Difficulty

[axefrog]

Surely the interpretations of the game dev are broadly irrelevant.

If your end goal is to enjoy the game then the optimal difficulty has nothing to do with the game dev's intentions for you.

It's the difference between playing Chess and using the Chess pieces to construct imaginary battles between castles and goblins.


My original point was that you're deliberately hamstringing yourself by trying to intuit what the game dev's intentions are and then proceeding to assume that the dev knows what's best for you in terms of your enjoyment - which is crazypants.

You are massively overcomplicating (and misinterpreting) what I'm trying to convey. I do not want to construct fancy bases and "safe" preconstructed scenarios of my own choosing. I would play in basebuilder mode if that were the case. Or Minecraft. Or something else.

I want a challenge. I don't however want it to be an insanely unreasonable challenge. I don't want to play in carebear mode. I'm ok with losses and bad things happening. I'm ok with being forced to think and strategize if I have any hope of winning. I'm ok with the unexpected. I'm NOT ok with Cassandra mass-murdering my citizens without me having any chance to rise to the occasion and defeat the challenge (even just barely) and then having some vaguely-reasonable period to recover.

So yes, I do want a challenge. Challenges are fun. Nailing all of my colonists to a wall and executing them is not fun. See the difference?

Optimal difficulty is a level of challenge that you'll have to work at to ultimately stay on top of, but which you'll feel rewarded and satisfied for having done so. You'll feel like the work you put in was worth it, and had a purpose. An optimal challenge takes into account what you're capable of and pushes you to the edge to try and defeat said challenge. That is what should be optimised for by the designer, and what should be implied by "100%". 100% isn't just some fancy big number put there to sound scary. It is relative to something, and in this case, it is relative to some baseline difficulty that has been optimised for by the developer. If you think it means something else, you need to back that up with something Tynan has said, because otherwise you're just just making assumptions and preaching those assumptions irrespective of how correct they are.

[SSS]

I agree that 100% implies standard, the difficulty the game was "meant to" be played at. However, again, not all games are meant to be played at a "normal" level. Take maniac shooters or "platform hell" games as extreme examples of this. One person's idea of optimal difficulty isn't going to carry across every game, which is why difficulty settings exist in the first place: If you don't like the dev's intentions, you can tweak it yourself.

That being said, I do agree that the challenge is fairly one-dimensional and harsh at this point when played at challenge or above. Hopefully this will be alleviated when more types of challenges and goals are incorporated beyond "keep colonists mood high" and "survive overwhelming odds against raids".

[Boboid]

You are massively overcomplicating (and misinterpreting) what I'm trying to convey.

What you're really trying to convey is that you misinterpreted the difficulty variables and as a result think that the entire system needs to be revamped to align with your particular preferred perceptions.

What I'm attempting to point out is that if you take in this new information and incorporate it into further decisions about difficulty settings you will have a more enjoyable game experience.


You couldn't handle 100%, try something easier. It's that simple.

[rtiger]

No, it really isn't. The only real difference between the difficulties is how long the storyteller will wait before crap hits the fan, and none of then take into account any losses your colony has suffered, let alone get a chance to recruit more before dozens more enemies drop in to raid your base. You might fight off a wave of a dozen raiders with your 10, but lose 3 colonists in the process. Then the next wave is 16 and you only have 5 that are not dead or in medical. Get the picture now?

Constantly fighting overwhelming odds is not fun.

[Cazakatari]

Boboid, I just don't feel like you're actually trying to help as much as bully with your point of view.  Axefrog has been reasonable in his observations, you haven't been nearly as accommodating with yours

To the topic on hand, I'm pretty sure Tynan doesn't want the difficulty of the ''storytelling'' to be about increasingly massive numbers of raiders/mechs/supersquirrels.  Strategy and tactics help greatly but I'll agree that at the "100%" difficulty you basically have to resort to what I think are immersion breaking exploits to have any chance of survival that doesn't involve massive RNG. 

Unfortunately that is the only way for the storytelling to give any difficulty imo, because aside from physical attacks there isn't anything the storyteller throws at us that endangers colonies, even in the harsh biomes to a large extent.  With time I hope there will be plenty of other events that will be a significant stress but not necessarily involve hordes of raiders etc.

