This is a compilation of my thoughts on a range of topics, originally prompted by the thread on the RPG Codex and the request by Kevin at Larian for feedback. Since this was originally written by me on the RPG Codex, I will not primarily be monitoring this thread, but rather my original post there. However, I'd like to implore Larian to actually trudge through said thread, since it is in the nature of the 'dex to pick apart and analyze even the smallest flaws, even on the things we love, but also to relentlessly dump on the unacceptable. If you can get past the coarseness, it is a gold-mine of feedback and suggestions.

These are my thoughts regarding some of the issues discussed, why they're issues (when not obvious), and thoughts on what can be done about it. I'm just listing these based on the small list of suggested issues mentioned earlier in the thread, and while other things have been discussed, most other issues have been simple matters of taste, or things that are literally unfixable at a fundamental level (there is no way to get the money back that was spent on voice acting, anyway). These vary greatly in their size and relevance, as will commentary on them. Some are minor, and some are really major in how they affect the game.

There are also very specific issues in the game, such as how specifically Necromancy seems to have been somewhat shafted, or how incredibly fucking broken Bone Widow is, how oil surfaces penetrate anything but completely non-magical fire surfaces somehow target magic armor, or the apparent tendency towards massive over-compensation, will not be addressed. They're simply too specific and this would quickly simply devolve into a laundry-list of unconstructive bitching, like a list of unforgivable grievances straight out of the book of grudges. It's an imperfect world. The intent of this episode of "Issues as according to me" is to examine things that are actually fixable or discuss things that are systemic in nature. Not every single flaw in the game, big or small.

I do this, as stated, because I feel it necessary to do so, lest bitching remain simple bitching. I genuinely enjoy most of D:OS2, I absolutely adored D:OS1 and think it was one of the best co-op experiences I've ever had in a game, and while I doubt that D:OS2 will reach that same level simply because it seems to be based a lot less around the idea of co-op (I will forever lament the fact that there appears to be no points of contention or places in the game where there's an exchange between player characters on what to do next, and that the contentious resolution system (Rock–paper–scissors), however silly it may have been, is gone forever; it was a nice piece of shared storytelling, whereas now the focus appears to be on an individual story that you just happen to share).

There's really no way to TLDR this.
  • The AI behaving retarded at times, for no discernible reason.

This one is actually quite minor, but frustrating, because it is so consistent. By far the most common issue is that enemies often tries to run away from players with Opportunist, suffers an Attack of Opportunity, lose AP, and proceeds to commence the attack they were seemingly planning all along - often a touch-attack spell or even a melee attack. Neither of which they would've needed to move for to begin with. And this happens a lot. Sometimes, they break away from someone with Opportunist, only to run to someone else (that may also have opportunist) and attack them - then, the next turn, they break away from the character they are now engaged with, suffer an Attack of Opportunity, and runs back to the one they first ran away from. Sometimes this happens even over environmental effects.

Now, someone might argue that "the AI doesn't know that you have Opportunist, nor should they!" - and that's a fair point, and one I happen to agree with, but it is already an established fact in D:OS2 that the enemies are psychic, as evidenced by their use of everything against Glass Cannons, or how they immediately target undeads with healing spells. I personally believe that the AI should treat all opponents equally unless they have cause to do otherwise, meaning that until they know that someone is undead, they won't target you with healing spells, and until they know that you seem to have no armor at all, they should treat you as if you do, and if someone runs away from an Opportunist and immediately suffer an Attack of Opportunity, everyone involved should know that that man does Attacks of Opportunity, and stop trying to run away and get pelted again-and-again.

It's absolutely jaw-dropping. In many, many ways, the AI is actually fairly clever, and sometimes (but rarely) they do very clever things - although this is somewhat restricted by another issue that will be discussed later. While there's been a lot of discussion about these AI issues, make no mistake that the AI is a lot better than it was in D:OS1. Sometimes, the enemies even use barrels or multi-stage setups, and sometimes they even try to cleanse themselves of effects. All of this is well and good, but this also means that when the AI does something stupid, it really stands out, and it stands out every time. The AI can - provided that they are considerably outnumbering the player - move things around and set up smart tactics at the end of one turn, only to have another AI-controlled creature literally run around in circles at the beginning of the next round.

