gurugeorgey

I like the look of that system thematically, but who knows (until it's tested) how it would play out in a videogame. Fundamentally, the problem with RPGs is that the idea of creating a person by assigning attributes is sound, but the attributes need to be abstracted and distinguished more carefully, and more attention needs to be paid to how attributes can influence each other. IOW, some attributes should be the resultant of other attributes. Also, some attributes can improve with exercise (and exercise of other attributes) while others are more or less fixed and can only alter magically. BUT THEN, you have the whole problem of how that pans out in a videogame, in which the developers have to make it hard to make a gimp build. My solution is that choice for the attributes of the person and choice for the class have to be tied together in some way. I think the flaw in all the systems hitherto is the the idea of having a) a "pool" of points, and b) a choice of attributes and choice of class be choices you can make separately. They have to be tied together. IOW, every class has a "graph" with a "band" of required-attributes-for-functionality, the total pool of points is fixed and the same for every player, and most of them are assigned to the class within that "band", but there are a few points left over for free choice. You either choose a person (a bunch of attributes representing a person) and the class those attributes would be suitable is chosen for you, or you choose a class, and the attributes come with it. You then have a few points to play with, within that "band", for flavour and rp. That way, you can't make a gimp (of that class), but you can shape the character somewhat, either for rp flavour or for min-maxing, but only within the lower and upper bounds of the "band" of functionality. Any attribute going beyond that "band" in the upper limit, has to be attained either by exercise (if it's an attribute that's improvable by exercise), or by magic. tl;dr What's needed is a setup where you can't make a magician who's dumb as rocks and strong as an ox, because such a "mage" would never have gotten into mage college and been able to become a mage, but you can make a mage who's still functional as a mage, but was the laughing stock at mage college for being the dumbest mage in college, and can also lift heavy backpacks, but isn't quite as strong as an average warrior. Simlarly a physically wimpy "warrior" would never have survived long enough to become a warrior. But you can have a particularly cunning or perceptive warrior, who's still functional as a warrior but has an edge in conversations.

They ought to have it so that choosing your class automatically chooses a particular "landscape" of attribute levels, but then you have leeway to put a few points up or down here or there for particular flavour. I mean, if you're talking any sort of realism, a character would not have gotten to become a particular class without having those attributes as part of their nature. A physically weak "warrior" would have died quite early in life, a dumb "mage" wouldn't ever have gotten into mage college. But you might have a mage, who's a mage (i.e. still somewhat effective as a mage, passing skill checks for magey-ness), but a bit stupider than other mages, and a bit more physically strong. Or, you could just min-max and put extra into those stats that reflect the class speciality and dump the others - then you'd have a particularly intelligent mage, who's particularly physically weak. Then the game ought to track and reflect those choices you've made (i.e. present you with dialogue options to suit) (i.e. if you have a somewhat stupid but physically strong mage, then the game should shape itself a bit around that - maybe give you a dialogue check that that would allow you to physically intimidate someone). It's really the whole idea of choosing random attributes for a random person and then choosing a class, that's borked. i.e. it's the disconnect in choice there - you either choose a class (and thereby automatically choose the attributes that go with it), or you choose a random person with random attributes (and thereby automatically choose a class in which those attributes will be effective). But you can't choose both separately, it just doesn't make any sense. The other approach would be to, yes, create a random person with attributes you want, but have them discover a class that suits their attributes in the course of gameplay - i.e. they become a class through the course of the game by discovering what they can and can't do with the attributes they have.

I think it all just depends on difficulty. On Normal, the game pretty much lives up to the idea of "build how you like, even according to rp characteristics, and you'll still be viable". So long as you pay some attention to the "recommended" stats, you're golden. But on Hard, and especially on PoTD, I think you will have those two "paths" someone mentioned above, as optimal - CON, PER and RES for tanks, MI, INT and DEX for damage dealers, everything else dump.

I do think it makes sense from a sim point of view, that you'd be mostly sneaking cautiously through a dungeon, I don't think it's a problem at all that that's mandatory for dungeons. But I think they could have taken a bit more care with placement of purple items out in the world. They're in enough totally random places that the completionist is going to be tempted to fast-sneak throughout the game, whereas if it were clear that such items would only be in places that logically made sense (e.g. abandoned campsites), then you'd only go into "hunt the purple" mode in places where it made sense to do so.

