Over at Bio Break there is a post looking at procedural generation and the reasons why the technology, and implementation doesn’t appeal to him. My comment looked like it was going to get a bit to big so…. posting here for prosperity.
Honestly I do see where he’s coming from. What I enjoy about many of these theme parks I have played lately is that they are all rather well constructed. You have defined zones and characters, stories are well placed and it leads you through the game at a reasonable pace. There is always the next goal nearby to complete, the next quest hub near the last quest. World design that leads you on but also showcases certain special areas and activities. These are a game type that is run by a defined vision, guided by metrics and it is the hand-crafted aspect to this that makes it all work.
Procedural generation just wouldn’t work within this design. The secret world, one of my favourite narrative experiences works only because of this detailed and polished developer crafted world. Many others are like this too. Adding the randomness of procedural generation to these mmo’s would most likely break the experience, goals and the design are what keep us progressing.
I do think there is a place for procedural generation though. In the place of Crowfall I think this design works perfectly. You have many maps of varying rule-sets that each have their own individual world make-up which would make it that much more exciting to explore alternative worlds. I love exploration, just wandering around the world and seeing what I can find – thumping was my favourite activity in Firefall after all and the promise of having many worlds to explore for varying resources. To learn the terrain and objective placement is rather enticing to me. The fact that these constantly reset as well means an endless reason for me to do this activity I enjoy as resources will need to be found again, enemy encampments scouted and then, just random exploration to see the beauty of the world.
Having the same prepackaged set-up would make these multiple month-long campaigns rather boring after a time. Case in point, GW2. Having the same map/s will only last you so long. Eventually everyone knows the complete layout in their sleep, the strategies to engage in at every turn. The meta gets stale, and you don’t want that. Having it change after a time means having an evolving strategy over time, a campaign won’t just involve exploration of terrain but forming new ideas, new plans, new strategies for conflict. It creates a far more dynamic experience.
What it will also allow is creating worlds that are far bigger, and this I think is a perfect fit for the next generation of sandbox mmo’s. Land mass is an extremely limiting factor on a game, for the regular themepark it doesn’t matter too much but for sandbox design, and player driven worlds having the amount of land creates far more options to players and strengthens certain mechanics. Terrain and travel time begin to matter, allowing players to specialise in it.You can make the impact of players that much greater as well, impact on resources since there will be more around, the ability to make their mark with housing and such, and just controlling land becoming a cost benefit analysis… you can only survey and control so much a ta a time. There are a lot of benefits to having space.
The technology has come a long way now too. It isn’t just the boring terrain and basic placement you might expect. Their can be incredibly complex algorithms at work on all aspects of the game. The elevation of terrain and grading, the styles and texture of the land. how it is populated with assets like trees, rocks and such. Sometimes I think these programs do a better job of creating interesting but far more realistic spaces as well. Just looking at games like minecraft and terraria you have huge differentiation in the lay of the land but a realistic scope to the areas generated. There are some complex systems too like the interconnected tunnel networks that get constructed. It isn’t just terrain either as this can extend to the way it places pre built assets like walls, buildings, towns, as well as creatures. I believe this system is the main reason behind these titles success, it adds inherent replayability.
We aren’t quite there for the voxel tech just yet but I am incredibly interested to see how far it develops and the ways it becomes integrated into our mmo’s
9 thoughts on “The Procedural Phenomenom”
Comments are closed.
Healing the masses 
Sometimes Syp strikes me as a bit hysterical. Procedural generation of game assets is a tool, nothing more.
It’s not a bloody ethos or an artistic expression or even a sodding manifesto. It’s a goddamn tool.
And like any tool it can be wielded for good or ill.
It is indeed a tool and one you can rely on a little or a lot.
Also meant to say that, like a lot of things in game development it more depends on what you want to create as to how you use it
I gotta say, im super interested in Crowfall. I have major hopes for it like I haven’t for other MMOs. I think they’ve brought some of the right ideas together and I lvoe that the worlds only last for a few months. THATS REALLY AWESOME.
As for procedural world generation, I’ve seen it work well in a couple of games. Mount and Blade is one of them. It literally almost makes no difference for the main gameplay, but it adds a nice touch to a fresh campaign.
I definitely like the pitch too, and the world’s resetting and being remade is kind of like that table flip moment in monopoly. Everyone wants it to happen
Enjoying your website.
“Procedural generation just wouldn’t work within this design. The secret world, one of my favourite narrative experiences works only because of this detailed and polished developer crafted world. Many others are like this too. Adding the randomness of procedural generation to these mmo’s would most likely break the experience, goals and the design are what keep us progressing.”
As you agreed above, its nothing more than a tool. I don’t see why using it extensively and building on top of the structure that comes out of it isn’t reasonable. Imagine if I drew a map at random for you and I asked you to tell me who lived there. Now that would be a fun way to co-op a story!