Proceedings: GI 2016

Usability and Performance of Mouse-based Rotation Controllers

Steven Rybicki (University of Cape Town), Brian DeRenzi (University of Cape Town), James Gain (University of Cape Town)

Proceedings of Graphics Interface 2016: Victoria, British Columbia, Canada, 1-3 June 2016, 93-100

DOI 10.20380/GI2016.12

  • Bibtex

    author = {Rybicki, Steven and DeRenzi, Brian and Gain, James},
    title = {Usability and Performance of Mouse-based Rotation Controllers},
    booktitle = {Proceedings of Graphics Interface 2016},
    series = {GI 2016},
    year = {2016},
    issn = {0713-5424},
    isbn = {978-0-9947868-1-4},
    location = {Victoria, British Columbia, Canada},
    pages = {93--100},
    numpages = {8},
    doi = {10.20380/GI2016.12},
    publisher = {Canadian Human-Computer Communications Society / Soci{\'e}t{\'e} canadienne du dialogue humain-machine},
    keywords = {3D User Interfaces, 3D Rotation, 3D object manipulation, Arcball, Discrete sliders, Two-Axis Valuator, Usability},


Rotation controllers are used to interactively orient models in many important applications in 3D computer graphics and visualisation. Unfortunately, previous studies do not provide clear guidance on which rotation controller to use in a particular situation, either because they assess performance measures and rotation tasks in relative isolation or because they do not achieve statistical significance. In this paper, we present the results of a broad quantitative user experiment (n = 46) to compare the three most prevalent rotation controllers (Arcball, Two-Axis Valuator, and Discrete Sliders) according to both speed and accuracy across two classes of tasks (orientation matching and inspection). While we found no significant differences between Arcball and Two-Axis Valuator, Discrete Sliders were found to be significantly more accurate for simple orienting tasks (a medium to large effect), but slower across all tasks (a small to medium effect, median approximately two seconds). Thus, a Discrete Sliders controller is better suited to situations where finegrained accuracy is valued over speed and in other instances, e.g., inspection, either an Arcball or Two-Axis Valuator is appropriate.