Upvote Upvoted 320 Downvote Downvoted
1 2 3 4 5 ⋅⋅ 27
Comanglia's Config / FPS Guide
posted in Customization
31
#31
-3 Frags +

Q: What is the best OS for FPS?
A: Technically speaking Linux, but for most users it's actually Windows 8.1

Windows 8.1 is big poo

Q: What is the best OS for FPS?
A: Technically speaking Linux, but for most users it's actually Windows 8.1

Windows 8.1 is big poo
32
#32
3 Frags +
DamnEasyQ: What is the best OS for FPS?
A: Technically speaking Linux, but for most users it's actually Windows 8.1

Windows 8.1 is big poo

If you don't know what you're doing (or you try to use 1000Hz WMOs). Though I still use windows 7, resource wise though 8.1 uses a good bit less and you would notice a 3-5% boost in performance between the 2 OS's

flyingbuddyHey I was using one if your older fps configs before and decided to try out your fps config for toasters here to see if it makes a difference. I really like it but it seems to be quite blurry. Do you know the command/s to make it less blurry :3

"mat_viewportscale 0.5"

The value for that command is .5 to 1.
.5 means textures will be rendered at half resolution, 1 means full. the closer to 1 you are the most clear the game will look, the closer you are to .5 the less gpu power you would use, which does actually matter for people using their processors "gpu" (I'm unsure of the technical term for it)

[quote=DamnEasy]Q: What is the best OS for FPS?
A: Technically speaking Linux, but for most users it's actually Windows 8.1

Windows 8.1 is big poo[/quote]

If you don't know what you're doing (or you try to use 1000Hz WMOs). Though I still use windows 7, resource wise though 8.1 uses a good bit less and you would notice a 3-5% boost in performance between the 2 OS's

[quote=flyingbuddy]Hey I was using one if your older fps configs before and decided to try out your fps config for toasters here to see if it makes a difference. I really like it but it seems to be quite blurry. Do you know the command/s to make it less blurry :3[/quote]

"mat_viewportscale 0.5"

The value for that command is .5 to 1.
.5 means textures will be rendered at half resolution, 1 means full. the closer to 1 you are the most clear the game will look, the closer you are to .5 the less gpu power you would use, which does actually matter for people using their processors "gpu" (I'm unsure of the technical term for it)
33
#33
1 Frags +

.

.
34
#34
-15 Frags +

is there a way to make dx81 as pretty as dx90 without the fps loss?

is there a way to make dx81 as pretty as dx90 without the fps loss?
35
#35
1 Frags +

r_avglight 0 is blocked now, so idk maybe you want to take it out of your config so that it looks cleaner

r_avglight 0 is blocked now, so idk maybe you want to take it out of your config so that it looks cleaner
36
#36
12 Frags +
seiterseiter1is there a way to make dx81 as pretty as dx90 without the fps loss?

no nigga thats how the fps is gained

[quote=seiterseiter1]is there a way to make dx81 as pretty as dx90 without the fps loss?[/quote]
no nigga thats how the fps is gained
37
#37
0 Frags +

which one nets more fps, dxlevel 80 or 81 ?

which one nets more fps, dxlevel 80 or 81 ?
38
#38
-1 Frags +
emkawhich one nets more fps, dxlevel 80 or 81 ?

In short because I'm bad at explaining :c, dx 81 gets the most fps and makes everything look slightly worse where as dx9 makes everything look better but less fps

[quote=emka]which one nets more fps, dxlevel 80 or 81 ?[/quote]
In short because I'm bad at explaining :c, dx 81 gets the most fps and makes everything look slightly worse where as dx9 makes everything look better but less fps
39
#39
0 Frags +
flyingbuddyemkawhich one nets more fps, dxlevel 80 or 81 ?In short because I'm bad at explaining :c, dx 81 gets the most fps and makes everything look slightly worse where as dx9 makes everything look better but less fps

thats not what he was asking, dxlevel 80 or 81 not 81 or 90.
im not sure but ive heard 81 gives more but someone correct me

[quote=flyingbuddy][quote=emka]which one nets more fps, dxlevel 80 or 81 ?[/quote]
In short because I'm bad at explaining :c, dx 81 gets the most fps and makes everything look slightly worse where as dx9 makes everything look better but less fps[/quote]
thats not what he was asking, [b]dxlevel 80 or 81[/b] not 81 or 90.
im not sure but ive heard 81 gives more but someone correct me
40
#40
1 Frags +

Dxlevel 80 gives slightly more fps but dxlevel 81 is far more stable if I remember correctly.

