Interview Project Eternity Interview @ Irontower

[evdk]

Really not on-board with the no-dialogue skills idea. Dialogue skills can work well, just look at the trial in NWN 2. The simple key to it is that the dialogue skill option shouldn't be the auto-win option but instead just be that: another option which may have different outcomes from the other ones, some better, some worse. (Though generally slightly better, of course.)

Actually having a successful dialogue skill check fuck you in the short or even better long run would be incline in my opinion

EDIT: which is something I think did happen in my AoD demo grifter playthrough.

 

[piydek]

Great interview. Really like the way Sawyer thinks/approaches things. This game looks to be good.

 

[Vault Dweller]

Really not on-board with the no-dialogue skills idea. Dialogue skills can work well, just look at the trial in NWN 2. The simple key to it is that the dialogue skill option shouldn't be the auto-win option but instead just be that: another option which may have different outcomes from the other ones, some better, some worse. (Though generally slightly better, of course.)

Actually having a successful dialogue skill check fuck you in the short or even better long run would be incline in my opinion

EDIT: which is something I think did happen in my AoD demo grifter playthrough.

Dialogue checks should help you get what you want, but what you want (at the moment) may or may not be in your long term interests. At least that's how I see it.

 

[Castanova]

Josh is right. Dialog skills should be in the dustbin. Dialog is about human interaction on an intellectual level and your goals/abilities in a given dialog are governed entirely by context. Some examples:

Intimidation. Why is this a single number? When you actually try to intimidate someone their reaction to that depends on (a) their sanity (b) their perception of your combat abilities (c) their perception of whether you will actually follow up on your threat (c) their perception of their ability to escape (d) whether they want to help you or not regardless of your nasty behavior toward them (e) whether there are outside factors preventing them from helping you, despite your attempt at threatening them (f) so on and so forth. Only (b) is a factor that doesn't change depending on who you're talking to. How can this be governed by a single number on your stat sheet?

Diplomacy/Persuasion. Why is this a single number? Similar to Intimidation, maybe even more so, diplomacy is all about context. Sure, some people are better at diplomacy in general. And this is more of an internal character trait than being based on how scary they look. However, countries are constantly interacting with each other through diplomacy. And, in the end, a country is going to do what it wants to do. People are similar. Diplomacy is about convincing somebody that a course of action is in their best interest when they initially aren't convinced. How can this NOT be a contextual action? Your character would need to be privy to all kinds of information beforehand in order to solve a tough diplomacy problem. Simply having a sufficiently high single number on your character sheet doesn't cut it.

I could go on but you get the point. Social interaction is intellectually contextual and therefore extremely difficult to scale across a large number of encounters. Combat is physically contextual and therefore very easy to scale across a large number of encounters. No game will ever solve this problem without incredibly advanced AI systems or manually writing/designing/iterating on each and every social interaction in the game such that they are deep and believable without relying on a stat sheet, basically creating a massive CYOA.

 

[hiver]

coughDX:HRcough....

 

[Anac'raxus]

DA2 did not track reputation. It tracked your character's personality.

Depending on the time they spend developing their reputation system, there may not be a mechanical difference. Would a "diplomat" reputation be global of faction specific? Are factions going to be omniscient? Will there be more than three archetypes perceived by the factions? Will reputations (beyond enemy/ally) have any meaningful affect on which rewards are available to characters?

 

[Alex]

Castanova

I think I see where you are coming from. In fact, I suggested that this kind of thing should be considered in dialog gameplay a page ago or so. However, I don't think this kind of thing should be handled by context only. I men, when we get doen to it, almost everything is context. Skill with swords is knowing how and when to move. Skill with traps is knowing what and where to look for. Skill with magic can be a whole lot of things. And frankly, I think those things could do with more involved systems, rather than being abstracted to simple skills that are just rolled.

But just doing away with the skills misses the opportunity for the system to help us identify the strengths and weaknesses of each character. If a certain character is good with the covert type of threatening (for whatever combination of factors that are in the character sheet or not), it can create interesting situations to explore how such combination can use its strengths and how it will deal with its weaknesses. Whereas if the character is as good or as bad as the player makes him, such exploration can't really happen. It is like the phobias from RoA. You could have game based reasons for people to run from undead, but it can be interesting too to have people who are specifically horrified of them, and see how you deal with such people.

