How did LBNL get interested in this?
One of our sons--Nathaniel Mills--wanted to build a gaming PC. When we saw the potential power requirements we turned it into a scoping study that eventually grew into the current project. He is first author on the original journal article. Energy Wire wrote a nice account of the project's inception.
Who is funding this work and does the industry have input?
What is interesting about this work?
It's a rare example of a specific end-use domain that has been unstudied. There is potential to make a big difference. While the energy use of conventional PCs has been falling for some time (thanks in large part to energy policies), that of gaming PCs continues to rise, as do the number in use. Understanding this incredibly complex energy end use makes the work is quite a grand challenge.
What do you think are the most effective ways to cut energy use by gaming PCs and why?
There are three main "prongs": hardware, software, and user choices. On the hardware side, the graphics card ("GPU") is the greatest single node of energy use, but power management and improved power supplies and CPUs are also important. We have seen an enormous range in energy use per unit of workload (e.g., frames per second rendering rates - see Fig. 6 here), and so are very optimistic in that regard. On the software side, we are confident that given games can be designed with greater efficiency in mind without compromising user experience. On the user side, there are many untapped operational fixes (such as turning off unused slots in the motherboard).
What are the biggest challenges in this work?
The full magnitude of the problem and the diversity of efficiency opportunities are barely if at all on gamers' radar. The same can be said of energy and environmental policymakers, utilities, and other key stakeholders. So there is a large learning curve to address. There are also no widely accepted benchmarks of energy use or test methodologies, and so this is the place we must begin so as to have a common framework for measuring and talking about and describing energy use and savings in a consistent manner.
Have you found any surprises in this work?
Many. One example is that we initially thought that virtual reality would bring with it some intrinsic energy savings (since the displays in the goggles are so small), but have learned that the computational requirements of the graphics cards (a shift from 30-60 frames per second to 90-120 fps) can increase power requirements, although not in every case (see Fig. 8 here). Advanced "Foveated Rendering" can relax the fidelity in the periphery of the VR view range, and thereby reduce energy use significantly. Another other major surprise is how much energy is used in data centers for cloud-based gaming as well as the Internet infrastructure connecting those servers to the gamer (see our answers to various common questions about that below). We also found that some gaming systems use almost as much energy in idle or navigation mode as they use during active gameplay.
Does the design of gaming software influence energy use, and, if so, what can be done there?
Indeed. Attention to software design can yield big energy benefits. In one example, the majority of games - especially with first person or role playing style games - game play is structurally designed with “Save Points.” Meaning game story or game scoring progress can’t be saved between the save points without losing some hard earned benefits. As result, often gamers will simply load up a game menu, which serves to pause the gameplay exactly where the character is Save Point or not. While in this state, the gaming system is often drawing full power since the system is in Active Mode (this is more true for Consoles than Gaming PCs). It’s probably not uncommon for Gamers to leave their system in this state for long periods for restroom breaks, food breaks, etc. Better software design could reduce this impulse among Gamers or possibly provide software hooks for the gaming system manufacturers to detect unusually long game menu pauses in order to auto-save and transition to a low power hibernate mode.
In 2014, PS4 introduced a "Keep applications suspended“ functionality within standy-mode. This is a good step in the right direction but doesn’t seem to be universally functional on all games, needs to user initiated, nor has it been implemented on other console platforms.
Other examples are easier image resolution and rendering quality (directly correlated with power draw) setting menus or heads-up readouts providing real-time power draw feedback to the user so they can be better informed of current energy consumption. Aside from these specific examples, the basic point we’re making is that the creative space of the game developer community has not exhibited an energy efficiency awareness ethic up to this point. We’re betting this highly creative community can come up with some good ideas.
From what you know so far, what kinds of efficiency policies and programs might be most effective?
As this market segment is virtually untouched by the energy policy community, a wide array of possibilities exist. These include component and perhaps even game labeling, full system labeling, consumer information, utility financial incentives (e.g. rebates), and R&D into improved technology. If voluntary strategies don't suffice, policymakers may look to mandatory efficiency standards, although that would be quite technically problematic to implement.
How has the gaming community and industry responded to the research?
We've had the whole gamut of reactions. The industry was very supportive during our work and some shared useful information and provided good review comments on our draft reports. Many consumers react to brief news summaries without taking time to read the actual reports (we're heard "TLDR" more than once). Some gamers tend to compare their particular machine and personal utilization to our "average" one and balk at any differences. Of course, no one person is average but averages are relevant for extrapolating to large scales. Lots of people assert that it's not cost-effective to save energy in gaming, but this is a red herring insofar as some things can be done that cost nothing, other things will have a reasonable payback time, and others will yield non-energy benefits such as noise and heat reduction that are certainly valued by gamers. Some people are in disbelief that energy use can be reduced without taking a user-experience hit, but the data show clearly that this is quite possible.
From comments in the blogosphere, it seems that some gamers don’t much care about the environmental impacts of gaming - what’s one thing you would say to them?
Some gamers say that they simply don't care about energy or environment and that gaming is a necessity rather than an option. We are actually not advocating that people stop gaming or even reduce it, only to look at ways of specifying more efficient rigs. Even if environment isn't regarded as important, inefficient gaming PCs also release tons of heat and are noisier than efficient ones, so, it's actually a win-win proposition to improve efficiency ... not to mention saving money. The younger set may count on their parents to pay energy bills, when they become aware it will be a shock. Others care, but think our attention is misplaced. It's easy to dismiss any particular use of energy as being insignificant, and easy to point to "big polluters" elsewhere. The reality is that energy use is highly diffuse and there is no magic bullet or single type of use that will solve the problem. The inconvenient truth here is that gaming computers use more energy than previously thought, with the more power-hungry ones consuming as much as four efficient refrigerators. And gaming PCs are about the only piece of equipment left that has had a free ride in terms of no energy labeling, standards, or incentives for improvement ... this double standard will no doubt soon be resolved.
Are there free/cheap things that gamers can do to monitor their energy use, and if so, is there a place for crowdsourced data?
Crowdsourcing would be a great idea. We don't know of any sites that do it at any scale. However, before this could be done meaningfully there would need to be standard, accepted methodologies for measuring ... otherwise we will just have a lot of inconclusive apples-and-oranges data. Developing standardized methods is one of the goals of our work (report here).
What single piece of advice would you give to someone building a new gaming PC?
Carefully do the math to determine your total cost of ownership (purchase + operations). A $2000 machine could use again that much energy over just a few years.... If you're in a place with high electricity prices the operating cost can be particularly prohibitive. Take up the challenge of improving efficiency as a cool project to geek out on on! Check out greeningthebeast.org for lots of practical tips based on our research.
Lots of different questions have come in about cloud-based gaming - more here.