Larian Studios Forums Baldur's Gate III Baldur's Gate III - Suggestions & Feedback Larian: It's ok to let chars always win/lose rolls

Larian,

I've played this game far more than is healthy for a pre-release, and I've really enjoyed it. There are a few design decisions you've made, though, that I think seriously detract from the game. Mostly, these center around ways you have "homebrewed" D&D 5e. However, rather than just making a "PLZ FOLLOW 5E GG NO RE" argument, I thought I'd give some reasoning as to why it's important in these cases:

I. Skill Rolls: Automatic Failures and Automatic Successes
These need to go. In any worthwhile progression-style game, there are challenges that are either too hard for a character to currently complete, or so trivial that an advanced-enough character should automatically succeed. D&D 5e recognizes this. In 5e, there are only two times that a roll can be a guaranteed success or failure:

• Attack Rolls
• Save vs Death to avoid dying

That's it. And that's ok.

Consider a character with dexterity 8 and no slight-of-hand proficiency who is trying to lockpick a DC 25 door. Why should they automatically succeed on a natural 20? This character should either try to bash the door down, or they should find another character with the skillset required. At the same time, consider a character with 20 dex, guidance, and slight-of-hand proficiency wearing the smuggler's ring. Then they should never, ever fail the DC 5 lock picking check. They are too good. The roll is beneath them. It is extremely frustrating to see a failure when it neither mirrors the 5e rules nor what people experience in real life. Would the Lockpicking Lawyer ever fail to open a consumer-grade masterlock padlock? No, he wouldn't. It's beneath his skill. As a player who raises a champion to be explicitly good at a specific task, it is extremely frustrating to see my <whatever>-skill-focused champion fail at a menial task. It is not fun. It makes me wish I had save scummed beforehand for the explicit purpose of bypassing this unavoidable failure. And make no mistake: A 1-in-20 chance, also known as 5%, is a huge auto-failure/success chance. Most importantly, it completely removes agency from player decisions.

II. Ground Effects
I know Larian loves their puddles of death. It's a tradition. However, these puddles hurt the game right now more than they help. For example, let's examine the level 1 spell Chromatic Orb. In 5e, it does a flat 3d8 damage + 1d8/spell slot level above 1st. The only difference between the orb types is the damage type. It's designed to get around resistances or trigger various effects if hitting an applicable target. However, Larian has changed it. Now, it only does 3d8 sonic damage. The other types are all 2d8 and leave a puddle. This does a few things:

• Reduces the up-front damage of the spell
• Makes it unusable (or less-wisely used) with allied party members in melee
• Greatly increases the power of the spell in certain circumstances

For example, compare the power of doing 3d8 damage to doing 2d8 damage and knocking a target prone. There is no comparison. In almost all cases that aren't "killing blow," prone is much more valuable because you provide advantage on melee attacks, reduce the enemy's movement speed on the following round, and remove the enemy's action. In fact, 2d8 damage and prone is more powerful than most second-level spells. 3d8 + prone certainly is.
On the same ticket, if a player wants to fire chromatic orb into a melee between friendly and enemy characters, they are forced to use the thunder orb. Otherwise, they risk burning, poisoning, knocking prone, reducing armor, or shocking all members of melee, not just enemies.

Additionally, puddles don't currently go away. As a general rule, if it's not burning (and even sometimes if it is), we have De Beers puddles: They are forever. If allowed to exist at all, a puddle should only last a few rounds. Maybe 1 round + 1/spell level. So a first level spell would have a 2-round puddle -- just enough to get some use, not enough to turn an entire section of the battlefield in a death pond.

There's also the problem of inconsistent and unpredictable puddle creation. For example, Melf's Acid Arrow is a balanced spell in 5e. It has guaranteed damage to offset its attack roll. In this game, though, it also creates a huge acid puddle giving -2 armor to all characters in the puddle. And again, the puddle never goes away. In some cases, this makes Melf's unusable since it might get Lae'zel or other party members killed in melee. In other cases, it makes Melf's extremely more powerful than it otherwise should be since all of your ranged attackers will probably not miss targets in that area. And unlike chromatic orb, Melf's baseline damage has not been adjusted to compensate for this buff.

