MReed

Well, lets remember that Chapter 5 is also a major Hub. You're in the Underdark. It's massive. There are dungeons within dungeons. There's an open area, And a town, and a little village, and a bajillion side quests that will take you in and out of all of these places for hours and hours and hours. And then there's chapter 6, where you return to the world map. (just like chapter 2, 3) So lets recap: The, open, non-linear chapters in BG2 are: Chapter 2, Chapter 3, Chapter 5, and chapter 6. That's practically the whole game. What's the gripe again? Side note: We see a lot of people complaining about how BG2 is starts off all open and free and then pushes you into a linear main quest for the second half of the game....almost as if they dropped the ball or something. And then, we see these same people complaining about how illogical the main quest is (you gotta rescue Imoen. It's Urgent! But....But.... Lets do Athkatla first, and get our stronghold, and save trademeet, and explore the Umar hills. And Kill Fiirkrag. Then lets go after Imoen (hopefully she's not dead yet!) Which I find bizarre. People, BG2 is an RPG, remember? It's supposed to be up to YOU to act on this Urgency or not (depending on your motivations). And, interestingly enough, the entire game's progression and pacing begins making SO much more sense if you DO role play the game. Here, next time you play Bg2 try this: 1) Raise 15,000gp as fast as you can. As if it's an emergency. You can do this in the Copper Coronet alone (ie. without even leaving the slums) 2) Do Lindvail's (or Bodhi's) quests 3) Go to spellhold and rescue imoen. 4) do chapter 5 5) Take your time and Explore the world in chapter 6. 6) Take care of Irenicus. ^If you do that, you'll find that there's nothing wrong with the narrative structure at all, or the quest volume pacing (or whatever Josh's strange gripes are) Um, during Chapter 6 there is more urgency than in Chapter 3 -- you and Imoen have both lost your souls and the game tells you that this will kill you "soon" and Iranacus and Bodhi are currently attempting to commit diecide and ascend to godhood themselves. But other than that, no reason not to take your time...

tl;dr: Kiting is indeed a degenerative mechanic (making meleers obsolete). The AOO / engagement mechanc being implemented is even more degenerative. Um, yes, and your point is? If you choose to play a class oriented about ranged combat, then yeah, it is almost never a good idea to have them engage in melee (and vice versa, of course). The goal is to force the player to need friendly meleers to avoid this problem, which is a good thing in my book. Having a single strategy (kiting) that is ~100% effective at preventing meelers from doing their thing is degenerative game play -- it renders melee oriented classes irrelevant for both the party and the enemy. That's pretty much a text book definition, actually. When you add the fact that kiting requires enormous amounts of micro-management, especially with multiple party members then I'm not sure why anyone would defend this as a desirable component of CRPG gameplay. My objection to the mechanics being discusses is not that it eliminates kiting (which I think it probably do quite well), but my belief that the monster's AI will be totally incapable of dealing with it in a vaguely intelligent way. Based on this belief, I suspect that this mechanic will lead to combat where the AI enemies meleers are still utterly incapable of threatening friendly ranged combatants (because they are always engaged by friendly meleers first, and the AI is incapable of disengaging effectively [in a way that allows them to reengage with ranged combatants]) while the player is able to get meleers to engage AI ranged combatants with a high degree of reliability (by either overriding the [dumb] pathfanding to avoid engagement, or by disengaging in effective ways). Basically, I believe that this mechanic successfully eliminates kiting by providing a much more effective way to achieve the exact same result -- the only difference will be that is far easier to exploit, with less to go wrong. After all, if the player tried to run a party with 6 ranged combatants against 12 AI meleers under 2E D&D rules, the odds are fairly good that the player would eventually make a mistake (or be bitten by a pathfinding glitch) and be forced into melee combat with disastrous results. Thus, most people play IE games with 2 dedicated meleers, 2 backup meleers (to cover "leakers" or kiting failures), and 2 full-time ranged combatants. The very weak AI (out of the box) also helps. Given the anticipated effectiveness of AOO / engagement model being implemented, I'd expect PoE parties to run with 2 (maybe 3 -- depends on encounter design and how quickly fighters get multi-engagement abilities) full time meleers, with the remainder being full time ranged combatants. Heck, if you are able to couple this mechanic with natural bottlenecks (in 95% of combat encounters), you might be able to get down to 1 full time meleer and 1 part-time, the rest being 100% ranged -- but I think the odds are against this.

