Larian Studios Forums Baldur's Gate III Baldur's Gate III - Suggestions & Feedback Josh Sawyer about the DICE in a low level D&D + HP bloat is no solution

https://twitter.com/jesawyer/status/1326718682412720129


"RNG can be frustrating in general, but the core die mechanic and skill values of low level D&D don't help matters."

I have strong criticism over how Sawyer destroyed casters on nwn2, found the Pillars game very lackluster but was a good game to bring the party based isometric CRPG back and made possible for games like Pathfinder Kingmaker to be made. That said, Sawyer is 100% right about dices on low level 5e game.

For eg, a epic level fighter on nwn2(3e) would easily have a base attack bonus above 30 on nwn2 and multiple attacks per round, so he is mostly likely to hit even an ancient red dragon(AC = 39). And missing would't be a problem since he has a lot of attacks per round.

Now on 5e, the bonus to hit are much smaller and the low level focus make missing extremely common.

Larian, instead of giving hidden bonus or making the RNG more akin to a Bell curve, not a flat distribution, decided to lower AC and implement a high HP bloat in the second D&D edition with the highest HP bloat. 5e only loses to 4e on hp bloat. It only makes all spells fells worthless and the combat extremely slow.

Or could just made the game a higher level. Lets be real. We are adventuring into the underdark on chapter 1, on BG2 you only enter in underdark on chapter 5 and the game starts at higher level. The companions also looks too accomplished for a lv 1 char. If the game started at lv 5, fighters would have 2 attacks per round. And missing often a much smaller of a problem.

Agreed. I'm glad loaded dice will be an option but I will be clicking it off as soon as it arrives. One of the things I've liked best about the game is the d20 and thrill of waiting to see if the roll comes out. And the generous distribution of inspiration points makes me less likely to use that f8.

Starting out at higher level -- something consistent with the stories of the NPCS -- would be the best solution.

Pillars of Eternity was such a successful series that microsoft had to buy Obsidian after the failure of PoE2.
Let's say we start at level 5, what then?
They still have to make the goblins stronger, unless we want to make the game child's play (increasing the number of enemies is not the solution that will work for this game).

The issue is not just the dice and the results, or just the "rng" as if it exists in a vacuum.

The issue is also that the game doesnt provide enough diversity in its consequences to ameliorate for harsh and sudden outbursts of "bad roll". Thats why its a bad roll - that the players save scum - to avoid.
Large majority of instances where a dice roll is performed in BG3 only have an optimal and negative result in how that specific quest evolves from there.
The moments of actual failure that provides other ways to play as very rare and most failure lead only into outright negative consequences.

In that sense it doesnt matter how you calculate the "rng" if the only result is binary.

In smaller side plot situations that is not necessarily bad and can provide some engaging difficulty and necessitate to think outside of the box, but larger more crucial important events of the narrative - that can be played through in the EA - mostly end up as different degrees of straight up failure, either to affect the plot in some specific way or to get important information. The Druid - Tiefling refugees plot stands out as the worst example of this because its badly written and designed from the ground up. It even forces the player to fall on one or the other side of its binary nonsense by forcefully suppressing quite ordinary options players otherwise have. The problem in the trial is not the that dice rolls are high or difficult, its that the options and the results of both failure and success are completely incoherent insultingly simplistic garbage.


The other side of this coin is that "rng" is not supposed to be just random.
The character skills, leveling up, advancements and any gear and items and information the player can find - not only should, but must affect the outcomes. So it cannot just be fing "random". As if the problem is only between two binary options, having complete "randomness" or its opposite, complete determinism.

And that itself is a problem coming from far deeper issues we have as biological organisms that evolved on this planet. That stupid limitation, the tendency to think in opposite, mutually exclusive binary extremes.
Thats the reason why almost all discussions about "rng" devolve into two opposing sides each stupidly holding for their own binary extreme.

To be clear, higher level is the OP's solution.

