Expect The Unexpected

I’ve mentioned our issues with game animations a few times in the past, and to be honest: I didn’t except animations to cause us that much work in the project. I’m glad we’ve found a professional animator in the team, but there’s still problems to solve before characters are done.

In fact, now we have finally managed to export animations in the project – but there’s another problem. Now the animations work when run alone. When we combine two different set of animations to one character, it won’t work. We are checking out the Max hierarchy and trying to find out what might be causing the problem when the game engine loads the additional animations. Since two files aren’t working, next we are going to put all the character animations in one file – and let the engine splice them properly.

There are unexpected moments in game production, and while it’s almost impossible to know what might unexpected might happen – it’s good to expect that something won’t work as planned.

Expect the unexpected, and when that happens – find the solution.

Juuso Hietalahti


  1. I agree that the production pipeline is a key factor. The higher iteration rates you get the better. One thing underestimated is the production experience that one needs to design such a pipeline. That’s why even in commercial products it’s sometimes not optimal. It’s even worse in some new indie products (at least now).

  2. From my experience I can tell you : this will not stop. The 3d art pipeline is a bugger, always. Even if you think you have it worked out properly, something happens and nothing works anymore. Prepare to reexport and reimport at last half the 3d art around 1 month before master.
    If you don’t have too, thank everybody you can think of.
    Oh and if you get into the critical phase, forbid all artists to upgrade the modeler, install any new plug-ins (or code some themselves) or in any way change anything at all in the software ;-).
    And as a warning to everybody out there who starts as an indie. Work out your art pipeline first. You have to find a artist – modeler – game engine combination that works. Before you even code one line ! Because it is easier to change the engine, than to find artists who can work (well) with an alien modeler. And it is easier than finding a modeler software, which can work with your engine. There are 2.5 professional modelers out there Maya, Max and XSI. The later has smaller fan base and is less well connected. And almost only one cheap alternative, which is Blender, one of the most-hated pieces of well-used software. Yes there are others. But anything else and you have to read the small print very carefully to find out how to connect this to a more or less tested engine with a broad user base. You need the broad user base of the engine, because this helps to assure you that is is relatively bug free and works.
    Test the setup by getting a textured cube from the modeler into the engine (if you need 3 days forget it right there and then). If this works, try out animation. This should be part of your decision process for an engine.
    Oh and if you want to have a successful indie game company, do not try at all cost to reinvent the wheel and write the graphic engine for yourself, except you know exactly what you do. The art pipeline only gets worse.

  3. Maybe you should write a postscript to your article on “differences between 2d and 3d game production” :)

    I’m just hassling, I’m sure you’ll get to the bottom of it. I hate bugs sometimes, especially ones that take more than half a day to fix.

Comments are closed.