Enabling Games for Power

by Rajshree Chabukswar and Jun De Vega

Abstract

PC gaming has traditionally been associated with desktop systems due to the performance requirements of game applications. As new laptop capabilities become comparable to desktops, some gamers began to consider these as gaming platforms. As the gaming industry adapts to mobile computing, resources like battery life become significantly important for system usability. To determine the most power efficient configuration, this paper analyzes the power consumption of laptops while changing different game applications settings like graphics, resolution, FPS, and more. The study focuses on how the different game settings can provide a more efficient power configuration with a battery-powered laptop.

Motivation/Goal

The goal of this paper is to determine different game settings that will provide some power savings for a notebook. This software exploration may provide opportunities for power optimization.

Power Measurement Methodology

Measuring power usage of individual components in a mobile platform is not a trivial task. Various tools exist to provide a high level estimate of the power consumed by a mobile platform. While these tools may be sufficient in understanding the high level power consumption of particular platform, they do not provide fine grained details on specific components. A more accurate although more invasive method of power measurement is to use data acquisition (DAQ) tools where specific hardware components are instrumented, and a more granular power measurement can be logged. The following setup lists the platform details used for this analysis along with the power measurement methodology.

Figure 1: Test Setup

Test Setup

The Target PC (Napa/Yonah) has a special motherboard with built-in sense resistors. For each target component (i.e. CPU) all sense resistors are wired and soldered at both ends before being connected to a module attached to the NetDAQ unit.
The NetDAQ has modules that are attached with individual wires to the Target PC and measures the current and voltage drop across the sense resistors. The NetDAQ is connected to the Host PC via a cross-over network cable as shown in Figure 1.
The Host PC can be any IA32 system with Windows XP and the NetDAQ logger. The logger collects the measured current and voltages and calculates the average power (W). The sampling interval we used for our entire analysis was 25ms. The platform power measurement does not include the LCD.

SETUP 1:

Hardware

Fluke NetDAQ® 2686A
Target PC: Intel® Core Duo/2GHz Yonah, Jamison Canyon CRB, 2x512GB DDR2, 40GB SATA 5400 rpm (2.5” mobile), nVidia 7800, CD/DVD drive, WinXP Pro SP2.
Host PC: Any IA32 system

Software

Test Applications (different applications used)
NetDAQ Logger: Fluke* DAQ Software v2.2

SETUP 2:

Hardware

Sony* VAIO, Centrino Core Duo 2.0GHz,
Extech* Power Analyzer

Software

Extech Power logger
3 games applications (2 first-person shooter and 1 real-time strategy). The first person shooter games typically run on high frame rate; while real time strategy games don’t scale the frame rate up.

Power Consumption Profile

This section shows the impact of various LCD brightness levels on platform level power consumption with an idle system.

Figure 2: Idle system LCD brightness

As indicated in Figure 2, a Sony laptop has ~5W (32%) power reduction when using low brightness (0 of 8 bars) as compared to high brightness (8 of 8 bars). The data indicates that reducing screen brightness can help save power. The power saving percentage may change when the system is busy running workloads.

This next section examines power consumption data for individual games. The next section describes power consumption data when running each game on the following two configurations:

Intel Core Duo Sony VAIO Laptop with nVidia 7400 graphics card
Intel Core Duo Engineering Sample with nVidia 7800 graphics card

The game settings that are described here include the following:

Impact of game resolution on power consumption: The games support various resolution based on system configuration and game design
Impact of game quality settings on power consumption: These options include ‘Textures’, ‘Shaders’, ‘Shadow Effect’, ‘Smoke Effect’, ‘Water Effect’ etc. For example, ‘Shaders’ provide 3 quality settings high, medium and low. Shaders usually determine final surface properties of 3D object. It can include details on lighting effect, shadows, reflections etc and how these things would impact the object when displaying to user. The details on how much pre-processing is needed (finer level details) can be scaled by selecting high, medium, low shaders.
Impact of various frame rate on power consumption: This study demonstrates impact of power consumption by capping frame rate to lower values.

