Page 2 of 3 FirstFirst 1 2 3 LastLast
Results 11 to 20 of 28

Thread: Using the Dota performance test

  1. #11
    Basic Member
    Join Date
    Apr 2015
    Posts
    2
    ok but it doesnt help, all settings low, render set to 80%,vsync turn on and my fps 60 and drops during all game, if settings set to medium or higher i`ve got almost stable 40

  2. #12
    Basic Member
    Join Date
    Oct 2016
    Posts
    8
    http://pastebin.com/raw/DUUZhQhf
    any other suggestion to improve my FPS besides buying new computer (parts)?

    i tested Underlord since spamming his Firestorm ability will give me lowest FPS
    Last edited by ios; 02-28-2017 at 12:32 PM.

  3. #13
    Basic Member aveyo's Avatar
    Join Date
    Aug 2012
    Location
    EU West
    Posts
    2,718

    Post benchmark.cfg alternative - fixed .zip links!

    Pausing the game to execute perftest.cfg is misleading specially when it comes to particle performance, since the load drops significantly (-50%?!).
    Particles are exactly what hurts the most any low end system, and it's unfortunate that lowering Effects Quality only does a lame cl_particle_fallback_base adjustment, not affecting Shrines or Cosmetics at all.

    Anyway, more representative of the DOTA 2 real-world performance would be to test a particular demo file live - one that really puts a load on your system, like MatchID: 2844309781
    I've been providing a benchmark.cfg alternative for a while now (even before perftest.cfg, using ingenious pseudo-wait). Last versions use MatchID: 2844309781, and does 2x live tests, plus the static 7x tests in OP on a demanding scene.
    If any of you are interested, details are quoted below:
    Quote Originally Posted by aveyo View Post
    benchmark.cfg
    Code:
    ////--------------------------------------------------------------------------------------------------------------------------------
    ////
    ////     AveYo's DOTA 2 BENCHMARK.CFG v3.1
    ////     Previous D-OPTIMIZER V2 cfg engine based script got this job done using clever pseudo schedule / wait cmds
    ////     Now Valve finally added exec_async and a suite of useful commands featured in their perftest.cfg
    ////     This is a quick mix of the two scripts
    ////
    ////     1: remove .txt extension and copy this file to \Steam\steamapps\common\dota 2 beta\game\dota\cfg\benchmark.cfg
    ////     2: change Dota 2 launch options (-dx11 / -dx9 / -vulkan / -nod3d9ex / -nosound )
    ////     3: open Dota 2 and go to Watch - Replays - enter 2844309781 in Search for MatchIDs then download it
    ////        credits for the matchid goes to dev.dota user: jstq
    ////     4: to start, enter into console:  EXEC_ASYNC BENCHMARK
    ////     Zip file including replay available at: https://1drv.ms/f/s!AhH0pZaDBUFEgU6MJbUMugaQZuFw
    ////
    ////     Demo overlay and spectator menu cannot be closed by script anymore, feel free to do it manually in the Warm up phase
    ////     Try not to press any keys or move the mouse in the red Running performance tests phase ( takes ~2 minutes )
    ////     Results are pretty printed to console when finished
    ////     Can be saved to \dota 2 beta\game\dota\condumpxxx.txt file after entering into console: CONDUMP
    ////
    ////     v2 Changes:
    ////     - initial short warmup now uses normal demo playback
    ////     - removed demo_goto usage (caused Source2Bench.csv not writing fps number - it was reported fine in console)
    ////     - timedemo now runs uninterrupted for more reliable results
    ////     - added particle usage reporting
    ////     - improved ui feedback
    ////     v2.1 Changes:
    ////     - added a better alias filter to facilitate resetting timedemo_start / timedemo_end when done / intrerrupted with FF
    ////     - added pimpmuckl's demo alternative
    ////     - added a few more stats to the console report (video memory usage, sound cache usage)
    ////     v3 Changes:
    ////     - adapted Valve's perftest.cfg for more specific / grouped stats and extra machine info
    ////     - single replay load with several warm up stages implemented for more accurate and reproducible results
    ////     v3.1 Changes:
    ////     - split into live test (repeated once) and 7x static tests, overall duration reduced
    ////     - added 'Interpreting the static test results' by gdrewb-valve
    ////
    ////--------------------------------------------------------------------------------------------------------------------------------
    //// Last updated: 21:40 UTC FEB 28
    
