AnjyBelle

I wondered, too, but if you look here http://forums.obsidian.net/topic/81379-20-dlc-ai-discussion/ zero replies. Since I posted this, I watched the AI under different conditions. IMO it needs a total rework. Examples: The Paladin should use the talent "sworn enemy" if set to "aggressive". Sometimes she uses it, sometimes not. If she uses it, she seems not to distinguish week enemies from strong ones. For me it looks like a totally random use of the talent. The cleric set on "support" uses the talent "...radiance" hardly, even if more than one companions are under 50%. Rangers use their talent "wounding shot" also randomly and - like the paladin - seem not to use it on the hard enemies exclusively. Often I notice, that if a fight is over it is indeed not, because one of the companions ran away to fight something FAR away from the group, which results mostly in a defeat. As long as the one who ran off is "dead", the fight won't finish and I am forced to go to the "dead" and continue the fight there. This is especially annoying, when my party is heavily wounded (= health bars are under 50%) and I would need to rest. I believe that this happens more often in dungeons, where the space is too narrow for all to reach their targets. But - as others reported - it happens, too, out in the open. A good place to check this is the first level of Od Nua. Enter and fight the mob in the room southwest. Proceed to the room in which the looters are. Here my "disruptor monk" runs always further southwest and triggers the beetles there instead of fighting the looters. IMO the "special roles" like monk-disruptor or a rogue set to "cautious" and a cleric set to anything else than "aggressive" are not really working correctly. A pity for me, because I really like a functional AI like we had in NWN 1+2 (Tony K. !!!) and in DA:O. As someone stated in a Steam-thread: The AI is capable to manage fights against easy mobs but fails miserably when things get tough. @ Synfrei Clear answer: No. When they get charmed or confused they hardly run away. Instead they just stand there and mostly attack the own group. I've never seen one running away. In my example above (looters in Od Nua) is no enemy caster to charm or confuse someone in my party.

Thanks for taking care

Thanks for the links. Here is a savegame: http://s000.tinyupload.com/index.php?file_id=00228705844517632918 My "research" so far: 1) Precision enhancement because of a better weapon quality applies after or during the first fight 2) Talents like "Hope Eternal" apply the boni after the first enemy was hit by the equipped melee weapon I hope, we see a hotfix soon.

Output Log: http://s000.tinyupload.com/index.php?file_id=81468415937651527020

Seems to be a bug. I wrote it here: http://forums.obsidian.net/topic/81646-precision-bug-in-201/

I noticed, that the precision value won't change, when I equip a fine weapon instead of a normal one. Example: My leveI 2 Cipher has a normal hunting bow with a precision of 30. Now I found a "fine hunting bow", equipped it, but the precision stays at 30 Same on my Barbarian, when I equipped a fine weapon instead of the simple one. Did I miss something like new balance stuff or is this a bug?

As I wrote: It was VERY tedious and I could only win because of the three figurines I found in Defiance Bay. And the chest is right beside him? OMG....

So, there is this "Drunken Orlan" sleeping. I can sneak up to him, take "his key", go away or kill him. When I try the latter, he wakes up and - although weakened - is still hard to defeat with my current level 9 party. I finally won, but what now? If I "go away" and attack someone else in the camp, who is in a radius of 100 m around the drunk, he wakes up and wastes my party almost solo. I realize, that this camp is not meant for levels below - let's say - 12. Nevertheless, I could win against some of the packs there when I pulled them carefully. But it was tedious with most of my partymembers down. My question: Where is my advantage, when I take "his key"? I scouted the camp with my rogue and found a chest, which requires 15 mechanical (!) but was obviously something else and not for the key. Should I just ignore the drunk, walk away and try to avoid a fight with him? Any other insights?

Rightclick beside the character (or the entire party) on the ground and kind of a compass rose will appear. Now turn the chosen characters in the direction you want.

Since the beta-2.0 I play with the AI. In my eyes it was certainly a tremendous step forward to implement this in PoE. Nevertheless, IMO there is a great need of finetuning it. Examples: The map to the east of Dyrwood Village, where the Ogre's lair and the dragon egg are located. Right after entering the map, a few steps to the north is a large pack of these annoying beatles. While five of my custom party fight the pack, my monk-disruptor runs down south, where the next pack lurks and pulls them into the ongoing battle. Besides, how would she know? The southern pack is over the river, hidden in the fog of war. The same in the dungeon with all these cultists. Instead of fighting the visible ones, someone of the party (mostly monk or rogue) run into the next room to trigger even more mobs. Sometimes someone just runs elsewhere and sometimes one ore more of the party are "glitched", unable to fight, just "sliding" aimlessly over the ground. In generally I think, the AI works, although some of their decisions are hard to understand. I often see the barb not using her talents, or the pally not engaging with her flames etc. My settings: Everyone to "aggressive", use per rest abilities enabled, and aggressive auto attack. Only the monk is set to "disruptor" plus aggressive. Current party: Pally (main), barb, monk, chanter, priest, rogue. Your opinions and experiences?