Since everything that has to do with the AI is fairly esoteric, I cannot fathom why this is, or what can be done about it, but, fundamentally, the AI should be reluctant to move away from someone with Opportunist, and they should be similarly reluctant to move on dangerous surfaces. In D:OS1, this wasn't a problem, and the AI had no trouble avoiding running around on dangerous surfaces. Therefore, I have a suspicion that this has something to do with the armor system, and I have consistently seen enemies with very little magic armor run straight into blazing infernos, only to - of course - catch fire two steps in. At the same time, the AI being able to tell exactly how much damage moving a certain distance in dangerous terrain would cause them would be equally silly - just like the psychic nature of enemies in regards to the nature of the various allied characters (undead, glass cannons, etc).

And for all that is holy, tell the AI to stop trying to run away just to get a set distance away from you before they engage you in melee or casts a touch spell. I suspect that this is due to some order that they have to get in range, with "in range" defined as the maximum distance of whatever they're trying to do, and they're ignoring the fact that they are already close enough. I'm probably wrong, because it seems too simple, but that's the impression I have of what they're doing. I've seen enemies die to this on multiple occasions, and it's.. I don't want to sound entitled, but they really shouldn't be doing that; it's simply unacceptable. And I don't mean that as some kind of childish demand from my side, I'm saying that you, Larian, must agree with me, from your own point of view, that an enemy literally killing itself in front of me instead of attacking is unacceptable.
  • The laughably bad defensive abilities.

Another issue that I have a feeling can largely be chalked up to the binary armor system, because they seem to exist on some kind of principle, but are essentially vestigial since the binary armor system rendered the original defensive abilities meaningless or non-functional. The new ones, however, in what appears to have been an inability to conceive of new defensive abilities to match the new combat format of binary armor and outcomes, aren't just conceptually lackluster, they're simply bad. Now, I won't pretend like the defensive abilities of D:OS1 were amazing, or that the system was some perfect gem, or that I ever put a ton of points into any of those, but they all had a purpose and while a min/maxer might argue that they were a waste of points, it was never a complete loss to invest in them, and again, they still interacted with the overall system in a meaningful fashion.

Defense abilities in D:OS2, however? Can they really say the same? Leadership didn't even use to be a Defense Ability, but seems to have been moved under that header in an attempt to salvage the category. What's worse, the bonuses aren't necessarily bad, a bonus to dodge and a bonus to resistances, but at the same time, the range is restricted to 5 meters. 5 meters. That's practically "melee combat only" in D:OS2, but for whatever reason, Leadership is an "ability" Conjurers start with. A min/maxer might argue that you can move those points around - and you should, absolutely, because Summoning is the only ability that has a capstone reward at 10 points (which is a bad idea for many reasons, but that's another discussion entirely) - but from a design perspective, the preset is encouraging the use of an ability set (or, as I would call it, skillset) that will practically see no use.

Perseverance is even worse, conceptually, but only because the armor system is bad, and Perseverance has the potential to take it from "bad" to "borderline psychotic". But more so than that, Perseverance is simply bad. It allows you to restore Magic Armor after recovering from Frozen or Stunned, and Physical Armor after being Knocked Down or (for some seemingly arbitrary reason) Petrified (which does Earth damage, targeting magic). The only one out of those that happen consistently is Knocked Down, and at best it gives you a small breather from being affected by the same thing(s) again, but unless you're facing tons of enemies, it is unlikely that armor will actually remain long enough to matter, and if you are facing tons of enemies, it will be stripped anyway. Perseverance appears to be an attempt at create a relevant defensive ability in the context of the binary armor system, in lieu of other potential abilities that can boost defenses, but because of how the armor system functions (i.e. there are no other defenses), there is no way to properly balance this or make it relevant enough to matter within this context.

