Power Consumption Model of a Mobile GPU Based on Rendering Complexity

Vatjus-Anttila, Jarkko; Koskela, Timo; Hickey, Seamus

doi:10.1109/ngmast.2013.45

Cited by 15 publications

(11 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Similar to previous work [VAKH13], we observe that GPU power consumption follows an inverted exponential function between a minimum P m and a maximum P M power, as the rendering load increases. Given a rendering configuration s and camera parameters c , we thus propose the following power consumption model:

…”

Section: Power Prediction Modelsupporting

confidence: 87%

“…Instead, we aim at predicting power consumption using only rendering information, in order to obtain a model directly related to scene complexity. Vatjus‐Anttila et al [VAKH13] proposed a model for GPU power consumption taking into account the contributions of three different primitives separately (batches, triangles, and texels), and combining them as a weighted sum. Different from this approach, our model includes render passes, takes into account all primitives simultaneously, includes the number of fragment shader invocations instead of texels as a better predictor of power consumption, and adapts in real‐time to changes in the scene.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

On‐the‐Fly Power‐Aware Rendering

Zhang

Ortín

Arellano

et al. 2018

Computer Graphics Forum

View full text Add to dashboard Cite

Power saving is a prevailing concern in desktop computers and, especially, in battery‐powered devices such as mobile phones. This is generating a growing demand for power‐aware graphics applications that can extend battery life, while preserving good quality. In this paper, we address this issue by presenting a real‐time power‐efficient rendering framework, able to dynamically select the rendering configuration with the best quality within a given power budget. Different from the current state of the art, our method does not require precomputation of the whole camera‐view space, nor Pareto curves to explore the vast power‐error space; as such, it can also handle dynamic scenes. Our algorithm is based on two key components: our novel power prediction model, and our runtime quality error estimation mechanism. These components allow us to search for the optimal rendering configuration at runtime, being transparent to the user. We demonstrate the performance of our framework on two different platforms: a desktop computer, and a mobile device. In both cases, we produce results close to the maximum quality, while achieving significant power savings.

show abstract

…”

Section: Power Prediction Modelsupporting

confidence: 87%

Section: Related Workmentioning

confidence: 99%

On‐the‐Fly Power‐Aware Rendering

Zhang

Ortín

Arellano

et al. 2018

Computer Graphics Forum

View full text Add to dashboard Cite

show abstract

“…e design objective is dded GPUs which bove energy models ters to estimate the reted by application [1] built an energy cteristics: number of heir proposed energy n of the graphic ically deducted 45% esis that 50% of the ue to the back-face Mochocki et al [10] rent stages of the 3D on, frame rate, level fect the 3D pipelines ystematically design s between high-level . Then, we build a t only requires highr real-time graphic 1] mostly focus on e events by profiling A set of important ache misses, shader they are trained by a (*1 in Fig.…”

Section: Methodsmentioning

confidence: 99%

“…Thus the result is linear as expected to the number of vertices with the slope reflecting to the complexity of the shader program. The time spent on vertex shading ܶ ௩௧௫ is shown in equation (1), where ݊ ௩௧௦ is the number of vertices of the mesh data and ‫ܥ‬ ௩௧௫̴௦ௗ is the complexity of the given vertex shader program.…”

Section: A1 Vertex Shadermentioning

confidence: 99%

High-level energy consumption model of embedded graphic processors

Huang

Chung

Huang

et al. 2015

2015 IEEE International Conference on Digital Signal Processing (DSP)

View full text Add to dashboard Cite

Embedded graphic processing unit (GPU) is an indispensable component in enabling real-time rendering and graphic applications on mobile devices. However, embedded GPU consumes a considerable energy [1] which is critical for battery-operated devices. To understand the energy consumption of a graphic application, conventional approaches suggested the energy model based on hardware performance counters. However, those low-level energy models are mainly derived from GPUs of desktop computers, and they cannot be applied to embedded GPUs directly and the low-level models are less intuitive from a programmer's point of view. In this study, we consider a high-level energy consumption model for embedded graphic processors, and then we can estimate energy consumption of a graphic application based on graphic attributes of a scene. We conduct a number of experiments on real platforms to validate the proposed model. Our experimental results demonstrate that an average energy estimation error rate of 7.30% can be achieved.

show abstract

“…In addition, compressed textures also require less RAM memory, and consequently, require a smaller number of memory accesses in the mobile chipset. According to Vatjus-Anttila et al [86], use of compressed textures typically results in four-six times lesser use of RAM memory. They concluded that from the standpoint of the mobile chipset, the use of JPEGs (i.e., RGB data) resulted approximately in 5-15 % higher energy consumption than the use of compressed textures.…”

Section: Data Type Transformationmentioning

confidence: 99%

Optimization Techniques for 3D Graphics Deployment on Mobile Devices

Koskela

Vatjus-Anttila

2015

3D Res

Self Cite

View full text Add to dashboard Cite

3D Internet technologies are becoming essential enablers in many application areas including games, education, collaboration, navigation and social networking. The use of 3D Internet applications with mobile devices provides location-independent access and richer use context, but also performance issues. Therefore, one of the important challenges facing 3D Internet applications is the deployment of 3D graphics on mobile devices. In this article, we present an extensive survey on optimization techniques for 3D graphics deployment on mobile devices and qualitatively analyze the applicability of each technique from the standpoints of visual quality, performance and energy consumption. The analysis focuses on optimization techniques related to data-driven 3D graphics deployment, because it supports off-line use, multiuser interaction, user-created 3D graphics and creation of arbitrary 3D graphics. The outcome of the analysis facilitates the development and deployment of 3D Internet applications on mobile devices and provides guidelines for future research. Graphical Abstract

show abstract

Power Consumption Model of a Mobile GPU Based on Rendering Complexity

Cited by 15 publications

References 7 publications

On‐the‐Fly Power‐Aware Rendering

On‐the‐Fly Power‐Aware Rendering

High-level energy consumption model of embedded graphic processors

Optimization Techniques for 3D Graphics Deployment on Mobile Devices

Contact Info

Product

Resources

About