Impact of game resolution on power consumption

Figure 3 shows the effect of power consumption on the Sony laptop with an nVidia 7400 graphics card. The data here represents power scaling, hence lower is better. For example, the blue bar (for game A and B) show baseline/default power consumption data scaled to 1.0. The other bars are scaled accordingly. For game A, with a resolution of 800x600, the power consumption is ~5% less as compared to baseline; when using the 640x480 resolution, the power consumption is ~8% less as compared to baseline.

Figure 3: Game resolution and average power comparison on a Sony laptop

Figure 4 shows the impact of resolution on power consumption with a high end graphics card. This analysis is done on an Intel Core Duo system with an nVidia 7800 graphics card. As indicated in the graph, going to a lower resolution saves power compared to running at a higher resolution. Also note that due to system configuration, the game B now has only 2 options for resolution selection.

Figure 4: Game resolution and average power comparison on an engineering system

Impact of quality settings on power consumption

Figure 5 demonstrates the power consumption impact of changing the game quality settings on a Sony VAIO laptop with the nVidia 7400 graphics card. The blue bar represents baseline/default power consumption data which is scaled to 1.0. The baseline quality settings change with system configuration and graphics card. The 2nd bar represents settings all the quality options to low. As indicated in the graph, setting quality options to low saves power across all 3 games.

Figure 5: Video options and average power consumption on a Sony laptop

Similarly, on Intel Core Duo engineering sample with an nVidia 7800 graphics card, there is some power savings when switching to lower quality settings as shown in Figure 6.

Figure 6: Video options and average power consumption on an engineering system

Impact of various frame rate on power consumption

In this study, platform power consumption was analyzed with various frame rates. On a Sony VAIO laptop, Game A runs at 20 frames per seconds as baseline (blue bar), the maroon bar represents 15 FPS and beige bar presents running at 10FPS. Similarly for game B, the baseline is 60 FPS (blue bar), maroon bar represents 30FPS and beige bar represents running at 20 FPS. As indicated in Figure 7, the platform power consumption is reduced when capping the frame rate at lower value.

Figure 7: Frame rate and average power consumption on a Sony laptop

Figure 8 represents a similar study done on Intel Core Duo engineering sample with nVidia 7800 card. Here, Game A runs at a much higher frame rate as compared to the Sony VAIO due to availability of high end graphics card. The game A runs at un-capped frame rate as baseline (blue bar), and the maroon bar represents 60 FPS and 30 FPS. Game B runs at 60FPS (blue bar), the maroon bar represents 30FPS, and the beige bar represents running at 20FPS. As indicated in the graph below, a considerable amount of power is saved when running at a lower frame rate.

Figure 8: Frame rates and average power consumption on an engineering system

Summary and Recommendations

Enabling for power is different from traditional performance enabling since software behavior is potentially altered via different execution paths when optimizing for power. Intel has a product titled 'Intel® Laptop Gaming Technology Development Kit (TDK)' which offers functionality to detect current platform status such as when a power source is changed from AC to Battery and when network connectivity is lost. Games can use the APIs provided by the TDK to detect the current system status or track changes to system status and alter game execution in order to save power when limited resources are available.

When running on battery power, here is a list of few features to consider for longer battery life and play time.

Reduce display brightness
Turn off a wireless card if not in use
Reduce the frame rate by capping it to lower value
Reduce game resolution to lower value
Reduce game quality options to lower values

Game developers may choose to offer various modes during the game play

Performance optimized: offering high performance as primary factor when playing game
Power optimized: offering to scale down some of the features as discussed above to enable for longer play time.

Users can then select the most appropriate mode based on their needs.

About the Authors

Rajshree Chabukswar is an Application Engineer working on client enabling. (Mobile Application Enabling group) Prior to working at Intel, she obtained an MS in Computer Engineering at Syracuse University, NY.

Jun De Vega is an Application Engineer in Intel’s Software and Solutions Group, working on application tuning and optimization for Intel® architecture. He supports enabling of ISV applications on Intel® Mobile and Desktop Platforms. Contact him at rodolfo.de.vega@intel.com.