From my perspective, I look at it like this: Assuming that different builds are possible, and that those builds have unique capabilities, then (by definition) exactly 50% will be "better than average" and 50% will be "worse than average" when builds are compared to one another with some common criteria ("combat effectiveness" being the relevant term here). It is almost certain that the builds will be distributed along a standard bell curve, with the vast majority being within one standard deviation of the theoretical "average build". This means that you can't simply eliminate bad builds from the game -- if you identify the bottom 25% of builds, and modify the mechanics to make these builds "better", all you've really done is push the average build "up", reducing the interval between "bad" and "good", but there are still bad builds available, they are just different bad builds. The only way to truly eliminate the risk of a player having a poor experience / having to restart the game due to make poor selections during level up is to altogether eliminate the functional distinctions between potential builds. While I doubt that Sawyer intends to go this far, I certainly haven't seen anything from him that says where he is planning on drawing the line -- and that concerns me. To use an analogy -- "poverty" is very much defined in relationship to your personal experiences. In the US, someone living in a poorly maintained apartment, with running water, who gets 1200 calories a day in food is considered to be living in poverty. In other areas of the world, though, that same person would be considered to be "average", or even "above average" -- after all, the have shelter and clean water.

Using a figurine (or other summoning device) wouldn't be kiting, now would it? At the start of any engagement, I'm assuming that all refresh counters start as "OK to attack" (or 0, if you prefer to look at it that way). Therefore, each character / monster can make an attack (or move) as they wish, thus the logic above. However.... Once the monster has made the first attack (in step 7) if the archer can successfully disengage then there is a potential for kiting -- to avoid it the monster's AI needs to be configured to force the monster to stand still after the disengagement for however long is required to satisfy the refresh timer, and only once this has happened restart the pursuit. This will, of course, allow the archer to further open the range, making re-engagement far more difficult. If the monster simply immediately restarts the pursuit then the situation you describe will occur (the monster will get within melee range, then be force to stand still for some amount of time, during which time the archer can withdraw out of engagement range). Having monster's stand still after disengagement to satisfy the refresh timer would produce... Odd looking combat, to say the least.

How does recovery matter anyway in a Kiting situation where the pursuer will also be moving? Sawyer can say whatever he wants, but the only ways to prevent player kiting is to either a) design a totally unnatural system where everyone on the battle field is rooted in place once combat starts unless they've got spells or talents that let them move or b) incorporate copious amounts of developer cheese and make all enemies move twice as fast as the player, or c) punish kiting by making all attacks against the kiter be insta-kills. As your link above indicates, however, all Sawyer has succeeded in doing (and even then only in theory) is to make Kiting a little more difficult to carry out than it was in the IE games. No, I'm pretty sure that the sticky engagement / AOO mechanics will suffice to make single person kiting impossible, even with a ranged attacker. The scenario goes like this: 1) Archer shoots at (& hits, for the sake of simplicity) the monster -- archer's action refresh timer starts (at, say, 2 seconds). 2) Monster moves towards the archer, attempting to engage in melee combat. 3) Archer retreats (at the ~same speed as the monster). 4) Archer stops and attempts to make another attack -- nothing happens, because he still needs to be stationary for 2 seconds before his attack can begin. 5) Monster continue to move towards the archer, and is now able to close the distance 6) Archer's refresh timer expires, and the attack takes place 7) Repeat the above steps until the monster (making up "2 seconds" worth of distance each attack) closes to melee range. As soon as the monster gets within melee range, he gets to make an attack -- his first of the engagement (he has only moved up to this point), so no refresh timer applies. 9) The archer attempts to disengage -- but when he does, the monster immediately gets another attack (with a bonus) -- if that attack hits, then the monster establishes engagement (guaranteeing another attack if the archer attempts to move away again) and prevents the archer from moving at all for some period of time (say, 0.25 seconds) , allowing the monster an opportunity to reclose the range. * The AOO attack fires even if the monster's action refresh timer is active, and doesn't reset it either. * Any character can take an unlimited number of AOOs against a single target in an arbitrarily short interval -- feats / perks / class abilities that increase engagement limits simply allow infinite AOOs against more than one foe during some interval. 10) If the archer does manage to break engagement (either via a special ability, or simply via persistence), then repeat the above steps again -- except that the initial range is likely to be close enough that one cycle will suffice to get back into melee range. Now, the above assumes a couple of key things: 1) The pathfinding for monster's is 100% accurate -- if monster's can be decoyed into taking bad paths around battlefield obstacles, then kiting may be possible. 2) The monster's AI fixates on a single target, and won't deviate from it no matter what else happens on the battlefield. Otherwise, kitting can occur by bouncing the monster between two archer's. 3) The player isn't able to lead the monster into engagement range of a friendly melee combatant -- if he can, then the monster will suffer AOOs if they try to disengage (although engagement limits may come into play here -- if the friendly melee combatant is already engaged with another monster, then they will need a special ability to allow them to make AOOs against the archer's foe) I suspect that the practical impact of this mechanic will in fact be to achieve some combination of "A" (combatants, once engaged in melee, can't practically move away without special abilities) and "c" (moving away from melee combat is never the right decision), and I don't think it is a good idea as a result -- but it likely will make kiting impossible, or at least far, far more difficult than it is in other games.