Finally, this is very similar to the discussion after an editorial Roleplayer posted here once, isn't it?

 

[derpherp]

Really not on-board with the no-dialogue skills idea. Dialogue skills can work well, just look at the trial in NWN 2. The simple key to it is that the dialogue skill option shouldn't be the auto-win option but instead just be that: another option which may have different outcomes from the other ones, some better, some worse. (Though generally slightly better, of course.)

Also, I think dialogue skills must be worked on as something challenging, instead of simply a skill option in dialog. For example, suppose you have the intimidate skill. Most games have the equivalent of, in a table top game, saying "I intimidate him". Having the ability of intimidation doesn't mean that you just use it, like if it was a mmorpg skill with a cooldown and a mana cost. ou need to figure how you intimidate a person. What is he afraid off? What does he value most? What kind of setting do you use to make your threat? Are you overt? Do you imply more than you actually say? Do you show your strength in your threats?Depending on who you are intimidating, who you are, and what consequences you are willing to stomach, the answer should be different.

In this way, dialog skills can become an actual part of the gameplay. You would need to find more about an NPC before you cna really threaten him. And if you don't know enough about him, you might end up in an undesired situation. Push someone's buttons too far, and they may surprise you. Make a covert threat to someone dull, and they may ignore you, while a very overt threat to a weak noble may leave him a useless pile of nerves. People frequently suggest some abstract combat like gameplay for conversations, but I find this can be really bad, as having fleshed out conversations is usually where you flesh out characters and convey the mood of the setting. Instead, I think an adventure like gameplay, where you have to explore, use your skills and try to get into the character's heads to understand how they think would fit very well here.

So basically what they started doing in Alpha Protocol? That was the vibe I got when Sawyer was saying dialogue skills belong in the dustbin and describing his way of doing thing with reputation and choices whose impact(short and long term) isn't necessarily 100% clear. Alpha Protocol's implementation was limited by the demands of the spy genre somewhat, which kept the game fairly linear and cinematic, they could flesh it out more and integrate it into the world in Project Eternity, at least theoretically.

"Sounds great, but I wonder if he can pull it off." basically summarizes Sawyer. He's like Molyneux for cynics.

 

[Alex]

Maybe, except without taking away the initiative of the player. AP relied on a railed story where you could affect stuff, but not really determine who your character is or where he is going within the story. I was thinking more along the lines of how this was done in the Neuromancer adventure game, but with a lot more of NPCs and having your interactions with them as an open thing, rather than as pieces of a puzzle that must always be solved in the same way.

 

[Vault Dweller]

Intimidation. Why is this a single number? When you actually try to intimidate someone their reaction to that depends on (a) their sanity (b) their perception of your combat abilities (c) their perception of whether you will actually follow up on your threat (c) their perception of their ability to escape (d) whether they want to help you or not regardless of your nasty behavior toward them (e) whether there are outside factors preventing them from helping you, despite your attempt at threatening them (f) so on and so forth. Only (b) is a factor that doesn't change depending on who you're talking to. How can this be governed by a single number on your stat sheet?

Nobody argues that real life is a lot more complex than the most detailed character sheet and that a single number defining your ability in any given field, be it sword-fighting, science, or intimidation, can't approach the real life's complexity. It's done for convenience. However, just because having two sets of numbers (your skills vs the NPC's difficulty level, which encompasses a, b, c, d, e, and f) isn't sufficiently complex doesn't mean that it should be eliminated entirely. It's a case where something is way better than nothing.

Diplomacy/Persuasion. Why is this a single number? Similar to Intimidation, maybe even more so, diplomacy is all about context. Sure, some people are better at diplomacy in general. And this is more of an internal character trait than being based on how scary they look. However, countries are constantly interacting with each other through diplomacy. And, in the end, a country is going to do what it wants to do. People are similar. Diplomacy is about convincing somebody that a course of action is in their best interest when they initially aren't convinced. How can this NOT be a contextual action? Your character would need to be privy to all kinds of information beforehand in order to solve a tough diplomacy problem. Simply having a sufficiently high single number on your character sheet doesn't cut it.

