Telling which of your weapons is more powerful in Baldur’s Gate 3 is a bit different from most games. BG3 leans into the Dungeons & Dragons 5e source material heavily, which means everything works off of dice rolls. In combat, these dice rolls are invisible, but each action the player takes is based on random numbers generated from a d4, d6, d8, d10, d12, or d20 die.
Since the damage potential in Baldur’s Gate 3 is so randomized, it’s hard to tell which weapon is better to equip. Most games have a flat number or range for how much damage a weapon can do per attack (and player level can factor into it as well). Here things aren’t so clear cut.
How do I tell which weapon is strongest in Baldur’s Gate 3?
Each weapon in Baldur’s Gate 3 will have a stat that tells which die is rolled to calculate damage. In the early access version of the game, the weapons we’ve run into so far can have their attack power based off a 4, 6, 8, 10, or 12-sided die.
BG3 doesn’t necessarily have weapons that are more powerful than others. Sure, a sword that uses a d6 die has more potential to do damage than a bow that uses a d4 die. However, the law of averages means that over time, the sword’s advantage over the bow won’t be significant. Instead, weapons are better or worse than others based on their qualities, weapons skills, and range (though fighters will want to concentrate on raw damage output). Some rare weapons may let players use special attacks, which makes them better than a common variant that does not.
There are some exceptions to this rule, though. A few weapons calculate damage based on two dice. The Sword of Justice, for example, is a rare two-handed melee weapon that uses 2d6 dice, which can give a player a maximum role of twelve. Only a few arms in the game calculate damage like this in early access, and they are the most desirable.
Other factors play into choosing the best weapon in Baldur’s Gate 3, like damage types. However, at a basic level, the higher number dice the weapon uses, the more damage potential it’ll have.
