Espionage in DA

It's simply a matter of accumulating time with spies placed on AI planets. The more turns you spend with spies on assignment, the higher your espionage level will get and the more likely you'll be to steal techs.

OK makes sense, the manual could have used a sentence on that feature though!

And what about planet subversion, do I have to plaster a particular planet with spies to have it riot potentially? Look logical given the first feature.

Thanks for the answer.

That's largely just a matter of sabotaging the appropriate buildings on the planet--morale, influence, etc. Sabotaging farms in particular softens planets up for invasion nicely, since it causes many of the people to starve.

That might be true, but I've played this game long enough to tell you that even with the present system of espionage in dark avatar which was a slight improvement, there needs to be a whole new system and a better one.

For example, in the previous example of putting spies on farms, you have to put them and keep them there for a long time. Again on other buildings that produce morale or influence, you always constantly have to have a spy on them until they finally achieve some kind of a result of maybe a riot. Remember, if the planet's morale is very low, the AI will decrease the planet's taxes just so it dosn't go into a riot. In turn this will affect their economy, so if you look at the big picture, it always comes back to economy and your spies aren't really affective because they get nullified too easily and meanwhile you end up spending more money on spies where you could of kept that money to persue a better strategy?! One good thing about it is if you concentrate on one planet and constantly put spies on different buildings and invade that planet so that your spies won't go to waste. At least you have conquered a disadvantaged planet with inferior bonus's. I guess it might help with the invasion in some ways.

Actually the spy on a farm only needs to be there for a single turn. You will see the planet's population drop down immediately. You can plant the spy the turn before your transport gets there. Not only does this drop the AI's income but it makes it easier to take in an invasion, just think about a maximum of 6000 enemy tropps (except on capital planets).

Your spys only get nullified if the AI alreday has one waiting. As much as it costs you to get more spys it costs the AI the same. So who has the stronger economy to get more, faster is he key. BAsically i use them on the 3 capital improvments and on some wonders. But im not sure if that does anything.

Does anyone know if placing a spy on, say, the resturant will reduce the AI's influence? or, can i place a spy on the Universal translator a turn before doing some trading to get better trades? I know it lets you plant them, but does it have the effect of nullifing that improvement?

Great ideas!

It should be visible with the restaurant. Do your borders change?


I want to ad that spying is particularly efficient if you coordinate it with the computer. I just played vs 2 races for a very long time and their spies kept adding up on me. I didn't have any chance to remove them.

The more spies you plant in one turn the longer it takes for the computer to remove them (unless it has a lot of spies in reserve which the AI doesn't seem to do.

If your opponent manages to build one spy per turn (which won't happen)
1 spy=1 spy turn -->1 turn per spy
2 spies= 3 spy turns -->1.5 turns per spy
3 spies= 6 spy turns -->2 turns per spy
4 spies = 10 turns per spy -->2.5 turns per spy
.
.
.
20 spies = 210 spy turns -->10.5 turns per spy

So spying on one civ at a time with all you got is advisable!

I'd recommend keeping a small reserve (3, 5?) of defense spies unless you already have overwhelming economic superiority. But then I'm obsessed with making sure no AI ever gets even Low intelligence on my empire.

Or just sacrifice 1 tile per planet for the counter espionage building to negate the need.

I had a feeling it was like this too, just wasn't sure but you confirmed it for me.

One thing that makes no sense to me is that without some level of espionage you can't examine/view an enemy planet. But if you want to place an agent - you somehow magically instantly know all of their improvements. How exactly does that make sense?! Knowing what's on a planet is pretty critical info, especially if one wants to invade and is considering which types of invasions to use (like whether to use some that damage structures).

It's also pretty amazing how the spies get from your planet(s) to AI planets anywhere in the galaxy instantly. I guess the "teleport ultradestructive and unbalancing secret agent" tech doesn't show up on the tech tree. You should need to have a ship somewhere near a planet (and that'd still be stretching it) to place an agent on the planet or have a trade route with the planet to place an agent. Something other than magic.

It's also (well beyond) silly to be able to wipe out a planet's population instantly by putting a spy on a farm. I can't believe this isn't a bug or unintended. You could theorically kill billions of an enemy with one spy at zero risk and with zero effort over the course of invading lots of planets. Bleh, utter cheese.

It's also kind of silly to only be able to raise your general intelligence level against an AI only by doing sabotage. It'd be nice to be able to just place an agent for intelligence gathering purposes without having to "attack" a tile and draw attention.

Overall, I don't find the DA espionage system to be all that great. I just build spies to nullify enemy spies and respond to events that involve spies, which is expensive enough. The DL system was more nebulous but fit the overall abstraction level of the game better and didn't involve elements of cheese.

Yes, I agree with all that you said here, that is why we need a re-worked espionage system. Even if some would counter-argue your statements just by saying that the time and spending that is taking for your agents to train, is in fact the time that is taking into account the damages incured including travel and sabotage. In other words, by the time you train an agent "stardock" is saying that you already earned the "cause and effects" of that particular spy unless it gets nullified.

I agree too. This is also a very capitalistic approach (well I'm one in real life, don't mistake me ) to have all features boils down to who has the best economy. One could have imagined something more fun for the creation of spy networks eg.

The espionage bonus only affects how much you can spend with spies, not their efficiency or chance of being caught. Also, the amount of cash needed to train spies builds up very quickly, so after training a dozen of spies, you would need several (more than 10) turns at maximum spending to have a single spy. Now, I recall the replies above about the issue of one-time drop of spies. It is not worth to wait a LOT of turns to train a single spy after having no one left (later in the game), drop him/her in a planet and in the next turn receive the unfortunate message: the ******* Empire neutralized your agent at ******** planet (I know this isn't the right message)? How is the formula that calculates spies' readiness and/or their soaring cost? This is an intended feature to make espionage gradually harder to prevent exploits?

I agree to that, the cost increase for spies (or the time needed to train one if you keep your espionage spending maxed out) is too steep.
I will often go for maximum spending on espionage early on (costs only a handful of bc per spy + you can take quite a bite out of the early growth of one AI), then almost stop spending until my economy is prospering, then return to an espionage spending of around 20%.