So its been a while since we updated this blog…yeah I know it seems like we say that every time.. Sorry about that..
Anyway during this period of absent updates we have been busily working away on a multitude of game related projects, from optimization to new content creation.

The docks level has had its poly count considerably reduced, without impacting the visuals I may add. Its also been adorned with voice over’s from the talented folk over at the digital bard  (no need to read those pesky subtitles .. Unless you want to of course). We’re also taking steps to up the fun factor of the level, so look forward to some interesting game play improvements in the not too distant future.

The cave troll has been receiving some much needed T.L.C. Unfortunately he seems to have intimacy issues, and since we can not afford therapy  you’d better prepare to engage with a much more formidable opponent. On top of his new found aggression, he’s been joined by an equally deadly companion. If you think this all sounds a bit unfair on Sintel, fear not as her combat system is currently being updated.

The bandit model is now game ready, you can check out a time-lapse of his transformation from high to low res below:

Geoffrey has been undergoing a complete revamp and many new characters are in various stages of development.
Below is a time-lapse of one of the new characters head being created:

Continuing our efforts to make Sintel the game look as good possible we are also experimenting with new material set ups, depth of field and even screen space ambient occlusions.
One of the new material effects can be seen below, its still a work in progress but we think it shows promise:

OK so I may have been stretching it a bit when I put real time SSS in the title. As  I’m sure you’ve realized it’s  a fake SSS effect, no subsurface scattering calculations going on. However it is real time and does not have any discernable effect on frame rate. (The slow frame rate in the video is due to the video capture)

Finally I know a lot of you  want to know when we will be releasing the demo. I’m afraid we still can not give you a firm date, as there are still many things we wish to add and improve.


As some of you may know we released a pre-demo version of the game back at the start of January for entry into the blender summer game contest. We are proud to announce that we came second in the contest and would like to thank every one who took the time to play the pre-demo and vote.

The demo itself was quite hastily put together. Although we have been working on the game for quite a while now, most of our time has been spent on research and design; meaning we had very little in the way of completed assets. As a result the main bulk of the pre-demo was created in the short time between Christmas and new years, culminating in a very long nights work which saw myself and Jon working right up until the deadline on new years eve.
Below are some in game screen shots and some production images from the pre-demo:

While we are reasonably happy with how the pre-demo went considering the timeframe in which it was put together, it’s no where near the standard we are aiming for. Many of the game play ideas which we are really excited about have not yet made an appearance.

The official demo release will include an updated and more polished version of the levels displayed in the pre-demo, along with a host of new features/characters/objectives. Which if all goes to plan will result in a longer play through time and most importantly a more fun gaming experience.

Its been a while since we posted anything on the blog, so Jon asked me to give you a quick overview of our modeling workflow, and kindly let me show you some of the stuff I’ve been working on.
Below is the first iteration of the bandit character (an enemy Sintel will have to face during the game).

Because the bandit was the first character I had made for the game I was not totally certain about the style I’d try to create him in. As a result he has been subject to quite a lot of change during his creation. Even as I write this, I have only just finished re-modeling  his head (below is a time-lapse of the process).

Below is just a quick mash up of the new head on the bandits body. Its only temporary but I wanted to check how well it fits. Seeing it altogether often gives a better read of the character; I’ve already seen a few things id like to go back and tweak .. But that’s a post for another day me thinks.

Heads in Sintel the game have some unique properties: some main characters like Geoffrey have their own unique mesh; however, many of the NPCs have interchangeable heads. We hope this system of “switching” heads and other accessories will enable us to fill the game with a variety of different looking characters whilst keeping unique asset creation to a minimum. To make head creation easier and to allow our system better flexibility, all NPC heads are created from these two base meshes, modeled by Noah.

The work flow used to create the in game characters varies from model to model and often includes a bit of back-and-forth between myself and Noah. Generally however I will start a character in blender by creating some very simple box meshes. From there I’ll either use the sculpting tools in Blender or export out to ZBrush. I use the sculpting tools to rough in the character and try to get a feel for the type of person/animal/monster its going to be. I often work with a variety of different reference images, from which I attempt to take the best bits.

The fact that I often don’t have a clear idea for the finished character is one reason why I like working with simple geometry. The lack of solid/defining topology allow lots of different ideas to be tested quickly and easily. Once I feel I’ve got the overall shape down pat ill move on to defining muscles and other secondary details. At this point I have to make a decision whether to continue working with the simple geometry, or to create some new geometry which better defines the form. Again the decision made here varies a lot, but as a rule of thumb I usually create new geometry if the object has clothes or a lot of separate accessories, and stick with the basic geometry if the character is relatively simple.

If I do create new geometry, I always try to keep individual parts as separate objects (the bandit for instance is split into 11 objects). This can be advantageous  for a variety of modeling tasks as well as normal map creation.

Once I’ve finished creating the character and the rest of the team have had a chance to look it over and give their comments and critiques, I send it over to Noah. This is the point where the in-game character is born. Noah builds geometry over my sculpt maintaining as much of the form as possible whilst keeping the poly count low. Once this re-topology step is complete we have both a high and low poly version of the asset. From this point, all that’s left to do is bake down all the tasty high res detail into the lowpoly mesh.

And, voilà, we have a completed in game character. Well, minus texturing, rigging, weight painting, animation, game logic….