The Eldritch Zookeeper is a game where you play a zookeeper, cursed to run a zoo full of inter-dimensional monsters. It's a cross between those old Bullfrog games, like Theme Hospital and Theme Park World, and more modern survival-type games informed by procedural generation. Players build fences, open boxes containing monsters, and try and stop visitors from being eaten by them. I'm sure that will always go well.

The best explanation is to watch the newest devlog below:

Video Devlogs:

• Devlog 5 - Take a Monster Home Today!
• Devlog 4 - New Graphics Everywhere
  
• Devlog 3 - Monsters Have Feelings Too
• Devlog 2 - Navmeshes and Pathing
• Devlog 1 - Angry Monsters

Screenshots:

Been working on all sorts over the last couple of weeks. Building a new fence now happens in a fancy 'blueprint' mode:

Also animated another monster called the Gastroquin, which hops around like so:

Monsters have their own boxes (so you can tell what is inside a box now) and they pop out with a little particle effect:

Been working on some of the game interface work, and inventory management:

New video devlog! I talk a bit about the four monsters that are currently in the game and that have their basic animations. Watch!

And I did another devlog! Lots of fun changes lately.

And... another devlog. Was thinking through how best to implement a shop into the game, allowing the player to buy items and upgrades, and this is my working solution so far.

"Run a cursed zoo, badly." is kind of the best game pitch ever. I love zoo simulators, occult creatures, and comedy.

Thanks Space Mace!

And I've done yet another video devlog. This one features explosions, stampedes, death. Hooray!

And another devlog! Two in one month.

My last one involved explosions and visitors dying, and in this one it... doesn't get any better for the poor humans. Just a lot of burning. The visitors aren't terribly clever and wander into fiery danger quite frequently, so perhaps it's their fault?

The fire visuals need some more work, most obvious is that the fire doesn't cast any light, and so even when the whole zoo is ablaze, it all looks quite flat. I think I need to concentrate on a more substantial update to the lighting in the game for the next update.

Happy New Year everyone, let's celebrate with another devlog, as we have done for millenia.

Stripping out and throwing away a monster design and redoing it completely took a total of four days, which was pretty good. Makes me far less scared to do this sort of rework in the future. Though of course, I hope I don't have to.

Ice Cream! Shops! A new game dev video:

Larger buildables like shops have been coming for a while, I think the first related lines of code I wrote back January, but it's taken a while to flesh out to the point where there isn't just a grey block sitting in the world. It's a complex piece of work! Take queuing, visitors have to line up, and move forward in the progression, but visitors can get pushed out and have to get back in line - it's pretty hard to not have would-be shop patrons fight over a single queue position, and lots of tweaking is required to resolve those sorts of conflicts.

A stretching piece of work like this really deepens the whole building side of the zoo management, and it's really easy to extend to however many other structures (drinks, umbrellas, zoo maps, toilets) that have queues I need, they just need different models and textures. The hard behind the scenes work is done, in other words.

Happy to answer questions!

Another game log!

One thing I haven't been able to do for a while is test how the game works at scale, so I've put some time into making sure performance is at least 'acceptable' with ~20 monsters and ~150 visitors. My current definition of acceptable is above 30fps, and I'm at around 40 with that sort of load. That's improvable, I think, especially with all the work Unity themselves are doing recently with the Job System, ECS and the Burst Compiler.

If you're interested in performant code in Unity, definitely take some time and read up on that, have a go at a few experiments. The API isn't quite finished yet, but it's quite exciting to think that Unity as a whole is going to be moving towards a faster, data-driven model.