    //// BACKUP CVARS? ( ASSUMING IT SHOULD DO THAT, IT'S ONLY PARTIAL )
    push_var_values
    
    //// LOADING REPLAY GUI HINT
    hideconsole
    activategameui
    top_bar_message "[Dota 2 Benchmark]  Loading replays/2844309781, please wait a minute.." 0
    sleep 5000
    gameui_hide
    
    //// LOADING REPLAY OPTIONS
    sv_log 0
    sv_alternateticks 0
    dota_spectator_use_broadcaster_stats_panel 0
    dota_spectator_mode 0
    dota_spectator_hero_index ""
    cl_showdemooverlay 0; // still broken, no other command usable to hide demo overlay
    demo_usefastgoto 1
    @panorama_show_fps 1
    //demo_pauseatservertick 104200
    
    
    ////  REPLAY TARGET                                                                                 DEFAULT: replays\2844309781
    alias #replay_target "playdemo replays\2844309781";  // reused by LIVE TEST 3
    #replay_target;
    
    ////  REPLAY LOAD TIME - DEFAULT 60 SECONDS SHOULD BE ENOUGH ( COULD USE 30 FOR FASTER SYSTEMS OR 90 FOR SLOWER ONES )
    sleep 60000
    //sleep 30000
    //sleep 90000; // sleep can't be aliased..
    
    
    ////--------------------------------------------------------------------------------------------------------------------------------
    //// CFG INTERNALS: CONSOLE OUTPUT / CANCEL / D-OPTIMIZER BASED SCHEDULER ENGINE REPLACED BY EXEC_ASYNC SLEEP
    ////--------------------------------------------------------------------------------------------------------------------------------
    //// CONSOLE ECHO ALTERNATIVE WITH ON-THE-FLY COLORS! - FINALLY VALVE (RE)ADDED ECHOLN SO USE THAT INSTEAD
    alias #echo_off "log_flags Console General +donotecho | grep %";alias #echo_on "log_flags Console General -donotecho | grep %";
    alias #color_none "log_color General C0C0C000 | grep %"; // invisible color - text can still be selected/copied/pasted
    alias #color_default "log_color General 00000000 | grep %;alias #color  #color_default";//ip "";// using echoln instead of ip for newline
    alias #echo "#color_none;echoln;#color";// basically sets invisible  color, prints newline (v2: ip = ) then restores previous color
    //// COLOR PRESETS! - WORKS THE SAME WITH BUILT-IN ECHOLN
    alias #color_cyan "log_color General 00FFFFFF | grep %;alias #color #color_cyan";
    alias #color_yellow "log_color General FFFF00FF | grep %;alias #color #color_yellow";
    alias #color_green "log_color General 00FF00FF | grep %;alias #color #color_green";
    //// FILTER
    alias `~` "";
    //// DISABLE ALL CONSOLE SPEW SOURCES FOR A CLEAN SCRIPT OUTPUT ( VALVE PLZ GIFF SINGLE CMD FOR IT )
    log_flags Console +DoNotEcho | grep %
    log_flags General Developer DeveloperConsole Panel Panorama  PanoramaScript VScript VScriptDbg VScriptScripts CustomUI  CustomGameCache CustomNetTable +DoNotEcho
    log_flags AnimationGraph AnimationSystem AnimGraphManager AnimResource  Assert "BitBuf Error" BoneSetup Client "Combat Analyzer" CommandLine D3D  Decals Demo DeveloperVerbose DotaGuide DOTAHLTVCamera +DoNotEcho
    log_flags DOTAHLTVDirector DOTA_CHAT DownloadManager EmitSound  EngineInitialization EngineServiceManager "Entity Dump" "Entity Load  Unserialize" "Entity System" Filesystem GameEventSystem GCClient  GlobalState HangWatchdog "HLTV Server" Host +DoNotEcho
    log_flags HostStateManager IME InputService InputSystem InstantReplay  LOADING MaterialSystem MeshSystem ModelCombiner modellib NavMesh  NetworkClientService Networking "Networking Reliable" NetworkP2PService  NetworkServerService +DoNotEcho
    log_flags NetworkService Particles ParticlesLib Physics PostProcessing  PostProcessPipeline RenderPipelineDota RenderPipelineVr RenderService  RenderSystem ResourceSystem SaveRestore SaveRestoreIO Scaleform  "Scaleform IME" ScaleformAS +DoNotEcho
    log_flags ScaleformParse ScaleformScript SceneSystem SchemaSystem  SchemaSystemUtils Server ServerLog SignonState SndEmitterSystem  SndOperators SoundOperatorSystem SoundOpGameSystem SoundSystem  SoundSystemLowLevel SpawnGroup SplitPacket +DoNotEcho
    log_flags SplitScreen Steam SteamDatagramClient SteamDatagramServer  SteamUnifiedMessages ToneMapping ToolGameSimulation TypeManager Vfx  VguiCallQueue VolumetricFog VProf VR WeekendTourney Workshop  WorldRenderer +DoNotEcho
    