That's not really how it works. Each class has its own accuracy rating (and each monster does too). That accuracy is used whether it's attempting to charm you, trip you, punch you, or shoot you. So lets say that your monk has 72 will, and the 'shroom has 52 accuracy. That means it will miss you 35% of the time (meaning no effect), it will graze you 35% of the time (meaning you're charmed, but for half the duration), and it will hit you 30% of the time (you get charmed for the full duration). So even with accuracy that's 20 less than your defense, you're still going to get charmed 65% of the time. Ok, I see my misinterpretation of INT. Nevertheless, IMO this is something to be looked at. 65% success rate for a common mob just doesn't feel right. Please, I realize that this is not a big issue, because if the mushrooms charm my party or not, the fights are easy to win, especially when you "pull" them away from their stationary buddies. Perhaps I am still thinking too much in categories like spell-DC, D20-saves and how many times my bards / sorcs / wizards in a D&D game missed with their charm spells (80% ?)

Well, you see it just the other way around: The consequence would be, that every mushroom has a 15 higher accuracy than my deflection. As posted above, my monk has 72 will-"deflection" without her personal buff (the one which can be switched from physical to mental protection). If I understood correctly, then it is INT which influences the accuracy of spells? If so, then we have lots of Einstein-like mushrooms in PoE Whatever it might be, a mushroom shouldn't have stats like that. Nothing against the occasional hit, but they hit always.

But besides that the corresponding spells of e.g. the Griefing Mother seem to have a a noticeable miss-chance. It's always hard to tell, because the "battle fields" are somewhat unclear. Sometimes I see an enemy with a green circle, mostly I see no results at all. Unlike my leader I've never seen the leader of an enemy gang being charmed / confused etc. So, they seem to be - as always - resistant to these kinds of spells.

What I mean about these charms, confuse-spells and the like is the impression, that they seem to be unresistable, no matter how many sides my dice have. I scroll through the log after a fight but never found an entry like "Enemy X attempts to cast <charm, confuse etc> on companion Y; Companion Y (D100 - whatever...) failed / succeeded to save..." Even the Griefing Mother with VERY high INT is victim to these spells on a regular base. IMO there is something ...intransparent.

Perhaps I'm thinking too much in AD&D-categories when it comes to saves during battle. Nevertheless, I wonder how exactly this is calculated in PoE. Example: My monk is currently level 11. She has a base INT of 10, buffed with items and from an inn it may reach 14-15. She wears all these rings of protection and has as a trait "mental fortress" or the like. Her current defenses are 69 / 82 / 82 / 72. She seems indead to be unstoppable and goes down hardly. But... ...every idiot mushroom and it's mother can "turn" her (charm or whatever it is called). Of course the same happens, if an enemy spellcaster has the corresponding spell. I don't wonder about the rest of my party, but the leader with these def-stats? But maybe I am totally wrong and her def-stats are not really good or I don't understand the mechanics at all. Perhaps someone could enlighten me, if there is even a will save against charm-spells and if yes, how is it calculated. Don't get me wrong: I'm not whining like "omg....my monk can be charmed...I gimped her, I must start over..BAD EVIL OSIDIAN !!!" Nothing against an enemy cipher who succeeds with his spell over my monk's resistances. Such is life. But these mushrooms? When I get close, they are dead with one or two hits. They qualify perfectly for the title "annoying critter" and shouldn't be more than a short nuisance. Weapon damage If I understood correctly, then it works like this: When a weapon has several dmg-types like the great sword, then only one is applied on hit, in particuar the one which hits the lowest res. What happens, when I add a third dmg-type by enchanting the weapon? Will the elemental dmg hit always or only then when the target's highest weakness happens to be agains this element? Example: A weapon has slash / pierce and I add fire. The target's lowest res is against pierce, so the pierce attack would hit. Because the enemy is very vulnerable to fire would both pierce and fire hit?

Same + them turning against me during a manually solved assault of bandits.

AFAIR the M&M-games I-IX were never "turn-based" per se; initiative, speed etc. determined a character's turn. But perhaps you didn't notice that. And of course - the game changed over time. From 6 - 8 party members the group shrinked to 4. But nevertheless, the player had full control over them. I agree - there were simplifications in the later sequels in regard of skills, but never down to a level which would nowadays be called "dumbed down". But still - all sequels were full scale RPGs, with nice playtimes (50 hours+), with an abundance of items and stuff and the possibility for individual party development. Thus, no party was like the other. Tons of quests and sidequests, many towns, no level cap (at least I never came close) and... and... and You think, that the example is "weird"? I don't care. Erm, can you check your quote please? I didn't say that, I think you meant to quote C2B! Oh... sorry. Of course you're right I meant Post #31

AFAIR the M&M-games I-IX were never "turn-based" per se; initiative, speed etc. determined a character's turn. But perhaps you didn't notice that. And of course - the game changed over time. From 6 - 8 party members the group shrinked to 4. But nevertheless, the player had full control over them. I agree - there were simplifications in the later sequels in regard of skills, but never down to a level which would nowadays be called "dumbed down". But still - all sequels were full scale RPGs, with nice playtimes (50 hours+), with an abundance of items and stuff and the possibility for individual party development. Thus, no party was like the other. Tons of quests and sidequests, many towns, no level cap (at least I never came close) and... and... and You think, that the example is "weird"? I don't care.