First, throughout history, some diplomats, even in the same countries (like post-revolution France) were considerably better than others. Talleyrand is a good example. Much disliked by everyone for his legendary amorality, he served five different regimes (which wasn't easy being both a noble-born and a priest - not the most popular backgrounds in France at that time) and every attempt to replace him with lesser people who were, of course, "privy to all kinds of information beforehand", ended in failure or poor results.

Second, diplomacy is a bad name for the skill. Persuasion is a much better one. As you probably know, I worked in sales for a while and know the industry well. I can assure you that you can line up tell well-spoken, pleasant and sufficiently charismatic people of similar intelligence, train them for a week, send them into the world, and observe vastly different results. I've seen people who simply had an uncanny ability to talk people into buying shit from them (i.e. establishing instant trust, overcoming their fear of making a bad decision, convincing them that they need whatever it is they were selling, weakening their positions on the issue and making them doubt it, etc) and they did it in a very effortless way. I've seen people who couldn't sell if their lives depended on it, despite being trained, being social and outgoing, being good talkers, etc.

So, again, it's more than a single ability, it's 3-4 abilities working together, but in an RPG a single skill will suffice.

I could go on but you get the point. Social interaction is intellectually contextual and therefore extremely difficult to scale across a large number of encounters.

Having a single skill and different difficulty levels representing various abilities (one NPC can be hard to convince because he's knowledgeable and his positions are well defined; the other can simply be distrustful of everyone - the outcome in both cases is the same, the PC needs to be very skilled to overcome the arguments in one case, the distrust in another). Additionally, one can do what we did in AoD: add supportive skills/stats - INT/lore in the first case, charisma in the second.

Also, the context is easy to replicate by tying lines to what you did, what you learned, specific conversations, etc.

 

[FeelTheRads]

Yeah, what VD said. Sawyer isn't even proposing a system to replace dialogue skills. He's just going his usual way of hacking out whatever he thinks doesn't work and replace it with some shitty thing. In this case, what in the hell is what he proposes good for? What does it improve? There have been games that tracked what you did and also had dialog skills. So he just removes skill checks. DERP?
Just like with his retarded overbalanced system you'd end up with barely any difference between playthroughs. Oh my, now I chose the asshole option instead of the good boy option! Why, I didn't play anything like this since the last Bioware game!

 

[Infinitron]

Yeah, what VD said.

Hmmm, I don't believe VD has actually argued against Josh's design decisions ITT.

 

[hiver]

I kinda doubt he may be thinking that bioware alpha brotocol dialogue wheel - may be something anyone would want to see as presentation of this new system - i think that has been clearly communicated to everyone.
someone better check, if just to banish that ghastly image.

 

[St. Toxic]

Classic case of a 'bad implementation vs no implementation, which is better and why?' straw-man. Dialog skills are there to differentiate between the experience of a character who is well versed in projecting their will on others by peaceful means and a character who, due the lack of social graces, is more likely to harm his case with every spoken word. Like trying to convince the gate-keeper of your urgent business and letting you in; the dialog 'puzzle' aspect of choosing the 'right option' shouldn't even enter into it without a high enough speech-skill. If you can't make your case with words, either do some additional tedious task, sneak in or fight your way in. Caught at the scene of a crime? Well, if you don't have the sense and the required skill to talk your way out of jail, fight the cops, bribe them, sneak away or try your luck at the station. There wouldn't be any point in having all these options if it all hinged solely on the player's ability to simply figure out what the options are -- just look at Dishonored.

I also think a reputation system alone is a poor replacement for a speech skill. I mean, I'm guessing we're talking about collecting rep-points rather than specializing in +/- reputation to decide whether you want to buddy buddy with criminals and cutthroats or priests and high nobility, which is kind of cheap. At the same time, once you've split the skill-pools (which is a horrible, horrible idea) and you're doing a party-based RPG (which makes split skill-pools even worse) you'll inevitably cover a broad selection of skills, and if speech is there then few players will opt not to tag it on at least one character, which is almost like not having it in there in the first place. Maybe a good counter to this problem would be a combat penalty to any character who has tagged 'diplomacy' skills, making perma-death of that char more likely? It would be pretty gruesome to lose your single diplomat somewhere mid-game, so I kind of like the idea. Or what about a mirror opposite, where taking damage in combat lowers your speech skill (or, really, any 'secondary' skills), thus hinting at a worsening head injury?