    ////--------------------------------------------------------------------------------------------------------------------------------
    ////  BENCHMARK MAIN - SYSTEM INFO, THEN 5 x 30 seconds performance tests
    ////--------------------------------------------------------------------------------------------------------------------------------
    
    //// PERFORMANCE TESTS GUI HINT
    activategameui
    top_bar_message "[DOTA 2 BENCHMARK]  Running performance tests - ETA: 2 minutes" 1
    sleep 5000
    gameui_hide
    
    //// CONSOLE LOGO
    clear
    #color_cyan
    #echo_on
    echoln ;// DispatchAsyncEvent spew in the General channel is so lame..  killing it would also kill echoln.. question is why is set like that?
    #echo_on; echoln "  _______             ______    ______    ________    __   ___  ___   __   ________   _______   ______   "; #echo_off
    #echo_on; echoln " |   __  \           /      \  |   _  \  |        | |   | |   \/   | |  | |       /  |   ____| |   _  \  "; #echo_off
    #echo_on; echoln " |  |  |  |         |  ,~~,  | |  |_)  | `~~|  |~~` |   | |  \  /  | |  | `~~~/  /   |  |__    |  |_)  | "; #echo_off
    #echo_on; echoln " |  |  |  | AVEYO`S |  |  |  | |   ___/     |  |    |   | |  |\/|  | |  |    /  /    |   __|   |      /  "; #echo_off
    #echo_on; echoln " |  `~~`  |         |  `~~`  | |  |         |  |    |   | |  |  |  | |  |   /  /~~~, |  |____  |  |\  \  "; #echo_off
    #echo_on; echoln " |_______/           \______/  |__|         |__|     |__| |__|  |__| |__|  /_______| |_______| |__| \__\ "; #echo_off
    echoln
    #echo_off
    #color_default
    
    //// CONSOLE MACHINE INFO
    #color_yellow
    #echo_on
    echoln
    echoln -------------------------------------------------------------
    echoln [DOTA 2 BENCHMARK]  Machine info
    echoln -------------------------------------------------------------
    echoln
    #echo_off
    #color_default
    log_flags Console -consoleonly | grep %; // allow sys_info to be logged to file
    #echo_on
    sys_info
    #echo_off
    log_flags Console +consoleonly | grep %
    
    //// CONSOLE TEST LABEL
    #color_yellow
    #echo_on
    echoln
    echoln
    echoln -------------------------------------------------------------
    echoln [DOTA 2 BENCHMARK]  Running performance tests - ETA: 2 minutes
    echoln -------------------------------------------------------------
    #echo_off
    #color_default
    
    //// TEST ACTIVATE ( Valve's proposed method relies on cl_showfps - it  should have lower overhead than D-OPTIMIZER V2 based Vprof )
    @panorama_show_fps 0
    fps_max 999
    cl_showfps 2
    