In my opinion, if Larian wants spells that create surfaces, they should make a new line of spells that do reduced damage but create the desired surfaces. "Create Acid", et al. On the flip side, they should stop trying to "fix" the 5e spells which are quite balanced in 5e, but extremely broken in BG3. I mean, a cantrip that knocks opponents prone because there's blood underneath them? How is that ever a good idea?

Conclusion
I really am enjoying this game. I made this post because I see things that could be, in my opinion, fixed to make it even better. I'm really excited to see what the full version has in store for us next year.

I just hope that fireball doesn't also create a giant burning flame puddle so that my evoker does damage to the party even though he has spell shaping.

Plenty of worthwhile RPG systems have critical misses and critical successes. A lot of systems care about consistency of what rolling specific number means in all cases. 5e cares only for those two specific things you mentioned. But why? They never offered an explanation AFAIK, why that specific distinction was made. Why is lock-picking failing 5% of the times different from an attack failing at the same rate? How can a lvl 20 Fighter miss when trying to cut a piece of paper?

Seems arbitrary to me.


A character that has just 5% chance to succeed at task shouldn't attempt it, especially if he can do it only once. And you are right he should try a different approach. If they suceeded, doesn't mean that they were right to try. And a character that has 95% chance to succeed at task given is the right guy for that job. But sometimes even the right choices might not result in success, both in DnD and in real life. I



Athlete failing at running.

Humans are fallible, there are plenty of examples of skilled specialists failing when under pressure to do a thing they should never fail at. Sometimes people forget details or words that they should definitely remember. Sometimes a tool might break or a hand suddenly cramp. The 1 in 20 might be too high but because of the nature of d20 it is the only way to represent it. And for roleplaying reasons it's better to have this mechanic, either at the table or in video game.


It does not. If a task, that you are trying to accomplish is important and crucial enough to warrant a worry about failing it 5% of the time, there are some ways to give yourself advantage reducing that risk to 0.25%. Or on a failed roll you may decide to use your inspiration pool. It does nothing to remove agency, it actually enhances it. And as a player you can even enable weighted dice, so that 1 in 400 failure does not happen.

In usually prefer flat checks in computer RPG - that said Disco Elysium used skill rolls extensively and it felt great. Josh Sawyer (Obsidian dev.) mentioned it in his reponse to BG3 checks. I think I too prefer bell curve, as extreme successes and failuer are rare instances indeed.
https://twitter.com/jesawyer/status/1326718682412720129?lang=en

Eh, when it comes to puddles, I am quite sure there are already mods to remove those from certain spells. I don't like how puddles are implemented - they adhere to D:OS2 ruleset, not D&D ruleset, feel out of place and don't play nice with other mechanics, like concentration.

They also really mess with evokers. Even though an evoker is supposed to be able to shape their invocation spells to avoid damaging allies, the giant burning puddle on the floor says otherwise.

With all the respect to Mr. Sawyer that tweet is misleading. Better way to illustrate would be this - when a character with +7 to lockpicking tries to open an:
- easy lock he will succeed 19% more times compared to times he would succeed without those modifiers.
- medium lock he will succeed 64% more times.
- hard lock 116% more times.
- very hard lock 700% more times.

I think we are all still holding out some hope that Larian will listen to this in the time they have left before release.

I think auto success or fail should be based more on common sense. Like I'm playing a Rogue/Thief with a background as a criminal. How is it possible for her to fail at a level 5 lock? Sure it is rare, but its still there and feels really dumb. With that said I also really don't like rolling in combat. I know its a D&D thing, but I hate missing an attack when I am standing right next to someone. I'm always like, "How the f**k did you miss that!? Are you blind?"

