Editorial David Gaider of BioWare on multiple paths, choices in CRPGs

[Saint_Proverbius]

Tags: David Gaider; Neverwinter Nights

<b>Ibbz</b> informed me that over at <A href="http://nwvault.ign.com/">NW Vault</a>, there's a <a href="http://nwvault.ign.com/index2.shtml#newsitem1044676870,53423,">news post</a> which is a highlight of something <a href="http://www.bioware.com">BioWare</a>'s <b>David Gaider</b>, design guy, said about why it's not worth it to make multiple plot choices in CRPGs. Here's a taste:
<br>
<br>
<blockquote>Don't believe me? I speak from personal experience. The main reason why this is true is the following: when you start adding radically different plot paths into a story, the complexity of that story increases exponentially. So there too does the probability of bugs cropping up in said story increase exponentially. Thus you spend more time to do less. As an example: in an average day of work here at Bioware, I average about 3,000 to 5,000 words of dialogue written. If, however, I am working on a complex character (a major NPC like, say, Aribeth or Aarin Gend in Chapter 2 of the OC), that speed slows down to 2,000 words or less in a day due to the time spent on arranging the structure.</blockquote>
<br>
<br>
But then again, if you have five years and a peak work force for the project of seventy five people, Dave... After all, <a href="http://www.interplay.com/fallout2">Fallout 2</a> didn't even take a year to make, and it offered what you're talking about with less people working on it!
<br>
<br>
PS. Someone needs to look up the definition of <i>exponentially</i>.
<br>
<br>
Thanks for the heads up, <b>Ibbz</b>!
<br>

 

[Araanor]

This is a bit more interesting than the usual stuff, it's an opinion, not some hype.

Radically different plot paths? A good start would be to make the dialogue less fake. Nothing like picking between four different alternatives and realising they're all the same when you get the reply.

Yes, it's more work. But not only does it add replayability, it actually gives the feeling that your choices matter as opposed to just watching what happens. Interactivity. Quality over quantity.

 

[Saint_Proverbius]

Well, the thing is, BioWare loves to claim they have these paths in their games, yet they never do. If you're not going to put these choices in your game, maybe you should stop advertising the game as if you are allowing them.

Honestly, given the size of BioWare and how much time they get on projects, him claiming it takes more time really doesn't cut it at all. Troika allowed you to join the bad guy in Arcanum, and they didn't have near the development time nor did they have near the size of the team that BioWare has, so basically it sounds to me like he's making an excuse on why they make mediocre CRPGs.

 

[Crazy Tuvok]

wow. what a load. has this guy even played another CRPG? as St. P pointed out one good recent example to refute *everything* he claims is Arcanum. after reading this, posted by an employee of Bioware, I am really shocked they even bother calling their games RPGs as they clearly don't give a flaming crap about the genre.
fuck them. go Troika.

 

[ljw1004]

The person who needs to look up the definition of "exponential" is Saint Proverbius. What Dave Gaider wrote is precisely correct, follows the definition precisely, and is one of the most common examples that mathematicians would give for an exponential increase.

(i.e. at every point, if you have a big choice like whether or not to side with Irenicus, then most every subsequent dialog has to be able to reflect the choice you made. Therefore, the size of each dialog in the game is 2^n, where n is the number of choices. i.e. an exponential.)

 

[Saint_Proverbius]

The person who needs to look up the definition of "exponential" is Saint Proverbius. What Dave Gaider wrote is precisely correct, follows the definition precisely, and is one of the most common examples that mathematicians would give for an exponential increase.

(i.e. at every point, if you have a big choice like whether or not to side with Irenicus, then most every subsequent dialog has to be able to reflect the choice you made. Therefore, the size of each dialog in the game is 2^n, where n is the number of choices. i.e. an exponential.)

Actually, every choice there is a dialogue tree path for it. Furthermore, he even states in that post, which is rather apparent you didn't bother reading, that if you have two choices, you have two plot arcs. That's a linear relationship, not an exponential one. If it WERE expontial, you'd have four arcs instead of an extra arc per choice. After all, 2^2 is 4, last time I checked.