    //// LIVE TEST 1. CURRENT USER SETTINGS         [DEF]  D-OPTIMIZER PRESETS:       Potato  Low  OPTIMAL  Med   High  Ultra     @
    ////--------------------------------------------------------------------------------------------------------------------------------
    //dota_portrait_animate 1;                   //   [1]  Animate Portrait             0      0      1      1      1      1      O
    //r_deferred_additive_pass 0;                //~  [1]  Additive Light Pass          0      0      0      1      1      1      M
    //r_deferred_simple_light 1;                 //   [1]  World Lighting               0      1      1      1      1      1      L
    //r_ssao 0;                                  //   [0]  Ambient Occlusion            0      0      0      0      0      1      U
    //r_dota_normal_maps 0;                      //   [0]  Normal Maps                  0      0      0      1      1      1      M
    //r_dota_allow_parallax_mapping 0;           //   [0]  Ground Parallax              0      0      0      0      1      1      H
    //dota_ambient_creatures 0;                  //   [0]  Ambient Creatures            0      0      0      0      0      1      U
    //dota_ambient_cloth 1;                      //   [0]  Ambient Cloth Simulation     0      0      1      1      1      1      M
    //r_grass_quality 1;                         //   [0]  Grass                        0      0      1      2      3      4      O
    
    //r_dota_fxaa 0;                             //   [0]  Anti-Aliasing                0      0      0      0      1      1      H
    //r_deferred_specular 1;                     //   [0]  Specular                     0      0      1      1      1      1      O
    //r_deferred_specular_bloom 0;               //   [0]  Specular and Light Blooms    0      0      0      1      1      1      M
    //dota_cheap_water 1;                        //   [1]  !High Quality Water          1      1      1      1      0      0      H
    //r_deferred_height_fog 1;                   //   [1]  Atmospheric Fog              0      0      1      1      1      1      O
    //r_dashboard_render_quality 1;              //   [1]  High Quality Dashboard       -      -      -      -      -      -      -
    //r_dota_allow_wind_on_trees 0;              //   [0]  Tree Wind                    0      0      0      0      1      1      H
    
    //r_texture_stream_mip_bias 1;               //   [1]  Texture Quality              2      1      1      1      0      0      L
    //cl_particle_fallback_base 4;               //~  [2]  Effects Quality              4      4      4      2      0      0      M
    //cl_globallight_shadow_mode 2;              //   [2]  Shadow Quality               0      1      2      2      2      3      L
    //mat_viewportscale 1.0;                     //   [1]  Render Quality             0.8    0.9    1.0    1.0    1.0    1.0      L
    //fps_max 60.0f;                             // [120]  Maximum frames per second   60     60     60    120    144    240      M
    
    //// Reset live scene
    demo_resume
    dota_spectator_mode 3
    dota_spectator_hero_index 0
    demo_goto 102900 pause
    sleep 2000
    dota_spectator_hero_index 2
    stat_dropdown 0
    demo_resume
    #color_green
    #echo_on
    echoln LIVE TEST 1. CURRENT USER SETTINGS
    #echo_off
    #color_default
    //// Let new state settle.
    sleep 1000
    cl_resetfps
    sleep 30000
    #echo_on
    cl_printfps
    #echo_off
    
    //// LIVE TEST 2. CURRENT USER SETTINGS
    //// Reset live scene
    demo_resume
    dota_spectator_mode 3
    dota_spectator_hero_index 0
    demo_goto 102900 pause
    sleep 2000
    dota_spectator_hero_index 2
    stat_dropdown 0
    demo_resume
    #color_green
    #echo_on
    echoln LIVE TEST 2. CURRENT USER SETTINGS
    #echo_off
    #color_default
    //// Let new state settle.
    sleep 1000
    cl_resetfps
    sleep 30000
    #echo_on
    cl_printfps
    #echo_off
    
    //// STATIC SCENE SETUP
    demo_resume; demo_pauseatservertick 104081; demo_goto 102800
    dota_spectator_mode 3
    dota_spectator_hero_index 0
    sleep 1000
    dota_spectator_hero_index 5
    stat_dropdown 0
    sleep 30000
    
