Let's be honest - this kind of RNG, even when not wonky, has no place in a computer game. Purely luck based failures can be fun in a tabletop game where you can laugh about the result with friends, especially if the DM is good at describing failures, and if a player is creative they can always try to bounce back from a failure or somehow use it for a good character moment. But in a computer game it's just "oh, another miss. Combat lost. Guess I have to reload. Again." "Oh, I failed an intimidation check, even though I put plenty of skillpoints in it. What fun." It's all terribly tedious.

Oh, I wholeheartedly agree. It just incentivises save scumming, one miss with a 95% chance? Alright, unlikely but it happens. 3 of those back to back or seven of those within 15 minutes of playing, yeah, I'll just reload.

It makes sense for them to have implemented it like this since it's a video game adaption of a table top game, yes, but I prefer everythiing to be skilled based. Have 80 points in that skill to increase your armour by x points or increase your weapon efficiency to make crits more likely. This way you know that when you fail something it's your own fault and not just RNG being a bitch. :P
Having everything determined by dice rolls in the backgrounds leads to frustrating situations like knocking an enemy down only to miss every consequent hit...which happenend twice today.

I have a love-hate relationship with this type of game, I love turn based, tactical RPGs, but so many of them rely heavily on RNG which I loathe.

What you see as the target is already the modified value. So if the target is 13 and you have +4 to intimidation (or persuasion) then you will see 9 as the target. Roll 9 and you are successful, roll 8 you fail.

However, the system is bugged at the moment. I was watching a stream and the target was 3. She rolled 4 at first and FAILED. So she loaded and this time she rolled 2 and the game counted that as a success. And yes, she used the same character to talk.

And I agree with the dice rolls. It can be really frustrating. I don't like THAT much luck in video games.

Well, I for one always thought that RNG in CRPGs sucks. Like I described, With all the mitigating factors present in a tabletop game removed, losing because of bad luck is simply not fun, and even winning because you're lucky isn't much fun either. I don't care how seemingly everpresent it is CRPGs, it's a carryover from tabletop that, for me, contributes nothing fun to a computer game. Yes, an RPG can still be a lot of fun despite RNG. And I don't know, maybe to some people rolling high on a virtual die when playing with themselves is fun. For me - the less luck a game requires, the better.

Fair enough, everyone has their own preference. Personally, I find the dice factor helps to make it feel like an authentic recreation of the tabletop game. I wanted to point out though that disliking dice rolling seems like more of a criticism of CRPGs in general, rather than a criticism of this specific game.

You mention 'winning or losing' because of the dice, but I don't think it's always that black and white. In some of the situations where you are making a skill check, for example, it's not so much about 'losing' if the roll fails, but just that something different happens. I kind of like being able to try a certain approach, without knowing it is guaranteed to succeed, and then having to perhaps try something else if it doesn't.

Another thing to mention regarding dice rolling in D&D is that, I don't know about 5th ed, but in 3rd you are often allowed to try skill checks more than once for the same action. If you are in combat, a failure will often just cost you time in some way, as you have to try it again, but is not necessarily an irreversible failure. Outside of combat, the dice roll is actually removed entirely for situations where you could try something over and over with no consequences for failure and you are under no time pressure. They allow an assumed roll - 15 or even 20 in a situation where you could just keep rolling the dice until you get what you need.

Now, I don't know if Larian have scripted that into BG3 to that level of detail, but I would be interested to know.

About 'save-scumming', which some people have mentioned, I think it depends on what sort of player you are. If you are the sort of player that will repeatedly re-load until you pass a skill check, then I can see how the dice factor would be seen as a more of a nuisance. But, perhaps not so much for someone who is wiling to just go with the dice and see what happens.

Obviously, if you get 'TPKed' in combat, then you have to reload. Although, I'm playing DOS1 atm and I'm finding that running away from a fight to regroup/replan is a totally valid strategy. I'm almost at the end of the game and there are very few fights I have reloaded and I can probably count the number of times I've been TPKed on one hand. Not aiming this at you personally, Breja, but I get the impression quite a lot of CRPG players are somewhat allergic to just running away when necessary.

It is about how the player feels and that hugely depends on the chance vs result.

