[360, PS3] Crysis 2

Does anybody have a clue on which part controls the Virtual Point Lights? I want to play with the Global Illumination once the PC demo arrives :)
 
The PC config has 'e_GINumCascades', but other than that there doesn't seem to be any control in the cvars.

WOW, great to know that there is something to tweak at least. What numerical value is listed for the number of cascades? Crytek used 3 cascades in their tech document. See Figure 4 for a graphical representation. I hope there is also some way to change the grid resolution.

http://www.crytek.com/sites/default/files/20100301_lpv.pdf
 
Last edited by a moderator:
Hmm, just downloaded the demo, can't say I'm impressed - I've played 5 matches and they game has crashed twice crashed - this is a demo? Or an alpha?

The graphics aren't great, technically they might be but I see no point in having real time global lighting & shadowing when there's ridiculous amounts of shadow (and other object) pop-in everywhere. Everything also looks washed out for some reason which makes it look generic and the weapon effects are poor, grenade explosions especially look very low res and weak.

The gameplay itself is also pretty generic, the shooting itself isn't much fun, there's something iffy about the aiming too, very inconsistent. The controls are also strange, they tend to accelerate too quickly if you move the stick too much causing your view to snap off to the side.

I highly doubt people are going to stop playing their BFBC, COD, Halo & KZ to play this instead.
 
Last edited by a moderator:
I saw there was around 540,000 on the leaderboards so far.

I think it's gonna be pretty popular. Crytek put massive effort into this.
 
Another thing:
r_Height = 704
on the ps3 code. It means 1280x704p it's real resolution?

If the measured res for the 360 is correct then the real resolution should come from somewhere else in the files. Perhaps that height means the PS3 would scale whatever it's real resolution is to a 1280x704 window (8 pixels top and bottom of the output)?

I hope to play this tomorrow, had to wait for bandwisth cap to reset. Though I understand the demo will eventually be taken down sometime?
 
Last edited by a moderator:
If the measured res for the 360 is correct then the real resolution should come from somewhere else in the files.
Or it could be hardcoded into the 360 build. That would also explain all the missing variables (like the settings for anamorphic flares, particle refraction, etc).
 
Looks like 360/PS3 version is what would be classed as medium for Crysis 2 if going by class numeration since 2 is medium (possibly modified sub category values), 3 high and 4 very high in Crysis/Warhead.

Also in comparision it ends up from low to very high with most on medium ballpark if compared to Crysis default configs.

Btw a lot in there matches pretty good with my assimilation config for DF article. :cigarr:

Use program to do text comparision, PC Crysis medium settings in same matching cvar layout as the Crysis 2 config file. Though cant call 100% equal or such even if same numbers for PC first Crysis game vs console cvars as it might be better/worse effect depeding on tradeoffs or optimisations and improvements done to CE3.

Code:
--sys_spec_ObjectDetail=2

e_lod_ratio=4
e_view_dist_ratio_detail=19
e_vegetation_min_size=0.5 
e_decals_life_time_scale=1
e_max_view_dst_spec_lerp=0.5
es_DebrisLifetimeScale=0.6
e_cbuffer_resolution=128
ca_AttachmentCullingRation=120
e_terrain_occlusion_culling_max_dist=130

--sys_spec_Shading=2
q_ShaderGeneral=1 
q_ShaderMetal=1
q_ShaderGlass=1
q_ShaderVegetation=1
q_ShaderIce=1
q_ShaderTerrain=1
q_ShaderShadow=1
q_ShaderFX=1
q_ShaderPostProcess=1
q_ShaderHDR=1
q_ShaderSky=1
q_Renderer=1
q_ShaderWater=1


r_SSAO=0
e_terrain_normal_map=0
r_EnvTexUpdateInterval=0.075
r_TexturesFilteringQuality=1

--sys_spec_VolumetricEffects=2

r_BeamsDistFactor=0.5
r_BeamsMaxSlices=32

--sys_spec_Shadows=2

-
r_ShadowJittering=1
e_shadows_cast_view_dist_ratio=0.45
e_shadows_on_alpha_blended=0
e_gsm_cache = 1

--sys_spec_Texture=2

r_ImposterRatio=1.5
r_DynTexMaxSize=60
r_DynTexAtlasCloudsMaxSize=24
r_DynTexAtlasSpritesMaxSize=16

--sys_spec_Physics=2

e_cull_veg_activation=30
es_MaxPhysDist=100
es_MaxPhysDistInvisible=15
e_phys_ocean_cell=1
e_foliage_wind_activation_dist=10
p_gravity_z=-9.8
-
-

--sys_spec_PostProcessing=2

r_DepthOfField=1

--sys_spec_Particles=2

-
-
e_particles_quality=2
e_particles_max_emitter_draw_screen=8 (bigger particle limit)
e_particles_min_draw_pixels=1 (lower better)

--sys_spec_GameEffects=2

g_ragdollMinTime = 8.0
g_ragdollDistance = 20.0
g_battleDust_enable = 0

p_max_substeps=5
-------------------
sys_flash_curve_tess_error=4
r_HDRRendering=0
-
-
r_HDRBrightOffset=6
r_HDRBrightThreshold=3 
r_HDRLevel=0.6
-
r_EyeAdaptionBase=0
r_EyeAdaptationFactor=0.5
r_EnvTexResolution=2 (lower worse)
r_TexSkyResolution=0
r_DetailNumLayers=1 
r_DetailDistance=4

r_TexBumpResolution=0
r_TexResolution=0
r_TexturesStreaming=0
-
r_TexturesStreamPoolSize=256
r_ShadersUserFolder=1
r_GeomInstancing=1
r_refraction=1
r_ShadowBlur=0
- No SSAO under medium settings
- 
-

