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

I've spent 30+ hours on previous version, trying to do every quest and read every dialog. I enjoyed it a lot.
This time i decided to avoid quests, and kill npc and test hard difficulty, just to enjoy tactical fight.

I didn't enjoyed it at all because there was no challenge. The game was too easy in previous version, it's still way too easy. Even in hard difficulty, it's a piece of cake. I didn't noticed any difference between normal and hard mode. Worst : it's even easier than previous version, because the loot is much better.

The only real improvement in this version is the increased AP cost of potions. Everything else sucks. And I encountered much more bugs than before.

I killed every living creature in cyseal : npc, sheeps, chickens, cows, everything (only cats were left alive because they are cute). When I left the city i had 80k useless gold and many epic weapons and jewels. What's the point ? Why npc are so easy to kill, and why do they drop so good loot ? Aren't city guards supposed to be strong ?

After the slaughtering in Cyseal, i killed some mobs in the land around, but it was so boring i stopped playing.

Do you really intend to release the game in june ? This game could be great, but balance and game mechanics really looks broken right now, and a lot of work is needed.

What balance improvements would you suggest ?

Same improvements i've already suggested in my first feedback for previous version a few weeks ago.

I played a mage/ranger hybrid and a warrior. Like in previous version, i think that melee are overpowered, mostly because melee weapon deal too much damage. Difficulty is ok at very low level with basic weapons, but once you start to loot some good weapons it kills challenge.
Some skills need to be nerfed too. "Phoenix dive" is awful : very long range teleport + good aoe damage and only 2 turns cooldown, it's insane. This skill does everything : you can use it for offense or defense, for movement or dps, and you can use it very often.
Some spells combo are very op, like "haste" + "oath of desecration" (especially when cast on melee characters who have the "raistlin" trait).
Crowd control is too easy : freezing, teleport, stunning... you can easily incapacitate a lot of mobs.
Some traits combo are still there and still op, despite the fact that many beta players reported this flaw : "lone wolf" + "raistlin" can kill this game.
Npc in Cyseal are too weak, citizens should be lvl 4, guards lvl 6-8 and merchants at least lvl 10. Right now killing merchants is so easy and give so much loot it's ridiculous.

Mobs AI is lame, but i suppose we can't expect better, it's a common flaw in most games. At least, they should compensate this with higher mob damage and hp (in hard difficulty).

Some people will find the game frustratingly hard...
Moreover, its is maybe easy because you don't drive forward, and comeback after some time in a place you "shouldn't".
Maybe things are balanced overall in the whole game.
But I agree there is severe unbalance in the beta. Wrote a subject about weapon balance and damages, I still think some primary combat unbalance comes from here. I couldn't get any proper figures, because I am probably right, but no fanboys here will admit it.
Too bad.
My suggestion: Don't make weapons with such a big differnce. Organise the numbers. think about balance before building the systems, it could help.

So for example:
1h weapons: 5-6 damage /AP. Maybe a little bonus here or then so that a mass isn't the same as an axe (stun chance, armor piercing chance, and so on...)
2h weapons: 7-8 damages /AP Still better.

Poor weapons: -1Dam, good weapons +1Dam.

Magical weapons: beside giving special bonus, +1 (weak magic) to +5 (high magic) to good damage, those last being hard to come by, and requiring high stats to wield.

This is only an example.

Because if you go from 5-6 dam to 15-20 damage for basic weapons, how can you balance anything ? Unless mobs HP follow the same increase of course, but then what ?

As someone who has also played more than 30h on the alpha/beta, mostly in co-op, I can say that all your points are indeed very true.
I'm not too concerned about weapons and skills balance, because I figure that must be one of the latest things to be done down the development pipeline and that it shouldn't be too hard to correct that course by altering some numbers... but I also share your worries about the ennemy AI. True, we haven't had the chance yet to encounter some higher level monsters whith a greater set of abilities, but the current ones do not offer much challenge.
It's indeed very easy for the players to alter the battlefield and stun lock mobs or damaged them over time while easily escaping their pathetic attempts at closing range... and the AI currently reacts poorly to those situations. You also rarely find yourself hindered by the same spells you cast relentlessly at your opponents, which makes it really unfair, to say the least... :P I really hope Larian has some surprises in store for us in this department, because it would really make a huge difference in the final game. Right now, combat is still fun but becomes trivial once you reach level 3-5.

