Gamasutra: The Art & Business of Making Gamesspacer
View All     RSS
October 24, 2014
arrowPress Releases
October 24, 2014
PR Newswire
View All
View All     Submit Event

If you enjoy reading this site, you might also want to check out these UBM Tech sites:

What ‘90s SSI Goldbox RPGs Can Teach Us about the Modern RPG (Part 2 of 2)
by Thomas Henshell on 08/21/14 03:37:00 pm   Featured Blogs

The following blog post, unless otherwise noted, was written by a member of Gamasutra’s community.
The thoughts and opinions expressed are those of the writer and not Gamasutra or its parent company.


Sometimes a new solution is just an old idea in a new package.

Last time, I covered how re-reading the Dragonlance novels got me thinking about a new kind of RPG.  One that is a realistic open world, procedurally-generated, and full of meaningful choices.  The question is: How can I build it? Is it even possible?

Shark Tank is a TV show where prospective entrepreneurs (fish) pitch their business ideas to industry titans (sharks).  After one of these pitches, Kevin (the grumpy Simon type billionaire on the show) asked, “Which is more important: The idea or the execution?”

Thinking for a minute, the entrepreneur nervously answered “Execution.”

“Right answer,” said Kevin.  And he proceeded to invest in the idea.

I had a solid idea, but I wasn’t so sure about how to execute it.  If I want to allow the player true choice, I have to address the cost of choice head on.  I found the answer in two places: another TV show . . . and the early ‘90s.

Cost of Player Choice

There is one reason why the same characters die at the same time both in my playthrough and your playthrough of Watch Dogs: cost.  Video game budgets now eclipse movies by a considerable margin—leading to less relevant choices.

Two aspects of cost affect choice: 

1) The industry has painted itself into a corner; everything must be a 3D model and fill a 1080p (and soon 4K) framebuffer. For your character to have a dragon mount, someone has to model, rig, and animate it. If your character speaks to you, someone has to record a voice actor acting the lines.  What if the designer wanted to add a Pegasus?  A griffon? A wyvern?  Four mount choices is four times the cost. So the narrower the player’s choices, the lower the cost. 

2) Branching paths lead to unplayed segments of the game. Let’s say it costs $1 million to create one hour of game content. Let’s also say we want to offer the player a 10-hour experience.  Imagine that our inspired game designer wants to have three totally different branches of the game based solely on player choice.  This means $30 million in cost, but the player only sees $10 million of it.  From a player’s perspective, the other $20 million might as well not be there.  That hurts when you’re footing the bill and don’t even know how many units you will be able to sell. Contrast this with a movie: Every dollar spent is experienced by the audience—a 1:1 ratio. This isn’t so with games: The higher the ratio of “experience vs spend,” the better the producer feels about the whole thing.

So the only way I see to offer players real choice is to greatly reduce the cost of choice in the first place.  Reduce it down so low that (from a cost perspective) it doesn’t really matter what they choose.

But would reducing this cost somehow lose the magic found in role-playing?  Fortunately, watching TV gave me the answer.


If you don’t know Abed, you should.

Community is a comedy TV show about community college.  Outcasts decide to form a study group, and hilarity ensues. In one of the later episodes Abed (a super geek) runs a D&D game for the group to cheer up a fellow student.  Whether you have played D&D or not, the episode is ridiculously funny. They managed to write something both my wife and I enjoyed, yet for entirely different reasons.  Beyond this, it is also a brilliant showcase of role-playing at its finest.   

Abed is the Dungeon Master (or GM).  He simply tells the context of what is happening, and asks the characters what they want to do.  There are a few dice rolls, but it is mostly just the characters making simple choices from the options Abed gives them.  Sid Meier said it so clearly: “Games are a series of interesting choices.”  It is so simple . . . yet incredibly easy to forget.  Something inside us seeks a more complicated solution.

Much of what I see in role-playing tabletop (be it the miniatures with grid map or rule books galore) and role-playing computer games (simply moving from A to B, FedEx quests, forever rearranging my inventory) is minutia, which gets in the way of the really interesting choices.   But what if a game is solely focused on interesting choices?

