wih

I guess this is left to your imagination. If you have righteous companions, you can imagine they are not seeing your mischief.

I wonder why didn't they introduce a simultaneous turn based mode, where you give your orders at the start of the round and then the round resolves synchronously in real time, but you can't change your orders during the round. Something like Combat Mission. This can be done without rebalancing the game. Almost everything needed for this is already in the game including ability to set round length to be like 2 seconds or 4 seconds or more.

I would want to see a possibility to play the initial part of the fight in turn based mode where you are able to switch to realtime mode when you see that you are winning already. Otherwise I am concerned that the fights would become long and tiring.

Deadfire is a different adventure than PoE 1. And the game is much improved. Those are the primary reasons to play it. If someone doesn't want to play it because of the need to start from level 1, they probably aren't the target audience anyway. I strongly hoped I won't be level 17 at the start of the game.

I'm starting to think that making PoE 2 a direct sequel was probably a mistake. It says to the potential new players: Don't buy PoE 2, buy PoE 1 instead. (and POE 1 is on sale, so no big money for Obsidian). DoS 2 wasn't a direct sequel so people weren't afraid they won't understand the story if they skip DoS 1. Also these RPGs are big. People buy PoE 1, then months pass before they finish it. Most don't even finish it. PoE 2 should have been either a prequel or set fifty years after PoE 1 events.

Paradox owns Tyranny.

Even if Deadfire is not selling all that well, I don't see why Obsidian should refrain from another "Figstarter" that can potentially give them another 2-3 million dollars so that they can develop yet another game (POE 3) with the already available engine and tools and IP. If they do that they will have three games bringing revenue for them and not for some publisher. There are fates worse than that.

POE has slow combat? How so? For me the main problem with POE combat was that everything happened lighting fast. I don't remember events in IE games unfolding so quickly. In POE things were uncomfortable for me - I had to pause all the time and then scroll back the event window in order to understand what hit me.

I would try the following condition to find a target for Slicken: Target: Allies in Melee Range > 1 and Target: Enemies in Melee Range > 0 NOT I haven't tried this, because I usually pick abilities that are enemy only, but it seems to me that this logic should select a group of 3 or more enemies. I don't think such casting would be very reliable but it is worth trying.

I think the difference between the Aggressive and Defensive is the range. In Pillars of Eternity I found that Aggressive was too aggressive. My characters ran to attack monsters that were too far. So I set them to Defensive. With POE II Aggressive seems to be working OK but Defensive setting is too passive. So now I set all my characters to Aggressive. It seems the developers just try to find some reasonable values for these settings, maybe they are even experimenting a bit. I doubt this is documented somewhere.

Ah, external tool would be great. I would be happy if I could write custom AI scripts by typing on the keyboard instead of mousing. Nice point about the abilities. I guess this limitation can be worked around somewhat. Save the game, level up a character, take an ability, script the ability, then reload the game. But it is a slow and unwieldy process.

I want to draw attention to the behavior editor user interface. For me, the customizable AI is one of the most important features of the game. It turns out that the custom ai files are in a custom binary format so the only way to edit the AI is through the ingame editor. And the behavior editor user interface is just Less. Than. Perfect. What do I mean? There is incredible amount of wasted vertical space in the behavior editor window. The most important thing in this window are the conditions - but you can see only about three or four of them at a time. I have a tendency to set many conditions and it is tiring to have to scroll so much just to be able to read them. Imagine working in an IDE where the editor window is so small that you can only see four lines of code. This would be simply cruel. There is no reason to force your players to scroll up and down all the time. Those green 'plus' buttons that sit under each condition - look at them. Because of them there is so much empty space on the editor that goes unused. Why don't align them horizontally with the other buttons and reclaim that space? A major pain point is when you need to insert a new condition at the top of the set. You have to scroll all the way down, click that plus button and then start moving the new condition up - for which you need to click the up button once for each condition in the list. But this is not all - most of the time when I click that up button, the list scrolls up or down on its own and I need to scroll back to reposition it and find the condition I am trying to move up. This gets harder and harder with each new condition. This is incredibly inefficient - and it can be easily fixed. Just introduce an insert condition button (similar to the duplicate button) so that the players can create the condition and have it appear right where they need it. Another thing that wastes vertical space are the column labels. There is no need to place the same column labels on each condition row. Just one label set at the top of the condition list would be enough. Or incorporate them in the tooltips that appear on hover. Vertical space is also wasted by the 'and' labels. They can be placed on the same row before the condition. There is currently not enough horizontal space for that, but this can be fixed (I write about this below). And while we are still talking about the wasted vertical space, I want to ask - why is the behavior editor window not maximized? The behavior set selector is also badly placed. It steals from the horizontal space that should be given to the condition list. This behavior set selector should be placed on the top panel and this top panel should move up when we scroll the conditions list down. Also, that behavior set selector should probably be a combobox. The game doesn't remember which condition set I am using. Every time I open the behavior editor, the top condition set is autoselected. This set is uneditable and undestructable. I never use it. I make my own condition sets and they are always below the default ones and I have to click and select them every single time when I open the behavior editor. Please Obsidian, do something about this. Fixing these problems does not involve changing the game mechanics - it is just a presentation. And you will save your players from developing RSI.

The ingame behaviors editing involves an awful lot of mousing and I have RSI from using the mouse. Is it possible to edit the scripts with just the keyboard, for example with text editor? Those files are contained in "c:\Users\%userprofile%\Saved Games\Pillars of Eternity II\CustomAIBehaviors", but they seem to be in some unknown format. Maybe they are compiled? Is there some tool that allows editing them or some other way?