Responding in order: * Because they wanted to make this particular type of game, but recognized that such a product probably wouldn't recover its costs. With Kickstarter, though, they don't have to recover their costs, so even if the game was a flop (and they were aware that this was a very real risk) they would only loss they would incur is opportunity cost associated with assigning valuable resources (mostly staff) to a project with less profit potential than other projects that they could have worked on. I suspect that the morale boost (the staff wanted to do this project), the gain in skills (the developers have learned how to work with Unity) and IP rights (both technological and otherwise) factored into this decision as well. * Including a 10% profit margin into the Kickstarter goal is possible, and for all we know they may have done this. I'd consider that to be very unethical, however, and I would never even consider backing a project that announced it was baking profits into the base Kickstarter figure. Now, salaries are just fine, and at the individual level that could be considered "profit", as is budgeting money for overruns (unallocated funds "just in case"), but my expectation (and, I suspect, the understanding of most backers) is that 100% of the money contributed via Kickstarter (less fees and backer awards) will be used directly or indirectly to fund product development. * They have added additional staff to accelerate PoE's development, based on Obsidan posts that I've seen. Much of this is additional manpower investment is in the form of contracting out specific development tasks rather than direct hiring, some of it is in the form of moving resources that would have otherwise been allocated to a different project to PoE, and a small portion of it is in direct hiring. You are correct that there are economies of scale in game development (as in most things), but my understanding is that the savings realized as a result of this are being re-invested in the game (by further increasing the scale, adding additional features, improving reactivity, additional testing, accelerating the development pace, and so forth) so that a break-even situation is created.

Thanks for the read. Although I've read similar ones in the past, it was still educational. But I don't follow your logic. The article you linked never mentions Johnnies being opposed to the concept of balance. I consider myself belonging to that category too, and I love balancing (did it for years on an RTS mod). You can always self-gimp yourself, no matter how balanced or unbalanced the game is. In a cRPG, play without mages (Keyrock's all might MM X party does this right now). Don't use multi-class or dual class characters. Only use melee weapons. The list is endless, you're not limited by game balance. -- Now that I'm writing about this, I recall that I played DA:O on Nightmare with the self-imposed challenge of never allowing characters to be downed. I had to win all battles with all party members standing. That was a fun one. If a game was fully (or statically) balanced then, by definition, there would be no functional difference between factions / classes -- in MtG terms, black would have just as much healing as white, because otherwise it is "unfair", or in CRPG would not have classes or races that had any difference beyond the cosmetic. Chess is an example of a statically balanced game. In such games the skill of the player is exhibited by their ability to win despite a level playing field. Now, dynamic balance is something that most people, including myself, favors -- but this means that different factions and different classes have strengths and weaknesses, and therefore there are "good decks / builds" and "bad decks / builds". In such a game the player can (if he/she is sufficiently skilled) ensure their success before the conflict even starts, due to (meta-game) preparation. In a CRPG, this could include everything from only using the "most effective" abilities, exploiting limitations in the AI of opponents, to taking advantage of outright bugs in the games rules.

This article might be relevant to the balance discussion: http://www.wizards.com/magic/magazine/article.aspx?x=mtgcom/daily/mr11b Yes, a MTG is a totally different genre than CRPGs, but the distinction between the player types is universal. "Johnny" type players are strongly opposed to the concept of balance, because part of the point of this personality type is to make something that everyone else thinks is "bad" into something that is useful. I fall into that camp myself.