What is the tiping point ? Why does it become so easy suddenly ? Maybe that would hint to some changes .

My tipping point was finding a decent weapon. Bfore = fights frustratingly hards. After = fights frustratingly easy.

If it is so easy to change the numbers and the balance of such an important aspect of the game, Why didn't Larian just do it, so the beta can test this ? Because there are more important things to change, I agree. Unless this is a real balance issue.

I disagree. Why would you want to kill everyone? I can understand that someone maybe wants to "try it out". But please, for a normal playthrough nobody would kill all the NPCs and merchants in town. If you want to kill all balance, immersion and roleplaying for yourself feel free to do so. But then calling for the devs to make that harder is ridiculous. This is not a "game to master" alone (aka Dark Souls). It's an RPG which means that there is roleplaying involved. That's a fundamental difference...

For the actual combat balance I agree though as I've said numerous times before. Some spells are overpowered and enemy AI is lacking. Loot balancing could be improved as well.

For me it's much more about skill diversity and thus variety of possibilities. Shitty weapon makes combat more tedious, but not really harder.
However, once you get a hold of "combos" like teleportation/blitzbolt/oil/flare and you also have an archer, for instance, you can easily stun ennemies, keep them at range when they're finally closing in, etc. and damage them from afar, while the poor undead swordsmen and zombies don't have that kind of arsenal to oppose you.

Granted, the amount of possibilities you get as a player in combat is also what makes the game's fights unique and so enjoyable, thus I don't think you want to limit the player more there. I'd say that's where resides the difficulty on making this game's combat both challenging and interesting in the long run.

I guess the people who had the hardest time with combat at this point were those who hesitated to diversify and built characters focused around one archetype, like they're used to in "classful" games, mmos and the like.

To be fair, some of the higher levels encounters might be more challenging if you don't have the right tools, but by level 6, your party should be able to tackle any kind of opponent, be it fire elemental, pirate ghosts and whatever its particular immunities etc. will be. I'm still confident they have much more in their pockets for the final release, though. (Those environmental effects seem nasty, I think starting in the next area after Cyseal "weatherman" is going to become a must have.)

As for the level of the citizens, and guards especially, in the city, I'd agree that it's maybe a little too low... but I also agree with LordCrash here : why would you absolutely want to kill everybody, and why is it a problem if it's easy to do so ?
Also, making really overpowered guards could be a problem for those few starting encounters near the gates where it's easy to draw the city watch in to help...

How so, drawing in the guard is a valid tactic, and these battles aren't the hardest... it's a good introduction. I don't think it would harm those encounters if guards became harder (and a bit more prominent, especially in the market region).

I had the same experience where fights prior to level 3 were balanced: they were extremely difficult, I had to run away from most of them, and those that I won required a great deal of effort.

Rant about the early game encounters being 'too hard':

As others have pointed out, the game immediately becomes trivial once you find your first mediocre-average weapon. However, the balance changes we need are much more than a simple number rework on weapon damage and monster HP as others have suggested.

If weapons were reworked so that the strongest and weakest weapons in the game did not have a huge difference between them, then both looting, crafting, trading, and gold all become useless. Why would I bother learning how the crafting system works or saving up gold to buy from vendors if I'm only getting a 5 point damage boost at the end of it? Imagine identifying all your loot if you knew that no significant improvement could come out at the end of it.

There needs to be a huge range between a random weapon you find and a great one you buy or craft in order for these major systems to function, so the difficulty of the game must be somewhat independent of the damage from the weapon you are using.

A great example of this is the Lighthouse fight, which seems to be widely praised as the most challenging and best encounter in the game.

The reason this fight is great from a balance standpoint is the reason why it is challenging:
1. You have to react to what the enemies do.
As soon as the fight starts, one of your characters has multiple dogs and archers attacking him. He will almost surely die the second turn unless you do something about it.

2. There is a significant asymmetry in the enemies that you are fighting.
The dogs are easy to kill but get resurrected every turn. The archers are hard to get to, but deal the most damage. The ghoul just needs to be killed immediately, but is protected by the archers.

3. The challenge is not killing the enemies, but in surviving while you do so.
Even if you have great weapons, you will still be at the risk of dying unless you play it out properly. Obviously good weapons make it easier, but some of the challenge still remains despite it.

Compare this to any of the orc, skeleton, or cultist fights.