Even the dialogue doesn't go up expontentially, since that's a case by case situation, and governed by what is required of that choice.

Let's look at his case for joining Aribeth and how hard that would be, and all the exponential stuff that would involve, shall we? Well, the prelude and the first chapter would be indentical to what was in the find game, since Aribeth doesn't turn evil in those. In fact, most of Act 2 would be largely the same. The only difference would be at the very, very end where you see Aribeth denouncing Tyr and joining the creator race. At that point, there'd be a little dialogue where the player is given the choice of joining or rebuking Aribeth for her evil ways.

Now, on to Chapter 3. There's two ways to handle the player has joined Aribeth side of things, and you don't even have to do both. They could impliment one or the other. The first way is the BRAZEN way, where you set up a camp location for the forces of evil, and you continue looking for the other worldstones, just like you would if you stuck with the good guys. The second way, which is easier, is that you establish a contact in the town that you take the real worldstones to and then swap them with fake ones to take to the Spymaster. Largely, Chapter 3 would play out in the same fashion as it does in the original game. The only things that would change would be who you take the worldstones to. There's nothing exponential about this at all.

Chapter 4 wouldn't need to change that much either. In fact, you could have it so the PC gets screwed over by joining Aribeth at the end of Chapter 3, which would make what goes on in Chapter 4 line up fairly well. You storm the bad guys' base, and kill them, only out of revenge instead of because you're a good guy. Maybe fit a few lines of dialogue to fit the revenge path instead of the goodie two shoes path, but that's it.

Of course, you could always reverse the scenario in Chapter 4, where you have to deal with the Spymaster and some cohorts of good who have shown up to stop you and Aribeth. This would mean a new module, but the area design would largely be the same, so you wouldn't have to make a brand new map. You'd only have to change the scripting.

Now, given how long they had this game in production, I don't think making two modules with different scripts would be that much of a task, and that's if they chose to do it the HARD way. However, that would have made the game actually have a bit of depth to it, and wouldn't have taken that much longer to impliment if they followed both easy methods of doing it.

Hell, if Jeff Vogel can make Genforge in less than a year, by himself, and allow you to join up with one or the other or both of the antigonists in the game, you'd think big, bad bioware could do it. They just didn't do it, and copped out with the same argument they used when they were asked why Beholders and cloaks weren't in the game, it's too hard.

 

[Vault Dweller]

I totally agree. Having multiple storylines distracts me as it's not always clear which path is the right one. I think that the developers should focus only on what makes the game fun like running around, killing monsters, and looking for loot. Everything else takes too much time and not fun.

 

[Rosh]

BioWare Moron and Fanboi (ljw1004) = gaoled, with ass as hat.

That was pretty cruel of you, Prov. I like it.

Of couse, it could be easy to point out that support for another plot fork could also take LESS time than though before, since you've already got most of the entities built. You just need to add in the path divergence for the certain conditions and go from there. It's one of the ways to create a bit more non-linearity and a sense of world immersion, like adding in solutions for Diplomat, Thief, and Mad Dog Fighter types. Such support could be done by having an item of information put on an NPC or in a nearby object, and also able to talk to the NPC for the information.

Such as many of the quests in Arcanum, Gilbert Bates, Schuyler (sp?) and Sons, etc. Unfortunately, such design abilities are far from the inept chimps at BioWare. While those are on a small case of path forking, it's not too hard (with a bit of imagination and talent) to do it on a large scale while making it good and save effort. It also explains why most of BioWare's games have jack shit for replay.

No wonder BioWare games are bloated shit. They have genuine morons working on them with little clue of game design. Let me guess...he's graduate from one of those "Gaming Colleges"?

