Experimental Approach of the Asymptotic Computational Complexity of Shaders for Mobile Devices with OpenGL ES
Keywords:Android, iOS, shaders, mobile devices, computer graphics, asymptotic complexity.
The usage of mobile devices and increasingly realistic graphics is emerging, but the graphics performance is still a critical factor in games. There’s more hardware restriction on mobile devices than on a computer. Thus, this paper proposes an experimental approximation of the asymptotic computational complexity of miscellaneous vertex and fragment shaders for Android and iOS platforms. The asymptotic complexities of the shaders will be analyzed based on number of instructions per second and rendering time metrics, depending on the number of polygons rendered. By means of the adjusted curves is also possible to compare the performance of the devices used in this work, which are the Nexus 4, HTC One, iPhone 5s and iPad Air. Besides, an automatic tool – that plots the data and uses the method of least squares to adjust the values obtained – will be presented, being able to estimate which curve has better approximation to the sampled data.
LicenseAuthors who publish with this journal agree to the following terms:
- Authors maintain the author's rights, under de Brazilian legal system, and retain copyright of the work. Authors license the work under a Creative Commons Attribution License. The authors grant the SBC Journal on Interactive Systems the right of its long term archiving and publication under this same licence, without restrictions of time and media, as well to make the required transformations to maintain the work archivable, accessible and of open access.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See The Effect of Open Access).