Larian Studios Forums Divinity - Original Sin Divinity Original Sin - Beta my feedback on hard difficulty

Well, in that case it depends on your own definition of skill (or mine of course). But I think you blame the wrong genre: first person shooters are not the same thing as QTE (what you describe here).

Anyway, that doesn't add to the discussion I fear...But bringing chess to the discussion wasa good idea: without enough elements of randomness D:OS and every other turn-based game would be just a different version of chess in combat - with probably inferior rules and mechanisms- which would make it a game I definitely wouldn't want to play.

As you would expect, I firmly disagree. But maybe I wasn't able to express myself good enough here: I don't want a game based on luck and randomness alone, not at all. I want a game with a clever combination of luck and logic. Luck is the spice in the meal imo. Maybe if I could use a picture: luck is chaos and logic is order. Each of them is bad but together they are king.


Where is the satisfaction of memorizing patterns and using numbers games to win a fight? There is quite some fascination to beat a human player in a strictly logical game like chess, I wholeheartedly admit it. Being more clever than the other one and having the better strategy (or the ability to think more in advance and knowing the rules better...).

But this fascination is hugely inferior in a SP video game against a pre-defined enemy AI. Pure logic and numbers games just make the whole experience boring here. There are only two possibilites to eventually prevent that: first, making each opponent unique with a unique fighting style, abilites and strategy (which is almost impossible in a game like that, but that's mostly the solution of Dark Souls for example) or second, add randomness to the forumula (which is the solution used in most games of its genre and even pen and paper games like Dungeons & Dragons).
Luck in that respect means that your strategy or tacitcs could fail. Anytime. There is no golden way to the win. There is no pattern to memorize (hello, Dark Souls...) but there is sure tactics and even strategy left. You just have to react what actually happens in batlle. Your mage failed his cast? Bad luck, adapt your tactics. The enemey failed to hit you? Fine, use the advantage.

And as I said, there is already "luck" in the game. You melee fighters can miss their targets. If you don't want any luck or probabilites in the game even your melee fighter had to hit with each turn. On top of that there shouldn't be even a range for weapons since 70-170 damage is also based on luck and probabilites. Just let them hit every time with 100 damage. I don't know what you think but to me that sounds incredibly boring. In a game like Dark Souls this is ok because you still have the action and to press the buttons all at the right time but here, in a turn-based game? It would be an inferior version of chess against a predictable enemey. Doesn't sound like fun to me tbh...


As I said before the outcome of a fight should rely on luck alone. But luck should be part of it for each and everyone, not only melee fighters.


But of course you don't have to agree with me. Maybe (or even likely) to some a turn-based game without any random component sounds like great fun. Praise the variety!

To go on a little tangent that's at least partly relevant to the latest discussion (and that will probably also start a huge forum battle that I won't be taking part in, since I've been there before), there are three old ideas still kicking around the CRPG world that need to be repeatedly killed with fire:

1) Enemy level scaling: thank God Larian hates level scaling as much as I do. It robs the player of the option to ever explore a tougher area and try to take on a bigger challenge - or on the flip side, to feel the satisfaction of actually getting stronger vs. enemies that could previously kick the player's tail - and essentially makes every... single... battle... the... same.

2) Hard level caps: especially when you hit the level cap significantly before the end of the game, again, this kills player development - which is much of the fun of a good RPG. It's a clumsy and lazy attempt to force "balance" on the endgame, which should properly be done by ensuring battles consistently require strategic thinking, while not stifling the player from exploring new equipment, skills and abilities.

3) Abuse of random factors (RNGs, etc.): all this does is encourage tedious save-scumming. As Arjiki explained, I want combat challenge to be based on the relative strength and strategic skill of the enemies and player characters - not from some lame RNG throwing off the entire fight. That's just cheesy, sloppy programming, IMHO.

OK, getting back off the soapbox now. There's some interesting discussion in this thread, anyway.

Thank you for the added explanation. I think I understand your point and I see where you are coming from.

I think there is a balance to be had. A game should have enough certainty in its mechanics for you to plan ahead with a reasonable degree of confidence. This can be achieved either through a high probability of success (eg. 90% chance to hit) or low cost of failure (it only cost you 3 action points to try to trip your enemy) ...or both.