I don't know. In a perfect world, putting points in speech would be a tough choice as it would mean not putting points in other skills which are just as useful to the character/party, if not more useful. As it stands, this kind of balancing is unlikely even from devs who seem obsessed with balance, and when you have multiple characters with 2x skillpools then building characters becomes less like choosing between door A and B and more like exploring the sewers of Tarant.

 

[Dorateen]

I've seen gamers who aren't interested in an RPG unless there are dwarves in it paging Dorateen

Good interview BTW.

Indeed. And what's with the comment,"Is it a Baldur's Gate thing?" As if there weren't dwarves and elves for decades before BG and since.

Glad to see that the boreal dwarves are properly designated as a sub-race, where they belong.

 

[Vault Dweller]

I meant "why put dwarves and elves into a BRAND NEW rpg? is it a Baldur's Gate thing because PE is being advertised as a BG-like game and targeted at BG fans?".

 

[Dorateen]

I see. However, I take a longer, historical view of these RPG titles. PE is not just a BG-like game, but a fantasy cRPG in a long line of others, which to me are an even further extension of PnP Dungeons & Dragons. I am more passionate about the hobby and the genre as a whole, than being just a BG fan. But I guess the comment wasn't directed at a player like me.

 

[Castanova]

Nobody argues that real life is a lot more complex than the most detailed character sheet and that a single number defining your ability in any given field, be it sword-fighting, science, or intimidation, can't approach the real life's complexity. It's done for convenience. However, just because having two sets of numbers (your skills vs the NPC's difficulty level, which encompasses a, b, c, d, e, and f) isn't sufficiently complex doesn't mean that it should be eliminated entirely. It's a case where something is way better than nothing.

I didn't say something isn't better than nothing. I said that a system that actually succeeds in capturing what socials skills are really about would not and could not boil a social skill down to a single number. Condensing social skills into a number means that when you, say, try to persuade someone of something, you're effectively participating in an extremely simple competition with that NPC. But trying to persuade someone of something is not a clearly defined competition. No exercising of social skill is a clearly defined competition. Combat, on the other hand, very clearly IS a competition which is part of the reason why even dreadfully simple implementations of stat-based combat can be "good enough" for a lot of players.

Sure, you can boil social skills down into a number-based approach so that you can have something rather than nothing, to try to achieve that same "good enough"... but then I have to ask: what is your overall goal? Because if your goal is to give the player the feeling of accomplishing challenges using social skills then you'll surely fail. If your goal is simply to quickly simulate the actions of someone using social skills to accomplish tasks for some other larger reason (perhaps your larger goal is to leverage video game technology to tell a choice-responsive story) then this method has a chance at being successful. But if that's your goal, you have to live with the fact that the naked mechanics of accomplishing the social tasks will never be rewarding, per se, like a challenging combat encounter can be. It will be only be rewarding in the sense that you get to progress the story that you're presumably interested in.

First, throughout history, some diplomats, even in the same countries (like post-revolution France) were considerably better than others. Talleyrand is a good example. Much disliked by everyone for his legendary amorality, he served five different regimes (which wasn't easy being both a noble-born and a priest - not the most popular backgrounds in France at that time) and every attempt to replace him with lesser people who were, of course, "privy to all kinds of information beforehand", ended in failure or poor results.

Second, diplomacy is a bad name for the skill. Persuasion is a much better one. As you probably know, I worked in sales for a while and know the industry well. I can assure you that you can line up tell well-spoken, pleasant and sufficiently charismatic people of similar intelligence, train them for a week, send them into the world, and observe vastly different results. I've seen people who simply had an uncanny ability to talk people into buying shit from them (i.e. establishing instant trust, overcoming their fear of making a bad decision, convincing them that they need whatever it is they were selling, weakening their positions on the issue and making them doubt it, etc) and they did it in a very effortless way. I've seen people who couldn't sell if their lives depended on it, despite being trained, being social and outgoing, being good talkers, etc.

