So when we finally got a working robot back from Aldebaran’s Nao Clinic, we put it on the field and set it off walking. To our horror it immediately started shaking as it walked. At first we were very confused; some were thinking, “What have we done to our code?” Others were convinced, “Aldebaran did it!” But after some quick timing of our processes, we found that the DCM was taking an unusual amount of time, 35 milliseconds to be exact. And that’s very significant if we want to be running at 33 milliseconds per image we get from our camera. The DCM essentially lets us control the robot’s body. It turns out that because of the DCM, we were instead running at around 75 milliseconds per frame. After some digging in the code, Jack found a few commands that told the DCM to wait for some number of milliseconds. And it happened to be 5, 10, 20 (5 + 10 + 20 = 35) hmmmm…. Thanks Joho…. On the plus side, we’re pretty sure we are the only team here that has doubled their frame rate in 15 minutes. Good start we think.