This is necessary for a game to have an element of strategy. To give an example of what I mean, imagine a skill that had a 20% chance to hit, costs 12AP, and deals 1000 damage. You can't count on it killing your target (because it probably won't), but you also can't plan around it not doing anything because you're out of Action Points for that turn and can't do anything about it. It results in a scenario where you just cross your fingers and hope to try again next time if it doesn't work.

On the flip side, you need some amount of uncertainty in order for a game to be interesting. If there was a 100% chance to hit and your weapon dealt the same damage every time, fights would before extremely tedious. You'd know ahead of time that you can kill each zombie in precisely 3 attacks, so your course of action becomes obvious. Given the chance to "fail" on any given step of your plan makes the game more interesting to play because you have to take into account the fact that something could go wrong at any given step and what you are going to do about it.

If there are too few random elements or they are too insignificant, then you know exactly how each fight will end before it even starts, making things somewhat anticlimactic.

Edit: I should clarify that my main objection to randomness in the previous reply was the notion that it should be prevalent enough to determine the outcome of fights. I do believe randomness should determine how fights are played out, and possibly determine the outcome occasionally in cases where it would have been a really close call to begin with. However, I feel you should be able to approach any encounter with 99%+ confidence of the ultimate outcome, just not exactly how you'll end up getting there or how easy it will be.

I personally feel DOS is in a good spot right now when it comes to balancing the two, but then again, I played a Rogue/Ranger without companions, so I never experienced the combat with spells. Maybe I would also feel that they need more randomness, but I am inclined to guess that I would be satisfied with the chance to burn/stun/freeze that they have right now.

Yeah, it was a reply to my "powerful spell/skill -> little chance to hit" post which wasn't all that well explained.

The basic principle of my combat design approach is that the chance to hit (or cast) with a skill or spell should be going down with the damage going up. Of course that adds randomness to the combat and depending on your own way to fight some encounters could indeed fail or succeed because of a random element but it would also add tacitcs and gameplay variety. That doesn't mean that I want to force any system on D:OS, it's imo just a working principle to be aware of...

I know that the combat in D:OS has its focus on elemental effects and combos and it's a oool feature. Let's hope that it will stay interesting for the whole game and won't become boring soon.

Lord Crash, any fight can be too hard/too easy if you're too low/high a level for it.

Being three levels below the enemies will make for a tougher fight, and there's no fight in the beta which is challenging at level 10, because no enemy in the beta is above level 8.




Are you suggesting that D:OS add some spells that if they succeed will screw a character unless there's a very specific recourse to curing them?

Okay, but on a scale of screwed from "just need a resurrect scroll", to "need a golden tongue from a winged basilisk which only appears by the light of the waning crescent", what do you think is the sweet spot?

I have an idea regarding difficulty levels: why don't make them partially dynamic? I can imagine a horizontal scale in the options with the classical easy, normal, hard, hardcore, but it could go on and on, so if we play different and the other parts of the world become easy and boring (just like in another thread someone killed the whole city and after exiting the orcs and undead were too easy to beat even in hard mode).

I know, one of the main issues is that the foes don't level up with the characters, but if there would be more difficulty levels, they could give boosts for the foes if needed. For example the next level after hardcore could add +50% to the HP of the foes, or increase their damage, or whatever. With this dynamic concept Larian could solve the issues because of the large freedom of the game, since this is the main factor, why it is so hard to balance the game.

They could balance the game on easy, normal and hard, (or even hardcore), and after that they could add such extra boosts.
The diff level could be changed anytime, so it would be kinda self-balancing.

What do you think?

Sorry if someone already mentioned this, I don't have much time to read the threads..

I think more interesting hard difficulty shouldn't just have an increased enemies level, HP, damage output etc., but also have these things:
-enemies use higher level skills (for example, not flare but fireball) and use them more often
-there should be more enemies in every encounter (but to maintain balance, experience gain for every encounter should not be higher than on easier difficulties)
-there should be more hazardous objects, like oil barrels, in enemy encounter areas
-negative effects, like being stunned, should last longer
-allow companions permadeath if they take damage well above their HP number, like, say, 200% of their HP number (also a good thing in general because it encourages players to try other companions)
-items in stores should have higher prices

Yes, this things you wrote belong to the normal balancing. The extra things I wrote are meant as additional difficulty. I kept this things simple so that it would be easier for the devs to implement.