 

[Spazmo]

I hate it when developers or fans say that a given feature isn't in a game because it would be too hard to implement. Hey, BioWare, you know what: I don't care how hard it is! Your job is to make a quality game with all sorts of cool shit in it. If you encounter a problem during the course of development, you don't go post on the boards and say that such thing won't be in the game because you couldn't get it working, you get your nose right back on the grindstone and make it work. I have a sneaking suspicion that the folks at BioWare were the kind whose rich daddies bailed them out of most of their troubles. "Daaad, this exam is too tough, could you bribe the dean?"

 

[Saint_Proverbius]

BioWare Moron and Fanboi (ljw1004) = gaoled, with ass as hat.

That was pretty cruel of you, Prov. I like it.

Well, if I thought I could have gotten away with, "Hey, ask yourself what has to change by joining aribeth when she gets her evil groove on, and how much beyond that would need to change?", I would have just done that. After all, that's pretty much where I started when thinking about writing that post. However, for the sake of nipping that short, I figured a more indepth approach was in order, so I went ahead and just answered that question.

In terms of the bare essentialls, you'd have to add some dialogue at the end of Chapter 2 followed by the setting of the YOU ARE NOW A BAD GUY flag. An additonal NPC that's in Chapter 3 with dialogue that's encountered fives times since there's the intro speech on what to do, followed by the dialogue at the end of each worldstone delivery and swap, and then the final, "Hah! Hah! We will betray you, player!" screwing over speech. You can actually reduce this to four bits of dialogue, if you wanted to do so, but there's really five events there. You'd also have to make it so the Spymaster accepted both the real and the gake worldstones, which is trivial. That's really all you'd have to do for the bare bones, "the player is evil" deal.

Of couse, it could be easy to point out that support for another plot fork could also take LESS time than though before, since you've already got most of the entities built. You just need to add in the path divergence for the certain conditions and go from there.

Also true. You wouldn't HAVE to make new models and NPCs for what I was saying. You could just take a normal NPC that's useless in Chapter 3 and script him up for the bad guy quests.

You could even use Aribeth as the evil NPC that does the swapping, as long as she's hidden away somewhere in town.

It's one of the ways to create a bit more non-linearity and a sense of world immersion, like adding in solutions for Diplomat, Thief, and Mad Dog Fighter types. Such support could be done by having an item of information put on an NPC or in a nearby object, and also able to talk to the NPC for the information.

Which BioWare doesn't have a clue on how to do. You'd think, given the single character nature of NWN, they would have figured out that having thief and bard/high charisma routes would have been manditory. However, they just decided to half ass things and toss in the obligatory henchman that covers the skills your PC may not have, then designed it as a party game.. Oh, and they hyped it up like it would allow multiple methods of completion, which it didn't do.

No wonder BioWare games are bloated shit. They have genuine morons working on them with little clue of game design. Let me guess...he's graduate from one of those "Gaming Colleges"?

Given the simplistic formula they use, it's actually shocking they are so bloated. Then again, most of that bloat comes from assets, like canned audio dialogue, rather than things of substance.

 

[Jed]

The full litany of excuses from Galder:

Quote: I agree that there is definately a diminishing return as you add more and more paths to a module. But I dispute your basic assumption. If I design two paths - the second path will NOT only be seen by those who replay it and if I design 10 paths - the 10th path will NOT only be seen by those who replay it 10x. Rather - if I put in 10 paths (and all are equally interesting) - each path will be seen by approximately 1/10th of the people on their first play through.
I would tend to think that that's not quite correct. Each path wouldn't get a similar number of adherents... this would be correct only if classes and alignments were played pretty evenly across the board. The largest chunk of your audience is going to follow a good-aligned, basically benevolent path. The second largest chunk (but well, well behind the first) is looking for an evil-aligned, so-malevolent-I'm-cool path. Everything past that is basically supporting a choice that a smaller and smaller portion of the player base is likely to take. I'm not saying that that option is therefore rendered worthless or wouldn't be extremely cool and gratifying for the player who does choose it, however, but the return on the invested time for its creation does lessen.