Even in a best-case scenario, it's extremely situational, and if you invest enough for it to even matter in those scenarios (so, 10 points for 50% restoration) you've irrevocably shot yourself in the foot. It would surely be fun when it does trigger, but by the time it does, you could've done something useful.

Now Retribution. Now we're really deep into the barrel of uninspiring, lackluster and useless. This thing deserves some kind of reward for it's uselessness and just how ill-fitted it is for this kind of game. A retribution-based ability is something you'd expect to see in a run-of-the-mill ARPG, and in such a game, it might even be useful - to an extent. But to have it in D:OS2 is just.. weird. And the ability is terrible. Not just in the applicable sense, as in it likely being the single most worthless place you can waste a point, with little-to-no battlefield application, but also conceptually, in a mechanical sense.

The reasons for this is actually quite simple. It is terrible as a choice for the player because it is entirely reactive, and completely dependent on the damage the enemy does to you. With 1 point in it, you take 100 damage, and you'll reflect 5 damage. The thing is, though, even with 10 points wasted into it, you'll "only" be reflecting 50% of the damage done to you against your enemies. And when has an opponent ever done damage anywhere even comparable to his health pool relative to yours?

Keep in mind that this is also in a game completely without defensive abilities beyond these three, and if you were to put those 10 points into, say, Single-Handed, you'd instead have +50% Damage and +50% Accuracy with one-handed melee weapons. Nothing that wouldn't absolutely destroy you will even react to the fact that you reflect 50% of the damage it does to you, and it certainly would not compete with a consistent +50% damage (and Accuracy!) that a weapon ability would net you regardless of whether the enemy is attacking your or not (such as being under Crowd Control).

However, even if Retribution would be buffed to 200% in returned and you had the defenses of a god, simply standing there until the enemies kill themselves by hitting you, it would still be conceptually ruinous, because you'd be committing one of the biggest sins in design that you can commit: you'd be rewarding the player.. for doing nothing. Just stand there, soak it up, and have the enemies kill themselves. In a turn-based, ostensibly tactical, role-playing game.

There's no way to actually save this ability. It should never have existed, and it should simply be removed. It's not just bad, it's conceptually and mechanically unforgivable.

I honestly cannot understand who thought that was a good idea, or good design, or how they even thought it could be useful. It deserves it's own little special layer of hell.
(Also, a completely separate issue, let me state for the record my enduring hatred of the D:OS franchises cemented tendency to systematically and categorically conflate the terms "skills" and "abilities" in a way no other game I have ever played does - "Skills" and "Abilities" are, everywhere outside of Belgium, apparently, the other way around. This messes considerably with my ability to uphold consistent terminology.)

  • The near-meaningless attributes and the slaughter of complexity & depth.

Now, I'm not going to pretend like D:OS1 didn't have any balance issues. That would just be silly. We all know it did. It was generally easy to simply pump one Attribute, maybe two, and that was it, and due to some other issues, largely absent in D:OS2 due to other changes, stacking modifiers to AP and Initiative was king - move first, move hard, and then move again.

But look at this. And then compare it to this.

It's depressing! It's downright pathetic! It's jaw-dropping how an attribute system can become so watered down and so simplified that it almost has no meaning or point to it's existence. This is Diablo 3-levels of oversimplification and dumbing down, with the exception that we pretend like there's any meaningful choice remaining, by still allowing the player to place their points.

"You are playing an X-type character. Do you want +5% damage to X, +5% damage to Y, +Health, or +1% Critical Chance? Or do you want to pay this Attribute Tax so you can have more skills?"

