Crysis..I am getting 27-32FPS 1920x1080...95% settings on VERY HIGH

K

Kraylon

Limp Gawd
Joined
Apr 27, 2006
Messages
135
Hey guys... i know there has been a lot of tak about the SHADER quality being the cuprit for low FPS but i may have to disagree.
I've found the following settings to give me playable FPS and still great Visual Quality. I know some of you may not believe me, but try it out for yourselves. By the beach where you have to destroy the satelite disk, i am getting 27-32FPS with this settings at 1920x1080.

My System:
E6600 C2D OC'd 3.6Ghz
2GB RAM
8800GTX Oc'd 636/2000/1500 using the tool from Gainward's website
X-Fi sound card.
Using Nvidia 169.04 Beta Drivers.
VISTA 32-bit
FRAPS is reporting 27-32FPS

I have every setting on VERY HIGH except for 2 settings
my Physics Setting is on HIGH
my Post Processing setting is on HIGH....everything else VERY HIGH
AA is OFF

If i turn AA on to 2x i get 19-21 FPS

I am not sure what i am losing by puting my physics quality and my post processing quality to HIGH...but with eye candy at 1920x1080 and 27-32FPS... i think i can live without it....

Give it a try... i am curious to see what results other people get.
 
Thats why its called beta, I don't see how their going to finalize their code, stamp out disks, and ship it in 14 days w/o it still being beta quality.
 
so why are the very high settings not avail for me on the demo? rig in sig.
 
I have almost the exact same setup except 4 GB ram and Vista 64. The man speaks the truth. Crysis beta is very playable if you lower post processing. I am not running the beta drivers either.
 
; default of this CVarGroup
= 4

r_DepthOfField=2
r_MotionBlur=3
r_sunshafts=1
r_UseEdgeAA=1
r_Flares=1
r_Coronas=1
r_colorgrading=1
r_GlowScreenMultiplier = 0.2

[1]
r_DepthOfField=0
r_MotionBlur=0
r_Flares=1
r_Coronas=1
r_UseEdgeAA=0
r_sunshafts=0
r_colorgrading=0
r_GlowScreenMultiplier = 0.2

[2]
r_DepthOfField=1
r_sunshafts=0
r_MotionBlur=0
r_UseEdgeAA=0
r_GlowScreenMultiplier = 0.2

[3]
r_DepthOfField=1
r_sunshafts=0
r_MotionBlur=1
r_colorgrading=0
r_GlowScreenMultiplier = 0.2

thats post processing 1 being low 4 high

i changed my very high setting to


r_DepthOfField=2
r_MotionBlur=0
r_sunshafts=1
r_UseEdgeAA=1
r_Flares=1
r_Coronas=1
r_colorgrading=1
r_GlowScreenMultiplier = 0.2

i run at very high settings with aa off and at 1680x1050 and i get 27fps avg. cutting motion blur off helps alot + it gets on my nerves
 
alphaqforever247 said:
u guys talking bout beta multiplayer or the single player demo?
Click to expand...


Talking about Single player demo. Before i turned down the physics and Post processing... i was gttin 14-15FPS with shader on Very High....now that i turned those 2 features down... this is a very viable solution for the DX10 folks out there who want eye candy without feeling like they are watching Crysis on Microsoft Power Point.
 
run the game benchmark for gpu, I was only getting an average of 19fps with those settings, sure it can hit 30's but once you get into areas with high poly count, the fps suffers.

For a good gaming experience, you don't want it to drop that severely when you get into high poly count zones or rough firefights with things exploding etc.

The "slow down" I've noticed at those settings makes me want some of the crack you are smoking thinking it's a good framerate...

also, shaders have more of an impact that post processing. Actually by going to high and removing the motion blur, the "lag" of the game is more apparent, rather than being hidden by blur.

I found dropping shaders, shadows, had more of an impact than dropping physics and post processing, actually I gained nor lost any at all by dropping physics down.
 
This is the truth. I just downloaded Crysis a couple days ago and got it running tonight, prior to the latest NVidia drivers the game was sluggish, downloaded these drivers... and I can play at Very High on everything but a few settings (the others set at High), I found the game actually to be a lot more responsive to drop some on the shadows, post and water (water seems to take so much fps for how little it gives back for visual quality). This is at 1900x1200~

I'm running 4gb of ram & an 8800Ultra. Now I need to go find out the info. to see actual FPS. :D
 
Maybe it is my style of play where I tend to not get into large firefights. I sneak around with the silencer and snipe the North Koreans to death. However, I did get into a fairly large skirmish today with 6 guys at the beach with explosions galore, and slowdown was minimal and very playable. For multiplayer I can see how 24 fps might be deemed unplayable, but for singleplayer, it really is an enjoyable experience to me.
 
newster said:
Maybe it is my style of play where I tend to not get into large firefights. I sneak around with the silencer and snipe the North Koreans to death.
Click to expand...