This has been discussed before (although not in a dedicated thread). Part of the problem is defining your terms: Net Revenue is what most people are talking about -- that is, the amount raised by selling the game less the direct costs associated with selling it. Direct costs in this case covers things like fees that need to be paid to Stream and Paradox, but doesn't include the Sawyer's salary (or any other person who works for Obsidian). Net Profit is what a few people are talking about -- that's the amount raised by selling the game less both the direct and indirect costs associated with the game. The costs for developers would be included in this number. In a conventionally funded project, "Net Profit" is the only number worth talking about -- if you don't cover your costs, then the investors who financed the game will incur a loss and will be less likely to finance future development efforts. With IKickstarter, though, things get blurry: the investors don't get repaid regardless of whether or not the game is a success or not, leading to the question of "is Net Revenue the appropriate measure of success for this game?" From a purely financial point of view, it'll be interesting to see how the IRS (US federal taxing authority) is going to answer this question -- normally, you can write off R&D expenses against the revenue generated from sale of a a product, but that's based on the assumption that you'll be repaying investors (which could be yourself, if you self-financed), and that's not the case here. Anyway, if Obsidian wants to receive conventional funding for a sequel, then they will need to have sales in the $4 million - $5 million range at the very least. If they only achieve sales of $1 million, then the game will be considered a failure by the financial community.

There will probably be plenty of documentation on file formats, and I suspect that the developers will respond freely (once the game is released) to questions / clarification requests about both how they've used the capabilities of the game engine and potential capabilities. But the release of any tools (even the tools that the used in-house to develop the game, "as-is") has not been promised at this point.

Based on the current statements made by Obsidian, it would be up to the community to develop such tools -- the funding for this project simply doesn't extend to developing these tools. And, to answer your likely objection: The tools that they use internally depend on the presence of commercial (not free) products. They may releases these import tools anyway (there was an active discussion about this some time back), but there is no funding available for developing modeling (or other tools) specifically to support the modding community.

I guess so... but then again, it shouldn't be hard for any decent modeller to create such backgrounds from scratch. With free 3D modelling tools and a powerful graphics card for rendering, any decent 3D artist can create convincing background images that could then be filled with a selection of destructable 3D objects that are used by PoE (crates, barrels). From what I've seen from the development screenshots, they are using an editor to apply basic-geometry cutouts to the 2D background image to simulate the dynamic light and create the pathing map. Assuming the community could get access to this tool (and also the tools used to create and script dialogues, npcs and quests), the modding possibilities would be almost endless. It might seem that this would be the case, but in practice... Well, look at the IE modding community -- the file formats are 100% decoded at this point, and have been for years, yet only one or two mods have ever been built that actually include totally new areas. There are a few that (illegally) leverage IWD content in BG, I believe, but 100% original areas is all but unheard of. In any case, Obsidian won't get in the way of anyone who wishes to create new areas (or other mods) to PoE, and will even provide some limited assistance (documenting file formats and the like), but have no plans to release any sort of tools that are designed to either modify the existing content or create original content for this game.

It is pretty clear to me that what Sawyer is saying here is "We won't include any segments of the game where you must have someone of a certain class (or even a specific companion) in your party to proceed with a critical path quest". For example, it is literally impossible to complete DA:O solo -- you must have certain companions in your party at certain times, because the game literally locks them into your party (I believe, for example, that Alister must be in the party at various points in the game). The most common reason for imposing such requirements is "We can't (or don't want to) write the dialogs / cutscenes to make since if character X isn't present". Less frequently, this is done by implication: certain abilities are needed to overcome a story obstacle that the player faces. An example of the later would be a game that included a "move massive boulder" spell and the critical path involved passing through a pass that as blocked by massive boulders. Note that this doesn't mean that side quests can necessarily be completed solo -- all that has been promised at this point is that the critical path quests can be completed solo. Now, it doesn't follow from the previous paragraphs that it won't be practically impossible to complete the game without companions. Indeed, it almost certainly won't be -- and, to the degree that it is possible at all, it will be because the players discover one or more exploits that the developers either didn't anticipate, and the developers may choose to patch the game to remove it. TL;DR: All Sawyer has said to date is that the critical path quests will be completable (in some form or another) by a single PC of any class / race combination -- it won't be literally impossible to solo the game, as is the case in some other RPGs. There is no reason to believe, however, that a solo character will be able to beat the game without using exploits of one description or another (such as the ones discussed earlier in the thread in regards to IWD2).