Dxlevel 80 gives slightly more fps but dxlevel 81 is far more stable if I remember correctly.
41
#41
0 Frags +
cinderDxlevel 80 gives slightly more fps but dxlevel 81 is far more stable if I remember correctly.sevennnflyingbuddyemkawhich one nets more fps, dxlevel 80 or 81 ?In short because I'm bad at explaining :c, dx 81 gets the most fps and makes everything look slightly worse where
as dx9 makes everything look better but less fps
thats not what he was asking, dxlevel 80 or 81 not 81 or 90.
im not sure but ive heard 81 gives more but someone correct me

Whoops, guess I can't read as well haha. Yeah I've heard dx80 has stability issues although I'm not to sure about it. Dx81 seems far better based on what I've heard.

[quote=cinder]Dxlevel 80 gives slightly more fps but dxlevel 81 is far more stable if I remember correctly.[/quote]
[quote=sevennn][quote=flyingbuddy][quote=emka]which one nets more fps, dxlevel 80 or 81 ?[/quote]
In short because I'm bad at explaining :c, dx 81 gets the most fps and makes everything look slightly worse where
as dx9 makes everything look better but less fps[/quote]
thats not what he was asking, [b]dxlevel 80 or 81[/b] not 81 or 90.
im not sure but ive heard 81 gives more but someone correct me[/quote]
Whoops, guess I can't read as well haha. Yeah I've heard dx80 has stability issues although I'm not to sure about it. Dx81 seems far better based on what I've heard.
42
#42
3 Frags +

this is amazing! The only issue I have is when I am scoped in with the sniper rifle I cant see the left and bottom crosshairs, I use this for lateral tracking and it has made it more difficult. Any idea how to fix this? Ive tried dx81,90, and 95 and its the same in all of them. Here is a screenshot of what Im talking about (taken in dx95):

http://i59.tinypic.com/11ils1z.jpg

this is amazing! The only issue I have is when I am scoped in with the sniper rifle I cant see the left and bottom crosshairs, I use this for lateral tracking and it has made it more difficult. Any idea how to fix this? Ive tried dx81,90, and 95 and its the same in all of them. Here is a screenshot of what Im talking about (taken in dx95):

[IMG]http://i59.tinypic.com/11ils1z.jpg[/IMG]
43
#43
0 Frags +
ComangliaMax FPS for Potatoes and Toasters // Also known as WOW LOOK HOW BAD THIS GAME LOOKS
https://dl.dropboxusercontent.com/u/92187841/myeyes.cfg
Launch options "-dxlevel 81 -full -w 640 -h 480 -console -novid" (remove the -dxlevel 81 for better stability with ALT+Tabbing)

I'm running on a laptop toaster but I can run the game at 1280x720 so can I bump the launch options up to that without everything messing up?
[quote=Comanglia]Max FPS for Potatoes and Toasters // Also known as WOW LOOK HOW BAD THIS GAME LOOKS
https://dl.dropboxusercontent.com/u/92187841/myeyes.cfg
Launch options "-dxlevel 81 -full -w 640 -h 480 -console -novid" (remove the -dxlevel 81 for better stability with ALT+Tabbing)



I'm running on a laptop toaster but I can run the game at 1280x720 so can I bump the launch options up to that without everything messing up?
44
#44
1 Frags +

Just change -w 640 -h 480 to -w 1280 -h 720

Just change -w 640 -h 480 to -w 1280 -h 720
45
#45
0 Frags +
Widespread_Panicthis is amazing! The only issue I have is when I am scoped in with the sniper rifle I cant see the left and bottom crosshairs, I use this for lateral tracking and it has made it more difficult. Any idea how to fix this? Ive tried dx81,90, and 95 and its the same in all of them. Here is a screenshot of what Im talking about (taken in dx95):

