Frayed Knights: Talking Too Much

Posted by Rampant Coyote on March 4, 2011

I’m overdue for some updates on Frayed Knights: The Skull of S’makh-Daon, the indie RPG coming later this year from Rampant Games which refuses to treat RPGs with the seriousness they deserve.

I’m still (after all this time) feeling like kind of a n00b when it comes to making a CRPG. I feel like I’m learning hard lessons that those who came before me have already learned, internalized, and understand so implicitly that they don’t even think about it. Though those hard lessons may very well be the reason that some modern, big-budget RPGs feel so streamlined and “dumbed down” – as these games grow in complexity and features, design concepts which are merely ‘complicated’ in a smaller game might become really nightmarish to manage.

This is purely conjecture on my part, but I can imagine this is a factor.

This is kinda delving deeper into a subject I’ve talked about before, but I tend to ramble about what’s on my mind lately, which has been consumed with “filling out” a lot of the world, NPC dialog, and lots of bug-fixing. A type of issue I keep running into in Frayed Knights 1 comes from two design decisions:

#1 – I decided to have conversations be full-fledged scripted conversations between all characters in the party, both among themselves and with NPCs, rather than the traditional approach of letting the NPC do all the talking in big blocks with a few one-line choices that will usually all be selected anyway. I can pull more characterization and humor out that way.

#2 – I have a lot of optional events or events that can occur in a free-form order.

These two design features have a somewhat complicated relationship. To understand this, you should recognize that traditional dialogs with NPCs in moden(ish) RPGs tend to be fairly generic and relatively context-free. The game simply locks out dialog options that no longer make sense, and the NPC spouts off blocks of exposition that only occasionally seem out-of-place given the current game state, and then only by player request. The player’s brain can fill in it’s own context to form the other side of the conversation, and even work itself into knots suspending disbelief when an NPC is talking about a dead villain as if he was still alive.

But when writing two-sided (or more-sided) conversations, having only one side be an active participant in the conversation sounds weird. Plus, I’m trying to make the dialogs at least somewhat amusing some of the time (though too many dialogs do get played pretty straight – there’s just not much I can do with them).

So for me, at least my style of writing for this game, this means the dialogs should be much more context-sensitive and refer back to the current game state and choices the player has made.

And combined with #2, this means a heck of a lot of variations of conversations, because while I do provide some choke-points in the game, I try to leave what I can pretty free-form and organic, which means players may totally screw up my imagined narrative progression ALL THE FRICKIN’ TIME, and my dialogs need to relate to that. And so I end up writing several variations of a full dialog navigation tree in a traditional system.

So – a hypothetical example quest (to avoid too many spoilers, though similar formulas abound) with some of the variants I might have to write. This is an easier, linear example… which gets more complicated as players might have A and C but not B and D, etc. :

Variant 1

John Enpicee: Greetings adventurers! I was wondering if you’d do a favor for me. I want the legendary Shoelace of Despair, which is found in…

Dirk: Oh, that thing? Already got it, and we sold it to a merchant for about fifty silver. If you hurry, he may still have it.

John Enpicee: Uh, oh. Okay, thanks. Bye.

Variant 2:

John Enpicee: Greetings adventurers! I was wondering if you’d do a favor for me. I want the legendary Shoelace of Despair, which is found in…

Dirk: Oh, we’ve got one of those…

John Enpicee: Oh, do you? Great. Um… can I have it? I will offer you two-hundred silver and this Sword of Infinite Destruction I’ve got laying around. Plus you can rest assured it won’t fall into the hands of somebody evil who might do terrible things with it…

Chloe*happy: Really? What kind of terrible things could someone do with it?

Arianna: Um, Chloe…

Chloe*neutral: Right. Nevermind.

Arianna: We’ll absolutely consider your offer.

Variant 3:

John Enpicee: Greetings adventurers! I was wondering if you’d do a favor for me. I want the legendary Shoelace of Despair, which is found in the Dungeon of Facepunch, found in the North Caverns of Anarchy…

Chloe: Oh, yeah, we’ve been there.

John Enpicee: You did? Oh. Well, did you happen to run across the Shoelace of Despair?

Benjamin: Uh… no?

Arianna: Not to my recollection.

Chloe: Is it cool?

Dirk*sly: Is it valuable?

John Enpicee: I will offer you two-hundred silver and this Sword of Infinite Destruction I’ve got laying around. Plus you can rest assured it won’t fall into the hands of somebody evil who might do terrible things with it…

Chloe*happy: Really? What kind of terrible things could someone do with it?

Arianna: Um, Chloe…

John Enpicee: Will you go back to the dungeon and see if you can find it for me?

Dirk: We will absolutely maybe consider it!

Variant 4:

John Enpicee: Greetings adventurers! I was wondering if you’d do a favor for me. I want the legendary Shoelace of Despair, which is found in the Dungeon of Facepunch, found in the North Caverns of Anarchy. In order to get into the dungeon, you will need to obtain the Lens of Myopia to get past the Maze of Macrame…

Benjamin: So that’s what that lens is for!

Chloe: I was wondering.

John Enpicee: Oh, so you got the lens already, did you?

Arianna: Yes, from the Mystic north of the river.

John Enpicee: Oh. I was totally going to tell you that. But hey, since you have it already, would you mind going into the Dungeon of Facepunch and getting the Shoelace of Despair for me? If you do, I will give you two-hundred silver and this Sword of Infinite Destruction. Plus you can rest assured it won’t fall into the hands of somebody evil who might do terrible things with it.

Arianna: Fine. Dungeon of Facepunch in North Caverns of Anarchy. Use the lens to get through the Maze of Macrame. Get you the Shoelace of Despair. Is that it?

John Enpicee: Pretty much, yes.

Dirk: Cool.

And so on….

Granted, the dialogs themselves aren’t horribly difficult to write, though polishing and finding and fixing typos make them a lot more work than they might appear. But hooking up all the logic to drive them – and then testing each of the possibilities, one-by-one, to make sure they all work right (even with cheat codes and / or save points to speed up the process) can be pretty time-consuming and prone to errors.

I still get a lot of holes (often identified by my testers) where they note that a dialog assumes something has already taken place that hasn’t, and vice-versa – particularly for dialogs written earlier in the development process when I didn’t have quite the handle on things I do now.

Where things get really crazy, I do tend to pull the same stunt of breaking the conversations into smaller, shorter pieces that can be mixed-and-matched with greater freedom. For Frayed Knights 2, I will definitely have to improve that process and make better tools for handling those conversation choices.

It could be worse, I guess. At least I’m not doing voice-overs for all the dialog.

These days, when I hear someone say that “it’s only text” or that writing dialog for an RPG is trivial, I want to tie them up with the Shoelace of Despair and throw them into the Dungeon of Facepunch. Even for traditional RPG dialog trees, it’s nowhere near as quick and easy as it looks. And for the way I’m doing it, it’s crazier still. But if I’d choose the same approach if I was starting over today… I’d just have been more clever about it. I’m tired of doing it the hard way.

But that’s what sequels are for, right?

Filed Under: Frayed Knights - Comments: 20 Comments to Read

McTeddy said,

I’ve actually been programming RPGs for a veeeeery long time. I think I wrote my first one when I was 11. 🙂

And even with over half my life as experience I learn something new every time. Whether it’s a new way of running a dynamic conversation system or a smarter way to implement items… I’m still screwing up and I’m still learning.

I know your pain, but I also know how good it feels when it’s over and you can see how much you’ve grown over a large project.

Keep up the good work!
Tom said,

Don’t worry, no one will read the text anyway 😀
Rampant Coyote said,

Hah – okay, Tom, that would have been funnier if I wasn’t so worried how true it will be…
Charles said,

Ooooh good thing you’re not underestimating this. I had the same problem in UW and quite frankly I didn’t go overboard with the variants. I found it very nervewrecking testing it all & remember Corwin still found a glitch that remained after release. The worst part is doubting whether or not you’ve actually really tested them all, and if there is one you’ve forgotten, you can be sure Murphy’s law is waiting around the corner 😀
Rampant Coyote said,

I’m actually at the stage now where the principle quest line is done – though I’m expanding on it a bit towards the end to make it more interesting. But I’m also fleshing out a lot of side-quests and optional content at this point. There are a lot of areas that needed more information and feedback to the players, so I’m writing all that up. I think I’m doubling my number of scripts from what I had only a month ago… and yet players will probably not even see half of it in a single playthrough!
Menigal said,

I always think there’s got to be a better way of dealing with conversations, but I’ll be damned if I can think of it!
My own experience of writing CRPG conversations is limited to a handful of NWN modules, but I found it to be simultaneously the single most interesting aspect and the most frustratingly time consuming. I always had to limit myself and the number of speakable NPCs just to finish what I did.
I don’t envy you testing all of the conversations in a game this size! I look forward to playing it, though. 😉
SteelRiverSavior said,

Great job in identifying this problem with RPGs. This, I think, is something big studios don’t really care about, because they build their games to have constant choke-points, or they make the quests so boring in the first place that it never really comes up.