Instantaneous effects, such as fireball, magic missile, lightning bolt, whatever are both easy to design graphics for and pretty much irrelevant to the discussion at hand. After all, such effects are trivial (either damage or healing) and the log is right there if you need it. The problem, though, is with effects that have a duration (whether buffs or debuffs). It is easy to come up with a dozen effects that fall into one of these two categories: * Poison (more generally, damage over time) / Regeneration (healing over time) * Blind / Improved vision (accuracy buff / debuf) * Increased strength / decreased strength (damage buff / debuff) * Mage Armor / Weaken armor (damage prevention buff / debuff) * Haste / Slow (speed buff / debuff) * Protection from <element> / Vulnerability to <element> * Spell turning or Protection from spells And so forth -- we could probably keep this going all day if we wanted to. All of these buffs and debuffs share one common element -- they have a lingering effect (that the player might want to remove, or at least take into account when making actions in the future), and many of them have no obvious visual effect to make it clear that the effect is active. Note that the combat log doesn't do you much good for these sorts of effects either -- depending on how busy combat is, and how long the effect persists, it might have been 200+ lines ago that the effect was put into place. Reskinning existing animations won't work here -- if you decide "Protection from fire" will be represented by a blue aura around the character, and protection from cold is represented by a red aura, what do you do if both are active at the same time? So, ideally, a unique animination should be applied to each buff / debuff (at least the ones that are compatible with one another), and that gets very expensive very quickly. Thus, my recommendation of providing a way to see a simple text list of active effects on anyone, at any time. When feasible, of course, animination should be provided for buffs and debuffs, but eventually you are going to run out of time / money / ideas...

I'm always worried that lack of visual fidelity, diversity and animation quality will utterly destroy indie games. Sorry if this seems superficial but if you don't have those things your whole game is ruined no matter how "deep" or expansive anything else is. This is 2014., not 1995., so let's all stop pretending it is, OK? This is very much your opinion -- it isn't one that everyone on this forum will share, and I (for starters) disagree. Good graphics enhance a game, but only if there is the fundamentals exist in the first place. I'd far, far prefer to play (say) BG2 than play DAI, even though DAI is clearly a graphically superior game.

One problem that I see with the combat log in the Infinity Engine games is that it doesn't tell you why a target is immune or suffered reduced damage. It really should say something like "Hit: Immune (Improved Mantel)" or "Hit: Immune (Innate Immunity)". One of the big problems with the IE games is not knowing that the spell hasn't been cast (which, as has been pointed out, is mentioned in the log), but realizing that the reason you can't do damage now is because of the spell that as cast 200 lines ago. Another UI issue that I'd like to see addressed is that there is no way to get a list of effects that are currently applied to opponents -- much less descriptions of those effects (if you can't use them yourself). You really should be able to right click (or hover, perhaps) over an opponent and see list of all currently active effects, which you can then click on to bring up a standard "spell description" page that covers things like duration and effects.

For the record (because its the polite thing to do, even if the thread has moved on): I concede that I was in error when I stated that Obsidian was not integrating community feedback into their decisions making process. On the current topic of the thread -- given the POV and size of characters in the screenshots that have been released to date... Doesn't this amount to arguing over a few dozen pixels (when breasts are visible at all, which will be no more than 50% of the time) should or shouldn't be there? Not a big deal to me either way, obviously.

The only change that I'm aware of that made in response to backer feedback was to replace cooldowns with "/ day", "/ encounter", and "at will" abilities -- and that was back while the Kickstarter was running, I think. I'm aware of other examples where initially announced systems have been changed (most recently, the resting system was altered to allow resting in most places, limited by the availability of rations, with bonuses available for resting in appropriate locations), but... Based on the developer comment, the change was driven by internal playtesting or discussion rather than specifically backer feedback (see http://forums.obsidian.net/topic/66100-death-dying/?p=1443390 -- note that the change occurred "in the past several months", so it clearly wasn't driven directly by the thread in question). Can you provide examples that meet the following criteria: 1) Something was announced 2) There were one or more active discussions (either here or elsewhere) where concerns were raised 3) Whatever it was that was announced was changed in ways that clearly addressed the concerns With links, of course. Honestly, it wouldn't shock me to hear that such examples do exist, because there are such a wide variety of mechanisms where the developers sometime interact with the community that I find it impossible to keep track of everything that is going on. Basically, I read this forum with some regularity, and I read the Kickstarter updates, but that's about it. So if Sawyer or another developer is having indepth conversations on Twitter / Facebook / interview threads on various websites, in other forums, and so forth it is entirely possible that there are examples where there is a clear relationship between community feedback and game changes. But if there is, I'm certainly not seeing it...