Alas, I agree 100%. No thought went into character creation and planning in the beta. This balance hopefully will change in the complete game, but I doubt it. Game is wrong from the very beginning, by having you choose between classes in a classless system. It says it all about the planning and thoughts so far.

I just wanted to drop a note that we're quite aware of the problems our system still has. Some of the problems you've been encountering are indeed related to systems not fully working, but that's but a temporary thing. A lot of work is currently being done on encounter scripting & balancing as well as character building and I expect that will make a big difference once we're finished with it. You can also expect fundamental changes in what the "class specific" abilities do (e.g. fire elementalist or way of the rogue)

Absolutely true.

I think the lighthouse fight and Braccus court of sceletons are the two best encounters in the beta - if their level is high enough. Imo both should at least consist enemies with level 8 or 9 or even higher. But I agree that both encounters offer some nice variety of melee fighters, rangers and different mage types.



Well, hard to say. The spells I meant were some kind of "perma-death" spells. That means that you had to decide whether you want to reload the whole encounter (because saving wasn't allowed during combat) or you want to proceed with the respective character staying dead (of course if your main character was hit it was game over). Some people who are friends of perma-death mechanics and hardcore roleplay choose the latter but many others choose the former.

Example: an enemy basilisk casted "turn to stone" on a party member. Now there were two possiblities: either you have a "turn to flesh" spell or scroll available to bring him back to life or you don't. In the worst case some area damage hits the stone statue causing it to burst into pieces which was guaranteed perma-death with no chance to resurrect.

Or another example: mind-flayers had some very nasty ablities to drain your intelligence. That means if they were able to hit you they reduced the intelligence of the respective party member permanently without any chance to cure it (intelligence at zero means death). Fighting against these (very powerful and dangerous) enemies without proper resistencies and/or gear was more or less a suicide mission.

Another thing were lasting effects. You could be cursed or diseased for example and the effects didn't go away just after a few rounds. You have to actively cure them, either by spells (which made them valuable) or by visiting a temple (which were often quite far away with many dangerous combat encounters on the way).

As I see spells in D:OS right now there is no really decision making what spells and skill to use apart from using something that the enemies has weaknesses for or low resistencies at least. In most cases you can just use every spell or skill without any trade-off for the one using it and the whole party. There is no possible negative effect to use your most dangerous and powerful spells and skills as often as possible. That's also true for environmental combinations. They work every time. There is no incentive to not use them apart from using them on enemies of the same environmental school. I wished there would be some kind of additional trade-off based on probablities which would make combat encounters more dynamic and far less expectable. Without that I could just use the same strategy over and over again (which is still true but the chances to suceed would be lower).

Cool, can't wait for the next update!

No, this is precisely not required. Skeleton archers are already vastly unbalanced for their levels. Their damage output, compared to their melee / mage companions is already far higher.

For reference, a fight that isn't mentioned much is the 4 stone guardians, which is badly designed because its difficulty relies purely on the party build you have. 100% two turn petrify x 4 will wipe a melee focused party; however, if you have summons to soak up the effect, you can then use ice wall [on stone enemies? slow at most, surely?] to freeze x2 and split the encounter into two parts.

It was an interesting fight, but I can see it being impossible with certain parties.

The issue at the moment, even without duping them, is that summons can double your party size, and the ice elemental's special is > anything else. Fire elemental gets... flare. Whoo-hoo. (From memory of the alpha, air got elec, earth got boulders?). Adding to this, melee chars can spam x3 trips / knockdowns in a single round [rush / bull (3/4AP) + crushing hand (4ap?)] which adds to the carnage.

In comparison, barring a badly placed ice elemental of my own dying, I think I was tripped / knocked down once during my entire play through. I wouldn't want the ice elemental to lose ice wall, but I would like to see certain enemies immune to freeze and the other summons have comparable utility. e.g. there's really no point to the witches' summon warrior spell.


TL;DR

At the moment, knock-down / freeze / petrify [many weapons have 10% chance on hit] are the mainstay of any fight. I suspect that once mobs start using AP / skills correctly this will change, and probably slaughter players.

Great to hear, lar!



And the basilisk can use this spell multiple times a fight?

I'm not a fan of roguelikes and permadeath myself, but I'll set that aside. One of my concerns with that idea is how will it work in co-op?

