Proteins are biomolecules present in living organisms and essential for carrying out vital functions. Inherent to
their functioning is folding into different spatial conformations, and to understand these processes, it is crucial
to visually explore the structural changes. In recent years, significant advancements in experimental techniques
and novel algorithms for post-processing of protein data have routinely revealed static and dynamic structures of
increasing sizes. In turn, interactive visualization of the systems and their transitions became more challenging.
Therefore, much research for the efficient display of protein dynamics has been done, with the focus being
space filling models, but for the important class of abstract ribbon or cartoon representations, there exist only
few methods for an efficient rendering. Yet, these models are of high interest to scientists, as they provide a
compact and concise description of the structure elements along the protein main chain.
In this work, a method was developed to speed up ribbon and cartoon visualizations. Separating two phases
in the calculation of geometry allows to offload computational work from the CPU to the GPU. The first phase
consists of computing a smooth curve along the protein's main chain on the CPU. In the second phase, conducted
independently by the GPU, vertices along that curve are moved to set up the final geometrical representation of