Thursday, May 24, 2012

Evolution of Slarty's Spawning Technique

Warning: This post may not be as sexy as the title implies.

I've been wanting to write a bit about my creature spawning strategies for a while, since I've put a lot of thought into it over the past few years. This is one of those things that can really ruin the magic, so I've held back. I'm hoping to go light on the specifics, and just focus on where I've been, what I've learned, and where I'm going.

The first spawns I ever created were for Metzetli. I started with the initial jungle area, with a goal to make "South Forest for 6th circlers". The spawns took heavy influence from South Forest, which contains a very varied spawner, with chances of some extremely difficult challenges (for the level). I added my own tweaks to this style by overlapping very quick spawners that had about a 90% chance of junk and a 10% chance of harder things, with slower spawners that focused on the beefy stuff. The element of danger is always there, with very quick spawns which usually result in 3 vultures, but could sometimes result in 40 rudremau. This got rid of the problem I saw with most hunting areas, where the retreat was either always dangerous (because the tough monsters spawned too fast), or always a cakewalk (because the tough monsters spawned way too slowly). Now it's usually fine, but you never really know what you'll find.

I used something like this style for a while, but I was really starting to question the idea of "junk" creatures. The idea was it allows weaker characters to tag along and still contribute, but that really doesn't scale. What is considered "junk" when you're fighting sazajas? Note the comical grey wyrm spawns in Tepui, that was a junk spawner, it didn't make sense, and it wasn't really fun. The Mountain Glen was a large scale experiment in having a lot less junk, and it was the first (large) area to get rid of my patented "dual spawner" system. This reduces the unpredictability, and lets me focus on a more consistent and targeted challenge. It was pretty successful, I think people enjoyed it a lot. Looking back I think Metzetli's spawns are still more fun than Mountain Glen, partially because I spent a lot more time on Metzetli spawners, but part of the problem was Mountain Glen was a halfway step. There was still junk, but there was less of it. And it largely used the spawning strategies of older areas which has a mix of "always junk spawners" and "always good spawners", which can make things predicatable and boring.

Still, I deemed the experiment a success, and moved even more extremely in that direction, practically eliminating the idea of junk spawns. I try to keep variance by creating lots of different spawner types with slowly increasing difficulty, then you mix the rate and amount of each to keep things interesting. The Lily Pond expansion was the first full on experiment with this style, and then Estuary after that was the biggest high level experiment. You can actually see the spawn strategy evolve as you go deeper into the Estuary, as I learned with each release. I still feel my newer areas are vastly more predictable than my older ones, which saddens me a little. I also think retreats spawn a up with tough monsters a bit too fast with this new style. It makes recovering from a tough situation too difficult, and too reliant on long chain runs.


  1. nice to get some of your thoughts into that.
    Well maybe you might have a look to the map of Arindal to get a clue how tough areas could be designed as well?
    In Arindal we had a lot of saves in different areas. In CL there are way less and also some "semi-safes" like in Glen-Spiders and Middle-Estuary (PF Entry with sometimes Monsters in it). There are also some huts that are partially save and partially not (Fenge) in the snells between the river and the 100c boat...
    But there are other areas (Tepui, Rainforest, Sarra) where there are no real retreat-spots especially not when talking about non-pathfinders. Also Paths into other areas could help there like they do already in some areas...
    Maybe it would be cool if some huts or caves can be built somewhere with the donation of Stones and Wood...

  2. I actually tend to hate safe areas, which is why most modern Clan Lord areas don't have many. With safe areas, I see a lot of situations that would be tense and interesting turn into a series of "fight a bit then retreat to a safe area". You'll find on Metzetli (the first experiment in the "no safe area" design) there are actually lots of good places to take breaks, if you're clever, just not many guaranteed places to retreat to.