Larian Studios Forums Baldur's Gate III Baldur's Gate III - Suggestions & Feedback Compilation of community feedback and suggestion - How to Improve combats in Baldur's Gate 3

As a Note, Longswords are Versatile and even list the one-handed and two-handed properties but players can't pick which one. It defaults to two-handed unless a shield is equipped leaving it impossible to dual wield.

Also for the Ranger class in BG3, the class is listed as having martial weapon proficiency but upon equipping a martial weapon, like the longsword you acquire during the tutorial, you receive no stat bonuses for the weapon. This means that even thought the class has the proficiency listed, it isn't actually being applied in game.

Also the entire system for how combat works could use a better description and explanation in the tutorial. What all factors into hitting and dealing damage? At 100% chance to hit on an elevated ground above an enemy, should in theory, hit guaranteed. Instead I've noticed the game has a chance to Critical Miss, which implies that dice are coming into play. I love that the dice rolls are being implemented and used, but if they are going to be then they should be done as 5e has it. The system is already built for it. There is absolutely no reason for percentiles to hit if the dice are being rolled. Also showing the roll, just like in ability checks, would be a major boon. This would allow the player to see what exactly is going on. Instead of percentiles you could just give the player advantage or disadvantage in certain situations and set ups.

SURPRISE ROUNDS. I did not see this specifically enumerated in the compilation post, so here's a suggestion to include it. Surprise round mechanics are currently unpredictable - sometimes when you engage from stealth your party will each get their chance to strike before joining the initiative order. Sometimes on the very 1st attack initiative is rolled, regardless of being in stealth or visible to enemies, resulting in a sort of "Reverse Surprise Round". This needs to get worked on.

Another problem with combat mechanics & initiative - when two or more characters "share" an initiative value, and you end the turn, it sometimes skips characters that have not yet taken their action. This needs to be reworked so that each click of the End Turn button ONLY ends the turn for the active character, not the next character who just happens to be on the same initiative value. This has resulted in me manually selecting the characters who share an initiative to make sure they all take their actions, which is unnecessary and wrong.

Completely against all these suggestions, expecially auto-increse to hit chance after miss (seriously?). Missing is not fun but is part of any D&D game ever released, expecially on early game. Hell, of any percentage-based rpg ever existed. That's why sources of advantage are an important gameplay component: to turn the odds in your favour. What's the point then of flanking, blinding your enemies, bless your party, using the help action. Eventually with all attacks hitting devs will inflate monster HP to prevent monsters from being ganked to death, because attacks can and will carry powerful riders on top of them, like smites or stuns to disable an opponent (and why not, increase spell chances because is not fun to lose a spell slot o special feature because the monster saved). Before we know we are going to spend more time grinding huge health bars with the same amount of resources

Im up for a general decrease to monster AC or flat increase to hit chance in an Easy or Story difficult level, but Normal mode should stay as D&D Core is. Dont fix somethin aint yet broken

I didn't write it but of course, all these suggestions are meant to be a part of difficulty levels or options.
Solasta recently announced that they're going to add an option to enable a "cheated RNG" (+x% to hit each time you miss), but of course D&D is still the basic balance.
I mean... Why not ? I'm fine with missing (a lot), but I can understand that some aren't.

Advantages are a huge problem in the game at the moment which drastically reduce the replay and the tactical value of the game.
As I wrote they should probably stick to D&D about advantages and why not, give players choices to increase our %to hit through something else (i.e +2 attack roll if you're higher or if you "backstab").

D&D have another defensive system called "cover" that increase AC.
BG3 could have another offensive system through high ground/backstab/... I'm fine with it, but it should be balanced. Advantages for highground/backstab is not.

Cover is a very nice feature, like flanking reward tactical gameplay and will definitely work around high ground bonuses (be if flat or advantage). Say a goblin is hiding behind a low wall, half cover means +2 AC. Getting above enough should negate the cover bonus while keeping realitively safe a ranged character


then an action rpg may be more suited for those who so vehemently hate missing due to how statistics based rpg works. As a personal experience giving too much flat modifiers to d&d 5e eventually breaks the balance, as this system is NOT designed around abuntand flat increaes to either AC or to hit

You're right and I like cover, but it won't happen because it would mean redesign all maps. We have to deal with backstab/high ground alone.