Community helped me see role-playing in its simplest, purest form.  I had to be reminded of what role-playing once was back in the late ‘80s and early ‘90s—when developers weren’t tempted to throw in the proverbial kitchen sink. 

SSI Goldbox (1988-1993)

In four-ish years, SSI released a total of 14 AAA (for the time) role-playing games.  Many people fell in love with Pools of Radiance (the first game)—but for me, it began with the Dragonlance themed Champions of Krynn trilogy.  It doesn’t matter.  They play similarly due to sharing almost identical game design and engine.

The SSI games had to do a lot with little computing power and even less memory.  Some of them ran on the Commodore 64—which for reference, only had 64 kilobytes of memory.  An impressive feat when you consider that there are dozens of types of enemies, equipment, treasure, settings, weapons, and spells.  Character creation alone has 9 races and 7 class types—63 possible combinations before you even get into multi-classing.

How could they pull this off?  Short answer: Text and imagination.

They used basic graphics to set the mood, then used text to feed the player’s imagination.  Did it work? You bet!  I and many others have extremely fond memories from that era.

Here are three examples of how these games played: 

This is the overworld map in Champions of Krynn.  You are the square. The circles are locations you know of.  You can get from your starting location to a quest location using the cursor keys in about 12 seconds.  Contrast this with all the running around in Final Fantasy or Skyrim.  Once you know where you want to go, decide to go, it takes tons of time to get there.  I believe deciding where to go next is an interesting decision.  Actually travelling there is not. 

This is an interaction with an NPC.  There is an image of the character you’re talking with (for context), and then text describing what they say or what is happening.  As an aside, I like how they made it even more minimalist by telling you to go to the included physical game journal to read the rest of the text. 

And finally, here is navigating a town. You decide the stores/locations you wish to visit by selecting the name below, and bam!—you are there and can start buying/selling.  Now contrast this with Skyrim’s Whiterun where I have to run from my house to the general store, to the castle wizard, back again to the center of town for the other vendors.  It’s just so much running around to execute a decision I already made minutes ago!  Ironically, Skyrim’s creators recognized this and put in a fast travel system.  Well, what I’m talking about is an RPG that only has fast travel with zero wild goose chases and absolutely no running around.

Pictures suppress imagination

When we start life, we cannot read. So what do our children’s storybooks have?  Big pictures and little-to-no text.  Later in life, we can read The Lord of the Rings (which is all text and a few maps)--but as our comprehension and imagination mature, books decrease what is shown and increase what you should imagine.  Yet AAA games have gone the other way; they spend millions to try to replace imagination.

Chris Crawford makes this point in his 2003 book, Chris Crawford on Game DesignHe states that chasing reality with graphics is a fool’s errand.  If the game world is super-realistic, the character models will break the suspension of disbelief.  If the character models are super-realistic, then the animation will break it.  If the animation is super-realistic, then the player choices will break it.  We will never replicate reality in a video game, so why are we trying so hard to get to a destination that’s impossible to reach?

“Good artists copy, but great artists steal.” - Picasso

If you thought I was only stealing from the ‘90s, you would be wrong.  Having acquired a taste for creative theft, I am plundering the following modern era games as well:

Wolf Among Us / Walking Dead – Branching storylines based on character choice.  Telltale’s games are well-made and have significant cost to their choices. Therefore, there are limited actual branches. I will be offering more choice than their current titles.

Mass Effect 1 / Dragon Age Origins – I found the ability to choose your allies and befriend certain NPCs compelling.  There is also some story branching.  I plan to take this further.

Skyrim – The most open world AAA game I have played.  I like that the player can decide to simply kill a random NPC—and the guards react.  And even if you get away, people still remember what you did.  I am stealing this core concept of openness with consequences and taking it further.

Dwarf Fortress – A hardcore indie game where you try to found a dwarven settlement in a procedurally-generated world.  Extremely challenging to figure out how to play, but worth it.  I have spent the last few months developing my own procedural world generation with Dwarf Fortress as a conceptual guide.

Red Shirts – An indie game about relationship management. It’s billed as “facebooking in space.”  This game showed me how complex NPC interactions can be achieved in a simple and fun manner.  Also as an aside, this is a great example of what a female game designer can bring to the table: totally new ideas where we all benefit.  I cannot imagine a man designing this type of game.

