Prototyping is hard

I’m building Syrthe with Dominique and we are spending a lot of time deciding the size and placement of all the needed structures.

Here is where we are at the moment:

Not very fancy, isn’t it?

And yet, this took a lot of time.
The ugly blocks at the bottom of the image are the bailiff castle and the cathedral, how do you get the sizes right?
Well, we’ve been timing how long it takes to run from one end of a building to the other, compared to real life structures, thinking about how pretty it will have to look and how playable it will have to be.
Running from the castle to the bridge takes over 2 minutes in a straight line: not exactly the kind of thing you can ask a player to do repeatedly, he’ll get bored pretty fast and will soon hate the game.

There is still a long way to go, the finished city will pretty much look like this:

Syrthe is taking shape

Compared to Neverwinter Night’s Aurora, designing a map under UE4 is incredibly hard. Dominique and I have been planning the city, sketching the layout on paper, sculpting the landscape to place the river, the central hill, the palace, the cathedral, the industrial district, the various neighborhoods, Emer’s tower, the cemetery, the bridge, the walls, the marketplaces and much more, but there is a long way from envisioning a city and actually laying down all the elements in a pretty, realistic and playable way. A very long way!

Fortunately, UE4 is a blast, it’s unbelievably complex and, at the same time, powerful.
Everything is coming along nicely and in a few days, or weeks, I have no clue at this point, it could be months, we’ll have a stunning photo-realistic medieval city as a starting point for our adventures. I already own many UE4 packages but still have to design lots of objects, including all the cathedral elements, I got nothing at the moment and will have to start from scratch. That will be cool though, it’s only a lot of work.

Development will go faster and faster.
I spent hundreds of hours learning the ropes of UE4, now it’s time to put all of that to good use. The beginnings are tedious but the more I’ll be using the tool, the more confident I will grow and the faster I’ll design the game.

I now have a pretty decent knowledge about pretty much every technical aspect of development except characters.
My knowledge encompasses dozens of expertise areas from simple mesh design up to programming pretty much any functionality I want, but characters and armor are still a big question mark. I know how to sculpt a creature, but fear it will be ugly, I can texture it, and again fear it will look like a sack of shit, I can rig it, and fear the rig will not work properly, I can animate it, but fear the animations will suck and, most importantly, I dread the moment I will have to string all the pieces together.
Anyway, that’s not for this year for sure, so my goal is to get the most beautiful city I can dream of and go on from there.

Behavior trees

I’m now setting up the first dialog, using Unreal’s behavior tree, which I have yet to discover. Some people like it, others hate it, at first glance it looks like an intuitive decision tree which could be used for conversations as well as AI.

This is going on the slow side of things though, I’m on holiday and spend some time on WoW with the family.

Ah! rigging…

I must confess I’ve been hit pretty hard on the head as of late, animations are a can of worms!
I thought I just had to put bones into my mesh to create the armature – or skeleton, or rig, all the same – and use any kind of pose editor to define all my poses, idle, running, attacking, dodging and so on.

Not so fast, young Padawan!
The easiest way to build a rig is to connect all the bones of an armature, which works fine but each bone has to be moved individually to create any pose.
That’s why IK – inverse kinematics – was invented which, in a nutshell, is a way to tell which bones move in conjunction with others.
It’s a fantastic technique, once an IK rig is correctly built, moving any part of the rig will move all other parts in a natural way.

The bad part is that I have – once more – to learn a new job to get all of this working.
The good part is that, as far as I can tell, animation is the last piece of the puzzle and once I am able to sculpt, texture, rig and animate a creature, I will have all the techniques needed to start building content.