    //// STATIC TEST 1. CURRENT USER SETTINGS
    #color_yellow
    #echo_on
    echoln
    echoln STATIC TEST 1. CURRENT USER SETTINGS
    #echo_off
    #color_default
    //// Let new state settle.
    sleep 1000
    cl_resetfps
    sleep 5000
    #echo_on
    cl_printfps
    #echo_off
    
    //// STATIC TEST 2. BASELINE MEASUREMENT
    mat_viewportscale 1
    cl_globallight_shadow_mode 2
    dota_cheap_water 0
    r_drawparticles 1
    cl_particle_simulate 1
    r_drawblankworld 0
    r_drawpanorama 1
    #color_yellow
    #echo_on
    echoln STATIC TEST 2. BASELINE MEASUREMENT
    #echo_off
    #color_default
    //// Let new state settle.
    sleep 1000
    cl_resetfps
    sleep 5000
    #echo_on
    cl_printfps
    #echo_off
    
    //// STATIC TEST 3. HALF RESOLUTION
    mat_viewportscale 0.5;//1
    cl_globallight_shadow_mode 2
    dota_cheap_water 0
    r_drawparticles 1
    cl_particle_simulate 1
    r_drawblankworld 0
    r_drawpanorama 1
    #color_yellow
    #echo_on
    echoln STATIC TEST 3. HALF RESOLUTION
    #echo_off
    #color_default
    //// Let new state settle.
    sleep 1000
    cl_resetfps
    sleep 5000
    #echo_on
    cl_printfps
    #echo_off
    
    //// STATIC TEST 4. REDUCED DRAWING
    mat_viewportscale 1
    cl_globallight_shadow_mode 1;//2
    dota_cheap_water 1;//0
    r_drawparticles 1
    cl_particle_simulate 1
    r_drawblankworld 0
    r_drawpanorama 1
    #color_yellow
    #echo_on
    echoln STATIC TEST 4. REDUCED DRAWING
    #echo_off
    #color_default
    //// Let new state settle.
    sleep 1000
    cl_resetfps
    sleep 5000
    #echo_on
    cl_printfps
    #echo_off
    
    //// STATIC TEST 5. PARTICLES DISABLED
    mat_viewportscale 1
    cl_globallight_shadow_mode 2
    dota_cheap_water 0
    r_drawparticles 0;//1
    cl_particle_simulate 0;//1
    r_drawblankworld 0
    r_drawpanorama 1
    #color_yellow
    #echo_on
    echoln STATIC TEST 5. PARTICLES DISABLED
    #echo_off
    #color_default
    //// Let new state settle.
    sleep 1000
    cl_resetfps
    sleep 5000
    #echo_on
    cl_printfps
    #echo_off
    
    //// STATIC TEST 6. UI ONLY
    mat_viewportscale 1
    cl_globallight_shadow_mode 2
    dota_cheap_water 0
    r_drawparticles 1
    cl_particle_simulate 1
    r_drawblankworld 1;//0
    r_drawpanorama 1
    #color_yellow
    #echo_on
    echoln STATIC TEST 6. UI ONLY
    #echo_off
    #color_default
    //// Let new state settle.
    sleep 1000
    cl_resetfps
    sleep 5000
    #echo_on
    cl_printfps
    #echo_off
    
    //// STATIC TEST 7. GAME ONLY
    mat_viewportscale 1
    cl_globallight_shadow_mode 2
    dota_cheap_water 0
    r_drawparticles 1
    cl_particle_simulate 1
    r_drawblankworld 0
    r_drawpanorama 0;//1
    #color_yellow
    #echo_on
    echoln STATIC TEST 7. GAME ONLY
    #echo_off
    #color_default
    //// Let new state settle.
    sleep 1000
    cl_resetfps
    sleep 5000
    #echo_on
    cl_printfps
    #echo_off
    r_drawpanorama 1
    