[Tynan]

I do think the game is a bit too much about slugging it out with raiders these days. I'm planning some changes that will change the entire difficulty matrix, hopefully in ways you guys will all like. These likely changes includes:

-Reworking the raider AI so they know how to get around killboxes. This will make them stronger, so we'll be able to reduce their frequency/numbers.
-Reworking the challenge scale system to do more than scale raider numbers. Scaling other things like crop harvests will provide a more evenly-changing experience as you shift challenge levels, instead of one that becomes more and more about optimizing a killbox.
-Obviously the previous thing will require rethinking all the difficulty settings, including which one might be presented as "100%" in a tooltip.

EDIT: That all said, there will probably still be a mode called "challenge", and it will still be about defeating a serious challenge. I hope the challenge will be broader than perfecting a killbox design, but it will still be very difficult or impossible to beat for most players. Some people want that, and that's what "challenge" means. I guess where I fell down in messaging is exposing that percentage figure.

It's an interesting game design observation just how much heartbreak can be caused by a single percentage number in one tooltip. If OP had simply been given the difficulty level labels, he'd likely have had an entirely different perception of that choice and an entirely different experience.

[CheeseGromit]

Very cool. Nice too see some official feedback. Sounds promising for how I like to play.

Interesting to see two sides of the argument. One relating to the difficulty nomenclature and peoples preconceptions of what it means vs the more practical, choose the difficulty that suits you regardless of what it's called.

[Boboid]

Scaling other things like crop harvests will provide a more evenly-changing experience as you shift challenge levels, instead of one that becomes more and more about optimizing a killbox.

Little bit worried about that - Crop harvests seem innocuous enough but scaling various game mechanics to difficulty can be a bit of a brain-chomper for players.

Sticking with Crop Harvests - It's annoying to have to memorize the effects of various difficulty levels on crop yields, if it's as simplistic as " X difficulty increases growth time by X number " then that's not too bad, but if potatoes yield 2 instead of 4 or if trees yield 30 wood instead of 60 it gets a bit frustrating to keep all that info together.

Combat difficulty has the same problems, if you make raiders limbs have 5 more hitpoints each or make them 10% more accurate it's doubly frustrating to increase the difficulty, because not only are you likely dealing with more but also simultaneously all your preconceptions are off.

X-Com EU and the original CoH both come to mind as examples of statistically scaled difficulty done very very wrong.

[Tynan]

Combat difficulty has the same problems, if you make raiders limbs have 5 more hitpoints each or make them 10% more accurate it's doubly frustrating to increase the difficulty, because not only are you likely dealing with more but also simultaneously all your preconceptions are off.

Yes, that is the main danger of this kind of system, which is why I want to keep the changed variables few in number and obvious in effect. E.g. it's not that complicated if plants simply yield 40% more food at an easy difficulty level, or the colonist default mood is 8 points higher. These effects are simple, centralized, and easy to understand.

I wouldn't do something like modifying raider health; it's way too obscure and makes no sense since raiders become colonists and can retain their same injuries.

[Mathenaut]

Combat difficulty has the same problems, if you make raiders limbs have 5 more hitpoints each or make them 10% more accurate it's doubly frustrating to increase the difficulty, because not only are you likely dealing with more but also simultaneously all your preconceptions are off.

Yes, that is the main danger of this kind of system, which is why I want to keep the changed variables few in number and obvious in effect. E.g. it's not that complicated if plants simply yield 40% more food at an easy difficulty level, or the colonist default mood is 8 points higher. These effects are simple, centralized, and easy to understand.

I wouldn't do something like modifying raider health; it's way too obscure and makes no sense since raiders become colonists and can retain their same injuries.

You want raiders to be more threatening, not just harder to kill. Right now, the raiders aren't very threatening at all and are particularly hard to kill, vs how vulnerable colonists are in contrast. Thus, the killbox design to optimize damage output at minimal risk. Alternatively, many requests/popular mods revolve around moar damage for turrets or better protection for colonists (who do more damage than turrets).