I don't think we have any opinions to give about who should play this game or not. That's Larian's job and it looks like they want to please people that don't like missing too often.

Not sure why a +2 bonus would break D&D's balance, but I'll be glad to hear it.
(That's a suggestion I read a lot on forums from D&D players to replace the balance issues of highground/backstab advantages).

Read about bounded accuracy online and the difference between 5e and 3.5.

Adult Black Dragon has 19 AC in 5e and 27 in 3.5 (Great wyrm in 3.5 has 42 AC and highest AC in 5e is Tarrasque with 25 AC).

Single +2 bonus doesn't break anything, but being able to stack more of them does. Stacking is a thing in 3.5e, bounded accuracy is a thing in 5e.

i do however have an opinion about not wanting a system designed around missing and making tactical decisions to improve my odds to be warped for increasing audience.


A single +2 is not going to hurt anyone, two of them already push too much the limits of 5e. Here is what bounded accuracy is about in practice and what it means



If larian wants us to hit more often they should work around baseline bonus: +1 to base proficiency or -1 to all armor/difficulty classes means you will hit more often without breaking progression or even better without reworking all features to avoid godlike scenarios. But again failure is a big part of D&D experience, why someone would get rid of that is beyond the gaming me.

EDIT: a 'D&D standard' difficulty can be made without all these modifiers for us tabletop hardliners, while all others can enjoy the 'Normal' setting Larian want to push so hard

TBH hitting more often belongs to easier difficulty levels like story and easy. If PCs get +8 (story) and +4 (easy) to all d20 rolls (both combat and dialog) and their spell/ability DC, that should solve the problem easily.

Bit of a pedantry note, but the pen & paper rules as written do require both weapons involved in 'two-weapon fighting' to have the "Light" property tag, which the longsword doesn't have. Shortsword, scimitars, etc., yes. So that is actually accurate, unfortunately. (p. 195 in the PHB, if you'd like a citation) The only way to get around that is the Dual Wielder feat, which idk if they've implemented?

Depending on your interpretation of the Dueling fighting style (and whether it can be used with a shield in the off-hand or it's meant to be Errol Flynn-style sword and flourishy bare hand), it could very much nerf using a longsword with that fighting style, though. So there is still a case for allowing one-handed wielding without a shield, I'm just not sure how you'd flag it so you can't dual-wield unless you had the feat, assuming they put that in.

(Separate argument over sticking to RAW vs. common house rules/QoL changes not pictured, though I'm sure it's ongoing in other places on the forums)

I think thats the best solution when incorporating a story and easy mode.

I got an idea thats may or may not be so good but its a break from the either hit or miss results of combat. Why not have glancing blows or attacks that land but have a penalty to the damage roll. So for example, target enemy has an AC of 16, you have to roll 16 or higher to get a damage roll yes as standard and anything lower than 16 is a "miss". But why not have a pocketed area of...

If roll (4 ) under target's AC, so 15 to 13 in this example, attacker still hits and rolls for damage, but is a "glancing blow" and damage roll suffers resistance penalty. So attacker rolls 15 on attack, gets the "glancing blow", rolls damage dice of 6 but only does damage of 3 from resistance penalty. Any character based resistance penalties would also stack. So if defender also had a resistance to the specific attack, it would bring that damage of 3 to 1 (rounding down from 1.5).

I like to role-play the term "missing" not so directly as in "you completely missed the target" but much more along the lines of, you rolled too low and really did miss the target, or you didn't roll high enough but did a "glancing blow" or such and such for the name.

This makes it so there's atleast a minimal flow to combat. Instead of a hard hit or hard miss. You never know, a few glancing blows maybe all thats needed to turn the tide of battle.

I have a problem, that whenever I change companions, the new ones aren't grouped and end up, where they were the last time, when out of camp. That is not always a problem, but when you are in the hostile goblin camp and the companions end up in the tea house, there is no way to unite them.