Oh, absolutely. Like you said, what makes a person good at selling is the combination of a ton of different factors. Same with being a good diplomat. But being naturally good/mediocre/bad at diplomacy in such a way that it can be baked into a single number or even a small set of numbers doesn't make exercising that diplomacy mechanically fun. It just means you've executed a simple simulation of a social skill. It's not mechanically fun to read a dialog and then choose the one marked [Diplomacy] because you're good at Diplomacy. It's not even mechanically fun to read a dialog and then pick the un-marked response which only appears when your Diplomacy is high enough. The result may be fun, if you're enjoying the responsive story, but the means are not fun. So, again, it comes down to the same question. What is your goal?

 

[Castanova]

I don't know. In a perfect world, putting points in speech would be a tough choice as it would mean not putting points in other skills which are just as useful to the character/party, if not more useful. As it stands, this kind of balancing is unlikely even from devs who seem obsessed with balance, and when you have multiple characters with 2x skillpools then building characters becomes less like choosing between door A and B and more like exploring the sewers of Tarant.

But why? A system of balancing skill points between social and combat skills certainly bears no relation to reality. So the only reason you'd want something like this is because you believe this choice makes the game more fun. Why? What's fun about splitting points between Diplomacy and Dodge?

 

[Infinitron]

Wow. I really was 100% right about Josh Sawyer: http://forums.obsidian.net/topic/62...t-iron-tower-studio/page__st__80#entry1288334

I want to say something about my high-level approach to design, whether the systems being described are dialogue, rest mechanics, or how you gain experience: the bottom line for any mechanic is how it affects the ways in which players play the game. I.e., after all of the theorizing, all of the speculation, and all of the strong statements of feeling on a mechanic, what matters is how people play the game.

So when I write that what Tim and I want to do is use quest/objective/challenge XP as the primary (if not only) methods of achieving XP, that means "want" will give way to "reality" if they are in conflict -- conflict in practice, not conflict in a forum discussion. When changing the system requires relatively little effort, there's not a ton of benefit to being absolutist over a year in advance. Moving from a class-based to classless system -- that's a big deal. That's something you decide and pretty much stick with. Deciding whether to give XP for monsters or not give XP for monsters -- that's not a big deal. That's easy to address, even late in development. Deciding whether people can rest at certain locations or they can rest anywhere is also pretty easy to address.

These things exist on a sliding scale of difficulty, implementation/adjustment-wise. We plan things so we can make the simple changes easily later on. Generally that means creating simple base layers of mechanics and adding in "adjustment" or tuning mechanics when the metrics/gameplay we see demands it.

 

[Volourn]

"DA1 is better than NWN, that's for sure..."

FUCK OFF YOU PIECE OF SHIT! TAKE THAT BULLSHITZ BACK!

 

[St. Toxic]

I don't know. In a perfect world, putting points in speech would be a tough choice as it would mean not putting points in other skills which are just as useful to the character/party, if not more useful. As it stands, this kind of balancing is unlikely even from devs who seem obsessed with balance, and when you have multiple characters with 2x skillpools then building characters becomes less like choosing between door A and B and more like exploring the sewers of Tarant.

But why? A system of balancing skill points between social and combat skills certainly bears no relation to reality.

I disagree. Not that a soldier can't be charismatic, naturally, which is a reasonable compromise between skill allocation. But dedicating your life to the pursuit of a single power or ability surely results in some neglect of other, equally viable, but less interesting, potential traits and skills. In terms of a speech-skill, it's more difficult to define exactly what is being learned and improved upon, unlike the expert use of a weapon, so it's easy to mistake it for something of an inborn attribute -- you're either a sociable, agreeable individual by nature or you're not. But what we're really talking about is psychology and the ability to manipulate people, bending them to your will. It doesn't mean you can't also hack them up in pieces with a sword afterwards, but maybe you'd like to bake some excellent pie out of the left-over bits -- but alas you never found the time to get into cooking.

So the only reason you'd want something like this is because you believe this choice makes the game more fun. Why? What's fun about splitting points between Diplomacy and Dodge?

