Larian Studios Forums Dragon Commander Dragon Commander - Beta Lone Trooper suspiciously good in autoresolve.

Sorry, but I do not get it how it should eliminate the problem. Send a lonely ninja trooper with a 0 % chance to win, choose to enter the battle instead of autoresolve, produce hordes of units on the RTS map - and your trooper (and all the support units produced) will kill kind a few enemy units before you are defeated (if at all). At the end, you will get the SAME result screen from which it seems that the lonely trooper has ninjad quite a few more powerfull units.



Absoulutely agreed.



Agreed as well.




Basically, you are right. However, you seem to oversee one important point. If you have a large chance to win the battle (say 80 %), than your enemy has necessarily a small chance to win the battle (20 %). So, if you are playing againts a human opponent, he would very probably like to fight this battle. And once the RTS is in progress, you will very likely suffer the same losses as you do if you now autoresolve this battle.



Absolutely agreed.




Why should different rules apply? I mean, it is still the same battle - no matter whether you choose to fight it or autoresolve? There is NO difference between campaign map battles and RTS battles - it is just the way how you choose to fight it. It is a bad balancing, I agree, but what is the point of eliminating the bad balancing for autoresolve while it is still present if you fight the battle yourself?

EDIT: What I mean is the following. Let us assume that we adjust the autoresolve in such a way how you suggest. So, you do not loose any troops and almost always win if you have an 80 % winning chance. Now, let us take the viewpoint of your human opponent. He has a winning chance of just 20 % and would loose if the battle is autoresolved how you suggest. Of course, your enemy is not very keen on such a prospect. Well, the most obvious choice for him is to select the dragon form. Then, the RTS battle begins and you are very likely to loose the same troops as by autoresolve which is implemented now (even if you win the battle at the end). And, I am pretty sure that this scenario will happen EVERY time there is a 80/20 percent battle. So, adjusting the autoresolve in the way you suggest is just to make it unfair for the player with a small winning chance and will force him to always fight the RTS battle.

EDIT NO 2: Adjusting autoresolve without changing the underlying RTS mechanics just does not make sense. Player who is unsatisfied with the autoresolve result (if it is decoupled from RTS as you suggest) can just force the other player to fight the RTS battle. And the result after this RTS battle will be pretty much the same as it is now with autoresolve.

I think EVERY battle has to be solved in RTS when playing against a human player???

So the whole thing is only important when you play against AI.

Well, the percentages then would have to be adjusted of course, but yes, I agree that this might just relocate the problem somewhere else.
Either one of the two ways to resolve battles might lead to exploits.

I wonder what might happen if building units in RTS would actually cost gold, just as when you buy it on the strategy map?

No, that is unfortunately not the case. In multiplayer each player has the possibility to decide whether he wants to fight or to autoresolve. And even if there is at least ONE person who wants to play as a dragon, all the other players are forced to enter the battle as a dragon. That is why adjusting autoresolve as you suggest would probably be unreasonable. If it does not mirror the actual outcome of an RTS battle, it just won't be used in multiplayer. That is because the person who has a disadvantage due to autoresolve will just probably almost always force the others to play the RTS fight. (See EDIT NO 2 in my previous post )

In the advanced options it is possible to limit the number of RTS fights to one per player per turn, so if there are multiple battles you have to choose which do yourself.

So I should be happy that having an overwhelming advantage only ended up costing me, the victor of the battle, 4.67 times the gold cost of the single unit?

(Assuming a 3 gold territory) It'll take at least 5 turns to get enough gold from that territory to replace the Hunter and Armour that the 3-gold Trooper killed. The enemy can replace the Trooper in a single turn with a single 3 Gold territory and do it again.

That does not exactly seem worth dancing a jig over.




The problem isn't that the lone Trooper strategy will single-handedly win the war. (Although it probably can, eventually.)

The problem is that 7 times out of 9, sending a lone trooper versus an army is a worthwhile venture, and only 2 times out of 9 is the lone trooper destroyed harmlessly. At least half the time or more, sending a lone Trooper against a vastly more powerful army SHOULD be a waste of resources for the side sending the trooper. The side who has the overwhelming advantage should not be punished with disproportionate losses 7 times out of 9.

If I am at a 97% chance to lose, facing 176 Gold worth of units with 3 gold worth of units, I think that even a 3% chance of victory is 3% too much.

