Mass Effect Tweak Guide
[Page 8] Advanced Tweaking (Pt.2)
Static decals are markings or posters in the game world which are placed on surfaces as decoration by the developers, and are not created by players. Setting this option to False removes these, and depending on the area the visual and performance difference may or may not be noticeable. In general it is not recommended that this be disabled for gameplay purposes.
Dynamic decals are the marks left by players when using weapons on the surroundings. If this option is set to False, while weapons may show impact effects initially, most weapons will not leave dark scorch marks or bullet holes on scenery. Disabling this option can improve performance in heavy combat, but it also reduces realism.
This setting controls the use of Dynamic Lighting, which is lighting that interacts realistically with the game world. Disabling this setting in Mass Effect results in a range of lighting glitches - wherever dynamic lighting is used objects will appear as flat black, so it is not recommended that you alter this.
This setting controls whether the Light Environment, namely the general lighting in a level, such as sunlight, creates shadows. When set to False, it disables many shadows similar to disabling the Dynamic Shadows setting, but it still does not remove the shadows for objects with fixed shadows, such as buildings and terrain. This may improve performance but also reduces realism.
If set to True, this option appears to enable a less accurate form of dynamic lighting, which may improve performance at the cost of image quality. In practice the visual difference is not noticeable.
This setting appears twice in this section for some reason. In any case it is meant to control the quality of the shadow filtering used (if shadows are enabled). Higher values should provide better quality at the cost of performance, but once again with testing I was unable to note any real differences in shadow quality using this setting.
This technique allows for improved image quality, as surfaces will have additional depth, bump and detailing. Setting this option to False can increase performance but in Mass Effect simply results in flat black textures wherever the lightmaps were used, so there is no reason to disable it.
This option controls the use of Depth of Field, which is an effect used to simulate the way in which objects within focus are sharper than objects outside our direct field of focus. It's primarily noticeable when zooming in on objects with a sniper rifle - objects in focus will appear crisp, while surrounding objects will be blurred. You can set this option to =False to gain improved performance, or if you just don't like this effect.
This setting controls the use of Bloom Lighting, which is a form of lighting that exaggerates the brightness and glow of lights, creating a slightly hazy, more atmospheric look. With this option set to False, Bloom is not supposed to be used, however it doesn't seem to affect the lighting at all. If you want to remove Bloom lighting you can set bUsePostProcessEffects to False (See further above), though this removes all post processing effects.
As with Bloom above, this setting controls Bloom lighting effects, and if set to True uses a more accurate form of Bloom. Disabling this may gain additional performance without a noticeable change in image quality. Once again, in practice disabling this and/or the Bloom setting above doesn't actually appear to diminish or remove Bloom lighting.
These two settings are supposed to control how the SpeedTree foliage rendering system works. However in practice if set to False, it doesn't appear as though this has any impact on foliage in most areas.
This setting relates to the streaming system used in the Unreal Engine. However it has no impact on texture streaming, so setting it =True should primarily result in audio being preloaded. This may result in a reduction in stuttering throughout the game, however in my testing it had no discernable impact.
Lens Flare is an effect seen when staring at a bright light source, such as a Sun, or gun muzzle flashes. There are halo effects or changes in light intensity when this happens. When this option is set to False however these effects are not removed in Mass Effect.
This setting enables or disables Floating Point Render Target technology, which is used to provide certain post processing effects, such as the ability to enable HDR and Anti-Aliasing at the same time. The main impact of setting this option to False is that it can alter the appearance of the game, removing certain post processing effects. However for the most part there doesn't seem to be any benefit to disabling this option.
This option controls the use of Trilinear Filtering, which can improve the appearance of textures. Disabling this option may slightly reduce texture clarity but can improve performance. Note that the general appearance of textures is more dependent on the MaxAnisotropy variable, as well as the Texture Detail in-game setting. You can enable Trilinear Filtering with Anisotropic Filtering to get a higher quality of Anisotropic, however this might increase stuttering and/or further reduce performance.
LOD Bias affects the clarity of textures and details, by determining how Mipmapped textures appear. Higher positive values make things blurrier and less detailed but can improve performance, while lower values, including negative values, make things clearer at the cost of performance, and can also introduce further aliasing (jaggedness and shimmering) to the scene. The SkeletalMeshLODBias setting relates to animations, while ParticleLODBias relates to particle effects. In practice the impacts of altering either of these settings is not significant, but if you want the best quality try using a negative value for both, such as -1.
If you have Anisotropic selected under the Filtering Mode in-game setting, the setting here controls the actual level of Anisotropic Filtering (AF) used in the game to make distant textures clearer. You can set a sample rate value ranging from 0 (0xAF, or no AF), to 4 (4x AF) which is the default when Anisotropic is used, to 16 (16xAF) which is the maximum possible. The higher the sample rate, the lower your performance in return for crisper, more detailed textures. Note that you can also enable Trilinear at the same time for even higher quality - see the Trilinear setting further above.
This setting allows you to control the proportion of the screen in which the main game image is shown. For example a value of 50 for this setting means that the main game image (but not in-game menus) will be reduced to only take up 50% of screen. There doesn't appear to be much of a performance benefit in doing this.
There are a range of TEXTUREGROUP settings in this section relating to the Level of Detail (LOD) of different aspects of the game world, objects, characters, and lighting among other things. These are essentially supposed to control the level of detail and quality of objects and textures as they recede into the distance. Each has a MinLODSize, a MaxLODSize and LODBias which you can experiment with to alter image quality. By raising MinLODSize as well as MaxLODSize, and lowering LODBias, you should be able to get better image quality. However in practice not only will the game crash if you use certain values, there also appears to be absolutely no improvement in image quality even when correct values are used.
In any case I encourage users to experiment with these, and if you want to know the best possible values for these settings, then refer to the ones provided in this file: ME_TextureLODSettings.zip (1KB), originally tested and provided by Martigen.
This setting should control Hardware Shadow Mapping, which when enabled uses your graphics card to render shadows. When this setting is =True, that means it has supposedly disabled Hardware Shadow Mapping, and thus your CPU may be undertaking shadow calculations. The premise is that by setting this option to =False, those with weaker CPUs but faster graphics cards can get a performance boost. In practice however, having tested this in a range of areas, I found no performance or image quality difference whatsoever, even though I have a relatively weak CPU (X2 4400+) compared to my GPU (8800GTS). However single-core CPU owners and those with low-end Dual Core CPUs may still like to give it a try and see if it makes a difference for them.
There are a range of sections here starting with [Stats.] - as far as I can tell, these are not gameplay parameters to be changed by users, they are engine statistics. For example changing the line Total Texture Memory=200 MB to =0MB has no impact on the performance of the game.
The next page continues the Advanced Tweaks.