Edit:even when I regroup before leaving camp, this bug happens. In this case it was Astaruon I needed, but that's just not happening. He was last in my group when we were at Ethels and so my main Char and he end up there, while Gale and Shadowheart are back were they should be

I hate to say it, but this will never be a 5e game unless they go back and fix the action economy. More actions is great, but there should be a consequence to your choices and it shouldn't be watered down. Hide, dash, disengage and dodge should be full actions, you make a choice to engage in these knowing you sacrifice your ability to attack or cast a 1 action spell in return. Later you can pick up feats or class abilities that can mitigate the action cost, but often at the cost of a class resource (spell slots, ki, limited times per long/short rest, etc) or using an ASI for a feat like mobility. Actions like grapple and shove are part of the attack action and should be treated as such, if you chose to attack shove should be an option there. If you have multiple attacks you can use one of them to perform a shove (or grapple). Doing so you make a choice, do I want to give up potential weapon damage for gaining advantage on the second attack or simply a chance at pushing the enemy off an edge. Making it a bonus action breaks the game by removing the choice and giving the player the ability to do it all, leading to the shoot and hide from elevation BS style of combat.

After seeing so many comments about Solasta on these forums I decided to check it out. I think Larian has a better story, better characters, and better artwork overall, but Solasta is far better in terms of combat, tactics, and being true to 5e rules. Solasta lacks my favorite classes, the ability to dual class, and so many features that I love about 5e that BG3 has, but I still had much more fun playing Solasta than I had playing BG3 (including some of Solasta's game breaking bugs). Where Solasta adds a custom skin to 5e mechanics to fit the story they are telling, BG3 has made a 5e skin and wrapped it around the Divinity engine and called it 5e (it really isn't). One is fun to play, one is a chore unless you play in a very specific style which trivializes most combats.

Honestly I'm hopeful that Larian pulls it together and makes a wonderful game, because I've already paid for it. However reading a number of the suggestions on how to improve the game have made me more pessimistic on whether this will ever amount to more than a crippled DOS in a 5e wrapper.

This is a great point.

Can you return Early Access games?

This game is so poor, the lack of engagement so obvious, that I honestly want my money back.

I think it's important for the consumer to know there are risks when buying a game in EA. There are risks when buying a game at full release.

I personally don't like returning items just because it didn't meet my expectations. I should have done my research. I return when it is defective. I definitely wouldn't return if I used the product for many weeks. That to me is unethical.

I agree when it comes to a full game, this is the first time I have bought an EA game, and honestly even with that knowledge, this is a poor showing. This is an Alpha, a tech demo.

Party Movement
5e Implementation of Combat
Resting and Resourcing
Race/Class (I can excuse this, as it is not a complete game.)

But worst of all?

There should be constant communication. Get a community manager, I dont mean get the Devs on the forums, but get communication going, every day. Ask for focused feedback on these fundamental pain points.

I paid full price for an Alpha test. Thats not cool at all.

Oh I agree, this game has not met my expectations. But, it's not an awful game. It's just not what I expected from a D&D game of a sequel to a legendary CRPG.

Just hope that due to the holidays, the devs have been silent. Or they are working on a lot of fixes so it will take a lot of time.

If there was communication, I wouldnt feel so bad, but I absolutely feel they took advantage of the name Baldurs Gate, and sucked people in.

Full price for this? Shameful.

You did so in full consciousness, in an informed and completely voluntary manner.
You have been repeatedly warned about the condition of the product and were fully acquainted with its shortcomings.

And now you're complaining about them ... I agree with you, not cool at all.

Only part of that is true.

I was aware it was EA.

I wasnt aware that the Combat system would be a partial (at best) implementation of 5e.
I wasnt aware that the rest system would be a mess.
I wasnt aware that the class/race options would be incomplete.
I wasnt aware that the movement system was a travesty.
I wasnt aware that they are implementing spell balance based on analytics.

EDIT: And I wasnt aware, that there was a complete lack of meaningful interaction with the development/design team for what is presumably Early Access, to get that interaction and feedback going.

My first time, and most certainly last time, of paying full AAA price for an 'alpha state' EA, I can tell you that much.