Connecting movement data to the Maassilo brand turned out to be harder than expected. I explored several directions, including the idea of text that compresses like a silo when more movement is detected, but most of these concepts were difficult to realise with the sensor I had. A small test confirmed another limitation: to use the sensor at full capacity and control the scaling of shapes in the way I had imagined, an older Windows computer would be required, which I did not have access to. This meant I had to work with what was possible rather than what I had originally envisioned.

Because tutorials for getting the Kinect working on my setup were scarce, I relied on AI assistance to help get things running. Most tutorials were built around older Windows versions, so ChatGPT helped me work around those limitations. I started by testing how the 3D scanner read depth, representing each depth point as a small block to visualise differences in space. From a tutorial, I learned how to convert depth data into a threshold value, similar to calculating brightness as I had done in the boundary analytics style, which made it straightforward to map depth onto shapes.

Using this approach, I designed a quick logo for the Maassilo so I could focus on testing rather than spending too much time on identity design. I created four sets of the logo in contrasting colour combinations and built a system where different keys switch the background colour, producing a dynamic, high-contrast result.