Sawyer has a different one if you follow the link. You could even respond to him because he's tweeting about this right now.

edit: that was in response to @rhobar121

Yeah sure,

I didnt adress that anyway. Higher level start would be a bad solution to a completely different problem, and it would not work at all, in fact it would only create strings of other problems.
Its like trying to apply a sticky tape solution to a software problem. Or a problem of chemistry and biology by banging it with a hammer.

I already wrote about this in my post about writing of the start of the game, and this is a consequence of trying to start with an epic bang - which clashes with several features of actual early gameplay.
No, we dont need to start the game on a farm - that would be the opposite extreme again.

But this specific start does need to be adjusted, improved and expanded to allow this early gameplay to breathe for a while. And the issues with so called "rng" rolls would be best solved by providing expanded options and C&C where getting a dice roll failure would not automatically be the worse option. A lot of smaller quests and situations have a really nice diversity of outcomes - but for some reason the most important quest lines - that should be standing examples of this - dont. And its these first major quest lines that should serve as largest attractions and examples of density of gameplay options. Thats what the people should be talking about when talking about BG3, not the small side quests.

Stats and proficiencies don't impact skills enough, that's for sure.

Someone who sucks at something can succeed while someone who is supposed to be "best" can fail at the same check.

The modifiers should be doubled at least.

Start at higher level was just ONE suggestion that I gave.

After a miss, giving +5 hidden bonus to hit, for you and enemies on next round is the best solution. It would make missing 4 times in a row mathematically impossible, and if you have 50% chance to hit, you will never miss 3 times in a row.

It sure is funny how combat in Solasta still has lots of missing, yet the missing doesn't feel like much of a problem. I think it's because turns go by faster, and so missing doesn't feel like as big of a deal, because you'll get another chance again soon.

I'll say that catering to people who moan about missing does not seem to have not been good for BG 3. Missing still happens, but because of the higher HP of enemies, hitting doesn't feel as good either, and the unintended consequences have messed up the balance.



Please explain your reasoning why the goblins would need to be made stronger and why there can't be more enemies.




That's kinda what Passive skill checks are. Most people have only heard of Passive Perception, but you can really do most skills as passives. A Passive skill check in 5e is simply 10 (median of a d20) plus or minus your modifier to that score. If you have +3 to Perception, you succeed on any Passive Perception checks of DC 13 or lower. If you have -1 Intimidation, you'll auto-succeed on Intimidation checks of 9 or lower.

The problem here is we're not really mad about RNG we're mad about the results. You constantly feel like you're losing for failing rolls and you're simply getting the plot you're owed for succeeding them. There are a lot of rolls that really don't change much, there are some that dump you into combat, and then there are some that straight up fail quests or give you really bad results (Nettie and Kahga).

I think Josh is getting too focused on the RNG of it, Disco Elysium would be fun to play with 1d20, 1d100, or even having that stupid crap from Undertale popping up for you to pass checks. The core issue here is how they're handing rolls, what they are making rolls, and what we get for failure. If all you do for failing rolls is close off story, and all you do for success is give me the story of course I'm going to save scum.

First of all, increasing the number of enemies significantly increases the waiting time for a turn. Already in some fights this is a problem, guess what will happen as we add more of them.
Even if they fix the AI, which is not so sure, the wait for the turn will still be too long.
The fights in turn-based games are always the most enjoyable when fighting a small group of strong enemies. However, large swarms of enemies are irritating because each of the enemies has their own turn.
What's even more of a problem with BG3 is that bad rng can make your enemies easily overwhelm you.

Secondly, if you increase the chances of hit, you need to strengthen your opponents or significantly increase their number (see point one).

Thirdly, if you give the player fireballs at the very beginning (let alone other classes), you automatically trivialize a large number of fights, unless you strengthen enemies or add more.