The effect it has on your performance, your ability to deal with the problems with which you are presented? The more tools in your toolkit, the less meaningful the choice when choosing between which tool to use -- it becomes more of a cosmetic choice. The less tools you have at your disposal, the more specialized they are, the more meaningful your choices become -- you don't have the option to fall back on a myriad of other potential solutions. Fair enough that this hardly ever seems to bother fans of, say, Bethesda titles and the like, where skills mainly act as bonuses while every character is a jack-of-all-trades, but it feels like something is lost. In my opinion, the 2-pool system is edging closer to this kind of design-philosophy, where 'secondary' skills are like bonus points, earning you a cookie or two but not seeing you through the game or forcing you to rely on a particular strategy.

But maybe that's just me.

If your goal is simply to quickly simulate the actions of someone using social skills to accomplish tasks for some other larger reason (perhaps your larger goal is to leverage video game technology to tell a choice-responsive story) then this method has a chance at being successful. But if that's your goal, you have to live with the fact that the naked mechanics of accomplishing the social tasks will never be rewarding, per se, like a challenging combat encounter can be. It will be only be rewarding in the sense that you get to progress the story that you're presumably interested in.

It's rewarding to see well-spent experience amount to something in and of itself. Other than that, [Intelligence] there's no reason why a successful speech-check has to rely entirely on the skill, and not strain the player's wits. Having the necessary skill may be enough to access otherwise inaccessible dialog-options, but getting the desired result could still depend on your investigative abilities outside of the dialog and intelligent decision-making once you're being persuasive. I know people around here hate Fallout, but I found it extremely rewarding talking the Master down on my very first play-through, especially as I hadn't expected that possibility at all and figured I had no other choice but to fight.

 

[Infinitron]

Re: Combat and non-combat skills

I'll say it again, you need to stop thinking about this game as a Fallout-like game and more like what it is going to be, a D&D 3E-like game. Different model of skills and skill progression.

 

[St. Toxic]

I'm not talking specifically about PE though, I've already said my piece on that:

At the same time, once you've split the skill-pools (which is a horrible, horrible idea) and you're doing a party-based RPG (which makes split skill-pools even worse) you'll inevitably cover a broad selection of skills, and if speech is there then few players will opt not to tag it on at least one character, which is almost like not having it in there in the first place.

I mean, it still cheapens the secondary skills, but if it's a dungeon romp with some fluff I guess it wouldn't matter anyway.

 

[hiver]

If anyone can enhance and improve our old well loved and tried dialogue system it would be these two guys.
There is no rule that states only one solution should be applied until the end of times.
The old school dialogue we all know and like works - but it also brings its own specific issues. Lets remember that one of the biggest complaints about AoD was how binary the dialogue-skill checks are - compared to combat.

I am of the mind we should see what these two newbies can come up with and i do say "newbies" with the maximum possible amount of sarcasm allowed by this universe.

Wow. I really was 100% right about Josh Sawyer: http://forums.obsidian.net/topic/62...t-iron-tower-studio/page__st__80#entry1288334

I want to say something about my high-level approach to design, whether the systems being described are dialogue, rest mechanics, or how you gain experience: the bottom line for any mechanic is how it affects the ways in which players play the game. I.e., after all of the theorizing, all of the speculation, and all of the strong statements of feeling on a mechanic, what matters is how people play the game.

So when I write that what Tim and I want to do is use quest/objective/challenge XP as the primary (if not only) methods of achieving XP, that means "want" will give way to "reality" if they are in conflict -- conflict in practice, not conflict in a forum discussion. When changing the system requires relatively little effort, there's not a ton of benefit to being absolutist over a year in advance. Moving from a class-based to classless system -- that's a big deal. That's something you decide and pretty much stick with. Deciding whether to give XP for monsters or not give XP for monsters -- that's not a big deal. That's easy to address, even late in development. Deciding whether people can rest at certain locations or they can rest anywhere is also pretty easy to address.

These things exist on a sliding scale of difficulty, implementation/adjustment-wise. We plan things so we can make the simple changes easily later on. Generally that means creating simple base layers of mechanics and adding in "adjustment" or tuning mechanics when the metrics/gameplay we see demands it.