After realising that we were better to pursue suspending the shelter on a single wall, we tried again, but we learned two things. We couldn’t understand why the structure seemed more unwilling than usual to create the form we wanted, and realised that this was actually the first time we had tested the suspension with ALL of the inserts in and attached. This was compounded by the fact that we were trying to add the 2 side tetrahedrons to force it to curve. However, the inserts had automatically created a weakness within the modules and it was performing differently to when it had originally been lined with solid cardboard and the addition of the two extra tetrahedrons was applying too much pressure on the shelter. We agreed that the most expedient and logical solution at this point (with crit only days away) was to remove D1 and D2 and lighten the load on the inserts and overall shelter. This design was constantly evolving through the making and testing but we knew that being open minded about it was the only way to be with a project like this!