Josh is of course right that minuscule influences of characters skills increase ultimately create the sense that character progression is almost worthless.
But of course, the problem isnt just in increasing their influence numerically. If the gameplay is more variable and less binary, then binary results of dice rolls dont create such extreme binary problems. You cant remove the binary result of success / failure through numbers alone - if thats all the content gives you anyway.

You dont want to make every situation complex and variable. Some situations must be binary to make any sense. But that should be a minority of situations in the game. Just like they are in real life, only obviously bent a little because a game is high fantasy or a similar quasi realistic setting. And the only way to do that is to respect the setting and lore and characters and then write-design the situations and their possible permutations around those features, as it befits each situation.
So far BG3 did a pretty good job with smaller side quests and situations. But the main plot thread needs some serious redesign and improvements in this sense. From Nautiloid content onward.



edit:

As for combat, the worst solution is to increse hp of enemies.
Increasing their number is far better solution - and no, it will not automatically lead into "more waiting" because you will be able to deal with those enemies quicker.

Overall, the problem of "waiting" exists only in minds of players who got too accustomed to real time games. It doesnt really exist in TB games as such because time does not run linearly in TB systems.
But, some minority always makes a lot of fuss about it - despite it not making any sense. Enemies turns shouldn't take too long of course, but the current effect of slowness is there only because the game is in very early stage.

Ive just fought all the goblins under the windmill. Every time the enemy would move i wasnt "waiting" - i was thinking how to recalibrate what i am going to do on my turn and a few turns after that, which continuously changed depending on what each enemy would do. TB is for tactical consideration, making plans from the start of the combat that work or not, reloading and trying different positioning, etc, etc, not for speed and action.

I just putting out there that I'm not upset by the RNG at all. I like it and I really hope any proposed solution to the RNG 'problem' is optional.

The d20 is one of the best things about BG3. I'm being entirely serious.

Okay, those are different issues than "RNG is unfun".

The issue with Nettie is that 3 DC 10 checks of the same thing, where any one failure is a complete failure is not a DC 10, but actually a DC 17, which is itself pretty high when you're only likely to have +5 total to your roll. Kahga is the same kinda thing: You have two options, one is DC 18 the other a DC 20.

The issue you're complaining about is that the high DC of checks with few check options feels like being railroaded onto the path of failure.





1) Part of the issue with fights feeling too long is because the enemies have higher HP, so they stay alive longer. Additionally, more enemies with fewer HP will be more susceptible to AoE attacks or HP-targeting spells like Sleep and Color Spray.

2) No, you need to DECREASE the current chance to hit, not increase it. Larian DID increase the chance to hit by lowering enemy AC, and buffed HP. This broke the balance for all spells which rely on Saving Throws, not attack rolls. (Such as Fireball.)

3) ...And the problem with this is what, exactly? Yes, if you have fireballs, you need to have fights which aren't trivialized by Fireball. I mean, "if you start the game at level 5 instead of level 2, some balancing will need to be done is a "no kidding" kind of thing.

Not only Solasta has way faster animations but you can reach lv 5/6 on solasta EA.

So, you can have two attacks per round and cast spells like fireball on solasta. As a lv 5 shock arcanist(homebrewed class), it can dish 9d6 damage at lv 5+1 dice due the subclass), you also don't fight enemies with high AC or HP on the beginning, so each hit counts.



My point is that if Larian believes that missing is too bad, they could

• Made 2d10 or 4d5 instead of d20, making rolls more akin to a bell curve (I know that d5 doesn't exist on P&P but a computer can simulate it)
• Made after a miss, a hidden bonus on next roll for you and enemies
• Putted enemies with low AC on the beginning
• Homebrew powers that increase chance to hit
• Start at lv 5 where you can attack twice per turn and cast less RNG based spells
• Implemented concurrent turns to speed up the enemy turns
• (...)

