- Date:
- Monday , August 10, 2009
- Author:
- Mark Warner
- Editor:
- Brent Justice
- Google +1

ARMA II Gameplay Performance and Image Quality
In case you weren't frustrated enough with tactical shooters, Bohemia Interactive has dropped ARMA II on a wary and waiting gaming audience. ARMA II puts the current generation of video cards to the test, this realistic and demanding game is a graphics hog. We’ll evaluate eight video cards running this game and let you know what kind of gameplay experience to expect.
Graphics Options in ARMA II
ARMA II is a PC exclusive title. There is no console version of this game. Generally speaking, PC-only games have more advanced graphics options than PC games with accompanying console editions. With ARMA II, that is certainly the case.
There are sixteen graphics options in ARMA II. Two of them are simple brightness and contrast/gamma options, and there are oddly two separate resolution options as well.

Visibility
The first option, visibility, affects the level of detail of distant objects. With a very low setting, objects will appear to pop into existence closer up. With a higher setting, those objects may appear earlier.
Quality Preference
The Quality Preference option allows the gamer to quickly set most of the other graphics options in just a couple of clicks. Use this option if you do not want to individually tweak your graphics options. We did not use this option during testing and opted to manually configure each setting for the optimal gameplay experience.
Interface Resolution/3D Resolution
These two resolution sliders are interesting. The first, "Interface Resolution" actually determines which resolution for which your display will be configured. The "3D Resolution" option controls the resolution at which the games 3-dimensional view is actually rendered. If the 3D resolution is smaller than the interface resolution, the 3D picture will be up-sampled to match the interface’s size. If the 3D resolution is larger than the interface resolution, the 3D picture will be down-sampled to match the interface resolution.
In practice, this has two interesting effects. If you choose a smaller 3D resolution than your interface resolution, you will see a nice performance increase at the expense of visual clarity. Since you are increasing the size of the rendered image, the resulting picture will be blurry. If you choose a larger 3D resolution than your interface resolution, you will achieve a supersampling anti-aliasing effect. This will reduce jagged edges on objects, textures, specular reflections, and everything else in the game.
Texture Detail/Video Memory
The Texture Detail and Video Memory settings are linked. You cannot have a higher Texture Detail setting than your Video Memory setting. Obviously, if your video card has a smaller amount of memory, like 256MB, you will want to experiment with lower options on this setting. And if you have a 1GB or 2GB video card, try out the Very High setting. Remember that memory bandwidth is as important for texturing performance as memory capacity. If the memory is too slow to efficiently sling large textures around, it won’t help you at all to have 2GB of the stuff.
Anisotropic Filtering/Antialiasing
The Anisotropic Filtering (AF) option has five options: Disabled, Low (2X), Normal (4X), High (8X), and Very High (16X). During our testing, we did not notice a difference in filtering quality using the in-game option versus the NVIDIA/AMD graphics card control panel options.
On AMD video cards, the Antialiasing option seems to enable a shader-based edge-smoothing effect which is completely ineffective at reducing jagged edges in the game. With an NVIDIA video card, it enables proper MSAA. However, it gives no way to accurately determine what level of AA is actually in use with the setting enabled. Forcing an AA option from the NVIDIA and AMD control panels did not work for us.
Terrain Detail/Objects Detail
The Terrain Detail option controls the density and distance of landscape objects such as trees, grass, and shrubbery. Lower this option if your video card is challenged for shader power. The Objects Detail setting is related, but controls the level of detail of objects such as corpses, vehicles, soldiers, and weapons. This option is also dependent on shader performance.
Shadow Detail
The Shadow Detail option controls the number of objects that cast shadows as well as the distance from which some shadows are drawn and the quality of the cast shadow itself. Lower this option if your video card is low on shader rendering power.
Postprocess effects
The Postprocess Effects option controls graphical features like HDR, depth of field, and motion blurring. Lower this option if your video card is low on shader rendering power.
Interface Size
The Interface Size option allows the player to increase the scaling of the game’s HUD. We found that this setting had no impact on performance and should be set to whatever your eye prefers. We used the Normal setting of this option throughout testing.
Aspect Ratio
The Aspect Ratio option should set itself correctly automatically when you choose your interface resolution. If the game does not choose the correct option for some reason, or if you want to change it manually, you can. Changing the Aspect Ratio setting does not alter the list of available resolutions.
Testing Scenario
For our testing procedure, we are playing the first five minutes of the single-player scenario entitled "Counterattack", in which Russian forces are tasked with retaking control of a city from rebel forces. The scenario begins with a ride in a BMP armored personnel carrier, from which our player is automatically ejected. We begin our FRAPS recording immediately following that ejection, while we work our way into the outskirts and eventually into the city heart. There is no "God mode" for ARMA II, and as a result, we were forced to run this scenario several times to get one good set of data. As it is a combat simulator, and not an "arcade" action game, death comes very easily in this game, which interrupts the rendering with a saved game loading sequence.