This was one of the biggest handicaps of BG1. Imagine you are playing a Level 1 fighter. You can single-handedly take down 10 goblins if the dice favors you OR you can keep dying against only 1. That made the players feel their characters didn't really matter. That they were only a slave of the dice. And they were right. But that didn't really matter later on, especially in BG2.

This is simply because of the low health you start with. When you are level 1, you start with about 10 health which can be killed with only 2 successful hits by a short sword (D6). However, when you are level 5, you have about 50 health and that would mean at least 9 hits with maximum damage which is very unlikely. That's when your class skills really matter.

D&D Online (which is based on D&D 4th E) solved this problem elegantly by simply giving everyone a 30 bonus health at the start. That way, you never fall victim to RND.

Of course, you can't solve this problem with skill checks simply because you only have 1 shot. It is not fun when your 18 CHA with persuasive bonus fail while your 6CHA character wins. As far as I know, this can only be "solved" in 2 ways.

1. You remove the dice from the equation and simply give +10. A lot of games, including NWN2 used this to simplify the game and less frustrating. Do I need to persuade or intimidate that guy? I need 16... So if I have +6 persuasion (from CHA and proficiency), I always win. If I have +5, I always lose. No reloading, no fuss, simple.

2. You add advantage rolls (roll twice) or lower the dice roll and make it 4D6 instead of D20. That lowers the chances of failure with good characters.

That is true. I agree there are varying degrees to which outcomes can depend on the dice and there is certainly a point at which an RPG can be too 'dicey'. Yes, BG1 (and 2nd ed D&D in general) was notorious for low-level combat being very dice-prone.

Also, in general, I'd say situations where very significant outcomes depend on the result of a single dice roll are indicative of poor game design. In a tabletop RPG, if 'winning' or 'losing' ever rests on a single dice roll, that would probably suggest bad DM-ing. It is better in general to have skill-check failures simply mean the player has to try something else; or loses a bit of time in a somewhat time-critical situation; or they fall whilst climbing and take a bit of damage; or they fail to find one clue to a mystery out of 8 that are available ... not outright win/lose stuff.

Again, I haven't played the BG3 EA yet, but if they are doing things like gating whole side-quests behind a single one-off skill check then yeah, I would agree that isn't great design.

As I mentioned in my previous post, it doesn't necessarily have to be the case that you only have one shot on a skill check. With my example of 3.5 edition D&D, there is a concept of 'taking 10' or 'taking 20' in situations that are not time-critical. In the case of persuasion, a failed check doesn't necessarily have to be final. There can be scope for different 'degrees of failure'. If the PC rolls a 1 or fails massively, yes perhaps the NPC becomes outright hostile, but if the roll only fails by a little bit, the NPC might say 'no' and the difficulty level for a second attempt could increase, giving the player the chance to try again. Or, if they are more patient and come back later, the NPC's disposition could reset to the baseline.

So, I don't agree that, at least in theory, skill checks need to be as black or white as outright success or failure. Of course, whether that level of nuance is coded into BG3 is another question.

The problem with Lords of Xulima's system is that there's one stat that affects how often your turn comes up, allowing faster characters to get more turns in. In a game where you get stat points to distribute at level up, this leads to one of two possible outcomes. Either:
* If you focus on speed, you end up much faster than the enemies, and the enemies hardly ever get a chance to act, making the game too easy.
* You *have* to focus on speed, or else you'll fall behind, and the game will be too hard. (This is the category that Lords of Xulima falls into, from what I understand.)

Essentially, this results in there being less choice in building your characters than if the developers hadn't done this, or had still imposed a rule disallowing second turns before everyone has acted once.