But the low AC / high HP is IMO the worst "solution"

1) Provided you have the initiative, even then you will not always be able to reach many enemies with AOE.
Depending on the rng, the fight can quickly turn into a one-sided slaughter or a long boring fight.
As for the fact that hp based spells are weaker, well, we're dealing with smaller groups of enemies anyway, so even if it only works on one or two enemies, it'll be fine. Besides, the casters are already too strong because of the unlimited rest. The last one is unlikely to change much (or at least not enough to make a difference)

2) In this case, we will probably not come to an agreement. I still think the chance to hit was needed, hardly anyone will have fun dying by a bad RNG. Missing attacks (especially multiple attacks in a row) can quickly become flustering. I put in an excerpt from an interview with Josh Sawyer from the PoE production days where he talked about missing (and how they limited it in poe). PoE was an RTwP game where it is not as noticeable as in full turn-based games. So, Larian's fears are not completely unfounded.

Hmm, I don't think it's actually that different between 3.5e and 5e once you're high level. AC is universally lower in 5e compared to 3.5e. Recall also, in 3.5e your subsequent attacks each take a -5 penalty (i.e. a lvl 20 Fighter BAB is 20/15/10/5)

A 3.5e lvl 20 fighter with +(35/30/25/20) AB vs. 39 AC = hit chance of 85% / 60%/ 35% / 10%

A 5e lvl 20 fighter with +14 AB (x4) attack vs 22 AC (Ancient Red Dragon) = 65% / 65% / 65% / 65%

Despite all the talk about bound accuracy, all it's really done at upper levels is change the meta of character optimization. Whereas in 3.5e, martials stacked AB and AC to infinity, in 5e your AB/AC bonuses are more "bounded", but you simply find ways to ensure advantage as much as possible. Now in BG3, things are out of wack because getting advantage is too easy to get. But that's more Larian's issue, not D&D 5e.

Everything is just more scaled down in 5e. The real consequence of that is a group of lower level enemies can pose more of a threat in 5e than 3.5e, since they might still have a realistic chance to hit you, etc.




While I do not love Larian's solution - what they have done IS essentially flatten RNG with a "Curve". They simply gave us easy access to the advantage/disadvantage system, which applies a more extreme curve.

I do not agree with this solution due to the side effects of such a change (i.e. it messes up tons of ability and classes), but reducing RNG does seems to be exactly where their mind is (they just did a subpar job implementing it).


People have exaggerated how much they've adjusted AC/HP (let's use Goblins for example). Are their slightly HP higher than what they should be in the MM? Yes, but it's well within the range of other CR 1/4 monsters in the game. Instead of Goblins, we could've gotten any of these CR 1/4 monsters:

• Albino Dwarf Warriors (Tomb of Annihilation) - 30 hp, 13 ac
• Kenkus (Monster Manual) - 13 hp, 13 ac
• Giant Frog (Monster Manual) - 18 hp, 11 ac
• Ash Zombie (Lost Mines of Phandelver) - 22 hp, 8 ac
• Abyssal Wretch (Mordenkainen's Tome of Foes) - 18 hp, 11 ac
• Blink Dog (Monster Manual) - 22 hp, 13 ac

There is more, but the point is, facing enemies with 12-15ish HP isn't game breaking. Spells are not balanced only around 7hp Goblin campaigns.

Their AC also isn't horrendously low (aside from the few 8s and 9s we see on the new spellcaster goblins) - since the base Goblin AC is 12. They only get to 15 if they are actively armored (+1) and using a shield (+2). Spellcaster Goblins are 100% Larian's creation (not in the monster manual).




There are merits to this idea, but I don't think Larian will go for it since starting at level 5 will be a daunting task for people new to D&D and 5e.

Having to commit to 5 levels without playing when you can multi-class will cause tons of problems for new players who barely know what each class does.

There is a large part of me questioning the wisdom of me commenting on a SorcererVictor thread, but I have never considered myself particularly wise, so I will go ahead and do so.

#1: What is HP Bloat?.