I try to give you an example: Years ago, the "Might & Magic" series was a big brand. Whenever there was a number behind "Might & Magic" (from I to IX), the gamers knew, that this is a sequel to the R P G-series, in other words: The player could be certain to receive exactly that what the name suggested: A full scale role playing game following the tradition of it's predecessors. Then there were the other "Might & Magic"-games: Heroes, Crusaders, Warriors, Legends, Dark Messiah etc.; all with the "... of Might & Magic" in their titles. Now the customer knew, that these games will be different from the R P G-series, what they were, indeed. These "offsets" were liked and disliked, but noone complained about being intentionally mislead by a package with a brand's name and totally different content. And here is OBSIDIAN's lapse: If they would have called it "Dungeon Siege: The 10th legion", I guess, you would see 50% less complaints on all boards worldwide. But they didn't.

I doubt that, really. When I played my first RPG back in the 90s - it was "Blade of Destiny", first of the ROA-trilogy - I wondered, when I finally can start to play, because the character creation of six party members took several HOURS. But as a result I identified myself with all six, from the beginning until the last act of part three (we could import the party into the sequels) many years later. The same later with Wizardry VII (W7) and many other games of that time. IIRC "Pool of Radiance II" and ToEE were the last games, in which this was possible respectively necessary. For Wizardry VIII I even dug out my old 486, because I REALLY wanted to import my W7-party! I agree, that today this kind of (lengthy) creation process could repel people who are new to the genre. Some might start to love it, others won't - for the latter there are always premade heroes. If someone like the CEO of OBSIDIAN says things like that, I don't believe him to tell us his personal opinion. IMO these people tend to express themselves like politicians: They justify their current work with insinuations about what the player wants or not. The truth is - again IMO - that the development of complex RPGs with excessive character creations, large skill trees and a polished combat system is much more expensive than the "streamlined" console-counterpart, not to mention the fact that it would be - at least - hard for a console to handle the subsequent complexity. I'd really like to see a remake of such an old school RPG with all it's features and additional "shortcuts" for the "casual player" (premade characters, auto level etc.). After this I'd bet, that most of the gamers would have adopted the NON-streamlined features, once they played the game with "shortcuts" and then with all it's features - for more than 200 hours, of course

+ 1. + 1

Ok, this last one. No offense meant, but I have the impression, you interpret too much Of course I can only speak for myself. When did I write things like "the legion is good"? I'd say, that the "new legion", refounded by my Avatar, Odo and the other persons are "good", because they fight against someone who ravages the land. "Good" in a certain way: Do I kill the "Gentlman"? That would translate to "lawful good" in old D&D-terms. I let him alive and make him work for the legion = chaotic good. The same with that Archon: I kill her = lawful good, because she killed civilians and seized a city. I simply let her go could be neutral, to send her away with a message for Jeyne again chaotic good. Perhaps you agree so far, when I say, that all these actions make me "good" in a certain way, but won't make the legion as a whole "good" or "evil". One might see the "new legion" as an organization with tendencies to "good", though, because for the moment they want nothing but to end this war. But... When I suddenly turn against them and betray anything for what I was originally "hired", then I would certainly be "evil", especially when I do it for *my advantage* with not the least interest, what consequences this might cause. This kind of discussion you can see on old Bio-boards, when people talk about the old AD&D-rules, the actions of their characters and the pre-built "baddies", like good old Irenicus and the like. A never ending discussion, I can tell you that.

1. Yes, you have to beat the boss. Like in JE. Except there you had a third option regarding her you can make. However you also have several options to deal with Jeyne AFTER the fight in DSIII. 2. No, at least not how someone would define "real" allies. 3. Jeyne was none of that. She wasn't treachurous nor was she necessarly evil. And yes, you can be in a certain way. You can lead the legion into many ways. Just in a lot greyer ways than black and white ones. Your last two questions make me wonder if you "got" everything that happened in the games backstory and the Legions role in it. No offense. I didn't say, that Jeyne was "evil". But I would be, if I'd change sides and do anything for my advantage at the cost of the legion. But I guess, this all leads away from the original topic. If you want to discuss this further, I'd happily respond to a your PM.

Ok, I agree... there ARE different endings in a certain way. Perhaps my definition of "different ending" is different from yours Let's see: We have to beat Jane, right? All of us, no matter what we "decided" to do before the last fight. And we have to be allies with the royals, right? We can neither change sides nor can we take Jane's place and be "the incarnation of treachery and evil", right? Remember Jade Empire? THAT was a different ending.

I'm aware, that - due to the the design of DS III - many things are quite impossible: 1) replay+ because characters are maxed out after the first run 2) Much longer gametime for the same reason 3) In depth character development with only 9 skills as a tribute to the consoles. In the end, all Katarinas (as example) of all players will be at least very similar 4) Despite some exceptions dialogue options have no impact; so every ending is about the same (perhaps an addon later will take some descisions into account) With the current concept anything beyond 30 hours would feel artificially lengthened. But IMO that does not necessarily mean, that this *new* DS-concept is good.