Skip to content

Instantly share code, notes, and snippets.

@mastercoms
Last active October 20, 2022 19:50
Show Gist options
  • Star 0 You must be signed in to star a gist
  • Fork 0 You must be signed in to fork a gist
  • Save mastercoms/44be52f6c070de1a7c2e8a0430e0b257 to your computer and use it in GitHub Desktop.
Save mastercoms/44be52f6c070de1a7c2e8a0430e0b257 to your computer and use it in GitHub Desktop.
mastercomfig 4.0.0
// '--- Render ---'
// '-- Threading --'
// These settings will take advantage of your CPU cores
cl_threaded_client_leaf_system 1 // Asynchronously add new renderables and compute blends for added
// renderables
//cl_threaded_client_leaf_system 0 // Use traditional behavior for CPUs with less than 4 cores
r_threaded_renderables 1 // Asynchronously set up bones on animated entities
//r_threaded_renderables 0 // Use traditional behavior for CPUs with less than 4 cores
mat_queue_mode 2 // Force multithreaded mode for the material system queue
//mat_queue_mode -1 // Let the engine determine the optimal mode for CPUs with less than 4 cores
// '-- Material System Queue --'
// These settings will take advantage of the material system, depending on how mat_queue_mode is set
//r_queued_decals 1 // Offload decal draws to the material system
// Beneficial when decal count is above 256 or you have 4 or more CPU cores
r_queued_decals 0 // Disabled by default due to crashes
//r_queued_post_processing 1 // Offloads post processing to the material system
// Performance improvement if your driver/GPU works with it
r_queued_post_processing 0 // Disabled by default because it usually fails to render the post processing
// effects in time for the frame render
r_queued_ropes 1 // Queue some rope rendering using the material system
// '-- Optimization --'
// These settings will optimize TF2 without having any effects to the user
mat_tonemapping_occlusion_use_stencil 0 // Do not use stencil buffer for histogram occlusion queries
host_flush_threshold 0 // Early finish to a Xbox 360 only flush task
r_norefresh 1 // Do not store a useless and unused frame time variable
r_occludermincount 1 // Force using at least one occluder
r_fastzreject 1 // Renders first pass of a scene with a z-buffer on the CPU and then has the GPU
// quickly render the scene in a second pass, with the visible parts of the scene already
// calculated
//r_fastzreject 0 // Disable for slow CPU and fast GPU
//disp_dynamic 1 // Uses a dynamic mesh for materials
disp_dynamic 0 // Disabled by default because of flashing textures
fast_fogvolume 1 // Use optimized fog rendering
r_pixelfog 1 // Use a pixel shader for fog if possible
fps_max 300 // Prevent a possible memory leak with fps_max 0 and set to 300 because some systems in
// Source assume an FPS limit of 300
engine_no_focus_sleep 0 // Do not limit FPS when the engine is not focused because the focus detection
// can be buggy on some systems or while streaming and may limit your FPS,
// without you knowing
//engine_no_focus_sleep 50 // Power savings while alt-tabbed out of TF2
mat_clipz 1 // Clips what is drawn for a performance improvement
r_entityclips 1 // Clips entity rendering according to a plane determined at each entity run
//r_entityclips 0 // Disable for slow CPU and fast GPU
r_frustumcullworld 1 // Cull on world draw
r_worldlistcache 1 // Cache some world rendering
r_drawopaquestaticpropslast 1 // Ensure that the modern and currently preferred rendering method is used
// '-- LOD --'
r_rootlod 1 // Limits how high the model LOD can be, does not offload rendering to CPU like many people
// believe. Set to 1 for the best balance between model quality and performance.
//r_rootlod 2 // Maximum performance for models
//r_rootlod 0 // Maximum quality for models
lod_TransitionDist -5000 // Fade map objects over a long distance to reduce pop in while still increasing
// performance
//lod_TransitionDist -5001 // Fade map objects across a shorter distance to partially reduce pop in while
// still increasing performance
//lod_TransitionDist -1 // Do not fade in map objects for maximum performance
r_lod -1 // Use r_rootlod to properly fade through LODs
// '-- Lighting --'
r_hunkalloclightmaps 1 // Allocate lightmaps in the hunk
//r_hunkalloclightmaps 0 // If a map's lightmap can't be put in the hunk, use dynamic memory
r_lightcache_zbuffercache 1 // Load z cache data with the map
@FritzyFritze
Copy link

holy crap this was only 4 years ago this is when i started tf2

@pkmnmstr98
Copy link

does this boost fps ?

@mastercoms
Copy link
Author

@pkmnmstr98
Copy link

@mastercoms do these work for garrys mod too ?

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment