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

As long as it helps to make the game better...

Now, I am quite excited about what this "convenience trick" is which apparently causes so much misery^^

The RTS is like a big game abstracting a task in the manner of a minigame. The particulars of an RTS battle (rapid construction of temporary factories spewing forth an impromptu army of local recruits that will not stick around for coming battles, all done at no actual cost since recruits are a commodity only for the duration of each battle) are about as logical as the commander Shepard (Mass Effect) joining dots and matching colours to break locks and computers.

For purposes of immersion, I make myself assume that what the player experiences is not what happens in the game-world, much like I did my best not to think that Shepard was expertly matching three quasi-tech-images from a scrolling selection.

I'm not entirely satisfied with that solution, but many wish for the RTS to function largely as it does today. Besides, I think a Larian rep said it wouldn't change. No wonder, if so; release is just around the corner, and the gameplay works, even if it doesn't tie in seamlessly with the rest of the game.

The strategy map works (although the AI might be too weak, not taking advantage of near-complete knowledge of what actions I might take on my turn, heavily defending countries I couldn't hope to attack).

The connection between roleplaying parts and strategy parts probably works. What I've seen looks good.

The idea that the main force in actual battles is not the standing army, but a spontaneously constructed one, does not work. In RTS it could pass, if one considers RTS a resolution mechanism rather than a simulation.

I'd let the autoresolve determine which strategy units fire upon whom, and what units are lost. Essentially, it wouldn't be "autoresolve", but "strategy processing". I'd let the RTS override two factors of the autoresolve: How long the battle is (how many attacks each side/active unit gets to make) and who wins. When the battle ends, the losing side retreats. If the side that won RTS runs out of units in autoresolve, it still wins the battle. Credit goes to the dragon, since the RTS battles are led by the dragon in person. (Strategy processing based on an RTS battle could actually bring up the dragon from the fourth volley or so.)

<RTS> Winning early or retreating early would aid in limiting casualties (on both sides).

<Strategy processing/Autoresolve> Having multiple units active simultaneously in strategy processing gives the larger army a real advantage, and actually processing the strategic units in a few steps makes the flow of the battle a real thing (losing units along the way isn't tailored to an initial roll from 1 to 100; every volley is fired under updated circumstances, by remaining units).

What I put here is probably not what Larian has envisioned, or what they have programmed, and it is probably not how they are going to approach the problems we have. Probably. But it is an idea, and I think it has potential to make the game more believable. I also think it could make the gameplay better, but that's far less clear, and mostly a personal preference. Also, sharing many aspects of resolution between autoresolve and RTS would hopefully eliminate some of the "which game gives me better odds" gameplay we have today. There would instead be a "With 16% presence my army will be wiped out (or survive against realistically astronomical odds), but my dragon might help them hold the ground". (They might still be wiped out, but your dragon could drive the other force away.)

All right, Elwyn. It's not conclusive, because I haven't played nearly as many RTS rounds in this latest session as I did autoresolve ones, I didn't reach the stage where I was sending lone troopers into other territories. But I did try some battles against lone troopers with larger forces. I did typically end up losing some units, but this is the important part: The units I lost were Troopers and Shamans. That's fine. Losing a Trooper and Shaman to another Trooper is well within the bounds of reason.

It's not really possible to do this test with the exact conditions of an autoresolve, because the RTS battles are now all 4 players, and autoresolves are not.

I also watched some autoresolves for other units. An Armour entered a country and was pretty over-matched, a 10% chance to win. It took out a Trooper and another Armour before dying. Again, completely within expected parameters.

As a programmer myself (not in the game industry though) I am curious about that as well.

Could be an interesting blog post...

That's similar to my favourite solution to this issue Sinister, albeit better explained and thought through. I don't realistically think it's possible at this juncture to introduce a whole new system like that, the game is releasing in a few days and it's not the kind of thing I'd expect from a post-release patch. Maybe if we got a Dragon Knight Saga style update but it still seems like a long shot.

Hopefully adding some sanity to the auto-resolution will be sufficient. If the option to join the battle yourself is limited (which it seems it is and it isn't depending on mode) then it doesn't need to mirror the RTS results that closely and some kind of balance can be maintained. At this juncture the game feels like it's approaching real greatness and every quibble seems important.

True....



No opinions on my "wild thoughts", anyone?

IMO it is an unnecessary simplification on the strategy map (and would remove much of the strategy and gameplay there), and would reduce the control you have with your starting units in RTS.

I'm afraid I have to agree, it would help solve the issue you mentioned but there would need to be a whole extra set of systems added to add depth back into the strategic map. In any event you'd need some different unit types.

Well, the classic Risk works quite well without different unit types. And it's one of the best board games of all times so it can't be THAT wrong....

But I agree that you would have to adjust some systems for that in order to integrate the new "simplistic" version into the rest of the game. But perhaps it would be worth a try on the green field, just to explore the possiblites and possible flaws and benefits.

At least there is one simple reason when designing anything: make it as simplistic as possible and only add complexity when it's needed. A simple system that works well is superior to a complex system that has its problems.....

@ Raze
I don't think that strategy would be diminished on the campaign map. Risk has a lot of strategy as well. The important thing is to know, which countries you want to defend, which you want to invade and what cards to play.
Building different units on the strategy map just for the sake of having different units without having any real and reconstructable effect for autosolving battles is a quite dumb idea imo.
I could think of a system in which you take the number of "unified, simple units" which you sent into battle and buy RTS units at the beginning of the RTS battle. For example: you have 10 strategy-map-simple-units and your enemy has 8. For these 10 unites you can know buy RTS units before the respective battle, so e.g. a trooper squad for 1 unit, a tank squad for 3 units or a bomber squad for 5 units.
This system would really improve the transition from strategy map to RTS because it would also work the other way round!!! When the battle is over your remaining troops on the RTS battlefield are measured and translated again in "unified simple units" on the strategy map (with one limitation: you can't have more units than before battle). Or your kill-death ratio is translated into simple units on the strategy map. Either way the transition would be improved and could be reconstructed....

As it is now, strategy map battles and RTS battles follow the same "combat and unit rules" but without being based on the same "basic rules" which is kind of weird. Without that there is simply no justification by design to use the same units on the strategy map AND in RTS battles. It's like measuring with two different scales by design. And in a sitution like that abstraction and differenciation might be the better way to cope with the situation...

Out of curiosity - are you guys playing the AI dominantly or against other human players ?

I hate to be rude; but there are something like 90 posts in this thread many of them long walls of text by LC and Stabbey - while I admit they are well posed and have lots of information could someone perhaps provide a summary ?

(btw I still kind of like the idea that the only troops you have are the ones you bring into battle )

Well, here you go...



Well, first of all since they announced that currently the autoresolved is bugged anyway, we might have to wait for the next patch, or need to accept that some points we discuss might be moot.
E.g. if the unit types afterwards factor into auto-resolve in a rock-paper-scissors kind of way, the composition of your army does play a role and is a strategic decision to make.

That being said, I think that such an abstracted system might indeed help with some issues.
E.g. you could erlate some factors in the RTS to the number of invading "unified troops":
A larger army would give more starting units, probably more recruits or even increase income or reduce build times/cost.

On the other hand, moving individual and distinct units (or rather squads) has some merit to it, as well. In RTS it is a difference whether you attacked with a bunch of fast hunters or some slow warlocks, as an example. Attacking with a well-composed army might even secure an outright victory against a weaker enemy, if you manage to rush him before he gets his defenses up.

Personally I'd keep the current system, but adjust it in some ways.
As you say, there is a certain dychotomy between RTS and campaign map.
It is possible to just ignore it and have fun playing the game, but for me at least a small itching in the back of my head remains.
I feel that in order to reconcile the two parts, the effect of map units on the RTS part should be increased. Since the latest beta, this can be done e.g. by increasing the recruit cost multiplier.
On top of this there might be other things that could be done:
* adjust available recruits or income rate by the size of the invading army - more troops, better support.
* make basic units available all the time, but restrict building of advanced troops (e.g. more than 10 research points) so that you can only build more if you brought them with you while invading.

Of course such changes would also mean that "turning" a battle that starts with the odds against you would be significantly harder.
Then again, I think that you should work on the strategy map to prevent something like this from happening in the first place.

Mostly against AI lately...

I tried to solve this problem in my latest comment above. In my system you would be able to "buy" units at the start of the RTS battle according to the number of "strategy map simple units" you sent to battle. So the strategy would shift but it would still be there, especially if you want to engage in RTS mode.

But if not, you wouldn't have to rely on a system which is directly "ripped off" the RTS mode and transplanted into strategy mode without ever really fitting. The rules on the campaign map would be different to the ones in RTS (which is also true now but with - which is weird - exactly the same units as in RTS) and there would be a clear difference between the modes with a better chance to create a reasonable transtion...

AI.
Sometimes 2vs2, sometimes everyone for himself.
2vs2 tends to lessen some issues since the RTS-maps are clearly made for that.

The geek would disagree that Risk is one of the best board games of all time...

http://boardgamegeek.com/boardgame/181/risk

Beyond that though I'd agree with you that it would be an idea worth considering and experimenting with, except that we're 6 days from release and going back to the fundamentals isn't really an option.

With the system as it stands you've got a lot of interesting decisions in what units to build on the campaign map. Troopers are cheap and effective, Hunters are fast and enable you to capture a lot of ground quickly. Shaman can preserve your forces so you don't lose so many in advantageous battles, Transports and bombers help you plan ambushes, Warlocks can effectively counter these manoeuvres.

It would take a lot of work redesigning the systems in such a way you still had a lot of interesting decisions. At this stage I think Larian are closer to fixing the existing system than would warrant stripping things out to that degree. In practical terms it's an impossibility anyway. If you were talking about Dragon Commander 2 I think it would have more merit.

Oh and I'm typically playing vs AI, my dexterity is nowhere near the level that I could play an RTS online and I wouldn't be able to fit in a whole campaign anyway.

LordCrash;
I was thinking there are a lot of cards that would be impacted, etc, by the change. It is true some of the depth could be added back again (perhaps not in a week, though).
Risk has a social aspect that wouldn't be present playing DC against AI or in single player, and... and I haven't played Risk in about 20 years, so can't really comment on the strategy.


Lar;
AI exclusively. Maybe if I eventually become competent at RTS I'll find someone with one arm in a cast to try multi-player with.


meme;

- troopers (or other weak units) can sometimes do much more damage in an auto-resolved battle than would be expected

- that's because there is a simulation of an RTS combat, so it isn't a straight match-up

- but they can do too much damage

- just like can happen in an RTS

- the auto-resolve should use different rules than an RTS battle, though

- no, it shouldn't

(insert a couple theories, conjecture and proposals, and repeat)

Lar - actually, that's a bug (an unspecified shortcut that presumably was fine when implemented, but at some point started causing unintended results)

I play mostly against the AI. Usually in Free-for-all, not team games.

In part because there are rarely any open games, and games I do find often have repeated "waiting for player..." lag timers and disconnects. I don't think it's the netcode, I have played on servers hosted in europe with long or non-existent lag-free stretches. It's probably the fault of the ISP's. I missed the last hour of the Review Code twitch stream because it was raining (and it was still down when the rain stopped).

I really should make more of an attempt to play online. I got embarrassed in my last match because it was so long between multiplayer matches that I forgot I should make Grenadiers first against humans and not Troopers.

Lord Crash, I'm not a fan of removing unit types from the strategy map at all in any way whatsoever. Removing all differentiation from Strategy map units would diminish strategy and diminish the game. When carving a statue, you don't reach for a jackhammer when a chisel will do.

This issue doesn't need a drastic reworking of the entire mechanics, it just needs tweaking so that the player with a vastly larger army isn't consistently punished for auto-resolving when facing a lone trooper and such.

Thanks Raze. So it is a bug that will be fixed !