    //// STATIC TESTS HELP
    #color_cyan
    #echo_on; echoln; echoln Interpreting the static test results (gdrewb-valve); #echo_off
    #echo_on; echoln * If HALF RESOLUTION FPS is significantly higher than  BASELINE FPS you should reduce the Game Screen Render Quality; #echo_off
    #echo_on; echoln * If REDUCED DRAWING FPS is significantly higher than  BASELINE FPS you should turn off High Quality Water and set Shadow  Quality to Med/Off; #echo_off
    #echo_on; echoln * If PARTICLES DISABLED FPS is higher than BASELINE FPS you should drop Effects Quality; #echo_off
    #echo_on; echoln * If GAME ONLY FPS is significantly higher than BASELINE FPS you should turn off High Quality UI; #echo_off
    #color_default
    
    //// TEST FINISHED
    #color_yellow
    #echo_on
    echoln
    echoln -------------------------------------------------------------
    echoln [DOTA 2 BENCHMARK]  Finished!
    echoln -------------------------------------------------------------
    #color_default
    grep . To save results into ; #color_cyan; grep . \dota 2 beta\game\dota\condumpxxx.txt ; #color_default;
    grep . file, enter ; #color_green; grep . CONDUMP ; #color_default; echoln; // multiple colors on a single line
    grep . To run benchmark again, enter short alias: ; #color_green; grep . BM; #color_default;
    echoln
    #echo_off
    
    alias BM "exec_async benchmark.cfg"
    top_bar_message "[DOTA 2 BENCHMARK]  Finished!" 0
    cl_showfps 0
    dota_spectator_mode 0
    dota_spectator_hero_index ""
    VT0; // Undo Video Tweaks
    disconnect
    sleep 5000
    top_bar_message 0
    
    //// RESTORE CONSOLE SPEW
    log_flags Console +DoNotEcho | grep %
    log_flags Developer DeveloperConsole Panel Panorama PanoramaScript  VScript VScriptDbg VScriptScripts CustomUI CustomGameCache  CustomNetTable -DoNotEcho
    log_flags AnimationGraph AnimationSystem AnimGraphManager AnimResource  Assert "BitBuf Error" BoneSetup Client "Combat Analyzer" CommandLine D3D  Decals Demo DeveloperVerbose DotaGuide DOTAHLTVCamera -DoNotEcho
    log_flags DOTAHLTVDirector DOTA_CHAT DownloadManager EmitSound  EngineInitialization EngineServiceManager "Entity Dump" "Entity Load  Unserialize" "Entity System" Filesystem GameEventSystem GCClient  GlobalState HangWatchdog "HLTV Server" Host -DoNotEcho
    log_flags HostStateManager IME InputService InputSystem InstantReplay  LOADING MaterialSystem MeshSystem ModelCombiner modellib NavMesh  NetworkClientService Networking "Networking Reliable" NetworkP2PService  NetworkServerService -DoNotEcho
    log_flags NetworkService Particles ParticlesLib Physics PostProcessing  PostProcessPipeline RenderPipelineDota RenderPipelineVr RenderService  RenderSystem ResourceSystem SaveRestore SaveRestoreIO Scaleform  "Scaleform IME" ScaleformAS -DoNotEcho
    log_flags ScaleformParse ScaleformScript SceneSystem SchemaSystem  SchemaSystemUtils Server ServerLog SignonState SndEmitterSystem  SndOperators SoundOperatorSystem SoundOpGameSystem SoundSystem  SoundSystemLowLevel SpawnGroup SplitPacket -DoNotEcho
    log_flags SplitScreen Steam SteamDatagramClient SteamDatagramServer  SteamUnifiedMessages ToneMapping ToolGameSimulation TypeManager Vfx  VguiCallQueue VolumetricFog VProf VR WeekendTourney Workshop  WorldRenderer -DoNotEcho
    log_flags ServerLog Workshop +DoNotEcho
    log_flags Console -DoNotEcho | grep %
    
    //// RESTORE CVARS? ( ASSUMING IT SHOULD DO THAT, IT'S ONLY PARTIAL )
    pop_var_values
    
    //// SHOW CONSOLE
    showconsole
    pastebin cfg mirror (always more up to date)

    onedrive zip file mirror including the replay file - updated
    github zip file release including the replay file - currently broken