Quote: This still raises serious issues about how worthwhile it is - but the distinction is relevant.
True enough. I suppose I should qualify my time estimates by saying that they relate to real choices only, as in fully-developed paths. Inserting fake paths so that the appearance of choice is there is also an option... and a time-saving one... it just doesn't hold up to scrutiny and, in the end, may end up disappointing the audience it is meant to please (which is to say the hardcore roleplayers who would notice).

Quote: Especially if you consider that a module that has a good path will appeal to group X while a module that has an evil path will appeal to group Y. And, although group X and group Y are not mutually exclusive groups - neither are they the SAME group. Thus - with every path you are increasing your audience.
Does the audience actually increase with these seperate paths? I'm not sure it does. Some might certainly be more pleased with the options they had available, and therefore more satisfied with the story overall... but at the same time we go back to the sense of value. The width of the plots is inextricably linked to the number of paths... if you can provide very wide plots but still maintain overall length you've got an ideal situation that will please everyone... and yet that will never happen realistically. There's also a niggling little voice in the back of my head which says that opening up that kind of breadth also opens up a can of worms, to an extent. Is it even possible to cater to every option possible? And if you start trying to, is it not possible that you invite criticism from the very people who these options appeal to? "Yes, I know you provided options A, B, C, D and E... but what I really wanted was F. Why didn't you provide that?" I don't know how strongly I feel about that, though. I do know that if I was a member of the community who put that kind of effort into a module and that was part of my feedback, I would feel pretty discouraged.

Quote: Note: It is quite likely that these calculations are different for the highly competitive world of freely downloadable modules and that David's arguement is 100% valid for commercial products where the decision to buy or not buy is based on a much wider list of criteria.
Yes, we're talking completely different ballgames, here. A commercial product (one of any scope, anyway) is not really free to be a niche product. On the other hand, though, there is the idea that a quality product will draw fans if it's good enough ("If you build it, they will come" kind of thing). Looking at the commercial track-record of those RPG's regularly touted as the best by the hardcores doesn't exactly fill me with confidence on that point, sadly. Try juggling those two concepts when you create a design commercially... heh, good luck.
Some of the concepts do apply even to community-made modules, though... especially the overall width vs. length idea. If I was making a community module and was considering this as a real option for my design, part of the real hurdle would be that implementing this in anything less than a half-*** measure would be a monumental task. Worth it, regardless? Hard to say. Maybe.

More:

Quote: And there is no validity in the notion that "two paths = double the work load".
Sorry to pick on this one point in your post, Torias, but on this one you are dead wrong... at least from a design perspective. Yes, true, you can re-use art and similar kinds of physical resources when creating multiple paths for a plot... but when it comes to design, if you are creating real alternate paths (like the ones you suggested at the end of your post... joining the NWN cultists and invading Neverwinter or supplanting Irenicus at the Tree of Life in BG2) the following is actually the case:

two paths = more than double the workload

Don't believe me? I speak from personal experience. The main reason why this is true is the following: when you start adding radically different plot paths into a story, the complexity of that story increases exponentially. So there too does the probability of bugs cropping up in said story increase exponentially. Thus you spend more time to do less. As an example: in an average day of work here at Bioware, I average about 3,000 to 5,000 words of dialogue written. If, however, I am working on a complex character (a major NPC like, say, Aribeth or Aarin Gend in Chapter 2 of the OC), that speed slows down to 2,000 words or less in a day due to the time spent on arranging the structure. And that wasn't even as complex an arrangement as, say, Sarevok from ToB or (and the memory still makes me shudder) the Phaere/Dragon Egg double-double-cross options in BG2's Underdark. Let's take an example from your list and look at it more closely: kissing off Imoen in BG2, joining with Bodhi and supplanting Irenicus. What would be required, design-wise, for that to happen?

1) Added dialogue in the initial stages of the game to allow the player to take a 'I'm not interested in Imoen' stance. 'I want Irenicus's power' stance would have to be heightened and introduced to dialogues where finding Imoen is currently used as a motivation.

2) Joining with Bodhi could be easily added onto the current plot path when allying with Bodhi's thieves. After she reveals who she is, a plot path would be needed to be able to convince Bodhi to join you and turn on her brother.