I like the RTS mode, I really do. I do not want to be punished for not wanting to play the RTS mode for literally every single battle.




Yes, yes it is. Ground units without anti-air cannot attack air units. Hunters and Armours can attack the Trooper, and Hunters are a counter to Troopers.




I'm confused. I thought that the whole idea behind the system was that you should be taking unit types, upgrades and counters into account. You should be building a balanced force.

Man, this stuff gives me a headache.....

ATM I don't see any good way to solve this problem tbh. There is just no good solution (at least I haven't found one yet) to the problem of the transition from "static" campaign map to "dynamic" RTS battle. Other than with Total War they just don't "fit".

Even from a design persepctive they don't fit. Think about that. On the campaign map you cannot build units everywhere, only in countries with a capital or war factory. But as soon as you go to RTS battle you can - oh suprise - suddenly build unit in this country regardless if there is a war factory or not. That's in itself inconsistent design. And it makes the production of units on the campaign map kind of "useless" because their weight against the units created in RTS is way minor.

This is a serious flaw of game design I haven't noticed before tbh. But now I fully understand some problems I had on the campaign map.

@Stabbey, you have written a few posts above that you tested a lonely trooper against an enemy army by doing autoresolve. Could you please make the same series of tests, but just enter those fights as a dragon now? Write down the losses of your enemy, average them and compare them to your test results with autoresolve.

If you find, that there is a huge discrepency between these two approaches (i.e., your enemy's losses against your lonely trooper are much smaller in the RTS approach compared to autoresolve, or vice versa), then there is indeed an issue with autoresolve. If, in contrast, the results are pretty much the same, then it is how autoresolve should work.

Well, the inconsistency in that you can suddenly generate troops in the RTS portion is pretty much insoluble. The relative ineffectiveness of units on the campaign map is easier to handle, though I fear not something that can be fixed before release.

The simplest method is to upweight units on the field. If a hunter unit translates to 5 units rather than 2 a decent sized force will easily stomp the enemy before he can build an army up. As a result losses will be much reduced and the autoresolve numbers can be shifted accordingly.

The other solution is to de-emphasise the impact the RTS battle has on the campaign map. If I send a trooper against a large force and actually win the battle I should still only do minor damage to his army, the rest should be forced into the nearest friendly square. Only if I manage to surround his forces should they be wiped out. Of course if I sent a similarly sized army of my own then a strong showing in the RTS map could wipe out his army at a cost of significant casualties on my own force.

I will see what I can do.

But that's apples and oranges. My problem has nothing to do with the balance of an actual RTS match. My problem is with the balance on the STRATEGY MAP.

Troopers are overvalued on the STRATEGY MAP. In the Autoresolve, for a series of 9 matches in which one sides entire forces was only 1 Trooper (a 3 Gold unit), the enemy took the following losses in gold: 14, 14, 9, 10, 0, 10, 14, 0, 8. That's an average of 8.78 Gold destroyed for each 3 gold Trooper destroyed.

And it's not a bunch of light units that are get destroyed first, if it was a Trooper taking out mixes of Warlocks, Grenadiers, Shamans, other Troopers, that would be easier to understand. They're taking out heavier, more expensive units.

My problem is how Troopers cost 3 gold to make on the Strategy map, but in autoresolve, they seem have the combat effectiveness of a 9 gold unit. It has nothing to do with how an RTS battle plays out.

That is in fact the same problem. There is no difference between a Strategy Map battle and the RTS battle during the campaign (I do NOT talk about the stand-alone skirmish mode): It is always the same battle, but you can only decide whether you would like to fight it yourself or autoresolve it. Autoresolve should be adjusted in such a way that it mirrors the outcome of the underlying RTS battle (if it is actually fought). For an average player the results on the screen which you see at the end of the battle should be roughly the same - no matter if autoresolved or not.



Yes, I completely see your point. My argumentation is, however, the following: Even if you fight an RTS battle where one trooper faces a huge army as starting units, on the screen which you get at the end, the huge army will very likely have quite substantial losses (because the lonely trooper gets support from units produced during RTS), so this is where the 3 gold versus 9 gold problematic which you mentioned comes from. A good autoresolve should take into account this mechanics - that is why you see the ninja trooper duing autoresolve.