[IMG]http://i59.tinypic.com/11ils1z.jpg[/IMG]

mat_picmip 2 causes those lines to disappear
have "mat_picmip -1" if you want those 2 lines to stay, (will also make a lot of other texture look better/more defined as well)

cinderDxlevel 80 gives slightly more fps but dxlevel 81 is far more stable if I remember correctly.

the difference in fps is extremely minimal, the biggest difference is more to do with certain particles don't render/render correctly on dxlevel 80

[quote=Widespread_Panic]this is amazing! The only issue I have is when I am scoped in with the sniper rifle I cant see the left and bottom crosshairs, I use this for lateral tracking and it has made it more difficult. Any idea how to fix this? Ive tried dx81,90, and 95 and its the same in all of them. Here is a screenshot of what Im talking about (taken in dx95):

[IMG]http://i59.tinypic.com/11ils1z.jpg[/IMG][/quote]

mat_picmip 2 causes those lines to disappear
have "mat_picmip -1" if you want those 2 lines to stay, (will also make a lot of other texture look better/more defined as well)

[quote=cinder]Dxlevel 80 gives slightly more fps but dxlevel 81 is far more stable if I remember correctly.[/quote]

the difference in fps is extremely minimal, the biggest difference is more to do with certain particles don't render/render correctly on dxlevel 80
46
#46
0 Frags +

another question, Im currently using an i5 3570k at stock 3.4GHz. Since TF2 is so cpu dependent, what kind of performance increase could I expect from a mild overclock to 4.0?

another question, Im currently using an i5 3570k at stock 3.4GHz. Since TF2 is so cpu dependent, what kind of performance increase could I expect from a mild overclock to 4.0?
47
#47
0 Frags +

Just because I'm curious, what exactly do you feel I messed up with in my config? And is it fixed now?

Just because I'm curious, what exactly do you feel I messed up with in my config? And is it fixed now?
48
#48
1 Frags +
RhapsodyJust because I'm curious, what exactly do you feel I messed up with in my config? And is it fixed now?

You're good now. Everything that I disagree with your config now is largely subjective and not really a big deal.

[quote=Rhapsody]Just because I'm curious, what exactly do you feel I messed up with in my config? And is it fixed now?[/quote]

You're good now. Everything that I disagree with your config now is largely subjective and not really a big deal.
49
#49
22 Frags +
r_occlusion 0 // disables ambient occlusion. 2% FPS boost

No.

r_occlusion 0 disables the occlusion system, which allows mappers to manually optimize their maps via the visibility system. If you were testing the effect of this cvar on a smaller, older, or simpler map, that's probably why you were losing fps. On any moderately complex map, you will gain fps by having this enabled.

sys_minidumpspewlines 500 // Basically the number of lines saved to a log file from console. No FPS boost but 15% lower FPS variance

Again, way off. This controls how many lines of the console are saved to the minidump created when your game crashes. This has no effect on the fps you get, and the only effect of increasing or decreasing it would be changing the amount of time it takes for your game to crash by a few microseconds.

mat_antialias 0
mat_forceaniso 0

Don't set these to 0. The minimum value of these cvars is 1, and setting them to 0 will just cause a material system reload = breaking mumble overlay/increasing game startup time. Think about it: mat_antialias controls the number of samples taken per pixel on your screen. If you took 0 samples per pixel, your screen would be entirely black :P
Reading directly from the game's code, the same hardcoded minimum exists for mat_forceaniso:

pConfig->m_nForceAnisotropicLevel = max( mat_forceaniso.GetInt(), 1 );
mat_max_worldmesh_vertices 512

Minimum value for this in code is 1024. Additionally, lowering this value increases how many chunks world geometry is broken up into, leading to more draw calls = more cpu usage. No reason to lower this from its initial 65535.

That's all I have for now.

[code]r_occlusion 0 // disables ambient occlusion. 2% FPS boost[/code]

No.

r_occlusion 0 disables the occlusion system, which allows mappers to manually optimize their maps via the visibility system. If you were testing the effect of this cvar on a smaller, older, or simpler map, that's probably why you were losing fps. On any moderately complex map, you will gain fps by having this enabled.