The way forward

I am taking the elements I think are worthwhile (the gold) from ‘90s era RPGs and bringing it to a modern context.  By making the game in 2D—aided by the strategic sprinkling of amazing art, betting on a well-written script, and engaging the player’s imagination at all times—I believe I have effectively reduced the cost of choice.  Therefore, I can greatly increase player choice and the world’s reaction to it.

Not only this, the game plays faster.  A boon to those of us with only a few hours a week to game because of other responsibilities – like kids :)

I get that requiring players to read and utilize their imaginations will limit the size of the game’s total player base.  I’m OK with this.  I toughed it out and figured out how to play Dwarf Fortress.  I left that game with some great memories.  Last year’s big budget SimCity did not reach the same heights—to put it mildly.

You may be wondering: If these ideas are so good, has anyone done something like this before?  And if so, were they successful?

Well I’m happy to say that I’m not the first or only person thinking this way.  There is another indie game that implemented many of these concepts in 2012.  Maybe you heard of it.  It’s FTL.

FTL is a sci-fi game where you guide your ship and crew from one edge of the galaxy to another in a certain time period.  At each jump along the way, the game presents you with choices.  It implements many of the ideas I’m talking about here for Archmage Rises:

  • simple 2D context graphics
  • text describing the details of what is going on
  • text for the choices the player makes
  • no flying around; just “jump, make a decision,” “jump, make a decision”
  • procedurally-generated universe so that it’s different every time you play
  • permadeath (you only have one ship and one crew; let’s see how far you can go!)

Conceptually, I suppose you can say that Archmage Rises is FTL with mages in a fantasy world.  This probably gives you the best idea of what Archmage Rises plays like.

If you haven’t already done so, let me encourage you to sign up for the Archmage Rises newsletter.  I am making a video tour of the current build; newsletter subscribers will be the first to see it!

Thanks for the emails, tweets, and posts.  Your enthusiasm for Archmage Rises will get this game made. Stay tuned.



Any questions on the game can be posted on the Facebook page.  I answer questions almost every day.

Archmage Rises Website: 

Twitter: @LordYabo

Related Jobs

Activision Publishing
Activision Publishing — Santa Monica, California, United States

Tools Programmer-Central Team
Crystal Dynamics
Crystal Dynamics — Redwood City, California, United States

Senior/Lead VFX Artist
Magic Leap, Inc.
Magic Leap, Inc. — Wellington, New Zealand

Level Designer
Magic Leap, Inc.
Magic Leap, Inc. — Wellington, New Zealand

Lead Game Designer


Kyle Redd
profile image
I don't think going a text-only route would be a problem for you, making a CRPG. Games like Wasteland 2 and Pillars of Eternity have shown that there's clearly an audience for that style of game, even today.

The quality of the writing will be scrutinized by that same audience, though. I don't know how many resources you plan on devoting to that part of the game, but I think it may be both more expensive and more time-consuming than you are expecting. Particularly for the fantasy genre, getting good prose that doesn't sound either too generic or too obtuse is a challenge, because of the amount of content that has come before you.

Good luck. I am happy to have subscribed to your newsletter and will be following development closely.

Thomas Henshell
profile image
Good point for sure.

I'm fortunate to have a professional writer on the team. That in and of itself doesn't solve it, but it's how serious i'm taking it.

We're using planescape torment as our measure of "good game writing".

On second read of your comment I'm wondering, are you suggesting that it is easier to have "good art" than "good prose"?

Kyle Redd
profile image
Well, that wasn't what I was thinking when I wrote the comment. But now that you mention it... Yes, I do believe it is easier to have good art than good prose in a game, particularly for RPGs, and even more so for the type of RPG you are creating.

That might sound ridiculous to most game developers. Honestly though, thinking of all the games you've played in your life, how many of them have made you think "Wow, this art in this game looks amazing" compared to those where you thought "Wow, the writing in this game is amazing"?

Maybe it's not that art is easier to create than prose, only that prose isn't taken as seriously by game developers, because they believe (correctly for the most part) that their target audience cares more that the game is visually appealing rather than if the characters and story of the game are both compelling and believable.