I noticed im doing the same thing, i dont know if its a good thing or bad thing. Seems like you die really quickly if you get into a firefight so you're forced to sneak around.
 
Slade said:
run the game benchmark for gpu, I was only getting an average of 19fps with those settings, sure it can hit 30's but once you get into areas with high poly count, the fps suffers.

For a good gaming experience, you don't want it to drop that severely when you get into high poly count zones or rough firefights with things exploding etc.

The "slow down" I've noticed at those settings makes me want some of the crack you are smoking thinking it's a good framerate...

also, shaders have more of an impact that post processing. Actually by going to high and removing the motion blur, the "lag" of the game is more apparent, rather than being hidden by blur.

I found dropping shaders, shadows, had more of an impact than dropping physics and post processing, actually I gained nor lost any at all by dropping physics down.
Click to expand...

Going to run the benchmark and see what the average is.... like i said... i was just running fraps...
 
Here are the results of my GPU & CPU crysis benchmark at the settings i mentioned above.
Again this is EVERYTHING on VERY HIGH EXCEPT Physics and Post Processing which is set to HIGH @ 1920x1080 resolution :) from what i understand anything over 20FPS is playable. as far as 27-32FPS, i was seeing those numbers on fraps which i noticed while running the demo fraps was reporting slightly higher numbers than the test itself. In any case with all 95% of the settings on Very HIGH including shaders, i think these are some decent numbers to put down at my resolution.

C:\Program Files\Electronic Arts\Crytek\Crysis SP Demo\Bin32>benchmark_gpu
Running GPU benchmark 1
Results will depend on current system settings
Press any key to continue . . .
Running...
==============================================================
TimeDemo Play Started , (Total Frames: 2000, Recorded Time: 111.86s)
!TimeDemo Run 0 Finished.
Play Time: 91.13s, Average FPS: 21.95
Min FPS: 15.31 at frame 136, Max FPS: 28.33 at frame 1002
Average Tri/Sec: -5738995, Tri/Frame: -261502
Recorded/Played Tris ratio: -3.51
!TimeDemo Run 1 Finished.
Play Time: 87.35s, Average FPS: 22.90
Min FPS: 15.31 at frame 136, Max FPS: 28.33 at frame 1002
Average Tri/Sec: -5834166, Tri/Frame: -254812
Recorded/Played Tris ratio: -3.60
!TimeDemo Run 2 Finished.
Play Time: 87.32s, Average FPS: 22.91
Min FPS: 15.31 at frame 136, Max FPS: 28.52 at frame 1000
Average Tri/Sec: -5845323, Tri/Frame: -255196
Recorded/Played Tris ratio: -3.59
!TimeDemo Run 3 Finished.
Play Time: 87.32s, Average FPS: 22.91
Min FPS: 15.31 at frame 136, Max FPS: 28.52 at frame 1000
Average Tri/Sec: -5879380, Tri/Frame: -256681
Recorded/Played Tris ratio: -3.57
TimeDemo Play Ended, (4 Runs Performed)
==============================================================
Press any key to continue . . .

C:\Program Files\Electronic Arts\Crytek\Crysis SP Demo\Bin32>benchmark_cpu
Running CPU benchmark 1
Results will depend on current system settings
Press any key to continue . . .
Running...
==============================================================
TimeDemo Play Started , (Total Frames: 1500, Recorded Time: 44.62s
!TimeDemo Run 0 Finished.
Play Time: 74.65s, Average FPS: 20.09
Min FPS: 11.90 at frame 819, Max FPS: 29.31 at frame 85
Average Tri/Sec: -20287290, Tri/Frame: -1009657
Recorded/Played Tris ratio: -0.71
!TimeDemo Run 1 Finished.
Play Time: 70.20s, Average FPS: 21.37
Min FPS: 11.90 at frame 819, Max FPS: 29.31 at frame 85
Average Tri/Sec: -21489608, Tri/Frame: -1005723
Recorded/Played Tris ratio: -0.72
!TimeDemo Run 2 Finished.
Play Time: 71.19s, Average FPS: 21.07
Min FPS: 11.90 at frame 819, Max FPS: 29.31 at frame 85
Average Tri/Sec: -21219002, Tri/Frame: -1007122
Recorded/Played Tris ratio: -0.71
!TimeDemo Run 3 Finished.
Play Time: 71.03s, Average FPS: 21.12
Min FPS: 11.90 at frame 819, Max FPS: 29.31 at frame 85
Average Tri/Sec: -21195326, Tri/Frame: -1003612
Recorded/Played Tris ratio: -0.72
TimeDemo Play Ended, (4 Runs Performed)
==============================================================
Press any key to continue . . .
 
Isn't most of the graphical eye candy between DX9 and DX10 all in the post processing?

So when you say you're running it at very high... you're not really are you.
 
You must log in or register to reply here.
Back
Top