Category Archives: 7DRL

7DRL 2017: Day 1

7DRL Challenge Day 1: Unfortunately GDC lay me flat on my back for five days straight through the weekend, so I’m getting a really late start on my challenge. I guess I’m be aiming for submission late Sunday to get as many hours logged as possible.

Today I was able to re-acquaint myself with the ugly-ass 7DRL 2016 codebase and temp sprites. <Shudder> All I had time for was to start working on road generation and take a step in the direction of transforming my combat movement into strafes, accelerations and so on. Still got lots of work to do. Still, fun to see things moving forward.

7DRL 2017: Westbound and Down

This is my third 7DRL, and this year I’m hoping to kill two birds with one stone by exploring some alternate gameplay concepts I first explored in last year’s entry, Auto FireAuto Fire introduced car combat and exploration of a (poorly-rendered) cityscape and in the ensuing year has been extended to 3D, with richer generation, an overworld countryside, and a variety of superior UI and tools.  None of which I’m going to use this week, see below.

This year, with Westbound and Down, I want to explore some of the “Convoy” style aspirations that I have yet to put into Auto Fire…  Traveling from town to town and taking on cargo missions between outpost cities in a post-apocalyptic western U.S.  Instead of exploration combat, the player must drive “blocker” for a convoy of trucks that is continually harassed by bandits and other road hazards.  Upgrade your car and convoy vehicles, hire drivers, maintain your stock of ammo and fuel, and take on loads with higher risks for greater rewards.

The tough part of this year is that even though I’ve progressed quite a bit with Auto Fire’s codebase over the last year…  in the spirit of making this a fresh 7-day effort, I’m working from the code from 2016’s 7DRL and seeing what I can construct within those confines.  I hope to create an alternate movement model (forced directional movement, with most of the maneuvering involving lane-changes and acceleration/deceleration), a regional highway map, a cargo quest structure, some interfaces for cargo and hireling loadout, and hopefully some FTL-style dialogue encounters which could lead to bonus salvage or ambush.  That’s a fair amount of stuff, but I’m hoping I’ve got enough to build from…  and if I have to hack up my code a bit to try something, that’s okay, I can take any successes and work them into Auto Fire later, hopefully the “right way”. 🙂

So Many Datafiles

Yes, Technical Debt is still rearing its ugly head.  One of the things that any procedurally-generated Roguelike has is a ton of different files that hold profiles that define how to generate cities, landscapes and enemy encounters.  And tables, so many randomized tables!

sceneinspectorDuring the 7DRL I found an expedient solution that worked for the challenge and a fair amount of time afterwards.  I baked data right into each Unity scene that I saved out, imagining that I could just make a scene for each type of scenario or terrain profile I wanted.  I could bake in components that had all the predefined information I needed and just load them as needed.  I could even drag-n-drop the appropriate prefabs for everything I wanted to spawn.  How simple.  Sure, it nagged at me that it wasn’t super extensible, but scenes were cheap to make and I was interested in how far it could get me.

Wellll, it turned out it was pretty far, but eventually it started to haunt me.  The more scenes there were, the harder they all were to maintain, even if all the common information was kept in Unity prefabs.  Oh god, the prefabs…  they are great sometimes, but they also can puke all over themselves if I moved files around or a metafile got invalidated somehow.  Also, any time I wanted to choose something randomly, it felt like I was writing new code to deal with it each time.

I also used the serializer for a number of structures, but there was always a desire to have more flexibility when reading data.

Anyway, I knew I needed to up my datafile game.  My friend Jim’s amazing RL Dungeonmans has something like 500+ datafiles holding anything from name generation to encounters to tile definitions, with weighted randomization tables and tables that reference other tables.  How slick!  He spent many years refining his data methods and he encouraged us to reuse his approach in our own games.

So last weekend I finally bit the bullet and built a datafile system around some of the same concepts and in the end my format is virtually the same as Dmans.  This way I can build a sector with a pretty flexible format:

defThing sector_basic
{
 class adSectorData
 scene Overworld

 biome Desert
 nametable sector_name_chart
 treasuretable sector_treasure_table

 music mus_desperado

 basic_city_table 1d2
 sector_outpost_chart 2d4
 sector_town_chart 2+1d3
}

And these tables have some handy reference capabilities (recursing through each table referenced) and weighting for randomized results:

defTable "sector_name_chart"
{
 #t1 "sector_place_types"
 #t2 "place_nouns"
 #t3 "sector_adjectives"

 "The [t1] of [t2]" 10
 "The [t3] [t1]" 10
 "The [t1] of [t3] [t2]" 10
}

defTable "sector_adjectives"
{
 "New" 10
 "Old" 10
 "Dry" 10
 "Frosty" 10
 "Winding" 10
 "Hewn" 10
 "Locked" 10
 "Winding" 10
 "Ancient" 10
     ...

…and bingo, my world generation becomes 10x more flexible and powerful.  I’m dyin’ to get back to the drive-shoot stuff, but this was so worth it.

namegen