3) Much of the end half of Chapter 4 is now altered significantly. The confrontation with Irenicus is altered completely. New dialogue needed involving how Bodhi helps you and the new confrontation with Irenicus. Completely new scripting for the alternate plot path.

4) Assuming that the PC (with Bodhi) would still go through the Underdark section, you're now left with nothing to do upon exiting the Underdark. One would have to assume that the Rynn Lanthorn is moved elsewhere (and Bodhi does not have it), else you've just shortened the game considerably. Assuming you are able to re-use Bodhi's lair for Chapter 6, you need all-new dialogue and plot scripting for this entire section leading up to the acquisition of the Rynn Lanthorn.

5) Once you've acquired the Rynn Lanthorn, you're probably going to need some new dialogue with Elhan and Ellesime (who both would recognize Bodhi, one might assume), but that's not serious. The plot leading up to the Tree of Life could be very similar. Alternatively, Bodhi's presence could cause the elves to be hostile... in which case you've got a lot of extra scripting to do to make all the extra fights feasible without losing story in this section.

6) The confrontation at the Tree of Life is going to be problematic if you want to use it as the final confrontation. Is the confrontation in the Abyss cut out or still used? Either way, you're going to need considerably more dialogue. Adding in a usurpation of the Tree of Life with Bodhi's help is going to have to be a completely new branching path... new dialogue, new scripting... most likely new art if you're going to make it the climax of the story (cut-scene, perhaps?) and a new ending.

So... besides taking an already-complex game and adding new layers of bug-ridden complexity into it to add these paths, you're also talking about a *lot* of extra work. If James were to hand me this, I'd schedule two months extra... minimum... and that doesn't include all the extra time needed for bug-fixing. That is time that just doesn't exist in a commercial schedule.
Would it be cool? Without a doubt. There are a lot of cool plots and paths that got cut from the BG2 mix... this wasn't one of them, but it could have been. The amount of work, however, is far greater than the scope of the plot would seem to indicate and should not be passed off so lightly.

I'm a student teacher at my Ju Jitsu dojo; there's these two newer kids, white belts, who never want to take falls in class. I tell them, if you don't want to take falls, then you don't want to train Ju Jitsu; if you don't want to do Ju JItsu--why are you here? If Bio doesn't want to do the work, why are they pushing their crap as CRPGs?

Jed

 

[Rosh]

I do have to see a point of his.

Given that BioWare's programming fellates donkeys, particularly in terms of AI and pathfinnding, it would be rather hard to work with such a sloppy base.

You can't stack bullshit that high. Well, you can, which explains Dave Galder.

But that's just (hopefully) a rare case.

 

[GreenNight]

I find that the difference between Bioware and Troika games is that bioware writes a story while troika writes a world. So then, if they what to implement multiple paths, their view of the problem is miles apart. Bioware thinks of duplicate effort, like making the chapter 73 for good guys, the chapter 73 for bad guys, and the chapter 73 for the monkeys. Troika, the way they see it, simply adds some new areas, script more the existant ones and let the world fence with you and your view of your character.
Even I would get tired of making the nth version of the same chapter just to implement another ending or path, but not of making another city or fortress for you to join another side. The first looks more time consuming than the second, even if it might not be so.

Just my 0.02€

 

[SexExitium]

 

[ljw1004]

Furthermore, he even states in that post, which is rather apparent you didn't bother reading, that if you have two choices, you have two plot arcs. That's a linear relationship, not an exponential one. If it WERE expontial, you'd have four arcs instead of an extra arc per choice. After all, 2^2 is 4, last time I checked.

I'm guessing that maths isn't your strong point, SP If you add one plot point (i.e. a linear increase) then it creates two arcs (i.e. doubles the rest of it). Whenever a linear thing causes a multiplicative thing, that's an exponential.