Actually, the problem with low level combat in classic D&D is a symptom of what I consider to be a rather fundamental flaw in D&D (all editions, I believe) and games with similar to-hit mechanics. Basically, in this system, we have the following:
* A higher level character is more likely to hit; furthermore, abilities that improve one's fighting ability (in 3e, this includes but is not limited to Weapon Focus feat and the Bless spell. Notably, being a higher level does not increase the damage dealt per hit.
* Armor makes the character harder to hit. Notably, this does not affect the damage taken when the character does get hit.
* HP scales with level, but damage does not.
Actually, the problem with low level combat in classic D&D is a symptom of what I consider to be a rather fundamental flaw in D&D (all editions, I believe) and games with similar to-hit mechanics. Basically, in this system, we have the following:
* A higher level character is more likely to hit; furthermore, abilities that improve one's fighting ability (in 3e, this includes but is not limited to Weapon Focus feat and the Bless spell. Notably, being a higher level does not increase the damage dealt per hit.
* Armor makes the character harder to hit. Notably, this does not affect the damage taken when the character does get hit.
* HP scales with level, but damage does not.

This, in turn, results in the following:
* For the mechanics to be meaningful, the miss chance has to be high enough so that the character has room to grow in combat ability. As a result, accuracy frequently has a baseline of around 50%. (Contrast this to a game like Dragon Quest where an attack might only have a 1/16 or so chance of missing a particular enemy regardless of the player's level.)
* Damage is a larger portion of HP at low levels. This results in the result of battle being more RNG dependent at low levels, while the law of large numbers smooths this out at high levels.
* Battles at high levels take longer. (This is ignoring the effects of higher level magic on the game.)
* The game does not scale well to high levels. We have attacks either having 95% or 5% accuracy unless the game is precariously balanced (assuming the rules about natural 1 and 20 are in play). To give you an idea of how poorly this scales, in 3e if using Epic Level Handbook rules, at level 4,000 a 1% difference in level can mean the difference between 5% and 95% accuracy; that's too much of a difference for such a small difference in level. (Contrast that with Disgaea, which isn't *that* broken at around level 4,000.)

That's actually a very good idea. It reminds me of the "trap checks" in NWN series. If you rolled more than 10 below the necessary check (might be 5), the trap would explode in your hands. If you fail only slightly, you could try again. That can easily be implemented on other skill checks just like you say.

I completely agree that anyone who plays this game and finds out how Speed works, ends up putting a point in it at every level up and it does feel almost mandatory or risk falling behind the monsters. If they make a LoX2 game, I don't know how they will solve that unless they make the other stats just as critical, making it so players have to balance it out instead always using the min-max strategy.

It's been a long time since I last played it, but didn't BG series also have a initiative roll? Maybe I'm not recalling it correctly.

Either way, I imagine devs have a pretty tough call to make when it comes to balancing combat mechanics. Hopefully BG3 has a good formula.

An initiative roll doesn't have the same problem.

With just an initiative roll, everybody gets 1 action each round; a faster character won't get a turn advantage over a long combat.

In a game like Lords of Xulima, however, a fast character will get more turns over the course of a longer battle.

LoX has other issues, to the point where I ended up dreading the level up screen, at which point I stopped playing it. (This game, I think, is also the game that made me realize just how bad skill point systems really are, though I did start to feel that with Dragon Quest 8.)

Never played the Dragon Quest series, so can't properly comment on that.

LoX is still fun to play, so far anyway. Maybe that will change as I progress and level up a bit more. What do you mean about how bad the skill system is? I know some skills are not properly balanced. For example, Arcane Soldier never leveling up Flaming Strike beyond level 1 (max 30) cause it oddly enough scales better in damage and power with 1 point in it does seem like quite the oversight by the devs.

It's a problem with skill point systems in general, how they force you to make permanent decisions that shoehorn your characters into specific set-ups, without room to experiment unless you start the entire game over (unless there's respec, which LoX doesn't have), and can result in you having to start over if you happen to have a poor character build.

Dragon Quest 8 also has that issue, but in that game it's compounded by the fact that, when a character levels up, you *have* to spend all of the acquired skill points before the game can continue (in the PS2 version, at any rate). I much prefer the approaches that earlier DQ games work, with DQ6 and DQ7's class system decoupling levels from skill learning, and DQ1-5 having simple systems that don't present you with choices.

This sort of thing is why I tend to be skeptical of RPGs that have skill point systems, and why I've come to prefer other systems, like the simpler system of older DQ games, or even systems that are like Final Fantasy 2 and 5. (Note that FF2's and FF5's systems are dissimilar, so it feels weird to lump them together.)