While there are more major concrete issues in the game, this thing is probably the one that makes me the most depressed, because it shows such a clear decline in meaningfulness, build versatility, and system complexity. It is really on a point where I feel that you might as well remove Attributes as a concept altogether, and simply apply +5% damage to the character, a health boost, and an increase in critical chance on a per-level basis, because none of this ultimately constitutes a meaningful choice; it's entirely and painfully obvious from the creation of your character to the end of the game where you'll be placing you points, and the near-obligatory points you practically have to put into Memory doesn't change that, but is rather to be considered a form of cheap attribute point taxation in order for you to be allowed to use more skills - not that you have to have that many skills, due to other issues in the game, since action points are very limited, cooldowns short and crowd control shorter.

I really don't know what else to say about this, and there's no simple way to fix this. The attributes need to do different things in different ways, and with so many parts of the game cut into in terms of depth, with no saves or resistances (beyond pure damage), straight-jacketed action points, and completely normalized ranges across the entire field for weapons (13 meters, I believe, whether we're talking crossbow, bow or wand), it's hard to determine what the attributes should actually do.

A lot of these changes shouldn't have gone in to begin with, especially not if they affect the core character mechanics in such a way that you eventually come to the final conclusion that "Oh, wait, what are these attributes supposed to do now that we've stripped out all the subsystems?". Someone might argue that it's "balanced", but balance for the sake of balance itself has no inherent value. Balance is only needed in the sense that too obvious options are stripped out to avoid one-trick-ponies or abuse, and forcing the player to think and adapt, deriving pleasure from outwitting or outsmarting the opponent or the system, strategically or tactically, but here, it has clearly come at the cost of meaningful player agency in terms of attribute distribution and build versatility - the very things that the concept of balancing is intended to support, because if it's not that, then what point is there to it?
  • The rollercoaster-ride of Talents.

Oh man. Should I take Five-Star Diner or Duck Duck Goose? Should I take Escapist or Executioner? Should I take Ambidextrous or Far Out Man? Should I take Elemental Ranger or Guerilla?

All of these questions are rhetorical. Some Talents are extremely powerful. Others are almost completely useless. To add insult to injury, some completely beneficial, while others constitute trade-offs. Escapist is based around the concept of actually losing a battle and fleeing, Five-Star Diner doubles the bonuses of food, but the issue with food was never the bonuses it gives, but the duration which it lasts, and why on Earth would anyone waste four Action Points to go into Stealth in order to do 40% extra damage? Either you spend 6 Action Points to do 140% damage, or you spend 6 Action Points to actually attack 3 times to do 300%.

Why do Ice King and Demon come with trade-offs in the way they do? If it trades 15% for 15%, the only real benefit is the +10% to a single resistance cap, which you'll be hard pressed to reach and that matters a hell of a lot less in a game where the armor system negates magic damage for a while anyway.

Why does Leech completely consume blood on the ground, when Undead don't consume ooze/poison? Is it just to make it incompatible with the single most thematically fitting combination you could do when you play a Leech, which would be that of a Nercomancer with the Blood Sucker skill?

Why are Slingshot and Far Out Man separate talents, when they essentially do the same thing conceptually?

Why does Ambidextrous have absolutely nothing to do with actually being ambidextrous or duel-wielding, and do you actually expect people to invest a talent in the concept of leaving their one hand completely free (foregoing both the more powerful bonuses of two-handers and the dual bonuses of having a secondary weapon or a shield)?

Why are The Pawn and Executioner mutually exclusive? Are you really so deathly afraid of people saving even a single extra AP and being able to kill someone? It's not an amazingly powerful combo, even.

So many questions. The Talents are a mess, ranging from amazing to questionable to functionally worthless. And what's worse is that this was true already in D:OS1, but it has somehow gotten even worse, with many talents removed or nerfed - some of them because the subsystems sadly no longer exists to support them (Courageous, Voluble Mage, Headstrong, Lightning Rod, Sidewinder, etc.) and others seemingly arbitrarily and without reason (Packmule, Sidewinder, Swift-Footed, Thick Skin, Anaconda, etc.).

What's left is a small number of Talents that vary wildly in usefulness, and are almost all universally applicable, with very little effect on actual build or character, with a few exceptions (Opportunist is obviously for warrior-type characters, etc.).