[code]sys_minidumpspewlines 500 // Basically the number of lines saved to a log file from console. No FPS boost but 15% lower FPS variance[/code]

Again, way off. This controls how many lines of the console are saved to the minidump created when your game crashes. This has no effect on the fps you get, and the only effect of increasing or decreasing it would be changing the amount of time it takes for your game to crash by a few [i]microseconds[/i].

[code]mat_antialias 0
mat_forceaniso 0[/code]

Don't set these to 0. The minimum value of these cvars is 1, and setting them to 0 will just cause a material system reload = breaking mumble overlay/increasing game startup time. Think about it: mat_antialias controls the number of samples taken per pixel on your screen. If you took 0 samples per pixel, your screen would be entirely black :P
Reading directly from the game's code, the same hardcoded minimum exists for mat_forceaniso:
[code]pConfig->m_nForceAnisotropicLevel = max( mat_forceaniso.GetInt(), 1 );[/code]

[code]mat_max_worldmesh_vertices 512[/code]

Minimum value for this in code is 1024. Additionally, lowering this value increases how many chunks world geometry is broken up into, leading to more draw calls = more cpu usage. No reason to lower this from its initial 65535.

That's all I have for now.
50
#50
7 Frags +

Youi should make a fps config! you seem to know exactly what your talking about and i would be willing to test it for you :)

Youi should make a fps config! you seem to know exactly what your talking about and i would be willing to test it for you :)
51
#51
7 Frags +
seiterseiter1Youi should make a fps config! you seem to know exactly what your talking about and i would be willing to test it for you :)

Been thinking about it, plus a plugin or two... looking forward to a free summer for once :P

[quote=seiterseiter1]Youi should make a fps config! you seem to know exactly what your talking about and i would be willing to test it for you :)[/quote]

Been thinking about it, plus a plugin or two... looking forward to a free summer for once :P
52
#52
0 Frags +

is it possible to make botkiller weapon head be shaky when playing on dx8?

is it possible to make botkiller weapon head be shaky when playing on dx8?
53
#53
1 Frags +
Forsak3nis it possible to make botkiller weapon head be shaky when playing on dx8?

set cl_jiggle_bone_framerate_cutoff to 1

[quote=Forsak3n]is it possible to make botkiller weapon head be shaky when playing on dx8?[/quote]
set cl_jiggle_bone_framerate_cutoff to 1
54
#54
5 Frags +
pazer
r_occlusion 0 // disables ambient occlusion. 2% FPS boost

No.

r_occlusion 0 disables the occlusion system, which allows mappers to manually optimize their maps via the visibility system. If you were testing the effect of this cvar on a smaller, older, or simpler map, that's probably why you were losing fps. On any moderately complex map, you will gain fps by having this enabled.
sys_minidumpspewlines 500 // Basically the number of lines saved to a log file from console. No FPS boost but 15% lower FPS variance

Again, way off. This controls how many lines of the console are saved to the minidump created when your game crashes. This has no effect on the fps you get, and the only effect of increasing or decreasing it would be changing the amount of time it takes for your game to crash by a few microseconds.
mat_antialias 0
mat_forceaniso 0

Don't set these to 0. The minimum value of these cvars is 1, and setting them to 0 will just cause a material system reload = breaking mumble overlay/increasing game startup time. Think about it: mat_antialias controls the number of samples taken per pixel on your screen. If you took 0 samples per pixel, your screen would be entirely black :P
Reading directly from the game's code, the same hardcoded minimum exists for mat_forceaniso:
pConfig->m_nForceAnisotropicLevel = max( mat_forceaniso.GetInt(), 1 );
mat_max_worldmesh_vertices 512

Minimum value for this in code is 1024. Additionally, lowering this value increases how many chunks world geometry is broken up into, leading to more draw calls = more cpu usage. No reason to lower this from its initial 65535.

That's all I have for now.