Why? What data do you have that proves this? Baldur's Gate sold 2 million copies. If this game has half that number of sales from fans of the IE games, Obsidian would be in a much better place financially then they've been for a long while. Even more to the point -- the whole purpose of the Kickstarter (I thought) was to make a game that was highly unlikelly to be a mass market success because it catered to the whims and desires of a small "niche" audience of players. A project like that could only be funded through Kickstarter (or similar means) as people who actually expect a financial return on their investment would see it far high of a risk for far to small of a return due to the small known audience. Instead, as we now know, the actual purpose of the Kickstarter was to fund a game intended to be a mass market success that would normally be funded by publishers, but the publishers weren't interested. The opinions of the backers are totally irrelevant to the design of the game because they aren't and never were intended to be more than a small slice (say, 5%) of the total market for the game.

A nice thought (well, for me), but I just double checked the description, and it includes: "When an incoming spell targets the caster, the Reflection attacks the enemy's Will." -- unlike the D&D equivalent, the reflection effect can and will fail. In fact, successful reflections to occur less than 50% of the time if spellcasting classes have high will saves. Removing the effect will almost certainly be possible, of course, but I doubt that it will be worth the trouble in the vast majority of cases, which is the textbook definition of a "soft counter".

OK, hmmmm... Each encounter has a "fence" around it -- when the player (whether in combat or not) crosses a fence, all buffs are expired, but "per encounter" abilities aren't reset. Beat that.

However, given the focus that Josh has placed on eliminating pre-battle buffing mechanics, the logical solution to absolutely stamp this out is to simply cancel all ("standard" -- the kinds of buffs that the player can apply anywhere) buffs at the end of each encounter. This feature, combined with sharp limits on the abliity of the player to enter into combat without opponents present, should achieve the desired result. I don't think it is worth the cost, but then my preferences aren't particularly important to Sawyer.

If this is the case and multi-battle buffs exist, then the (highly dedicated) player will ensure that a minor combat encounter occurs immediately prior to any major encounter to save time in the major encounter. This would also have the benefit of being able to use "per encounter" buffs effectively twice in a single encounter (once in the trivial "pre-battle" encounter and then again, if necessary in the major battle). Also, if combat mode isn't a toggle then it will be impossible for the player to attack non-hostile enemies, including other party members. This would certainly be unusual in a western CRPG.

Thanks for the cite (I did read it back when it was new, but that was quite a while ago, and I had forgotten most of the details. If disengagement is not a feasible option without the use of special abilities (because each successful AOO aborts movement and the opponent will be able to reliably reengage) then my concern still exists: The player will be able to limit the AI to a single (losing) strategy far, far too easily. On the other hand... If the above only applies when the fighter is engaging a single target, then two (or more) targets should be able to reliably disengage by moving in opposite directions. To explain what I'm talking about: Fighter "F" is engaging two opponents,, "A" and "B", who are currently to the fighters "NE" and "NW" sides. "A" moves "North" and "B" moves "South" -- the fighter gets, and hits, with an AOO on both targets -- however, before the AOO occurred, a significant gap opened between "A" and "B" and the fighter must make a decision -- move "North" to restore his engagement with "A" or move "South" to restore his engagement with "B", but there is no position from which the fighter can do both. Therefore, one opponent "leaks" past (at a cost of two AOO's spread across two targets) the fighter. If the second scenario is an accurate description of how the engagement model works with multiple opponents, and the AI can be designed to take advantage of these scenarios when they occur, then... I'm still a bit concerned about whether or not the AI will properly prioritize the benefits of bypassing fighters to reach squishes, but that's a problem common to most, if not all, RPGs.

Has this one been confirmed? We haven't had the warrior class update yet, (although Josh has tossed some warrior skill snippets at us.) But I haven't seen anything yaying or naying an aggro mechanic. How do you figure this, exactly? The fact that your mage won't be able to cast a stoneskin spell when he wakes up in the morning to keep him protected all day, does not mean that there won't be multi-combat buffs. In fact, there will be. They've been confirmed. Josh has repeatedly talked about the various non-spell buffs (stronghold resting bonuses, for instance) that will stay active with you for several encounters. That's a valid point -- does this sound more accurate? "Buffs that the player can apply to themselves (vs. buffs granted for visiting a special location) in the field will last no longer than a single combat encounter. Otherwise, players will enter combat deliberately (with no opponents) in order to buff before key combats, which is exactly the behavior that the 'no-prebuffs' is designed to prevent".