Search This Blog

Tuesday, April 11, 2017

RogueLike Radio changed my mind about something in my "Let's Make a Roguelike" book!

Roguelike Radio Changed My Mind (and I'm only on the first year)

Still working on my how-to book, Let's Make a Roguelike with Construct 2

Here's an excerpt from my initial draft

...And here's a link to the Roguelike Radio podcast.

Apart from consuming everything I can about modern game design and programming patterns, particularly procedural generation, I wanted to uncover more of the mystery about what makes something a good Roguelike and gives players that "Roguelike" feeling.

So I decided to listen to EVERY episode of Roguelike Radio.

I have been immersed in books on game design, level design and the history of video games as well as every GDC or PAX talk on anything to do with Roguelikes I can find... and my favorite resource so far is the Roguelike Radio podcast.

I'm only on episode 18 now and just catching up with 2012.  The podcast is fantastic and even after the first season I'm beginning to unravel the mystery.

The podcast generally 2 or 3 talkers, Darren Grey, Andrew Doull, John Harris, with a special guest developers like Daniel Jacobsen, Nicholas Vining and David Baumgart of Dungeons of Dredmore, or Edmund McMillen of The Binding of Isaac.  They've even had the original creators of Rogue,  Michael Toy and Glenn Wichman.

So far I've learned a few things that make a Roguelike.  The speakers, generally aren't sticklers for turn based play or ascii graphics, but they definitely have things they like and don't like generally.

I know it when I see it

When defining what makes something Roguelike I am reminded of the phrase famously used in 1964 by United States Supreme Court Justice Potter Stewart to describe his threshold test for obscenity:

"I shall not today attempt further to define the kinds of material I understand to be embraced within that shorthand description ["hard-core pornography"], and perhaps I could never succeed in intelligibly doing so. But I know it when I see it..."

But one thing that comes up over and over is the idea that items and weapons should be there to change the way the game is played.  Not and I quote "Trading a plus 2 sword for a plus 3 sword."  This indicates a kind of linear player progression that doesn't really change the gameplay or add real variety.

Same S#$%, Different Numbers

I think it's something I noticed in other RPGs that made me bored and want to quit playing.  I call it the "Same s#$% more hitpoints" effect.

When I played Secret of Mana, a beautiful ambitious RPG for the SNES, after getting higher levels, I realized I was just attacking the same exact things but they just had more hit points and did more damage and I did more damage and had more hit points.  It was basically the same exact thing as the first dungeon, only the decimal point had moved.  Even the higher level spells didn't really change the game much.   They were like the low level spells but did more damage, or healed more... again same stuff different numbers.***

In RPG maker there's even a graph for this progression.

What's the point?  The above graph really soured CRPGs in general for me.  As Ed McMillen said in their interview, "There hast to always be a mystery."

*** This is given a strong caveat that Secret of Mana was in most every way a wonderful game and I got a lot of enjoyment out of it.  As far as SNES expansive for the time RPGs go it's pretty amazing, and was rated exactly that by IGN:  HERE 

So it's not so much THAT game, it's what I believe to be a flaw in the kind of game it is.  The potential for repetitive progression in RPGs in general... and I have run into that same problem in a wide range of games from The Bard's Tale for the Commodore 64 to Borderlands on Xbox 360.

I think there's a point in most any RPG including Dungeons & Dragons when the monsters just seem like numbers and it loses its fun.  Maintaining this fun and mystery is what a good Roguelike can do.
In a random review of Borderlands 2 I found, one blogger said it:

"Fiddlefarting over whether to sell Gun A that does X damage and has Y% of this effect or Gun B which does  X+1 damage but doesn’t have that effect but another is not role-playing...  ...the development curve and sense of progression in the game remains completely screwed up- it’s too long, drawn out, and rewards perseverance and grinding rather than good play and player skill-building. " --Michael Barnes'  blog

This is very close to the same things the Roguelike Radio podcast said in their Diablo episode.  "...If I fail it's because I didn't grind enough." (paraphrasing from memory).

So now that we've identified the problem, what is the solution?

I'm going to have to listen longer to find that answer, but one part of it seems to be having the player not just progress, but change.  In The Binding of Isaac, getting different weapons changes how the player looks and plays dramatically.  It's far beyond adding another +1 to your sword.  Different resources should add variety not just progression.  In a Roguelike, some form of resource management is a major part of the game, and how it is executed is as important if not more so than procedural generation.

So here is my new "big three" list.  Because it's important to have things in arbitrary 3 item lists.  

1. Procedural Generation
2. Permadeath
3. Player Progression  Resource Management

After one season of Roguelike Radio, I have changed my mind about my 3 most fundamental properties of a Roguelike.  Originally I thought that it was 1. Procedural Generation.   (still cool with that one) 2. Permadeath.  (It's important)  3.  Player progression.

But player progression isn't the best way of saying the "roguelike element" I'm trying to describe.

They described a better and I think more essential to the Roguelike experience game mechanic, and dedicated an entire episode to it:

Resource management.

First free clipart I found when I did an image search for "Resource Management."

The combination of things you collect and use should shape how the game plays out as much or more than the random layout of the dungeons or distribution of the monsters.

So I'm changing it to Resource Management, and I have much more to say about it, and I will improve the inventory and skill management parts of the book and sample program.

Yay research.  Only 4 more years of podcasts to go.

As far as the book is going I'm working on several generic dungeon making algorithms:  Traditional "Rogue" rooms and hallways,  Cellular Automata Caves, Agent based level carving, pre-made rooms in bitwise patterns... and more.

...back to work.


  1. I think the more appropriate term would be "player character progression". Roguelikes have an important aspect of "player progression", i.e., each death (or win) makes you a better player. Agreed that player character progression based on stupid increasing of numbers does not add much to the game, and I am proud of the system in Hydra Slayer -- as a general rule, bigger numbers are not better.

    HyperRogue is designed around an infinite world, and there is no standard resource management -- although the risk vs reward aspect of collecting treasures, and their unlocking of new lands which add more variety to the game, is important, and could be considered a non-standard form of resource management.

    1. Maybe that would work well too. The concept of "resources" is a much wider more diverse abstraction than I previously thought. Sometimes just the geometry you can get to and occupy is a resource. I'm learning all the time.

      I'll have to check out Hydra Slayer!

      Thanks for reading!

    2. Indeed -- geometry is an important resource in HyperRogue, for example, in Land of Eternal Motion you have to keep your distance from the dogs. I have read somewhere that hitpoints in D&D do not necessarily correspond to actually being wounded, but they could be understood as a general measure of health, stamina, positioning, etc.

  2. Hit-points are definitely an abstraction. A heavily aggregated metric of how much general stuff can happen to an entity before it can't take any more. It gets ridiculous with CRPGS when something starts out with 10 hit points then later has 50,000 hit points. Can the character swallow a nuclear bomb now?

    One thing I noticed from D&D was the idea that in most games with health you tend to perform full capacity performance until you lose your last hit point then... dead. So your states are 100% or dead. I remember being really shocked by the first Resident Evil when you got swiped by zombies a few times you started limping slowly and eventually holding your side in obvious pain.

    There's a really good GDC talk called "Made of Meat" on the subject.

    In real life I can assure you if I lose a single hit point I'm not at full capacity.

  3. I get a paper cut in battle.. .I'm like nope... you win monsters... I'm out. Keep your "orb of whatever... "