Here's a simple example. I start the game with one branch point, where you decide to be good or bad. That yields two options. Now I add a second branch point, where you decide whether to be red or green. That yields four options total: good-red, good-green, bad-red, bad-green. So the equation here is options=2^plotpoints. An exponential. Actually, I simplified it by only considering the end-states. If you sum up all the choices all the way through, you get 4^plotpoints I think, but can't be bothered to check since it's still clearly an exponential.

The rest of your tirade, SP, argued that the constant of proportionality was small in this case. That doesn't stop it being an exponential, it just makes it a slow exponential.

PS. I'm not a fanboy. I hate MP games, I dislike D&D, I hated the look of NWN, I've never played it and I'm not going to. But when a programmer understands maths, and he's criticised by someone who doesn't, then (as a programmer and mathematician myself) I have to step in. Well, logician rather than mathematician, but they're close enough.

 

[Section8]

Dave Gaydar sounds like a lazy fuck. Quality over quantity, fuckstain. It must be fucking horrible to actually put some effort into your work.

 

[Vault Dweller]

why are you here? If Bio doesn't want to do the work, why are they pushing their crap as CRPGs?

Right on. I was very surprised to read all this crap from a CRPG developer. If he does not like what he does, he should go and do shooters or someting similarly action-focused. He should NOT whine, bitch, and complain and try to convince everybody that his lame-ass games are great the way they are.

If Bioware was running out of time (after 5 fucking years :shock: ), say so. Admit whatever mistakes took place, and say you will try better in the future. Don't look for excuses as there are none.

Note to ljw1004: It's not about math, it's about a man who should not be doing the job he does.

 

[Saint_Proverbius]

I'm guessing that maths isn't your strong point, SP If you add one plot point (i.e. a linear increase) then it creates two arcs (i.e. doubles the rest of it). Whenever a linear thing causes a multiplicative thing, that's an exponential.

No, I'm guessing math isn't one of your strong points, since a linear function is defined as:

• y = a*x + b

a is what those math people call a coefficient, and it is multiplied by the x to get the slope, or rate of change, of the linear function. b is irrelevent here, but in case you get to algebra sometime, that's the y-intercept, which is where the line crosses the y-axis.

Exponential, on the other hand, is often a polynomial, or where the x is raised to a power to get y. It has nothing to do with multiplication by a coefficient at all. If there's not a power or a logarhythm, then it's linear, regardless of what value that coefficient in front of that x has.

The only exception to that is when a is equal to zero, then you have a linear case regardless of that exponent.

Now that we're done with our math lesson..

Here's a simple example. I start the game with one branch point, where you decide to be good or bad. That yields two options. Now I add a second branch point, where you decide whether to be red or green. That yields four options total: good-red, good-green, bad-red, bad-green. So the equation here is options=2^plotpoints. An exponential. Actually, I simplified it by only considering the end-states. If you sum up all the choices all the way through, you get 4^plotpoints I think, but can't be bothered to check since it's still clearly an exponential.

You're drastically oversimplifying what's going on and your math isn't even correct here. Take a case where you have three options through one quest. You don't have 2^3(which is 8, BTW) paths, you only have three paths through it, one per option.

Not to mention the fact that you don't have to make a completely new plot for every single option, which is the logical fallicy in your statement.

The rest of your tirade, SP, argued that the constant of proportionality was small in this case. That doesn't stop it being an exponential, it just makes it a slow exponential.

No, the rest of my "tirade" is an explicit case where it's most certainly not exponential. In fact, it doesn't double the work load at all, let alone raise it to any given power. Hell, I even listed what all would have to change to even allow that, in detail. So, either you're just really stupid, or you didn't bother reading it. I'll give you the benefit of the doubt, go back and read that "tirade", then tell me how that's remotely exponential.

But when a programmer understands maths, and he's criticised by someone who doesn't, then (as a programmer and mathematician myself) I have to step in. Well, logician rather than mathematician, but they're close enough.

You're apparently not much of a reader as well as a "mathematician", since I clearly stated that David Gaider is a designer, think "map making guy and scriptor", which is definitely not a programmer any more than a web designer is.