The solution here is both complex and very simple. First of all, if some things are changed for the better, such as the current armor system, the return of flanking, etc., many of the old Talents could easy be recreated and reinstated. But also, some Talents simply need to be buffed, or have the trade-offs removed from them, to actually make them solid choices. Other talents could be merged together, such as Far Out Man and Slingshot.

There's no reason I should go through all the potential fixes and things that could be done to each individual Talent, it should be fairly evident in each case. And for fucks sake, stop having Leech such the blood from the ground so you can actually use it with Blood Sucker (which should remove the blood from the ground).
  • Binarity of outcomes, predictability, and the armor system.

I'm unsure of where to even get started on this one. D:OS2 uses a binary system with two different types of armor, each of which determines whether something does damage to Vitality or not, and whether an effect of practically any type works or not, whether it's Taunt, Burning, Charm or Frozen.

First of all, it's completely unintuitive. You launch a big rock and throw it at the enemy? It targets magic armor and is absorbed by it, and only does damage to it, not to physical armor at all. You scream at someone and call 'em a cunt, trying to use Taunt on them? It literally bounces right off the shield, targeting physical armor. You launch a fireball? Magic armor. You throw a physical barrel of oil on someone and then throw a physical molotov cocktail on the barrel, creating a burning inferno? Soaked up by magic armor, and enemies can pass over it with little difficulty, without catching fire.

But that giant boulder you launched at the enemy, which was completely soaked up by magic armor? It spread oil on the ground, imposing the slowing condition, which automagically pierces all forms of armor, both magic and physical.

But the fact that it is unintuitive and feels arbitrary (which it isn't, it's actually completely consistent) is actually completely secondary. What it does is that it leads to a completely predictable combat system on any given turn. You know, with 100% certainty, whether something will work or not. There is absolutely zero ambiguity to it. Do they have armor? Your effects will fail. Do they not have armor? Your affects will succeed.

Any feeling of suspense is removed, and any risk taken with practically any action, making your actions in any one turn likely a given, without any considerations, hail mary's, or surprising turn of events. Simply put, the armor system completely negates any element of the Delta of Randomness. This is terrible design in itself, because that delta of randomness is one of the staples, I would say fundamentals of good turn-based gameplay, and the armor system breaks this completely on a systemic, tactical level.

So how can it be resolved? Short of reworking the entirety of the game and reinstate a save- or resistance-based system, there's not a whole lot that can be done. If absolutely determined to maintain the armor system at it's core in how it works right now, several things would have to change.

First of all, the intuitive nature of the system is more important than consistency. Whether something targets physical resistance/armor or targets magical resistance/armor should be determined by the nature of the spell or skill or item used, not by the damage type or the condition it imposes. A rock should target physical resistance/armor. A fireball should target magic resistance/armor. A molotov should target physical resistance/armor. And a lot of things - such as most surfaces or Taunt, should not stoppable by armor at all. The ability to resist them should either be by a flat percentage, or modified by something else entirely, such as Wits, or mental or magical resistances.

Furthermore, rather than having a system in which armor either completely blocks something or does nothing at all to block anything, the protection garnered could be a percentage based off remaining armor. If you have 100% magic armor, you could have a 90% chance to resist most spells or magic-based effects, as well as the magic armor soaking 90% of all magic damage, meaning the spell would only do 10% of it's damage to vitality.

At a theoretical 0%, the chance to resist would inversely be a mere 10%, and the spell would do full damage to vitality (since the armor is now completely gone).

While I would consider this approach imperfect, it would be a way to keep using the currently existing sub-systems and itemization and so forth, while constituting a considerable improvement upon the nature of combat resolution and the tactical layer of the game, which right now - for all practical intents and purposes - simply do not exist.
  • The Round-Robin Turn Orders.

