Game-o-Meter FAQ
GENERAL
Q: What is the Game-o-Meter?
A: Game-o-Meter is a service designed to estimate your computers gaming performance based on your hardware profile.
Q: What does Game-o-Meter need to know about me?
A: To be able to estimate your gaming performance, Game-o-Meter needs to run a brief hardware check. No personally identifiable information is checked. See our Privacy policy for more information.
HARDWARE CHECK
Q: What are the requirements for running the hardware check?
A: The Game-o-meter hardware check currently works with all major browsers.
Q: Why does my hardware check fail?
A: The hardware check is not foolproof and might not work correctly for all hardware. We strive to ensure that it works on as many hardware configurations as possible, but there are endless combinations of PC hardware out there, so we cannot promise 100% compatibility with every single one.
Q: What to do if my hardware is detected wrong?
A: Please make a post in the Site Feedback forum or email us at gameometer@yougamers.com and we'll try to resolve your issue. Please include the scan result link displayed under the results and your correct hardware configuration.
OVERALL PERFORMANCE ESTIMATES
Q: What is the scale of the bar graphic showing system performance in the Game-o-Meter results?
A: The bar ranges approximately from fairly ancient systems like a 600 MHz Pentium III with Geforce 2 GTS (almost empty bar) to current high-end gaming PC's. Note that the available graphics hardware performance has increased tremendously in the last few years with the absolutely fastest high end systems outperforming average gaming systems by a factor of 4-5x - a short bar for your system doesn't automatically mean that the system is unsuitable for gaming as many games still have very moderate system requirements and show a very short bar as the requirement. The important factor is if your system meets the listed requirements or not - Game-o-Meter will explain in detail how well your system should run the game after the scan.
Q: Is all hardware supported by Game-o-Meter?
A: Almost all viable hardware components are supported by the Game-o-Meter. Some of the hardware that is not supported may still be able to run some of the games featured on YouGamers.com, but performance is likely to be very poor. Data from server processors (Xeons, Opterons) or workstation video cards (NVIDIA Quadro, AMD FirePro / ATI FireGL) is very limited and we cannot guarantee estimates for all such configurations.
Q: What are the estimates based on?
A: Game-o-Meter performance estimates are based on Futuremark's industry standard 3D gaming benchmarks; 3DMark06, 3DMark Vantage and 3DMark 11. Estimation models have been calculated from Futuremark's 3DMark.com result database, containing over 20 million benchmark results from 4 million+ users all over the world.
Q: I'm not getting an estimate of my performance even if my hardware is detected correctly. Why is this?
A: This could happen if your processor and graphics card combination is so rare, that the 3DMark.com database doesn't have any data for it (Game-o-Meter will tell you this). You can help us improve by running 3DMark 11 and/or 3DMark Vantage (they're free!) and submitting the score to 3DMark.com.
Q: What is the performance estimate model like?
A: The Game-o-Meter performance estimation model is actually built from several linear regression models. To account for differences between different processors and graphics cards, a separate model has been estimated for every processor and graphics card combination for which benchmark data exists.
Q: Huh? Linear regression models? What does that mean?
A: I'm glad you asked! In linear regression, the estimate is calculated as a sum of predictor variables each multiplied by a constant factor.
This means that the estimate formula is of the form SCORE = b0+b1*x1+b2*x2+... The x's are variables that affect a systems performance, for example one of these variables is the processors clock frequency. The constant parameters b are estimated using the method of ordinary least squares to fit the available benchmark data.
These parameters also have a simple practical interpretation; they measure the average effect of the variable on the benchmark score. For example, if x1 is the processors frequency in MHz, then b1 is the increase in score when the processors frequency increases by 1MHz. So if we increase the processors frequency by 100 MHz, we should see an increase of about 100*b1 in the benchmark score.
Q: How reliable are the estimates?
A: There are many factors that can affect a systems gaming performance and the estimation models cannot take all possible factors into account. For a single system configuration, the actual performance level of most systems matching that configuration with correctly functioning hardware and latest drivers is within three pixels on the performance estimate bar graphic. It's important to keep in mind, that performance is always estimated based on 3DMark benchmark scores, actual performance levels may differ slightly from game to game.
Q: How are the performance estimates calculated for the minimum and recommended systems?
A: Minimum and recommended system requirements are often rather vague. The performance estimate shown on the chart represents a typical low- to mid-end system matching those specifications. Differences in performance levels between systems matching the given minimum or recommended system requirements so the estimates shown should be taken as rough guidelines. Generally even if your systems performance estimate is lower than that of the minimum system, but you pass all component checks, your computer can often run the game, but performance and gaming experience is likely to be poor.
Q: I have SLI/Crossfire, does the overall performance estimate reflect this?
A: The performance estimation model is equipped to handle SLI and Crossfire. If there's no increase in your performance with SLI enabled, it might be because we don't have enough benchmarking data fro
1f4a
m your hardware combination with SLI. Again, you can help by running one of our benchmarks and submitting your results.
Q: I have a dual-core or quad-core processor. How does this affect my estimate?
A: The benchmarks benefit from multi-core processors, so performance estimates are also higher for multi-core processors than single cores. Notice that while most games scale with multiple cores, there are still some that do not benefit from multiple cores, so your estimates might not totally accurately reflect actual performance. This can occur, for example, if you have a very slow quad-core CPU and the game requires a lot of performance from a single core.