 

[Rosh]

Point, set, don't even bother replying to save face.

 

[ljw1004]

[You're drastically oversimplifying what's going on and your math isn't even correct here. Take a case where you have three options through one quest. You don't have 2^3(which is 8, BTW) paths, you only have three paths through it, one per option.

*Sigh*. Let's try again. There are three options, as you say. The first asks: are you good or evil. The second asks: are you red or green. The third asks: do you help Irenicus or hinder him.

Now consider the end scene. How many variants are there? Eight:

(1) a good red hero helps Irenicus
(2) a good green hero helps Irenicus
(3) an evil red hero helps Irenicus
(4) an evil green hero helps Irenicus
(5) a good red hero hinders Irenicus
(6) a good green hero hinders Irenicus
(7) an evil red hero hinders Irenicus
(8) an evil green hero hinders Irenicus.

I.e. there are 2^3 variants of the end-scene. An exponential.

Except it's not quite like that, because the red-green encounter actually has variants itself (where a good hero chooses red, a good hero chooses green, an evil hero chooses red, an evil hero chooses green). That makes us have to do a sum of exponentials. But the result is still an exponential. So let's just say 2^n.

However, as you explained with your example, a major arc change doesn't have to change every single dialog. It need only change one in fifty dialogs, for instance. (although how one can call it "major" if it has such little effect, I don't see). That puts us at roughly 2^(n/50). Which is still an exponential.

PS. Yes, exponential increases are part of complexity theory, which was part of my course at university. I went on to teach it to undergraduates. The exponentiality of story arcs is a standard problem to set. I do know what I'm talking about.

 

[triCritical]

I think the whole exponential thing is rather misleading. I think the way Dave GAYder (thanks on that one section8), uses it implies that he is referring to an exponential growth. Arguing over what is actually exponential is not the point. Still it goes to show that Dave GAYder is not a very smart person because he feels that the only way to attack the problem is through brute force exponential work. I think the sign of a smart person is taking a tough problem and then finding a smart solution, something that he is obviously incapable of.

Anyhow, take for instance what Josh is doing in Jefferson. He is having a set of counters, thats right more then one, that essentially follow your parties reputation, on a microlevel, macro level and other levels. The idea is simple, give the player the standard set of options and have how the character solves the various outcome effect the various counters and set flags. It gives the feeling and actually accomplishes the feat of making different paths possible as well as cutting down on the work. Most importantly its not exponential. Obviously some events will be correlated with other events, but come on, your getting paid.

As far as exponential increase in outcomes, I am certain he meant dialogue only. And what David Gayder thinks is character development I think is a big waste of time. A character should develop not only through dialogue but through actions. Take Aribeth in the OC. NIce development, she's a paladin that does not do a god damn thing. Sure I had to sit through here annoyingly long dialogue's but her actions contradicted it. On the other hand take Killian Darkwater in FO1, his extremely small role, limited dialogue options and scripted events allowed for a much more developed character. Most importantly it left a slight vagueness, which allowed for the player to put their finishing touches on what kind of person Killian Darkwater is.

 

[voodoo1man]

Story branches

Thanks for the fundamental counting principle lesson ljw, but Mr. Gaider is indeed wrong here. He (and consequently you) presuppose that a branch earlier in the tree will have an almost 100% effect on all the later branching points in the game. In his contrived example, this is indeed true. Of course, not only is his example contrived, but it seems to be contrived on the basis of some Mexican soap opera plot (I admit that I have never played BG). He is working ass-backwards on the basis of his overly complicated story, making things needlessly difficult both for himself and his audience.

The key word here is emergence - GreenNight pointed out the difference between Troika's "worlds" and Bio's "story", and this is the core issue. Bioware's designers seem to think that an RPG is a choose-your-own-adventure book (but even those have much less than exponential branching growth) with some combat in between. Companies like Troika and Ion Storm, who are actually advancing gameplay and making good games, are taking advantage of the PC as a simulation medium.