This thing right here. I've pointed it out to multiple paste-eaters, and despite the very obvious and major issues, they simply hadn't noticed. It boggles the mind. And once you see it, it cannot be unseen. Most people assume that Initiative has a meaning in the game. They assume that it works like initiative usually works. They assume that the rounds in D:OS2 are operating on an initiative-based system.

I mean, it has to, right? I mean, if Initiative doesn't matter, then the mechanical benefit of Wits would be reduced to a mere +1% critical chance, the +2 initiative benefit for being human wouldn't matter, and all those initiative modifiers on items and equipment would be irrelevant? Surely, initiative must matter in a tangible fashion?

But it doesn't. D:OS2 operates on a round-robin turn system, where the only thing initiative determines is who goes first within a given group. And it is, hands down, the absolutely worst aspect of the entire game, and the single most indefensible design decision. To understand the issue, you must understand the implications of this, and to do that, you must understand how it works.

In D:OS2, allies & enemies always alternate their turns throughout a round. No matter who has the highest initiative, the opponent of that person will go second, followed by an ally, and then an opponent. Regardless of actual initiative scores. Let me give you an example:

Let's say your party of 4 has an initiative of 10, 11, 12, and 13 respectively. You are meeting a group of 8 enemies that all have an initiative of 1, 2, 3, 4, 5, 6, 7 and 8. So, markedly lower than even the lowest one in your party.

You will move in this fashion, bar any special circumstance (such as summons):

You (13 Initiative)
Them (8 Initiative)
You (12 Initiative)
Them (7 Initiative)
You (11 Initiative)
Them (6 Initiative)
You (10 Initiative)
Them (5 Initiative)
Them (4 Initiative)
Them (3 Initiative)
Them (2 Initiative)
Them (1 Initiative)

Now, that alone should be enough to make you raise your eyebrows and go "Wait, that's retarded!", but the issues with it goes far beyond that.

First of all, this completely invalidates Initiative as a relevant secondary attribute. That +2 Initiative humans get? Completely irrelevant. Wits? Neutered; it effectively only gets +1% critical hit chance, and the only reason you'd take it is because you maxed out either Strength or Finesse, or to find secrets, which only one character in the party needs to do - presumably the one that you want to move first in the turn order and use to initiate combat.

All those items with +Initiative? Useless. You used a Bucket for a helmet for much of the initial stages of the game? Makes perfect sense, because -Initiative literally does not matter, especially if you are the one to initiate combat anyway.

But these issues are secondary or even tertiary to the effect it has on the strategic layer of the game, the layer beyond the actions of a single individual character (so brutally savaged by the aforementioned armor system). The execution of plans and the evolution of the combat landscape during the course of a single round is essentially non-existent, and it affects both the player and the AI.

Because of the round-robin turn-orders rather than an initiative-based system, the battlefield is a constantly changing landscape, to the point where it's hard to actually determine what is going on, or make any plans.

A common concept that I would consider foundational to the very concept of turn-based as an enjoyable way to play out key resolution mechanics is the ability to think ahead and act upon the perceived development of the landscape as it is (and by landscape, I don't just mean environment, I mean it in the widest possible meaning of the word).

In D:OS2, that's simply absent. Or at least it seems to have been lost as a key source of enjoyment, because it is practically impossible to plan ahead, because a single turn later, the landscape may be completely different than from when you ended your previous turn. And then it changes during your turn, but is immediately undone the next.

Furthermore, you may not even WANT to put down water (just as an example) because the next turn, there's a guarantee that no-one in your party will be the one taking action, so you might actually be shooting yourself in the foot - but there's no way for you to know if you are.

You don't want to throw out a barrel, because there's almost a guarantee that it will blow up in your face - if there's even barrels around at that point (which is a big difference from D:OS1, where barrels sometimes would not even get used in combat, or simply not get hit by environmental or AoE effects, which is almost a guarantee in D:OS2).

