Larian Studios Forums Divinity: Original Sin 2 Divinity - Original Sin 2 - Suggestions and Feedback On the Topic of Initiative...

I've said this many times, but enemies have an absurdly high amount of initiative. Always going second unless you equip nothing but Initiative and Wit based gears undermines the power of characters and removes any advantage of going first.

So, I have a proposed idea to remedy this:
Either lower the initiative of certain enemy types... (I'm looking at you, zombies and undead. Your initiative is insane.)
OR
Similar to Miss and Crit chance, have each character roll an invisible dice to determine who moves in what order on each turn. Wits would obviously raise the lowest possible roll value.

Why a roll value? Because in any given circumstance, one person is going to react faster than another. Adding this slight nuance to combat would make each turn less of a predictable and chess-like process and more like the chaotic battlefield it is.

In addition, you could also implement a system for starting combat with an attack, letting the player have a free round for getting the drop on enemies.

I could see a roll being good. With the current system, either you have enough initiative to go first, or there's no point in having it at all. At least with a roll, there's always reason to have more initiative. And this is coming from a person who doesn't like RNG.

I think the point of enemies having high initiative is that the player gets rewarded for actually specializing in initiative. As you said, it's an all or nothing system so if you get everything with low initiative then there is no choice for the player. The unfortunate thing is that ALL enemies seem to have high initiative. There needs to be more a range of initiatives so that a few points in initiative feel just as rewarding as dumping everything into it.

That being said I've said it before and I'll say it again, the current stat system is wet garbage and needs to be completely reworked.

I like this....then again I liked RNG D20 mechanics in all my games.

Initiative Roll!!!! Let it be so!

The problem with initiative rolls in a game with save states is that, unlike in actual tabletop, the result of a roll that only happens once at the start of a fight tends to be meaningless. people either don't care, which means they probably didn't spec into anyways, or they get frustrated when they speced into something and are not seeing the benefit of it because they didn't get a roll they want. In a worst case scenario, it leads to a bunch of reloads and the initiative stat becomes the choice of convenience rather than a tactical choice...

I've actually been considering the whole RNG save scum = dump stat. Why not designated save spots ? Such as where one sleeps? It'd discourage scamming in combat

RNG save scumming could be solved by saving the random seed. Doing that would mean that, if the player reloads and does the exact same thing, they get the exact same results. However, as some people have proven to me in other threads, people REALLY want their save scumming to remain intact, so I think there'd be a lot of opposition to that.

Well it depends on when the seed would be made and in concerns to what.

Having the next ten rolls made in the middle of town would be agrivating.

Leave the option for save scumming but place the cost on the player to have to lose enough progress to be annoying.

Chest seeding should remain the way it is in EE for example vs classic.

Just depends on how and when

There's a difference in denying save scumming completely and simply discouraging as a way to break RNG and game mechanics

X-COM 1 and 2 are probably a very good example for avoiding save scumming during fights. The save every possible move at the current point of saving.

You want to institute an annoying, restrictive save system to solve a problem with Initiative rolls which do not in fact actually exist right now? Talk about using an axe to cure a hangnail!

First, even if initative rolls gets added, anyone who reloads for each and every fight until they get a great roll will only be hurting their own enjoyment of the game.

Second, who gives a crap if some other player save-scums in combat? If you don't want to save-scum? Then don't.

Oh? So they bulk up the save file and make the CPU do more work per turn but it's a very viable and comprehensive solution.

Lol now only if we could convince others that RNG D20 mechanics arent a "bad" thing

Well actually it was a comprehensive answer to RNG trivializing stats and such because as you said people hurt themselves when they can.

Second, initiative rolls would be an awesome addition if one can solve the RNG save scumming.

Third, if people should be allowed to save scum if they want why do you bring up them hurting their own enjoyment?

You can't have it both ways. You either care or you don't that they can break RNG.

All I'm saying is that I like RNG mechanics and this would help initiative. I then proposed a way to solve the breaking of RNG if people care enough to actually label that as a complaint

Yeah, I guess Divinity would need less effort from the systems. There are more possibilities per char, but in X-COM can move up to 6 guys in the same turn, here it's always only one. But if I get this write, they don't save the exact outcome of every move, only if an attack would hit or fail under current circumstances.

Some will always be against RNG anyway, the question is mainly, what will Larian think will be the best to solve current quite numbersome flaws of the system.

People can hurt their enjoyment of the game if they spend too much time saving and reloading. However, understand this: that is their own choice. I do not agree with the suggestion to remove the "save anytime, anywhere" feature because some people will choose to do a tedious thing.

Some people will pick up every seashell, crate. basket, and barrel and sell them for 1 gold each. That is a tedious thing, but I do not think that it requires a solution such as "baskets, barrels, and crates can no longer be moved".

Furthermore, there are many, many, MANY legitimate uses for saving anywhere, anytime, and completely banning all those legitimate uses because there are some less legitimate uses is ridiculous overkill.

Soooooooo, we shouldn't take save scumming into consideration when talking of RNG mechanics? And suggesting them?

Cause you did in fact mention it as a flaw but here you seem to be saying that saving anywhere is fine.

To be perfectly frank. I'm fine either way, but I tend to prefer the save anywhere feature. Like I said, I vote for RNG -> people say save scumming makes it moot and therefore bad -> I rejoinder that we can remove save scumming.

