Larian Studios Forums Baldur's Gate III Baldur's Gate III - Technical & Gameplay Problems GWF failed at 7/10

https://www.reddit.com/r/BaldursGate3/comments/jgbqic/gwf_fails_at_70/


I made 10 sequential attacks with Great Weapon Fighting Style which is supposed to reroll 1 and 2 damage dice.

Here are the results:

Attack 1: 1 damage

Attack 2: 7 damage

Attack 3: 2 damage

Attack 4: 9 damage

Attack 5: 2 damage

Attack 6: 2 damage

Attack 7: 5 damage

Attack 8: 2 damage

Attack 9: 1 damage

Attack 10: 2 damage

It is pretty unlikely but if GWF is working then it means 7 out of 10 attacks rerolled 1s and 2s again. Plus, out of 7 fails, 3 of them are consecutive fails.

Now, do you really think GWF is not bugged?

it is normal
The best focus - magic
spear is not best weapon

i have attack 6-10 with dagger all - but i wizard
str 16 is modificator +3 or 4 damage
i have stat 16 str and dex 9 con 16 int 18

i use spell sorcery and have 1 fast attack

WTF are you talking about?
Did you even understand what this post is about?

It's not currently rerolling the dice in my experience as well.

Spear is great when you flank .

You would need to reroll more samples cause even if it's unlikely, the 1s and 2s could still have been rerolled into 1s and 2s. Bad luck we call it.

Hey, as a stats fan I'm currently looking for possible issues with the RNG in general. You can read about it in this thread on the last few pages.
https://forums.larian.com/ubbthreads.php?ubb=showflat&Number=705432&page=1

I don't have a save with that feat at the moment, but would be very interested in testing this if you could upload one. Save games are stored under your documents folder ->Larian Studios -> BG3->Player Profiles -> Profilename -> Saves (they are but a couple megabytes in size). You could naturally also do it yourself, but ten is not a sufficient sample size.

If only there was a flanking mechanics in the game

BTW it is not a spear, it is a halberd.

It is a halberd, "mr.I use magic"

Until someone do atleast 1000 testing rolls, it's hard to say there is a "randomize" issue, but nearly everybody feel there is one so I guess it something to investigate.
Random doesn't really exist in a computer program, it's always related to a "real data", the common one is the actual time when the "random" number is generated, but depends on used code it can be easily badly done.

There is a big difference between feeling there is something wrong and the game having issues on randomness. This is the tricky part with probabilities.

[video:youtube]https://youtu.be/MEewLWDpscA?t=1507[/video]

Yeah, it's exactly what I said. We can't be sure there is a issue until real tests, but the fact a lot of people get the same feeling make the need of testing more relevant

Well, I still invite you to participate in the thread linked to above. Even if every contributor could record just 20-30 rolls, that'd be information. Overall we recorded roughly 700-800 attack rolls in that thread already. If there are bugs straight in the core RNG, they sure are not as obvious as people make it sound. This appears mostly bias, plus a possible confusion with no-damage rolls. In particular if you don't follow the combat log, those visually look almost like misses, when they are hits -- just hits that dealt no damage on the subsequent damage roll. I'd make this far more obvious in the visual feedback.

As for the OP, I take my request as a "no" then. But in my save I'm close to level 4 again, so may pick the feat myself to check. Could very well be a bug, however sample size.