And this goes both ways. The AI doesn't want to do these things either - unless there's a significant number of enemies, meaning that they do get to do several consecutive turns at the end of the round. The end result is that the idea of planning ahead or predicting the actions of your opponents are absent from the considerations in D:OS2 combat. The kind of set-ups that were so common and so integral to the enjoyment of the combat in D:OS1 is entirely absent in D:OS2, and combat in D:OS2 often devolves into "playing catch-up" and reactionary decisions on a turn-to-turn basis; and because of the armor system, the actions taken within those turns are entirely predictable and essentially binary, meaning that you know exactly what to do and what will happen at any one time within any given turn, removing any feeling of suspense or momentary hopefulness.

Before I got a chance to examine the game and notice the major issues with it, my SO was already all in planning-mode about how we should plan ourselves to be complimentary, and how if she was using fire magic, I should take something to use earth magic, so if I made some oil, she could set it on fire. But in D:OS2, that doesn't happen. What happens is that I throw oil, and the enemy either remove it or avoid it and set it on fire themselves. And this doesn't just affect the player - it affects the AI too. The AI has the capability to do these things, I see them do it, it's actually quite awesome. But if they were to (like me) throw out a barrel from somewhere in front of them, I'd drop it on their fucking faces the next turn.

The only time set-ups like this can be done reliable, to see a plan and see it take shape, to get that rush of "FUCK YEEAAAAH BURN YOU FUCKING CUNTS, YOU DIDN'T SEE THAT COMING, NOW DID YOU?!" is when you're at the absolute end of the turn order in a little clump. And 99 times out of 100, that's just going to be the enemy in one of those fights where they come busting in from every direction at the drop of a hat as the BBEG twirls his mustache and goes "Hohoho, you didn't see this AMBUSH coming, did you?".

There are no moments of "YES!" or "Aaaah, noooo!" in D:OS2 that isn't caused entirely by your own fault, or that comes unexpected, and there is no enjoyment in the procession of consecutive turns because there is no way to fulfil even a short-term plan. And both of those things are absolutely essential to practically all turn-based systems.

But it doesn't end here. No. It gets worse. There's not only a de facto inability to plan and execute, further hampered by the randomness of the evolving landscape in any given encounter, but round-to-round, the turn order can be perceived as essentially random. "How can that be? Weren't you just saying that they're utterly predictable?!" you might ask.

Yes, yes I was, and they are - to a point. You won't be the one to do the next turn, but you might not know who will be taking that next turn. You can probably figure it out, but it's by no means intuitive or sensible. You see, the turn order I described earlier is mutable. After all, enemies can die, right?

And if you kill an enemy, you might expect them to be removed from the turn order, right? Well they are. But someone else will immediately take their place. This leads to the fun, fun, fun situation of you killing a weaker opponent towards the end of a round, and in the next round - which is coming right up - the much stronger opponent has now taken his place, meaning that he may be moving before, say, the 4th person in your party, whereas if that enemy that had just died previously would otherwise have been moving in his stead - and maybe he would've died this round instead, or maybe even on his own turn by being on fire. But because you killed an enemy last round, the turn order gets reshuffled, and the much stronger enemy gets to move instead, and he might kill your 4th party member before he even gets a chance to move.

This has created the situation where you as the player, by killing an opponent, actually ended up in a much worse position than you would otherwise have been in if you had left the enemy alive. Yes, this is an extreme example, but this is still something that happens all the time in the game, even if you obviously don't end up losing a party member every time it happens. However, the fact that you can regularly screw yourself this way is completely absurd; taking out an opponent should always be something positive, bar special, narrative circumstances - the default should never be for the player to question whether he ended up in a worse position or not, or if he should simply have ignored the enemy until next round, maybe simply skipping taking any action, despite being maxed out on action points.

But wait. There's more.

For simplicity's sake, imagine a scenario in which you are alone against two opponents, and for whatever reason, they both have higher initiative than you. It results in the following turn-order:

Enemy #1
You
Enemy #2

Soon, you'll kill an enemy. Who're you killing? It better not be Enemy #1, because then the following happens!