But, I think the fact that you have identified this as a *solvable* problem, and have taken it into consideration, means that the game will have a much more interesting story to the player.
BenD said,

My angst over lacking a Windows machine to play this game on grows. 😉
Rampant Coyote said,

I would like to do a Mac port. I chose the engine I did SPECIFICALLY because it was Mac compatible. But I don’t actually have a Mac (or Mac programming experience) to use for porting.
DGM said,

>> “I try to leave what I can pretty free-form and organic, which means players may totally screw up my imagined narrative progression ALL THE FRICKIN’ TIME”

(evil grin)

You know what the best part is? By making me a playtester, you LITERALLY asked for it. At least, that’s going to be my defense in court when the time comes. 😉

But seriously, you’re right. Better tools for the sequels are probably an excellent idea now that you’ve done it the hard way and can better see what you need. I’ve never done dialogue trees before, but I almost feel guilty for all the different angles I’m hitting you from on some of these conversations.

>> “I would like to do a Mac port.”

It’s funny, but I was just thinking recently that it’s a shame you’re not releasing on consoles first. With a little tweaking I think your current control scheme would do alright on a gamepad. If I finally manage to nag you enough to get you to overhaul the controls, you’re going to strangle me if you ever decide to do a console release and I start bugging you to switch BACK.

>> “But I don’t actually have a Mac (or Mac programming experience) to use for porting.”

Here’s a crazy thought – once you get a mac maybe you could hire someone like Jeff Vogel to playtest for a week or two. His games are almost all cross-platform, so the crash course you’d get from him might be worth it.
David W said,

I have to admit, every time I start to debate whether I want this because learning a new character builder (especially one with as many choices as you’re offering) sounds like work, you come up with a post like this one which reminds me why I’m interested.
Modran said,

*sigh* I know your pain. I tried this with an NWN module months ago. A pain in the ass, but I know I read the texts every time. Even in WoW ^^.

DGM, you’re lucky you’ve been chosen :p !
Rampant Coyote said,

Lucky my foot! You don’t wanna know the crap he has had to wade through…
Modran said,

John Enpicee: To get the Holy Dung, you’ll have to wade through the sewers of the Dungeon of…

Dirk: Aw, Crap…

John Enpicee: How did you know it was the Dungeon of Awh’Krapp ?!
DGM said,

There have been some frustrating aspects, I admit. But honestly, part of me really enjoys uncovering and analyzing bugs. So it’s not as bad for me as it might be for others. If I could actually get paid to do this kind of work over the internet, I’d be a happy camper.

I will say that the “storm bat” business was bad, but that’s because it was entirely my own fault and my brain won’t seem to let me live it down. You know how you have your black triangles? I’ve caught myself thinking of similar mistakes as storm bat moments. Yeah, that one’s obviously going to haunt me for a while. (rolls eyes)
LateWhiteRabbit said,

I was working on a game just a few weeks ago, doing some design work and going over it with a professional developer.

I mentioned I was designing files and other clues in a level to fill the player in on backstory – to which the developer (someone who I have great respect for) exclaimed with a worried expression, “You aren’t having the player READ anything are you?”

“Well, I guess I could try and convey all the same information with visual clues only . . . .”

“They won’t be clues the player has to think about, will they?!”

“I . . . uh . . . isn’t that the definition of clues?”

“Rework the design. Players don’t won’t to read or think anymore.”

I picked up alcohol on the way home afterwards, in case anyone is wondering.
Rampant Coyote said,

Ye gods.

Okay, yeah, I expect your friend is pretty much repeating the mantra given to him by his employer and by the industry in general. Possibly a professional pop band is given similar advice, to make all their songs a metaphor for sex or something. It’s lowest common denominator stuff — adhering to the formula that generally reduces the barrier to sales.

But I don’t drink, and I’m sure I’d be tempted to pick up some alcohol on the way home after a conversation like that. This is what the mainstream industry has become.
LateWhiteRabbit said,

@Rampant

Pretty much. He certainly would be willing to read and think while playing a game, and is quite a cerebral thinker himself, but yeah, that’s the “industry line” he was repeating.

If your game doesn’t have instant appeal to a 15-year-old male demographic, it seems you have to promise your first born and an arm and leg (including sexual favors) to get it made in the current state of the AAA games industry.

That is indeed what the mainstream industry has become – a timid, lowest common denominator, market-driven piece of fluff. You have to be a old world god like Sid Meier, Will Wright, Tim Schafer, etc. to get a chance at an original design, and even then, one failure and the big boys will take their ball and go home to mark you out of their little black book.
Modran said,

*sigh*
Indie RPG News Roundup, April 2011 said,

[…] Development continues. Much to say, but not much of interest to people who aren’t knee-deep in development. But there’s been updates on how (and why) Endurance works in the game, changes to how the story and setting are presented, and the amusing pain in the neck that has been the Frayed Knights conversation system. […]