Well, OP is giving an opinion on the game and probably wants to hear other opinions. That these opinions tend to be from people who like the game shouldn't be surprising - because the majority of people who frequent this forum like the game. By calling them guard dogs you are devaluing their opinions. It seems that you like to see how everybody criticizes the game, but to what end? Nobody claims the game is perfect anyway. Specifically one point that OP makes - that a game is not a RPG if you can't play evil in it is weak. What if I make the claim that a game is not a RPG if I can't play a crazy person in it. I am actually replaying BG 2 now and I imagine my hero is slowly going crazy because of the Irenicus experiments. Well, that's somewhat hard because all of my possible options in dialogues sound perfectly sane. Deadfire gives an option to play evil though. And there are subtler things, which I like. In my playthrough I made evil things even though my character was supposed to be a good person. But I failed to get one faction's support and after that I didn't want to fail getting another faction's support, so I convinced myself that what they wanted me to do wasn't that bad or at least was justified... Which is how people do evil things in the real life.

Deadfire improves on this, though probably not directly the pathfinding. When a character wants to reach an enemy, the other characters now step aside to give way. For me, this practically solves the combat pathfinding problem. In PoE this caused the character to run in circles around the combatants which was very annoying.

I guess this phenomenon is additionally reinforced by the fact that CRPG lovers also need to finish their playthroughs of Wasteland 2 / PoE1 / Tyranny / DoS 1-2 / Tides of Numenera, etc. It makes sense that they are not in a hurry to buy Deadfire.

Yeah, it's unlikely we're going to see a reactive trilogy the likes of Mass Effect for a *long* time - you could tell from the interviews that they regretted the lengths that they had gone to by ME3 - Bioware had to create entire quests based on decisions small minorities of players made in ME1 (there are more M&Ms in a single bag than there are people who saved Kaiden) or for DLC companions only a handful of players were fans of. They pulled it off beautifully imo until their narrative cluster**** of an ending flushed the whole narrative retroactively down the tubes, but I think a lot of developers look at the reactivity of the ME series as more of a cautionary tale than an aspirational one. The more impactful the choices, the more mutually exclusive (and multiplicative) content you have to implement. Even if they wanted to, Obsidian doesn't have nearly the money needed to pull of anything close to that kind of reactivity over the course of a trilogy. It's extremely labor intensive and my guess is their telemetry showed that too many players made similar core PoE1 choices to really justify it. In addition, there are so many players who didn't even finish the first PoE. It is just not worth it to provide much continuity between sequels. Too much work for not enough payoff. I wonder what will Obsidian do with PoE 3. Will they have to write content depending on players decisions in PoE 1 and PoE2 combined?

To be fair, most of the times they don't appear to know it the moment it is done, but the moment you first return to their area. Which can be hours or days later. But word of mouth indeed travels quickly in these RPGs. NPCs can wait for you indefinitely to solve their quest but when you finally do - someone brings the good news to them faster than you can return to them. Maybe they use horses.

I'll try to defend the current system - not because I think it is very good, just providing some counterpoints. Reputation doesn't say that you are a cruel or a benevolent person. It has nothing to do with what kind of person you are. It represents what people have heard about you. Person A has heard that you are cruel because you have killed somebody in the past. Person B has heard that you are benevolent because you have helped someone in the past. Your score of 1x Cruel and 1x Benevolent represents that. Both persons can react to you - depending on what they have heard about you (or what is more important to them). With a single scale you would be zero on that scale (neither benevolent, nor cruel) and as a result reactions based on you being cruel or benevolent won't fire at all.

I would also want to see this as an option in the game itself. Make me able to customize my experience at the start of the game. Don't force me to choose between easier enemies AND lower count of them vs stronger enemies AND higher count of them. I have had this same problem with Icewind Dale. Years later such an option was added in the Enhanced Edition.

There wasn't such an option in the first game and neither in the second. You mean something like step advance? Or advancing the game while you hold a key?

This is not intended. Even if AI is active your commands should always have the highest priority. What you describe seems like a bug.

I'ts not hard to see that both this forum and reddit are filled with people disappointed in the game. And Obsidian, instead of holding its ground and focusing on minor changes and bug fixing, keeps overhauling it to try to cater to everyone. Completely randomly, not by releasing big changes with e.g. a DLC or an expansion pack. This is the result, and it's something I've said elsewhere some time ago: games like these are not meant to be balanced for months after the release, because it makes their sales suffer. I'm guessing if actual sales number don't convince them, nothing will. Pillars got through this phase by being something fresh. cRPG genre was a wateland at that point and Pillars 1 filled it, so it sold pretty well despite the developer's approach. However, as you can see, you only get one chance to get away with this kind of design philosophy. All I hope for at this point is that they manage to scrape enough to get POE3 going. And that they finally create an MMO game, where they can go bonkers with all the balance they want! Maybe I am as thick as Obsidian but I am still not convinced that this is the reason for the underwhelming sale numbers. There are so many possible causes, how are you sure this is the exact reason? And I don't think Obsidian is overhauling their design, maybe they are just trying to bring the game even closer to their original vision while at the same time reacting to the feedback.

Unfortunately they can't make that battle's difficulty scale gradually. Since this is an open world game, players can choose to do this battle at levels 14-20. If they scale the battle's difficulty down, then players will be able to win it at level 14 which will be ridiculous since this is the final battle. This is just meant to be the game's grandest challenge.

Someone on this forum explained how to mod this easily.