There is actually a flaw in the system behind the auto-resolve and we'll be working on it tomorrow. It remains a system of chance but the odds you're seeing for the moment are not by design. I lost 4 juggernauts to a grenadier yesterday, and I can assure you that nowhere in the specs a juggernaut loses from a grenadier

I fear so as well. And I even don't see a "perfect" solution for the long run yet....


Yeah, but that would probably cause serious troube for balance in RTS combat. What's the point of creating units on the field and conquering building spots if you can wipe out your enemy with your starting units in two or three minutes?


I have to say I like this suggestion somehow. You would still have to find a good mathematical equation how many units survive such a battle but it's a good idea which should be further explored. But there would be still the weird inconsistency that I won an RTS battle with perhaps a huge remaining army (if I'm THAT good) but only one single trooper remains on the map and the enemy army got shattered over the neighboring countries.....

It's that weird transition between strategy map and RTS I have most problems with....they seem to be deeply connected but in fact, they can't be "really" connected. You could translate pure mathematical outcomes from an RTS battle to the state on the strategy map but it doesn't really fit. The units are still different. You lost your troopers early in the RTS map and in the end you won the match with a huge army of bombers? Well, on the strategy map there will still be only on single trooper after the battle.... It's like two things which can't be combined properly and which are nevertheless pressed together. But I don't see a proper solution o address that problem...

Ok, good to hear. I'm curious with what you'll come up tommorrow....

Four pages of fairly heated arguments and Swen comes in with "Oh yeah, that's bugged." You have to laugh.

(One of my Autoresolves today had 2 troopers [one from Entrenchment] take out a [Mercenary] Juggernaut, but I left it out of the discussions because technically there were two Troopers.)

Kind of a moot point, but... to an extent, it should be better to build a balanced force on the strategy map, at least in your capital or countries you want to defend. The map might get a little cluttered if you have to have a dozen different units in any country near your front lines, and it would be a little tedious expanding your territory if you took the current amount of drag and dropping required and multiplied that by at least 4 or 5.

Of course that also assumes the auto-resolve would be separate from the RTS mechanic, which would produce a separate system that needed to be balanced, but since the auto-resolve and RTS use the same units, any changes to one system could cause issues with the other. You would also have situations where the same starting conditions would have significantly different results depending on whether you engage in RTS or auto-resolve.

Well, at least Swen recognized that a central element of the game is bugged before the got the review copy to the press......wait, no... :P

If you want, we'll leave it like this

I actually thought this particular debate was quite interesting - you guys made a lot of good points and to be fair, only today did we find the time to look into it. At which point we realized that an old "convenience trick" is now causing quite a lot of misery. To my juggernauts in chapter 3 for instance, which I absolutely need if I want to have a chance of winning!. Frustration is necessary from time to time to create gameplay, but not at this level

Hm, let me go crazy for a minute....

Perhaps the system could be fairly improved by removing all the different unit types from the stategy maps and make a real "Risk-like" approach? So you only produce troopers or "single units X" instead of the units which are translated from the RTS battles to the strategy map.

In that case your "single units X" would only measure the strength of your army and would be a real abstraction from the units used in the RTS battles. It would cleary simplify the autoresolve issues and reduce the transtion problems between RTS and strategy map. According to the strength of your "single units X" in numbers on the strategy map you would get a mixed bag of units in the according RTS battle depending on the map you plan and the tech level you have.

I always thought it a bit strange to use the same units on the strategy map and the RTS map because they depend on different rules (former can only be built in the capital and in countries with war factories, latter can be built in every country in RTS battle). Without a really "hard-working-stone-paper-scissor" system on the strategy map these different units don't even make much sense. The "one trooper to rule them all" problem is only the most simple thing you notice that there is a flaw in the basic system...If your system is based on probabilities and some random element so many different units are just somehow pointless. They make sense in RTS when you control them directly and use their abilites to counter enemy units. But without that direct control they are just "an unnecessary complex system" on the strategy map....

Just some wild thinking but I begin to like the idea. But maybe it's just me....;)

He, yeah.

Somehow I was wondering if it would boil down to this.
Anyway, let's see how it turns out.

Especially how autoresolve then compares to RTS - luckily since the various advanced settings are working now I can balance/adjust the RTS part somewhat to my liking.