Somewhat off-topic, but this seems like a good opportunity to once again advocate for different animations (or at least different words than "missed!") when you miss an attack. If you don't meet the target's AC on an attack roll, in reality you don't damage them because of some combination of:
- they dodged
- you hit their armor
- they blocked it with a shield
- you just missed
Currently, BG3 implies that everything falls into the last category, which can be disheartening and boring. You're not incompetent; you're fighting an enemy that is actively trying to (and at least sometimes succeeds in) avoid or block your attacks.

An extremely basic fix would be to simply change the words depending on the armor:
- "missed" appears if you roll less than 10 (roll + modifier)
- "dodged" appears if you roll greater than 10 and the enemy is wearing light armor
- "blocked" or "absorbed" or "0 damage" appears if you roll greater than 10 and the enemy is wearing heavy armor

There is expression in Czech: "I mistr tesař se někdy utne."
Google translates it as: "Even a master carpenter sometimes cuts off." so i can only hope its not some bullshit.
And i believe that is the case with nat 1 in skill checks ... so i kinda understand why Larian wanted to try it.

Dear Larian, we tryed ... and it sucks.
There is not much else to say, those 1 are droping too often to be rare enough to be even remotely concidered entertaining.
It just sucks.

---

As for the other part of the topic ...
I have mixed feeling about it to be honest ... i basicaly agree with everything OP said ...

On the other hand, imagining Chromatic Orb being 3d8 ... as it should be ...
but dont cost any rare material to cast ... as it should not be ...
i cant honest imagine much people who would take, well basicaly any other offensive spell at low level. O_o

Some tuning would be in place, that much is certain.

Chromatic Orb at 3d8 needs to hit with 78% rate to be better than Magic Missile. It’s not good enough to be on Wizard’s prepared list. Sorcerers can twin it so it is more useful for them. However, unless you abuse the current targeting oversight and wet condition, I still think that there are better options.

And in PnP you just need to hold a diamond in your hand, but the spell doesn’t consume it.

Not that hard to imagine imho since there are a lot of spells that require material components in DnD and getting rid of that tedious management part for the majority of spells in crpg's is one video game adaption/QoL that I'm actually all for. Besides, Chromatic Orb requires a small diamond worth 50 G to cast. It's not consumed, only required to be in your backpack(probably used as focus for the spell)

Edit: Damn it, Elebhra cut me to it :P
Edit2: Sorry Rag, I misunderstood you. (missed that important "t" on can't :P)

Magic Missiles are kinda special case ... especialy in BG-3 where with only two artefacts you can almost triple its potential damage outcome.

I was thinking about spells like Witch Bolt, or Acid Arrow.

Well. Witch bolt is more powerful, or atleast more economic, over a long fight since it can make continuous damage over several turns without wasting spell slots but is a bit underwhelming, especially in BG3 where few opponents will last the full duration of 1 minute ie 10 turns and concentration is easily broken plus that spell slots can be easily recovered after every fight.

Melf's though....yeah, it sucks in comparison except that it's guaranteed damage even on a miss.

Chromatic orb isnt OP at all at 3D8 if you compare the average damages of all other offensive spells.
(Inflict wound is 3D10, guiding bolt is 4D6, burning hand is 3d6 on an aoe, arms of hadar is 2d6 on an aoe...)

It's just a very good spell against vulnerable or resistant creatures but not especially a must have against common creatures.

2d8 + eventually surface damage is weak imo. Especially if the AI wasnt so stupid to avoid surface damage.

I have to say, I'm with the OP and with DND 5e on this one. If there is a need for "puddles", Larian could always introduce homebrew versions of the Create or Destroy Water spell, which has the added advantage that players have better control over whether or not they want to use such an effect. Also those spells could be balanced individually, which would probably be necessary. In fact, the DND rulebooks specifically state, that such a thing can be done if a certain spell doesn't quite fit your needs and if your DM allows for that. The spells in BG3, which have been changed by Larian to include puddles are better the way they are in the books in my humble oppinion.