A raider that shoots an exposed windmill/solar panel is more threatening than the same raider walking into turrets.

A raider with a riot shield (better than the shield generator, imo) is a bigger threat because of how turret targeting works. Even if such a shield provided little other protection and made him a weak melee combatant.

Two raiders, one with a shield, the other with an incendiary launcher, could complicate a killbox more than 20 raiders just mob-rushing in. Especially if raiders don't prioritize unpowered turrets.

It's also something that colonists could handle with much less effort, even with weak weaponry.

May also wish to reconsider general lethality of firearms. For the most part, they only really kill when they hit something vital, so they should be incapping more than obliterating people with protection.

[SSS]

Oh man, I can only imagine how stressing (in the good way) it could be to take down raiders in a modern sort of testudo formation, if riot shields were a thing. You'd be more likely to have to get up close and personal- which could turn ugly quite fast.

Raiders using strategies would make combat leagues more interesting/entertaining than it is now. I'm guessing it's a coding sink, though; programming even one strategy, when and where to perform it, when to stop performing it, and so on would take a lot of effort, I'd think.

It sounds awesome though. Maybe you could base an alpha (or two) around that sort of thing, Tynan? :3

[Mathenaut]

Well, there was a step forward with the personal shield concept, though that's an attempt to make melee viable on a per-pawn basis.

Equipping a riot shield (high bullet protection, can't carry anything else, does crap melee damage) on the first raider stepping into a killbox would draw most of the fire. Everyone else could function with only the change of ignoring  unpowered turrets. This means that instead of pounding on a turret until it explodes, they just attack the wire (or deactivate it since they're right there)

Doesn't require immense changes I don't think. If you want, you could make this 'tech-savvy' approach specific to pirates instead of tribals.

[Teague]

Peg legs will allow your legless colonists to move around, albiet a lot slower.

Trading system right now is a placeholder. It functions, and I am glad I have something instead of nothing.

The good replay value functions (procedurally generated landscape, rpg elements in pawns, random events, steep learning curve) are wasted if it doesn't challenge you / kill your first few settlements.

[Kegereneku]

Oh man, I can only imagine how stressing (in the good way) it could be to take down raiders in a modern sort of testudo formation, if riot shields were a thing. You'd be more likely to have to get up close and personal- which could turn ugly quite fast.

I actually use such formation against mechanoid from time to time, my shield wearer are on the front with sharpshooter just behind.
Of course it only work against concentrated/lone enemy and it take 2 time more manpower, depending if you attack with melee weapon or just want shield.

[axefrog]

Just thought I'd drop this in here.

Cassandra blessed all my shooters with an outbreak of malaria and then not a minute later, dropped a siege into an area with basically nowhere that I can get any decent cover from. I outfitted my two remaining healthy colonists with the best gear I could find laying around, then when I got to the siege had them hide behind a tree and a rock, and both were promptly taken out in succession - and i mean literally first shot taken on both - (one had his head destroyed, and the other was shot through the chest) by a single sniper with only 9 shooting skill.

Tynan, if you're still keeping up with this thread, it'd be interesting to know what sorts of plans with regards to building out the A.I. of the storyteller's own decision making process for what to challenges to present the player with, and when and how those challenges should be paced, particularly with respect to chronologically-adjacent challenges. I know you talked about making the way raiders behave more intelligent, and improving the types of challenges that we can be presented with, but I really can see a need to need to look at the logic and heuristics of how challenges are selected and paced. Would it be difficult to do something like build a score around each colonist's capacity to do a given thing, project that towards the base's general capacity to achieve a given thing, and also weigh that against those generalised base-level scores as they exist both before and after a given challenge has been met and triumphed against? Assessing the unbiased calculation of how well the storyteller expects the player to succeed against a given obstacle, as compared to how well they *actually* faired would perhaps allow further calculations to be weighted. I think there's a lot you could do with that idea anyway.

My experience of having the storyteller first incapacitate those who are capable of defending my base, immediately follow it up with a siege, and then one-shot kill both of my last-ditch attempts to fight off the siege, makes the blatant randomness of the current system quite visible.