Since you claim to be a programmer, the value of abstraction should be pretty apparent in helping to cut down design work. As triCritical mentions, even a simple counter (even something as simple as the good/evil meter in Fallout) is enough to provide a branching storyline with very little effort or side effects to cause bugs. Gaider seems to be bent on banging his head against the wall by accounting for every dialogue choice in the game.

This brings me to my next point. I forgot what the rule is called, but it goes roughly something like this - past about 5 developers, software actually takes longer to write than it would before. Gaider doesn't have the option of changing the storyline or adding features because he has a manager, and his manager has a manager. Likely it is impossible for him to implement something he wants or needs in the game engine. This is not the case in smaller game companies, where the average ability/motivation of each of the designers is also likely to be better (for reasons I won't go into here). So, even if Gaider was competent, he would more likely than not be unable to transcend Bioware's development and organizational methodology. Meanwhile, companies like Troika and people like Jeff Vogel can make deeper games in less time and with less effort.

BTW, is this the only guy doing writing on a given title? A 3,000 word a day average makes it sound as if there's a lot of banal dialogue ending up in Bioware games.

 

[Rosh]

ljw1004, congratulations on a flawed example, yet again.

Now, in the pursuit of acting as a BioWare apologist, you seem to be steadfast in defending their rather inaccurate claim. You (and the BioWare chimp) are going by the assumption that every path will be avilable to every type of character/story arc possible, and that it would be in such a stepped, equal progression. So much for story variety.

To put the progression in such a cookie-cutter manner would be exactly the talentless crap I'd expect BioWare of. You couldn't dissect Troika or Fallout in the same way, because they were designed, not clichéd-path.

Your example is also flawed to an extreme, because while it does superficially look like exponential growth, anyone who has actually developed and put together into branches (and who has a clue), could tell you that adding in a sub-branch or possibility an entire arc will not add in an entire exponential of complexity put into the development. In fact, it's rather easy with some practice. It's when you go into a stupid shopping list example is where you get hung up, nowhere even touching the practical reality.

Joining the bad guy at the end of Arcanum "exponentially complex"? Fuck no. It doesn't make playtesting, QA, development, etc. doubled by any means. Neither does a lot of other aspects that create numerous ways of playing through the game.

 

[Section8]

"...Dave GAYder (thanks on that one section8)...

I read the name and thought "Hey, this slur just writes itself!" Kind of like a Bioware "plot."

The key word here is emergence - GreenNight pointed out the difference between Troika's "worlds" and Bio's "story", and this is the core issue. Bioware's designers seem to think that an RPG is a choose-your-own-adventure book (but even those have much less than exponential branching growth) with some combat in between. Companies like Troika and Ion Storm, who are actually advancing gameplay and making good games, are taking advantage of the PC as a simulation medium.

Excellent point. If there's one uber design philosophy that should be held in higher regard over others it is "Create a game world, not a game" Designers with a programming background seem to "get" this. A game should be structured like a program. If all interactions are considered as functions, that's a good starting point. The player is essentially spending their time in the "main program loop", and can call an interaction at will, and every time that interaction is called it behaves in a manner consistent with any other calls to said interaction.

It's a hell of a lot better than writing massive chunks of nested if statements for everything possible interaction. That accounts for a great deal of the reasons behind why I can't stand BG, and think Deus Ex isn't nearly as good as it's cracked up to be. Reliance on explicit scripting. General cases and consistent game world rules are far superior.

 

[Deathy]

Now, in the pursuit of acting as a BioWare apologist, you seem to be steadfast in defending their rather inaccurate claim. You (and the BioWare chimp) are going by the assumption that every path will be avilable to every type of character/story arc possible, and that it would be in such a stepped, equal progression. So much for story variety.

I thin the funny part is that, in suggesting the exponential story lines, that BioWare chimp has essentially approached non-linearity in a linear fashion, and doesn't even realise it.

Weren't they looking for a dialogue writer? With that level of talent expected, even I (With no experience and average writing skills) could land a job at BioWare.