It's an odd thing because the painted look of portraits is a big part of the charm of these games, for me, but at the same time it is frustrating (granted, in a #firstworldproblems sense ) not to be able to have a portrait that matches the character you've created. Maybe for an expansion, Obsidian could commission a ton more portraits, and have the character creator START with portrait selection. That way, you create the 3-d avatar according to the "feel" you get from the portrait you choose at the beginning? Or, what about having a portrait creator similar the amazing character creation in EVE Online? That way you could have an incredibly rich and detailed 3-d creation for just the portrait (tons of options, backgrounds, lighting, etc.), then you take a posed "photo" of it? Then you model the actual in-game 3-d avatar as a simplified version of that?

... what the hell were going through Obsidian's minds when, having designed a totally new "engagement" mechanic, the first proper dungeon in the game is full of mobs that bypass that mechanic? I mean, I think engagement is actually a great mechanic, it works really well, and I'm sure introducing mobs that bypass it would be a great thing to spice the game up maybe a few dungeons into the game. But the first proper dungeon? What were they thinking? When you're just starting a game, everything is a mental fog, and it takes a bit of time for things to coalesce into some sort of sense. Engagement is different enough that it takes time to get the hang of. The first proper dungeon should actually train you in it, and to some extent, the way the Eothas temple is designed, with all the bottlenecks, that was obviously the intention to some degree. But then you plonk all those mobs that just totally bypass it and go for your Wizard or whatever? I'll bet you any money that this annoyance has single-handedly been responsible for putting lots of people off the game in the early stages. It very nearly put me off the game at the beginning - because it just turns into a chaotic scrum, when at that point, the game should be presenting you logic, so you can get a handle on things. But otherwise, yeah, fantastic game - story, art, music, sound, characters, combat, everything. And big as it is, I only wish it were twice the size! Looking forward to an even bigger expansion/game in the near future!

Is the new slider supposed to do something? Still getting no anti-aliasing on my Nvidia GTX 760.

still no anti-aliasing with Nvidia 760 - still pixellated, slider does nothing.

I actually agree with the OP, there's something fishy about Ciphers vis a vis mobs. All that positioning and going in with the tank etc. is great, and works a lot of the time, but sometimes you can't get that perfect positioning, and on those occasions when there are so many mobs they swarm around the front line, they will go for the Cipher, rather than the Wizard or Druid or Chanter - 9 times out of 10. Could still be a random result, but it's something I noticed myself enough so that this topic alerted my interest. Maybe it's something to do with a raw damage dealer or cc-er being higher on a threat table or something?

I like 'em, don't see what all the fuss is about. I quite enjoyed reading the stories.

The solution is to use spells to the extent that you need them. And try to use them from weakest-viable-first, saving "big" spells by preference, but using them if you have to. I think that's what the game expects you to do, and if you do it that way you'll be resting a reasonable amount of times, as rationed by camping gear. I think it's a good system, too much per encounter would cheapen the strategic layer and turn it into MMO style gameplay, and that's anathema to a game like this.

The idea is that if you spend time learning magic and spells, you don't have time to learn to practice sword fighting, and vice-versa - which is actually realistic (granted the existence of intelligence-based magic that you have to learn, and physical combat skills that you have to train). Why should that matter? Isn't it just a game? Well, yeah, it's a game, but it's also supposed to be a roleplaying game, which means that things have benefits and costs, just like reality in general. Wandering down path A precludes you wandering down path B. Choices and consequences.

Heh, I've registered an account to respond I think you're roughly right, but I can understand a lot of the complaints. The combat system is fun once you get the hang of it, but it's not fun out of the box for some noticeable number of people. I think to make it fun, the tutorial section would have to be a bit reworked, to introduce the concept of engagement in a more logical, step-by-step way, because it's quite different from what one is used to from most games, and especially from the old IE/D&D games. Personally, my favourite combat system of all time in this type of RPG was the turn-based implementation of D&D in Temple of Elemental Evil, and for the life of me I've never understood why that hasn't been developed further and iterated on. It was the perfect implementation of D&D. (I guess realistically, the reason ToEE never took off is because the game was buggy and unfinished on release - problems the Co8 modpack has solved since then.) But I am starting to get the hang of this game's system and starting to appreciate it for what it is, not for what I'd prefer it to be but isn't. Maybe it needs a teensy bit of refinement as you say (maybe making the ceiling for the penalty for disengaging just a little bit lower ), but overall, it's got its own charm.