Interesting

r_occlusion I must admit the only way I ever noticed a difference was the output into consol from doing timedemos, and the timedemo I use is on dustbowl, on the last point and the camera remains pretty much static the entire time.

sys_minidumpspewlines 500 Alright I was wrong on what it does, but the FPS variance DOES go down by setting the 500 to 50 or lower. and 500 is the default value.

mat_antialias 0
mat_forceaniso 0

You're right.

mat_max_worldmesh_vertices 512
I've always been unaware of exactly what this did, and every config I looked at had this significantly below default and usually was set to 512. In the time demo tests I did though I never did see a difference (visual or fps) between 512, 2048, 4096, or even the default 65535 so I assumed setting it to 512 didn't hurt anything.

[quote=pazer][code]r_occlusion 0 // disables ambient occlusion. 2% FPS boost[/code]

No.

r_occlusion 0 disables the occlusion system, which allows mappers to manually optimize their maps via the visibility system. If you were testing the effect of this cvar on a smaller, older, or simpler map, that's probably why you were losing fps. On any moderately complex map, you will gain fps by having this enabled.

[code]sys_minidumpspewlines 500 // Basically the number of lines saved to a log file from console. No FPS boost but 15% lower FPS variance[/code]

Again, way off. This controls how many lines of the console are saved to the minidump created when your game crashes. This has no effect on the fps you get, and the only effect of increasing or decreasing it would be changing the amount of time it takes for your game to crash by a few [i]microseconds[/i].

[code]mat_antialias 0
mat_forceaniso 0[/code]

Don't set these to 0. The minimum value of these cvars is 1, and setting them to 0 will just cause a material system reload = breaking mumble overlay/increasing game startup time. Think about it: mat_antialias controls the number of samples taken per pixel on your screen. If you took 0 samples per pixel, your screen would be entirely black :P
Reading directly from the game's code, the same hardcoded minimum exists for mat_forceaniso:
[code]pConfig->m_nForceAnisotropicLevel = max( mat_forceaniso.GetInt(), 1 );[/code]

[code]mat_max_worldmesh_vertices 512[/code]

Minimum value for this in code is 1024. Additionally, lowering this value increases how many chunks world geometry is broken up into, leading to more draw calls = more cpu usage. No reason to lower this from its initial 65535.

That's all I have for now.[/quote]

Interesting

[b]r_occlusion[/b] I must admit the only way I ever noticed a difference was the output into consol from doing timedemos, and the timedemo I use is on dustbowl, on the last point and the camera remains pretty much static the entire time.

[b]sys_minidumpspewlines 500[/b] Alright I was wrong on what it does, but the FPS variance DOES go down by setting the 500 to 50 or lower. and 500 is the default value.

[b]mat_antialias 0
mat_forceaniso 0[/b]

You're right.

[b]mat_max_worldmesh_vertices 512[/b]
I've always been unaware of exactly what this did, and every config I looked at had this significantly below default and usually was set to 512. In the time demo tests I did though I never did see a difference (visual or fps) between 512, 2048, 4096, or even the default 65535 so I assumed setting it to 512 didn't hurt anything.
55
#55
6 Frags +

Will you be updating the OP soon then in light of all this, Comanglia? Thanks!

Will you be updating the OP soon then in light of all this, Comanglia? Thanks!
56
#56
3 Frags +
pazer
r_occlusion 0 // disables ambient occlusion. 2% FPS boost

No.

r_occlusion 0 disables the occlusion system, which allows mappers to manually optimize their maps via the visibility system. If you were testing the effect of this cvar on a smaller, older, or simpler map, that's probably why you were losing fps. On any moderately complex map, you will gain fps by having this enabled.
sys_minidumpspewlines 500 // Basically the number of lines saved to a log file from console. No FPS boost but 15% lower FPS variance

Again, way off. This controls how many lines of the console are saved to the minidump created when your game crashes. This has no effect on the fps you get, and the only effect of increasing or decreasing it would be changing the amount of time it takes for your game to crash by a few microseconds.
mat_antialias 0
mat_forceaniso 0

Don't set these to 0. The minimum value of these cvars is 1, and setting them to 0 will just cause a material system reload = breaking mumble overlay/increasing game startup time. Think about it: mat_antialias controls the number of samples taken per pixel on your screen. If you took 0 samples per pixel, your screen would be entirely black :P
Reading directly from the game's code, the same hardcoded minimum exists for mat_forceaniso:
pConfig->m_nForceAnisotropicLevel = max( mat_forceaniso.GetInt(), 1 );
mat_max_worldmesh_vertices 512

Minimum value for this in code is 1024. Additionally, lowering this value increases how many chunks world geometry is broken up into, leading to more draw calls = more cpu usage. No reason to lower this from its initial 65535.

That's all I have for now.

Updated my config too, thanks for the info!

[quote=pazer][code]r_occlusion 0 // disables ambient occlusion. 2% FPS boost[/code]

No.

r_occlusion 0 disables the occlusion system, which allows mappers to manually optimize their maps via the visibility system. If you were testing the effect of this cvar on a smaller, older, or simpler map, that's probably why you were losing fps. On any moderately complex map, you will gain fps by having this enabled.

[code]sys_minidumpspewlines 500 // Basically the number of lines saved to a log file from console. No FPS boost but 15% lower FPS variance[/code]

Again, way off. This controls how many lines of the console are saved to the minidump created when your game crashes. This has no effect on the fps you get, and the only effect of increasing or decreasing it would be changing the amount of time it takes for your game to crash by a few [i]microseconds[/i].

[code]mat_antialias 0
mat_forceaniso 0[/code]

Don't set these to 0. The minimum value of these cvars is 1, and setting them to 0 will just cause a material system reload = breaking mumble overlay/increasing game startup time. Think about it: mat_antialias controls the number of samples taken per pixel on your screen. If you took 0 samples per pixel, your screen would be entirely black :P
Reading directly from the game's code, the same hardcoded minimum exists for mat_forceaniso:
[code]pConfig->m_nForceAnisotropicLevel = max( mat_forceaniso.GetInt(), 1 );[/code]

[code]mat_max_worldmesh_vertices 512[/code]

Minimum value for this in code is 1024. Additionally, lowering this value increases how many chunks world geometry is broken up into, leading to more draw calls = more cpu usage. No reason to lower this from its initial 65535.

That's all I have for now.[/quote]

Updated my config too, thanks for the info!
57
#57
-3 Frags +

I just personally want to say, that you should take a long, hard look at m0re's fps config again. Despite yours giving me higher FPS, m0re's is making the game a lot smoother for me.

I don't think you should acually measure everything based on FPS, since FPS doesn't mean smoothness in all cases.

I just personally want to say, that you should take a long, hard look at m0re's fps config again. Despite yours giving me higher FPS, m0re's is making the game a lot smoother for me.

I don't think you should acually measure everything based on FPS, since FPS doesn't mean smoothness in all cases.
58
#58
0 Frags +

I'm using the max fps for better computers and I have a problem I'd like to issue. (If anyone has any solutions)
You see if you play Mann-Power when you shoot your hook the usual grapple line is not visable and I wondering if this could be fixed?

I'm using the max fps for better computers and I have a problem I'd like to issue. (If anyone has any solutions)
You see if you play Mann-Power when you shoot your hook the usual grapple line is not visable and I wondering if this could be fixed?
59
#59
1 Frags +
MubbyI'm using the max fps for better computers and I have a problem I'd like to issue. (If anyone has any solutions)
You see if you play Mann-Power when you shoot your hook the usual grapple line is not visable and I wondering if this could be fixed?

Ropes are turned off in this config.

[quote=Mubby]I'm using the max fps for better computers and I have a problem I'd like to issue. (If anyone has any solutions)
You see if you play Mann-Power when you shoot your hook the usual grapple line is not visable and I wondering if this could be fixed?[/quote]

Ropes are turned off in this config.
60
#60
1 Frags +
MubbyI'm using the max fps for better computers and I have a problem I'd like to issue. (If anyone has any solutions)
You see if you play Mann-Power when you shoot your hook the usual grapple line is not visable and I wondering if this could be fixed?

change rope_rendersolid to "1"

[quote=Mubby]I'm using the max fps for better computers and I have a problem I'd like to issue. (If anyone has any solutions)
You see if you play Mann-Power when you shoot your hook the usual grapple line is not visable and I wondering if this could be fixed?[/quote]

change rope_rendersolid to "1"
1 2 3 4 5 ⋅⋅ 27
Please sign in through STEAM to post a comment.