However, it appears obvious to me that you do care about how your game is written, from all the influences you've cited. That's one of the reasons I'm looking forward to it.

Speaking of influences. The Dragonlance books are a good example of the sort of characters we don't ever really see in gaming. A persistent theme throughout the twins' storyline was Raistlin's deeply-felt resentment and envy towards his stronger, more well-liked brother (if I remember correctly).

That persistent undercurrent of anger towards those we love is a pretty common emotion that most everyone can relate to. But we almost never see that type of complexity represented in games. I can't think of a single title in which the protagonist demonstrated feelings of jealousy or envy towards anyone. And yet those are such common, normal feelings for humans to have!

There's been thousands of games where the hero set out to rescue his kidnapped girlfriend from his enemy, right? Why hasn't their ever been a game (other than maybe a visual novel or two) where the hero's girlfriend *left* him for his enemy? It's a part of life that we all have to endure at some point - The person we want to be with wants to be with someone else. But this sort of real-world personal conflict seemingly can't be replicated in games.

Thomas Henshell
profile image
Preach it brotha!

Matt Robb
profile image
The one big downside of only being able to go to places you know about is that you can't simply explore.

Wouldn't matter too much in a manually designed world, because they rarely put anything interesting out in the wilderness. Even Skyrim's wilderness just had "random hole in the ground #37" for the most part, especially since the game would generate a quest to send you to any spot you hadn't found yourself.

But in a procedurally generated world, no one really knows what's over that next hill or in that next grid square or what have you. Something to think about.

Thomas Henshell
profile image
I agree, which is why I'm taking the skyrim approach.

The world has a ruins. But you can't get to it until you have a reason to get to it. The game engine creates a quest for a location you haven't been.

Thomas Creutzenberg
profile image
Great article. And thank you very much for making me remember Pools of Radiance. I don't even dare to guess how many hours I spent here...
I have to agree with most of what you're saying, but I don't 100% agree on your opinion about travelling between location. I totally understand your point about how travelling can seem boring and no real decisions take place there. But in some games exploring the world and getting a feel for its size and feel can be a fantastic experience which actually influences your decisions later. Many JRPGs come to my mind. After the initial exploration though a fast travel system is normally nice to have to avoid travel-grinding.
The explorer type of gamer in the old 90's rpgs could explore the single levels, but I for my part never really got a greater feeling for the "space of the world" itself, so there was no place to create a lot of empathy. It was not necessary in these games back then, but for many games it totally makes sense and enhances the immersion quite a bit.
If you're going for a fast decision-heavy game though, then your approach seems to be absolutely right.

Thomas Henshell
profile image
I'd like to take this moment to bring up a good point raised by Thomas C. about the fun in exploration.
I learned a ton from Gregory Trefry's book "Casual Games"

He identifies 9 core things that humans find fun to do. One of them is exploration.

I have recently completed Watchdogs on my PS4. It has a huge "living" city for me to drive through as I go from mission to mission. Like Skyrim it has hooks for the A.D.D. exploration player to constantly be discovering crimes to solve, virtual games to participate in, etc. Personally I'm more of a main mission guy, but I have a friend who could care less about the main mission and is all about discovering neat side missions. I get that there are different kinds of players.

It appears in watchdogs that the designers put the main story missions purposefully far away from each other to force me to traverse the city and see the various districts. What this means is I spend a ton of time driving. TONS of time driving. And the driving just isn't that fun. It's kinda fun, it's super fun for the first 20 minutes, but really what is my motivation as a player? To get to the next mission objective. How do I get there? Driving in traffic just like in Toronto: tons of cars, lots of stop lights, basically the worst bits of real life. The driving is an annoying forced exploration of the game world in between goals I care about. The driving to me is just filler.

So in my design I have cut out the filler and there is no running around on an overland map. Just like FTL doesn't let you fly all around a solar system. This would leave you to think there is no exploration in Archmage Rises.

Actually, there is. It is exploration of a different kind. There are hundreds of cities to visit. There are hundreds or thousands of NPCs to converse with. The exploration is found in dialog choices rather than trying to find the entrance to a town.