Enemy #1 takes his turn.
You take your turn, and you kill Enemy #1.
Enemy #2 takes his turn.
Enemy #2 takes another turn, because it is now a new round and Enemy #2 at the top of the turn order now, because Enemy #1 has died.

Again, the player has boned himself by killing the opponent, through no fault of his own. Had he instead killed Enemy #2, the result would've been different, without the enemy essentially getting a free turn.

But wait. There's more!

You summon something. The summon moves right after you. But then at the next turn, it gets shuffled into the round-robin turn order, meaning that one of your characters can be pushed back as much as three turns, suffering additional attacks that it would not otherwise have done.

Summoning an ally, presumably done to improve your situation, might end up screwing you completely, and there's really no way to tell beforehand. The system punished you for buffing. This becomes very obvious as a Summoner especially, since Totems are extremely weak, but get shuffled into the turn order too, so that "lump" of potential "overflow" opponents could end up picking them off one-by-one in a single round before they can even possibly act in that round.

But wait, there's mooooooooore!

Because all Initiative does is determine your in-group turn order, gaining initiative is potentially detrimental. Why? Because if the only benefit of Initiative is between the player's own party members, it means that if a party member actually gets "too high" initiative, it puts him above another party member, and it might be disadvantageous for him to do so - such as if someone with Fire Skills gets bumped above someone with Earth Skills, meaning that if you rely on (in the extent you can rely on it in D:OS2; this is really just for the purpose of explanation, I know full well that this doesn't really happen in D:OS2, because of the armor system and the round-robin turn orders) the latter laying out oil so that the former can set it on fire, you can't do that anymore.

Another example would be a warrior getting higher initiative than the buffer in a party where you rely on the tactic of buffing the warrior before he moves. In fact, I know that there are players that consider initiative purely a negative stat for their main character, because their main character is a summoner, and relies on others putting down environmental effects before he takes his turn.

Can similar situations arise in an initiative-based turn-order system? Yes, but it is a lot more manageable, because if two party members are close to eachother in initiative, they at least do not suffer from the fact that an enemy will forcibly be inserted between them, further exacerbating the issue, and if someone gets higher initiative, it is still a net benefit vs. the opponents themselves, on average.

So what can be done about this? It's actually quite simple. Simply flipping the switch and make the game Initiative-based would be a tremendous improvement, and I know that D:OS2 can handle it, because that is apparently how it used to be, and in fact the entire game already appears to be geared towards using an initiative-based turn-order system, as evidenced by the modifier humans get, the initiative bonus from Wits, and the many items and equipments in the game which supports the notion that initiative is intended to matter and is intended to be a positive modifier.

But while that would be a tremendous improvement, it wouldn't be perfect, because it would obviously mean going back to the issues that prompted the round-robin turn orders to begin with. The ideal solution, in my mind, would involve rebalancing encounters to avoid massive stacking of initiative as an end-all be-all viable tactic, as well as to add a per-round element of randomness that, despite constituting a random modifier, would not be so strong as to undo the benefits of getting a higher initiative.

While obviously completely untested, my suggestion would be to flip the aforementioned switch and thus go back to an initiative-based turn order, and then, round-by-round, modify every participant's initiative by +/- (essentially either add or withdraw) 1d[total initaitive/2].

What would this mean? It would mean that someone with an Initiative would have a per-round Initiative range of 5-15. Someone with 20 Initiative would have a per-round Initiative of 10-30. Notice how the average is always equal to your base initiative, meaning that if you have 13 initiative, during the course of all your rounds, you will average out at 13. Your initiative still matters just as much as it did before.

However, it also does not guarantee you to go first in any given round. But at the same time, each round in itself will play out predictably (turn-wise), and you will never be punished for taking out an opponent, you will never be punished for summoning, and you can prioritize targeting. Obviously, up to two or three rounds should be displayed to you, with the modifiers for each participant already determined, so you can actually tell who moves when - and, as opposed to as how it is now, rely on that information.

I can only hope that Larian listens. Especially on the last topic. Because currently, it's balls.