Now either the initial complaint of save scumming making RNG moot is null in void or your against the more restrictive save at sleep spots.....in which case I point to the above X-com suggestion using seeding and such

Edit:
Also I'm very much the guy that picks up every single item on the ground, checks every single chest, looks for every single optional fight, spends hours crafting, and will save scum chests and carry gear just for lucky charm and switching out before opening a chest...hell Ill keep gear for every non combat ability so I have different gear load out

If there are issues with save-scumming and mechanics, my preference would be looking at ways to change the mechanics before even considering restricting saving.

I mentioned it as a flaw, but of the "Error between chair and keyboard" kind. There's a guy on another forum who mentioned that he really hated Tales of Zestiria's mechanics, and he let slip that he spent half an hour or more murdering 250 Hyland soldiers in the palace escape sequence to get their bestiary entry up to 5 stars (something to do with maximizing item drop chance or something)... for an enemy which is literally only fought in that sequence and never again. That's abnormal. That is what an abnormal person would do.

Some people are not right and will abuse some things mechanically well beyond the point which no normal person would do. That does not mean that you should aim your mechanics and design to try to (futilely) cure the abnormal person at the expense of making it worse for the normal person.

The cure should not be worse than the disease. It's just that damn simple.


And I'm speaking as someone who did some save-scumming in my Divine Divinity LP to check for spellbooks randomly appearing on bookshelves, and to try and get at least +60 magic and +60 health on the best equipment in the game (although that was largely because I was wondering how much I could break the game with absurd amounts of power).

I can agree with that. The needs of the many over the needs of the few, and whatnot.

I like the thought but this really just leads to people determining the easiest way to advance the seed and then scumming that even harder. It's mostly a single player game so arguments being made on save scumming are kinda meh, but to generate a fun user experience you typically want to avoid circumstances that reward save scumming.

There are instances where this is not the case... Dark Souls comes to mind right away with the game being centered around failure followed by immediately being tossed back into the action with you (hopefully) having more information to work with.

I don't really have any problem with RNG used in the right setting. In fact I play DnD, mostly Pathfinder, several times a week which pretty much boils down to rolling shitloads of D20s and seeing what happens. The difference between these two situations is subtle, I'll try to explain though.

DnD, at least in the campaigns I have always played in, is extremely flexible narratively meaning that things like character death, swingy rolls, and disastrous consequences are used as a tool to directly drive the narrative. This is only possible because the system is intentionally flexible and a human being (the DM) is there to make decisions on the fly. In a computer setting, creating narrative from rolls is MUCH harder because all of the situations and reactions have to be thought of and designed around in advance. This leads to a drastically less flexible system in which the character is either rewarded with success (advancing the game) or punished for failure. (having to reload or missing out on something they wanted)

It's not that a system like that couldn't work. Hell, if they made it a roll system right this instant without changing anything else it wouldn't be the worst thing in the world, but there are much better options that are honestly a lot easier to implement for such a structured experience. I hope that makes sense...

All this talk aboud D20 RNG brings me back memories of NWN1 and 2, where the result was hardcoded to always be to your disfavor (for example, always getting hit by an attack of opportunity whenever trying to drink a potion with low HPs, always getting hit with natural 20s when your AC was high and your foes' BAB was too low, etc.), which quite frankly caused huge amounts of RAGE amongst a lot of players. In short...using an RNG to "avoid save scumming"? The problem is that RNGs aren't that random and, what's worse, they can actually encourage save scumming.

As a programmer, I can guarantee everybody here that "random number generators" often aren't "random" at all, and that's because random number generators are based on algorithms. A "random" number generator generates numbers through mathematical formulae and/or precalculated lists, which basically equates to "roll as many dies as needed, write down the results, then assign those result to each and every subsequent roll". In short, they appear random, but are in fact predetermined, that's why they are called "Pseudo-Random Number Generators". Another key feature of PRNGs is that they are periodic - that is, the sequence will eventually repeat itself.

If you want a True Random Number Generator (TRNG), you'd need something to detect natural phenomena and then convert them to die result format, like a radio telescope tuned precisely to the frequency of the Cosmic Microwave Background produced by the Big Bang (since the time between radioactive emissions is truly random), a geiger counter and a numeric counter linked to a radioactive isotope (same as above), a sensor linked to a diode, or something like that. In short, to have an actual random number generator you need a random source (quite easy to find), and a device that doesn't introduce any bias (definitely not easy to find). Obviously, it isn't easy to accomplish with a gamer's computer, and that without even considering things like predetermined, hardcoded results (Like "enemy always acts first") that can easily be put in place. I'm not saying that Larian's developers are that wicked, but it has happened many times with other games, which makes me quite paranoid about it.

What I mean is: RNGs usually put your playthroughs in a predetermined, unavoidable path, and they can be circumvented with relative ease if you know how. You just have to change the variables and save scum until you've gotten the result you wanted.

Examples: someone here has already mentioned XCOM 1 and 2; in order to elude the RNG, having your soldiers move closer/farther enough would usually force the RNG to redo its calculations because the base parameters (distance, accuracy, etc.) have been altered. Save scum until the outcome is to your liking.

Empire total war/Shogun 2 total war/Attila total war: The outcome of agents interactions was usually predetermined. You could save and reload, which usually forced a recalculation of the outcomes, but it worked just once, after which the RNG would save the seed and force the exact same outcome over and over again. A solution to circumvent the problem? Have a very low level agent try any interaction; it will very likely fail, but it will force a recalculation of the RNG, thus giving your top agent another golden chance at wreaking some havoc. Reload/save/reload until the result was to your liking.

The only way to avoid such tricks would be to - you guess it - heavily restrict save scummings, which would bring us back to square 1. The rant is over.