"oops, the bad guy used a spell and now you are dead. Forever." It's no fun sitting there for an entire fight doing nothing because you don't have/ran out of/were the one carrying the very special and rare magic scrolls that were the only cure for that special status. (It's different than the more general Resurrect scrolls.)

D&D has a DM whose job it is to carefully guide events so that people are having fun. Sometimes that means fudging rolls, giving players little hints, or having critical items show up just when they're needed most. Computer games don't have that. I can't roll up a new Source Hunter mid-adventure and have him meet the party in the tavern. The systems are different and run in different ways, so you have be cautious what to borrow.




I agree - Skeleton archers don't need a buff, they're already extremely dangerous when there are two or more. Just visit the west cliffs if you don't believe me.

I would hope for such fundamental changes as to see them disappear altogether, they are sending the wrong messages in the classless system you are selling.
A rogue should be defined by its skills, as in a classless system, and not define its skills, has in a class system.

I really really really wish you can come up with a deep interesting character creation and evolution system, because it is a big part of any RPG, as you perfectly know.

I don't know for sure. But point is that you can actually prevent the basilisk from doing so by preventing it from casting or interrupting it. It's very unlikely that it will be able to cast this powerful spell twice during a combat encounter. That's really up to balancing. In D&D video games (let's take BG or ID as examples for its video game implementation again) you had to take care who was probably the most dangerous foe in the enemy party and how to take care of him as fast as possible, preventing him from doing serious damage. Casting times, interrupting mechanics, random elements, buffs, area effects and all sorts of systems played together to make that actually interesting and dynamic.


Well, roguelikes are a completely different genre imo. Perma-death is a mechanic to actually make decisions harder and roleplaying more impactful. I'm not a fan of the "hardest" form of it either, perma-death without saving (like in popular ironman modes). With saving before fights and after them available perma-death (aka the inability to revive fallen companions all the time) actually becomes an active roleplaying mechanic, leading to serious decisions, not by forcing them on the player but by giving him the opportunity to either continue and live with the loss or retry the whole thing (that might be a typcial roguelike mechanic, I admit it, but it doesn't make it the same experience here).
As for your question whether it would work in co-op: I don't know tbh. It was one of my biggest fears from the very beginning that mechanics and systems which are implemented to make the co-op experiene better could make the singleplayer experience "inferior" to what would be possible with two seperate experiences. I know that with the fundamental design decision to make the game a co-op/singleplayer cross-over with the basically same content and systems no matter how you play Larian has to make compromises. Compromises that hurt me personally as a singleplayer enthusiast tbh. So forgive me for trying to make the singleplayer component more enjoyable. I don't know if perma-death and stuff would make the co-op more or less enjoyable tbh. But I do think it would be worth to find it out though...


Sure, but I was talking about the D&D video game adaptations (namely Baldur's Gate and Icewind Dale here).
Of course it's hard to make such systems work if you miss basic requirements. Having a good number of companions or henchmen availalbe is crucial for perma-death mechanics imo to be actually able to replace dead party members. Right now they are missing in the beta but as far as I know there will be much more possible party members added to the final release so that won't be the problem. Nevertheless, D:OS would become in trouble due to its classless system. D&D games were based on classes and "full companions" which made their loss impactful. If you are able to just replace the fallen henchman with the next guy in town with the same talents and skills, the death becomes kind of pointless. So I fully admit that with the current systems planned for D:OS or already implemented many adaptations of D&D and stuff wouldn't work as they are supposed to work.

I disagree with your statement that you need a dungeon master for D&D video games though. That can also be done by clever game design, good writing and well meshed systems. A good game designer has to envisage what might happen in situation X and he has to create systems and mechanics to give the player options how to solve the situation - best by giving him decisive power and in the end, player agency. Bad game design would give the player an easy solution to every upcoming obstacle which actually discourages decision making and player agency. Examples are savegames during combat or even more obvious endless healing and resurrecting abilities. A party RPG without any fear of actual loss is imo wasting an opportunity to involve the player on an emotional level. You know, in Baldur's Gate your companions were that well written you began to really like them. If for example Minsc died in BG2 without the chance to resurrect him it would have meant a direct loading for me because I liked him that much that I wouldn't want to progress without him in my party. Every critique I expressed so far for D:OS is mostly based on my personal feelings. There is too little emtion in D:OS and the mechanics and systems don't add to them. Writing is too bland and short (although it's well written and often offers good humor), companions are too one-dimensional and systems don't encourage me to be emotionally involved. Don't get me wrong, I like D:OS very much, but actually not for being a deep RPG but for exploration. To me D:OS in its current state is more like Skyrim than like BG2 (note: not based on actual systems but based on my feeling and the motivation why to play the game after all!) and imo that's wasted potential...