    Why I believe this demo (Match id 2844309781) is a suitable choice (again, credits goes to jstq for suggesting it):
    - it's recent, official (i.e. complete demo stored on Valve's servers), and easier to get (no need to download it from filesharing services where it might catch golem knows what)
    - it's demanding, since it features Monkey King, Underlord, SF Arcana, and Rubik!
    - it can be used just as easy (without benchmark.cfg), with a few extra commands on the same line:
    dota_spectator_mode 3;dota_spectator_hero_index 2;timedemo_start 102900;timedemo_end 104200;timedemo replays\2844309781

  4. #14
    Basic Member
    Join Date
    Jul 2014
    Posts
    57
    Quote Originally Posted by gdrewb-valve View Post
    That happens even in a locally hosted bot game? You should have full control when you are running the server yourself (such as when you're running a local bot game)?
    https://www.youtube.com/watch?v=3qJBJoUC9h4

    I made a quick video just to show you what I'm doing exactly. Not sure if there is an error on my part or not, but I figured this would help.

  5. #15
    Basic Member
    Join Date
    Feb 2017
    Posts
    1
    Hi guys, I'm new here but I've been playing Dota for ~10 years including DotA 1. I bought my laptop ~4-5 years ago (yea it's pretty old) and it has AMD graphics card (AMD Radeon 6600M and 6700M Series, sorry if it's out-dated and offends your love for cards ) and a onboard card (Intel(R) HD Graphics 3000). I already set AMD card as high performance for Dota 2 but Dota 2 always runs with the Intel one.

    What should I do to force Dota 2 to run with AMD card? And will it be better? I got ~70fps ingame with absolutly no effects except animation portrait, normal map and 100% render, but the problem is after a short time in-game, fps drops to 10-15, makes the game spike af for 2-5sec then returns to normal.

    Should I turn off Intel card's driver to make AMD become the only card that running? I've not tried it because I don't have enough knowledge and I only have 1 laptop for both gaming and working so if it's fucked up, I'll die

  6. #16
    Valve Developer
    Join Date
    Jul 2015
    Posts
    57
    Doofyrob, what's in your autoexec? If you remove it and all of your launch options temporarily does that help?

  7. #17
    Basic Member
    Join Date
    Jun 2012
    Posts
    14
    Quote Originally Posted by gdrewb-valve View Post
    Doofyrob, what's in your autoexec? If you remove it and all of your launch options temporarily does that help?
    I'm having the same "[Console] SV: Cheat command 'exec_async' ignored." issue, only in lobby though, the command works well for me in demo and bot match. I've tried localhosted lobby, removing the autoexec, and launched the game only with -dx9 and -console parameters, still the same problem.

  8. #18
    Valve Developer
    Join Date
    Jul 2015
    Posts
    57
    Quote Originally Posted by MiL View Post
    I'm having the same "[Console] SV: Cheat command 'exec_async' ignored." issue, only in lobby though, the command works well for me in demo and bot match. I've tried localhosted lobby, removing the autoexec, and launched the game only with -dx9 and -console parameters, still the same problem.
    To confirm, you're creating a lobby and going into the advanced options and choosing "Local Host" as the server? Can you then do "sv_cheats 1" in the console?

  9. #19
    Basic Member
    Join Date
    Jul 2014
    Posts
    57
    Quote Originally Posted by gdrewb-valve View Post
    Doofyrob, what's in your autoexec? If you remove it and all of your launch options temporarily does that help?
    I removed all launch options, no change. (meaning no use of the autoexec, but here it is just in case it matters at all: http://pastebin.com/raw/wPwXgTyu )

  10. #20
    Valve Developer
    Join Date
    Jul 2015
    Posts
    57
    Quote Originally Posted by Doofyrob View Post
    I removed all launch options, no change. (meaning no use of the autoexec, but here it is just in case it matters at all: http://pastebin.com/raw/wPwXgTyu )
    autoexec is run automatically on startup so it may still be affecting things. Otherwise, you're creating a lobby and going into the advanced options and choosing "Local Host" as the server?

Tags for this Thread

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •