A Quick Look at our Testing Methods

The focus of this review is the comparison of some of today’s more popular hardware in relation to its performance with various settings in Age of Conan. One of the biggest attractions and complaints about the game is its system requirements which break the norm of conventional MMORPG graphics which cater to a wide range of systems. The scenery is similar to something one may find in a photo journal and the real world replication adds all the extras to fully captivate the player. With this comes the need for some very hefty computing power and in the past few weeks since release, there has been much talk about exactly what type of hardware is required to perform at optimal settings.

Due to the fact that Age of Conan is an Online RPG, there exists no canned demo or repeatable run through like those traditionally employed in the benchmarking process. While many harp on how time-demos are inaccurate in relation to real-world game play, as any scientist knows, a test must be done in a completely repeatable and controlled environment to produce accurate results. Unfortunately that is not possible in this case. We have done our best to ensure that the results are as accurate as possible and represent what a user should expect to experience using the same or relatively similar hardware and settings. At the same time we also stress that the results are only directly comparable within the confines of this review and will differ depending on multiple factors including a user’s hardware configuration and game factors such as server, latency, time of day (in game), location in game and users in game in your location

In all of our tests, we have set specific "waypoints" for our characters, all of which must be reached before the run is finished; though we occasionally do deviate from the straight line between them. The camera is positioned at an approximate 40 degree angle to the character and zoomed out fully to take in as much scenery as possible. In areas that incorporate weather and day/night cycles our tests were run at set times during the day across all hardware to ensure that all graphical variances were taken into account. Any and all abnormalities were retested three fold to ensure accuracy and each test was run a total of 5 times, taking the mean of the average frames per second. Interestingly enough, very rarely did any average FPS differ by more than 1-2 between runs, and when doing a quick recheck after all testing was complete, the results were accurate every time.

We consider 30FPS to be a passing mark in this game as often this is the point in which the human eye can no longer decipher a noticeable difference of refresh in games, and allows for seamless game play with very little or no stuttering.