As for the skill check problem. Please consider this:

In DND 5e there are certain situations, which make an automatic fail or success (on a natural one or twenty respectively) result in pretty ridiculous situations. For example Vampires have a trait called "Spider Climb", which allows them to climb difficult surfaces (like smooth walls) *including* upside down on ceilings without the need to make an ability check. They have this in their stat block, because an ability check (and possibly suitable tools) is usually required, naturally. So what if the player says: "I want to climb after the Vampire."? Well the DM would have the player make an ability check and have the difficulty of the check reflect the difficulty of the task. If the player was lucky and rolled a twenty, they would now be crawling along a perfectly smooth ceiling, without magic, tools or help, just like spiderman. What about swimming? DMs will often make players take a Strength (Athletics) check for that. What if the player wants to swim up a waterfall? With the auto success rule you'd have to allow the player to actually swim up a waterfall if they did roll the lucky twenty.

The same is true for low difficulty challenges, like that DC 5 lock. That specific example would be such an incredibly cheap and simple lock, that it would serve as more of a message stating "Hands off!", than an actual attempt to prevent unauthorised acccess. A rogue - especially one who is proficient with thieving tools - can open it as easily as if they had the key. Sure, other characters, who have never learned how exactly a lock is built (represented by a skill bonus lower than +4), may have trouble and even fail, if they roll a one, but the rogue is a specialist and simply cannot botch this.

TLDR: Yeah, I'd like the game to stick to the rules as written in these cases.

As for the "puddles" ...
I still believe that best suggestion that was introduced here was making toggable option for every spellcaster that would "sacrifice single damage die in order to create surface effect acording to spell element, for one round (or one round + one round per spell level)"


I dunno ...
Is it even fair to compare spells from different classes?

I mean single person should not have acess to both, unles it multiclass or take specific feats ... you could aswell say that nothing is really strong compared to Berserker in Frenzy with Great Weapon Master ... since your damage potential with regular +1 2H Sword is 2d6+10+1+2 twice per round. O_o

Same with AoE spells ...
It seems quite logical that if you wish to compare spells that can strictly attack single target, you will compare it either with other single target spells, or for single target ... meaning both AoE you mentioned is weaker. :-/

Not doing it would be nitpicking and I would just have to find other exemples to show that chromatic orb is not particularly an OP spell (eventually with higher level spells).

I won't do it, especially with you

So ... in order to show how Chromatic Orb "is not strongest spell" (for example for Sorcerer, since we have that ugly option to learn everything on Wizards) ...
All your Sorcerer need to do is "become Cleric instead" ?

Yeah that makes perfect sense.

Spells for different casting classes (except Warlock whose spellcasting system is different) are *roughly* equal in power. Otherwise, different casters would have different # of slots per day. So yes it is fair to compare spells from different classes.

Alternatively, here are other 1st level single-target damaging spells also on the sorcerer spell list:
-Catapult deals 3d8 bludgeoning but has a larger range (60+90 ft) and is a Dex ST instead of an attack roll
-Chaos Bolt does 2d8+1d6 of a random damage type but has a 1/8 chance of arcing to additional targets
-Ray of Sickness does 2d8 poison damage and can Poison a target
All of these seem comparable to Chromatic Orb's 3d8 damage of your choice in damage type.

Also, it makes absolutely no sense to compare a single-target spell's damage to the damage that an AoE spell would do to a single target; obviously you should assume that the AoE spell will hit multiple creatures (otherwise you wouldn't cast the AoE spell).

I agree with the OP that I’d rather have new spells to create surfaces than have them shoehorned into classic spells. I avoid surfaces in my games, other than “set flamable thing on fire” (which mostly happens by accident). Chromatic Orb may as well be Orb of Thunder as far as I’m concerned.

In theory, it’s cool how surfaces interact. Youtube has shown me tricks I didn’t suspect. Those effects, as illustrated by the OP, can be wildly out of balance. This makes me glad I haven’t engaged with that system so far.