I can just add that i almost completely agree with Arjiki.

The "difficulty" should not come only from crude increases in enemy damage or, god forbid HP, but from tactical considerations and more detailed adjustments and changes to the way enemies are designed, all of their stats and how they behave.

I already mentioned and suggested in other threads that undead should be much more resistant to specific kinds of damages - and susceptible to others.

For example, resistant more to slashing and piercing weapons and Air - Water elemental magic - while more susceptible to blunt weapons and Earth-Fire magic.

Thats already in the game but the difference is not big enough.
- of course, i have no doubt that devs will adjust and balance the difficulties of the game much more, just by themselves since - this is a beta, right? Thats kind of normal for betas...




As for OP suggestion of making the town residents much stronger - i would be completely against that, for reasons already mentioned but much more because they are not meant to be that strong.

The civilians and the guards or fighters of the Legion are meant to be weaker then the surrounding threats. And that should be maintained. Otherwise the whole situation would make no sense.
Why would Cysael be surrounded and besieged if a few guards or merchants can go out and clear almost all undead and orcs easily...

Lets not.



----

Oh, i dont know why everyone thinks Lighthouse fight is so difficult.
I mean, yeah... it is - if you go about it in usual ways. If you dont you can end it in a turn or two at the most.

EDIT: Fine.

In my ~opinion~

Permadeath which doesn't wipe your save is not significantly different than a game over screen. The typical player would just reload their last save. There's nothing impactful about that to me. In my personal opinion.

Permadeath isn't the way to make your choices meaningful. Quests with varied and continuing outcomes do, limited points available and no or costly respecs make choices meaningful.

In my own, personal, opinion, being a person who does not like permadeath in their own opinion, I do not personally think that permadeath is a good way to make combat choices meaningful. Maybe that's just because I don't really understand what your definition of a meaningful combat choice is, if it excludes quests and dialogue and such.


D:OS has you with two players because of the story, so it's not just because of co-op that there is no permadeath. Oh, and there's no need to actually try it to see if permadeath would make co-op less enjoyable, because the answer is YES in 72-point font.

The reasoning I am employing for that statement is that permadeath is either a game over or a reload. Currently, D:OS is not designed around/balanced for a permadeath mechanic.


***

I definitely love it when different difficulty levels have more impact than just in increase in enemy damage/health.

For example, in Serious Sam, a higher difficulty doesn't necessarily mean more enemies, it could also mean that instead of a uniform group of all the same type of enemy, there will be a group of mixed enemies that require different strategies to defeat.

I'm thinking specifically about the start of the second level. On Hard, a group that's only a large amount of kamikazis appears to charge you. On Serious, there are fewer Kamikazis, but a few Kleer skeletons are thrown into the mix too, which makes the fight harder because you have two enemies which have different ideal methods to defeat.

I think you should give my opinions the proper respect by marking your statements here as clearly your own opinions and not as they were a fact or something (I got at least the feeling while reading it...)

I told you about my own experiences with various party RPG games and I told you how about I felt in them. You cannot just wipe that away with a bland "the typical gamer would just reload". Maybe most people would do but at least you have the choice...

And I was talking about the impact of combat choices and not the impact of dialogue choices. These are two different issues or better two different layers of choice and agency. Permadeath was just one example of mine how to make combat more impactful. It's not an equation like "perma-death = better combat", not at all. All systems and mechanics have to work together to make the combat system impactful and worthwile.

Of course leveling and the danger of putting points in the "wrong" abilites or stats is again another layer of choice or meaningfulness. There is no indication that all these systems can't work together...


And I disagree on your statement that D:OS has two main players because of the story. I think that this decision was made after someone at Larian got the idea of a co-op game like we can see it now. There is no point in making a game about two characters for singleplayer, especially not in this case, as the game evolved to a real party RPG. And I think that if Larian is really honest about that they would agree with me here...