There are a bunch of orcs. Some of them are melee. Some of them are ranged. Some of them heal their allies a little bit. All of them have a high pool of HP and deal OK damage. There is nothing to react to. It doesn't really make much of a difference on who you kill first. The only "challenge" is killing them before they kill you.

Since there are no other elements at play, the fight just turns into a: is your weapon good enough to kill them before they kill you?

Just to list some easy to code ways to make current fights challenging in a good way, rather than a "gear-check" of boosting pure numbers. The idea behind these is that the player has to react to what the enemies are doing and is encouraged to kill the enemies in a certain order. Since the enemies they need to kill are way off in the back, they face the consequences of getting attacked the entire time they try to move to kill them.

1. Zombies leave a trail of poison wherever they run.
Makes you really consider movement. If the zombie follows you, you have more poison to deal with.

2. Add skeleton shamans that deal no damage, but summon a relatively weak zombie dog / skeleton every other turn and run further away from you whenever they are not doing so.
Do you try to chase down the thing dealing no damage to you right now, or deal with the actual threats?

3. Give certain enemies powerful defense reduction curses on a long cooldown.
If you charge straight in with Madora, you need to decide if you might just need to get her out of the thick of things and wait or come to her aid.

4. In all fights, buff the damage of ranged enemies, but give melee the chance on hit to cripple/blind/etc. for several turns.
Depending on the fight, you would want to take out one or the other first. Gives you a choice with significant consequences for how the battle will play out.

5. Add additional sleeping enemies to fights. They don't attack until they are attacked or turn 5. Every turn, they place a 30% damage and health boost on themselves.
Do you want to deal with the weak version now, or later?

6. Add shamans that deal no damage, but cast a long-duration curse or buff on someone every turn.
Are you going to try to end the fight fast before the curses make a significant difference, or will you actually need to kill that thing now?

7. Normal enemies (look just like any skeleton warrior, etc.) that summon enemies around them when they die.
Surprise! Makes the player react to what happens during the fight.

And so on and so forth.

What I feel is important about changes like these is that they do not require a massive amount of work on enemy AI. You are simply giving enemies a basic skill that fundamentally changes how the fight is played out. The resulting change typically forces you to make a decision that would yield a different result from one fight to another.

I'm not saying that we don't need a numbers change on weapons / enemy damage, etc. We do. Improved enemy AI would also be fantastic. But until Larian gives us meaningful decisions to make about how to approach fights by changing the enemies we encounter, any balance changes they do will never improve the combat experience. It will either be easy and boring (like it is now), or difficult and tedious.

I like your assessment of the Lighthouse fight and I agree with your rant about "realistically" early game.



I think one thing that can be done about the strength of weapons and the interest of players to pursue loot -- and that's already the case to some degree -- is not to scale them in terms of linear power (i.e moar dps) but also in variety and usefulness in various circumstances.


Weapons and armors that might not vary tremendously in damage or protection but provide different secondary bonuses, resistances, etc. are a good thing. For example, I held on to a pair of boots in the early game because it gave a nice boost to sneaking while I already found alternatives with 3 or 4 more ranks in protection.
It seems they already have introduced more of this type of items when they reworked the loot matrice and I think reducing the disparities between damage ranges but increasing their particular advantages like chances to stun/petrify/poison/freeze, types of damage, bonuses to talents or skills, etc. might do great things to both balance and loot.

I'd have an interest in paying lots of gold for a weapon that does not do a lot more damage than what I already have but would fit my character's build perfectly in terms of secondary characteristics and tactical possibilities... and I would probably hold on to it longer than a straight "improved damage" item.

That way, you might also want to level your crafting abilities to add some "fire damage" to your favourite sword in case you encounter skeletons immune to slashing, add some movement speed to your rogue's boots at the cost of reduced armor, add a stun effect to a mace wielded by a warrior with no other crowd control abilities, sharpen a low damage dagger to provide a a high chance of critical strike, etc.

Great analysis Arjiki. You've managed to put some things into words in a way that I haven't been able to.

Indeed, great ideas Arjiki.

I have to add my own rant, but this time to some players. Some people here look for the ultimate challenge that would allow them to tactically approach each little battle and learn it by dying multiple times. How would that be interesting and fun if every time you encounter an enemy you'd have to die 15 times before you figure out the best strategy? I don't find it challenging I find it stupid. But of course there are people who want this game to be a round based demon souls.