This fits my play style and therefore I can be a judge of if I've achieved it or not. What I see in AAA development is they always include stuff for a variety of play styles (the 4 core types of gamers). I think that is good and right on a big team. I don't want to sound like Phil Fish here, but "I'm only one guy!" Since i'm not that A.D.D exploration gamer I'm not going to design for them. Gotta keep it small and focused. :-)

Maria Jayne
profile image
I feel there is definitely a point where when you rely on visuals to tell a story you detach from that story. You stop thinking and imagining it and you just sit there slack jawed at the moving pictures and explosions. Imagination is the most powerful story tool and it requires such a tiny seed to grow, some people have very vivid/creative imaginations and benefit from a lack of visual explanation, some require it to be demonstrated to them because they don't.

The reason Baulders Gate, Planescape Torment, Fallout 1&2 etc are so loved are not because they are visually impressive, they were decidedly average even when they were new. They are loved by the people who took the tiny seeds of the narrative and grew them in their imagination. The characters,events and locations surpassed the sum of their code because of the imagination of the players.

Now I don't mean we should stick to text adventures, visual effects and action can certainly enhance a story, it's just often that becomes the only narrative and not enough is left to the imagination. Wherever the sweet spot is, it's behind the "next gen triple A photorealism" curve that is being force fed to us. We need to leave enough room for players to imagine some of what is happening, otherwise the narrative becomes impersonal and sterile.

By giving the player scope to imagine, you can create that more personal experience that can surpass whatever is happening on screen. First though, you have to discover where the line is on visual representation.

Abdullah Kadamani
profile image
I kinda disagree with you about the visuals on Planescape, Baulders gate, and fallout. Its true that they weren't technically impressive but they had interestingly designed character, creature, and in the case of Planescape, background designs. They weren't technically impressive but they were visually interesting. Tell me that any of you have forgotten what a super mutant looked like, even for a moment?

Joshua Darlington
profile image
Boredom leads to day dreaming. Its a scientifically proven fact.

So those that advocate under realized narrative as a trigger for user imagination should proceed with caution.

Michael Wenk
profile image
First off, this article made me happy in remembering Pools.


"Branching paths lead to unplayed segments of the game. Let’s say it costs $1 million to create one hour of game content. Let’s also say we want to offer the player a 10-hour experience. Imagine that our inspired game designer wants to have three totally different branches of the game based solely on player choice. This means $30 million in cost, but the player only sees $10 million of it. From a player’s perspective, the other $20 million might as well not be there. That hurts when you’re footing the bill and don’t even know how many units you will be able to sell. Contrast this with a movie: Every dollar spent is experienced by the audience—a 1:1 ratio. This isn’t so with games: The higher the ratio of “experience vs spend,” the better the producer feels about the whole thing."

The problem I see is that you are looking at it wrongly. The problem with making the 10 million dollar story is that it won't appeal to as many players as the 30 million dollar one. Marketing may think that all RPG players are the same, but that isn't true. Having a diverse story with many branches means you'll appeal to many different players, which means more revenue.

Nick Harris
profile image
Have you considered having no visuals, no text and no controls? The screen would be black, indeed you could turn it off if the audio was routed through external speakers. Something like Kinect would be used to hear questions and commands based upon the information the 'Dungeon Master' told you about your surroundings. If you fail to ask the appropriate questions, you never know about certain threats in advance, or walk right past treasure. Not everyone can organise a cosplaying Dungeons and Dragons group and what is appealing about using the ONE is that your friends who might have gone to different colleges or made lives in different states can arrange to get together on a Sunday and share an adventure furnished with their imaginations. If the DM can't be a real person doing halfway decent voices and tracking stats, then have the console take on that role.

It would resemble an improvised radio drama, not merely limited to D&D.

Matthew Bentley
profile image
This fits in with my theory of 'cognitive gaps':

But it doesn't just apply to graphics. It can apply equally to text, sound and other aspects of a game's cognitive layer.

John Gordon
profile image
Just want to say I love those old Gold Box games, and I totally agree that often "less is more". High production values are not a good substitute for solid design.

Look forward to seeing "Archmage Rises" when it's ready!