To even begin to have this discussion we need to define what HP Bloat is. Without having a definition of HP Bloat, we have no metric to determine whether or not HP is indeed Bloated in BG 3. In an attempt to speed the discussion along, I offer 2 potential definitions for HP bloat:

1. When the HP of enemies is different from that defined in the D&D Rulebook.
2. When the HP of an enemy is sufficiently high relative to the damage output of the party such that the fight becomes tedious.

Obviously, the second definition requires a definition for sufficiently high, which we can quibble about as well. I will throw 2 potential definitions of that at you, from an argument I have had with someone else over this topic, over in another thread.

1. If the monster's Hitpoints are double the stated value within the Monster Manual (MM).
2. If each combatant is able to act more than 4 times within a single fight.

If you do not like either of these definitions, feel free to insert your own and we can continue from there. Right now my discussion will take place assuming that one of these has been used.

#2: Why the deviation from the MM Argument is poor?.
The MM itself is a guideline, it isn't supposed to be taken verbatim. Even the people who wrote it will tell you, if your campaign needs a boss goblin, add a boss goblin with slightly higher stats. Furthermore, most of the Goblins within the game technically have lower stats than they potentially could in the MM. The MM allows for a 12 HP 15 AC Goblin, the average Goblin in the EA has ~11.75 HP and lower than 15 AC, if you average the HP across all goblins within the EA. If you distributed HP equally among all of the goblins, they would fit within the 2d6 HP range.

With that being said, if you are unwilling to accept any deviation at all from the monster manual, then I feel I am wasting my time argument with you and so I will bow out at this point here. Please do tell me if I should do so.

#3: The High HP Relative to Damage Argument.
If this is the route you want to take, then to some extent I am in the same boat as you. If a fight takes too long and combat feels non lethal, I also get very bored. My favorite example of HP bloat would be this fight here.

If this ooze is not bloated, I don't know what is.

So, if we go with the 2 potential definitions above, what do we find? Well, the first definition is not very good, because what matters is not the absolute HP of the enemy, but how long it takes to kill them. If an enemy had a base HP of 10,000, but a player did 100,000 damage with every hit, it would not matter that it had 10,000 HP, because it would still be dying in a single hit. This means that this is not a very good definition to use. If we go with the second definition, which does take that into account, you will find that almost every single fight in the EA can be resolved within the definition, unless you deliberately pull more enemies into a fight than is required for the fight to be completed (for example, fighting the entire goblin camp at once rather than just 1 group at a time). In my opinion, if you go with this definition, the HP within the game is not bloated at all, because the time to kill is very short. Most enemies within the game will die in either 1 or 2 hits if you fight them at their intended level and combat has a high degree of lethality to it.

Now, if there was actually bloat in the game, I would be in agreement but I strongly suspect that the people who are experiencing "HP Bloat" are either people who fall into that first definition there and just want to complain about any deviation at all, or they are people who are going into fights they are not ready for due to being under leveled, or they are not taking advantage of many of the character's features available to them which can dramatically shorten fights. Perhaps a lower difficulty would make them happy?

Epic level is not lv 20. Is above lv 20. 3.5e and 2e din't followed the "tiers" that 5e sadly brought from 4e. An lv 1 mage is a guy who casted the first spell today, a lv 20 is the guy who mastered the arcane arts and above lv 20, he broke all human limitations.

Also, you are ignoring that attributes, feats and gear has a HUGE IMPACT on Attack Bonus.

A epic level warrior would probably have a +5 weapon. Magical items and weapon focus, greater weapon focus, epic weapon focus and weapon specialization.

The same applies to spell resistance on 3.5e. Most casters would have spell penetration and greater spell penetration and some enemies on kingmaker like spawn of rovagug are hard to hit even at lv 20 with both feats. He has 42 SR if I remember correctly.

He is a hard enemy due regen, resistances, immunities, summons, powers, damage dealing, etc; not only due a hp bloat...

Am I in the minority? I feel like I had a normal distribution of misses, hits, critical misses, and critical hits.