Others though want to relax while playing the game and have a nice smooth progression, so that by the time they finished the game on easy or medium they would say - man, that was one of the best experience in years. Let me turn it up a notch and go for hard difficulty.

And there are also hotties that would want nightmare difficulty too. So what's more important? I believe both are important, but difficulty right now is a bit unbalanced. It seems that no matter the level of difficulty the first few levels are always the most challenging ones, and in some cases almost unforgiving. And then as higher level you get, as more weapons you find as easier the game becomes. Is that not normal?

I think the big thing here is that people over level and have nothing else to do but slaughter everything they can find I the map. I believe when we get access to "another 70% of the game" this will all even out. That's why I mentioned that it is important to let us test the rest of the maps so that we can collect a feedback for LS either anything needs to be balanced. Because no matter on which difficulty you play right now the Beta map everyone ends up destroying pretty much everything of the map by the time they are done exploring which is about level 8. If we'd have somewhere to go and flex our muscles before more powerful monsters the problem of us being overpowered wouldn't be so immediate IMHO.

I don't know what the problem with the lighthouse fight is...

I killed them all in three turns without any losses.

The fight against Braccus court of sceletons in the far east was much harder imo, the hardest in the game. But then I met them at level 5 and the lighthouse party at level 10. Guess why my experience is different than yours. For me the lighthouse fight was really a cakewalk...


But nevertheless I agree with most of what was said here about balancing in general.


Another suggestion:

Again me - as a fan of D&D and of the omnipresent mage as Stabbey might call me - I would vote for more deadly spells that have a low chance of succeeding. Not only for the player but especially for enemies. "Finger of death" or "entomb" are good examples from D&D. If they suceed one character has almost no chance to survive. Party members hit by deadly spells like "turn to stone" of "entomb" were even impossible to bring back by normal means or usual resurrect spells. The biggest con of these spells were that they were hard to cast and in most times the casting just failed. In that case the mage only waisted a lot of time, making decisions more meaningful and combat more determined by luck and dynamic behaviour instead of static "numbers games".

I know that some people hate games based on luck and percentages (this philosophical battle is even bigger in board and tabletop games) but personally, I'm a fan of that to a certain contect especially if it is connected in a meaningful way with other gameplay elements (not being able to save during fights is CRUCIAL to that!). And Divinity: Original Sin already has a chunk of percentages and luck integrated, namely for melee fighters. Every time you swing a sword there is a chance to miss the target. That - very very sadly - doesn't apply to spells and skills so far with is a HUGE hit to balancing. Everyone should be able or in danger to miss the target which means mages and rangers as well with every attack and not only melee fighters.



I fear that it is not even balancing we're talking about here. Balancing would mean that you have working system with the wrong numbers. But I fear that we have (partly) broken or non-functioning systems. In that case numbers don't matter because you would never be able to balance the game properly, to neither hardcore fans nor casual gamers... :-|

Having to die 15 times in each and every fight or having to spend some time in the game to learn some of its mechanics and the basic enemy capabilities is not the same thing. To me, the latter only means that the gameplay has some depth and, yes, that's pretty much what I'm expecting, and would very much like for it to stay that way. Now if people want to play it on easy so as not to suffer terribly through the most challenging fights, fine by me.

This is a good example, again involving the lighthouse : these guys have been playing the game for quite some time and know its mechanics well ; you can tell that the fight is a challenge for their characters, but they come up with a solution to each new threat, even though they have to consider their next move at each turn. It doesn't seem absurdly difficult to survive it either. I think in that regard the game has a nice vision of what balance to seek between fun and tactical involvment, now they just have to execute it well... :P

http://www.youtube.com/watch?v=dSDJw9mx8vc&feature=player_detailpage

@Clemens

To me a good turn-based game is NOT based on skill or learning the mechanics. "Skill" in that respect would only mean that you know how to beat the numbers. It's not about being fast or having good reflexes like in first person shooters. There is no skill in games like D:OS. To be able to react with what you have to a challenge ahead of you isn't based on skill but on basic intelligence.

As I said in my post before, I think that the game should offer a lot more randomness. A game like D:OS which is based mostly on number games and simple causal combat mechanism (actio->reacio) is somehow "dead" to me. You beat every enemy by the same patterns or different set of patterns based on their resistencies and stuff.

