. Some unknown share (future poll Q?) of players don't play to win, they just play, and one thing that can make a game more interesting is a variety of ways to bring it to a close. Sure, cobbling together an 'espionage win' could be an odd fit in a traditional framework with a dominant conquest-win POV.

But if you put on a role-playing hat for a moment, becoming 'the Arch-spymaster' can be just as satisfying a way to call a game over and the mechanics could be as simple as maintaining a large enough network for a fixed span, like the 10 turns you need to hold overwhelming influence in GC2 before the game is over. The thing only turns into a 'problem' if you insist on linking it to some kind of scoring/ranking system, and we still have no idea if Elemental will even include scoring. (I'm ignoring competitive multiplayer here, mostly because I don't grok it. Seems plausible to simply limit MP to the 'traditional' closure conditions that fit with a competitive viewpoint.)

Indeed. When I came up with the idea, I was thinking a lot of the Romulan Tal Shiar: when was the last time they were challanged in their supremacy!

Frequently, by the Obsidian order, not to mention Spock, who was secretely living among the Romulans...

Luck based irritants are not the same simply because a soldier also falls prey to luck.

When you have a battle between two armies, your skill guides the overall outcome.  You can protect important squads, sacrifice less important squads to do so.  In a game like Total War, your survivors gain experience and become stronger.  Luck isn't what determines your outcome, skill is.  The randomization serves to make things less uniform, not to remove predictability.  You have hundreds or thousands of units, all taking chances, it averages out very nicely.  Perhaps, while taking a risk, you lose your general or a family member and get boned by luck.  Typically, it's a minor factor.

When you have a spy, that has to do a hundred missions to become James Bond and take on entire empires, it's not the same at all.  You suceed or fail, on one bloody roll.  You then move onto the next mission where you repeat the process.  Your army succeeds or fails on many rolls in each battle, perhaps thousands.  There is no skill involved in nurturing your spy, it's pure luck because there aren't enough rolls to make it be determined by anything else.  It sucks.

If you remove the abstractions, and make it a true game of skill, where you have a normal unit with normal features, and sneak it through enemy lines to maximize the predictable and logical damage that unit will do by avoiding resistance, then you have something skilled.  The complexity is limited only by the limitations put on army interactions.  If you can pillage and steal, you can pillage and steal.  Sneaking a few scouts to the enemy library to raid it doesn't require a special espionage section.

Then you get to wish in one hand and shit in the other.  Massive micromanagement is the only way you'd get targetted, complex espionage out of a spy system without making it luck deterministic.  You'll go batshit insane managing the hundreds of spies you have wandering around the map.  Chance doesn't level out with a low number, especially in streak prone RNG's.

We'd be better off with a spy minigame where you ran around in a building, dodging guards.  There isn't any good way to abstract an action that would make for an in depth, several hours long rpg, into a five second activity.

I'm with psychoak on this one. There is simply no way to put something this involved in the game without making a micromanagement nightmare on anything but the tiniest maps. It's the same argument as true tactical ground combat in GC3 - it might be fun a few times, but it shouldn't eat half your game time. Every suggestion so far will result in people quitting in frustration before the game gets out of beta. Anything but a very abstracted system is a game breaker.

Give the DL spy system 10/10 for abstractedness. The TA system might be an 8, anything lower than a 7 on that scale will result in micro hell. It just makes large games too unplayable, and one of the main design decisions forthis game was to maximize the playability of truly huge games.