I think a good game with some "challenge" here should mean that you have to die in hard fights. Maybe two times, maybe ten times, maybe you make it in the first attempt. Since many stuff is based on random principle you can never know how it plays out in the end. That's the basic principle of many board games as well, at least the entertaining ones. They combine tactics and strategy with luck. A game which is only based on one of them is either too chaotic or too boring and expectable...

I completely agree. I personally hate games like Dark Souls because I feel the difficulty arises for all the wrong reasons: the player is not provided with adequate information and the mechanics are needlessly challenging. Some people are into that sort of thing, but it's not my cup of tea.

This is not the type of tactical challenge we are asking for. What I was proposing was a level of tactics in the fights where optimal solutions would naturally emerge. Just because optimal solutions exist does not mean that they would be the only solutions. Easy-Medium-Hard difficulty settings appear to be simple adjustment of the numbers of the HP and damage of enemies.

On the easier settings, I would hope that that virtually any strategy would be enough to win any fight. You should be able to just run into battle, attack the enemies in whatever order you want, and ultimately come out victorious. Granted, there would be optimal ways of going about things where you take less damage and win more quickly, but they would not be necessary.

On the harder difficulties, yes, you should have to approach fights with some form of good strategy in order to get through them. That does not mean that there's a particular grand strategy that you have to figure out for each fight in order to survive the encounter. No. The "optimal" solution should reward you with a relatively easy win. Any good/okay tactic should get you through, but maybe with a few potions used or a resurrect scroll. A bad strategy or lack of one should absolutely get you killed.


To a degree, yes. As your characters get stronger, they should feel stronger. By the end of the game, you should definitely feel powerful.

Ideally, you would have something along the lines of:
Level 1, you fight a skeleton. The fight isn't too hard.
Level 2, you fight a zombie. The fight is really hard, but you manage to get through it. Zombies are really scary and you avoid groups of them.
Level 8, you no longer run in fear from packs of Zombies. Cultists are your new nightmare now. Why are they so strong?
Level 15, Zombies, cultists, and skeletons are all things of the past. You can fight off hordes of them whenever those turn up.
Level 15, Demons are a hard fight, but not impossible. You still kind of want to make your way around that group of 5 of them, though.
Level 30, you can kill anything in the game. You walk straight through armies of zombies like they weren't there. Even the higher level enemies like Demons no longer pose a serious threat. Only the huge bosses matter now.

Players should not start out the game being able to kill every monster they encounter. There should be easy fights, difficult fights, and enemies that are impossible for you to kill that early on. Eventually, the enemies that are impossible should become difficult, and eventually be really easy even in large quantities. But, by that time, you should have new enemies that are hard, and new fights that you avoid because you are not ready for them.


I can see this being an issue, but that was not the case for my playthrough. I don't think I encountered a single difficult fight after level 3. At level 5, I was taking on fights with 10 level 8 monsters with ease, and it only went more downhill from there. I didn't actually end up completing the Beta map. I think I have ~60% of it explored.

Although I agree with you, I think Larian has already told us that the later maps will only make it into the game with its release.

Allow me to disagree with you - I think there is pretty much skill involved in D:OS: ability and skill to think strategically, to perceive the weaknesses and strengths of your enemies, to solve riddles and so on^^. I agree, those are based on basic intelligence, but not each and every player has this intelligence! Otherwise, everyone would also be able to play chess... which clearly is not the case. (In contrast, in my personal opinion: there is not real skill involved in first person shooters - they are based on brainless smashing of buttons and for me there is no skill involved in being able to hit the right button when there is a red prompting on the screen "press 'F' to finish your enemy")

As you mentioned in a previous reply, the argument is largely philosophical, but I could not disagree more with this.

Luck based mechanics that determine whether or not you win or lose fights are the most frustrating and unrewarding things you could put into any type of game.

There is nothing worse than going into a fight 5 times and die each time all while knowing you made all the correct decisions but just got really unlucky. Then, when you finally win the fight, there is no sense of satisfaction since you know that you would have probably lost if your 5% chance to insta-kill spell had not succeeded.

I do agree that a game with challenge should mean that you die in some fights, but those deaths should not have anything to do with luck. Whenever you go into a fight, you should be able to win 100% of the time if you make the right decisions, and all the information to form those decisions (enemy abilities, resistances, etc.) should be handed to you upfront, not through several deaths of collecting information.

I would much rather have games challenge me to think through my actions before doing them than challenge me to beat my head against a wall multiple times until I am lucky enough for it to give way.