-
-
r_WaterReflections=1
r_WaterUpdateFactor=0.1 (lower more updates per second)
e_water_tesselation_amount=6 
e_water_tesselation_swath_width=10
r_WaterUpdateDistance=1
r_WaterCaustics=1
-
r_WaterReflectionsQuality=1 
e_water_ocean_fft=0 
r_WaterReflectionsMinVisiblePixelsUpdate=0.05
-
r_WaterRefractions=0
-
r_WaterUpdateFactor=0.1 (lower better)
r_WaterReflectionsQuality=1
-
-
e_level_auto_precache_textures_and_shaders=0
e_shadows_slope_bias=4
e_shadows_max_texture_size=512
e_vegetation_sprites_min_distance=8 (lower worse)
-
e_vegetation_sprites_distance_custom_ratio_min=0.75 (draw distance mult, higher better)
-
e_gsm_range=3
e_gsm_range_step=3
e_gsm_lods_num=4
-
-
e_vegetation_bending=2
e_clouds=1
-
-
-
e_terrain_ao=0
e_dissolve=0
r_UsePOM=0
e_sky_update_rate=0.5 (higher better)
-
-
e_proc_vegetation=1
e_lod_min_tris=300
e_vegetation_use_terrain_color=1 (?, might be 0)
e_view_dist_ratio=60
e_view_dist_min=0 
e_view_dist_ratio_vegetation=21 
e_obj_quality=2
e_lod_min=1 (lower better)
-
e_vegetation_sprites_distance_ratio=1
-
-
r_PostProcessEffects=1
r_UseEdgeAA=0
r_MotionBlur=1
r_SunShafts=0
r_Glow=1
r_Beams=3 (3 means optimised volume beams)
r_Coronas=1 
- 
r_DepthOfField=1 (higher better)
r_ColorGrading=0 	
-
r_Rain=1 (higher better)
-
r_CustomVisions=1
-
-
-
-
-
-
-
-
e_dynamic_light_frame_id_vis_test=1
e_decals_merge=1
e_decals_hit_cache=1
e_decals_neighbor_max_life_time=4
e_decals_overlapping=0
e_terrain_lod_ratio=1 (lower better)
-
-
-

e_shadows_cast_view_dist_ratio=0.45
e_gsm_cache_lod_offset=3 (higher better)
-
-
-
-
p_max_MC_iters=5000 (higher better)
-
g_breakage_particles_limit=130
-
-
p_max_entity_cells=300000
-
 
Last edited by a moderator:
Nebula it's often been said Crysis 2 on 360 looks like a mix of settings anywhere from Low to very high compared to Crysis PC, what settings would you say are "low"?

Too me there doesnt seem to be anything really bad aka low, except maybe some of the pop in/LOD.

I also just noticed in a vid and then tried for myself you can still shoot down the few trees in the Skyline level (except for a couple in containers), the old signature Crysis effect, and AFAIK not really replicated on any other console (or PC) game besides something dedicated to it like Red Faction. Pretty cool.

But yeah I'm really liking the game. I'm going to make a serious effort at this in multiplayer when it comes out. I've barely touched the Bulletstorm demo because of this.

Also RE the PS3 version, people seem to be expecting the worse but somehow I dont think Crytek is just going to push out a piece of garbage there. They have 600 employees and this is one of the highest production values games yet seen, so again it would just be weird if the PS3 version was terrible. I'm half expecting to be pleasantly surprised.
 
Last edited by a moderator:
Download and played for a while, reached rank 5 fairly quick :) Gameplay is quite enjoyable. It is very similar to Call of Duty in concept but I think there's enough to seperate it.

Graphically in spite of the complaints of image quality it's not a major issue. Pop in is far more noticable. I also see some physics bugs like a body being impaled through a wall (legs one side, head the other), and other bodies landing in generally strange places. Also saw a pot acting like it was a tree branch, swirling and swaying with the wind.

The things I don't like is the menu UI (too dark, too many different colors, weird design), and I had to raise the default brightness because the gameplay is rather dark.

Overall it left a positive impression on me, definitely on my radar now.
 
Roughly low or quite below medium regarding these settings. Value in () is Crysis low value. Some of them have major perfomance impact. Doesn't mean it's horrible though. For example reflection update rate will be for nice looking reflections while low reflection quality is simplistic and less for medium but still.

r_DynTexAtlasCloudsMaxSize = 8 (24)
r_DynTexAtlasSpritesMaxSize = 8 (16)

e_water_tesselation_amount=1 (vs 360, PS3 has 6 like medium going by config) (1)
Set tesselation amount

r_WaterUpdateFactor = 0.15 (low 0.5, medium 0.1, the 0.01 is for very high, not global default)
Distance factor for water reflected texture updating.
Usage: r_WaterUpdateFactor 0.01
Default is 0.01. 0 means update every frame

r_WaterUpdateFactor=0.25 (low 0.5, medium 0.1, the 0.01 is for very high, not global default)
Distance factor for water reflected texture updating.
Usage: r_WaterUpdateFactor 0.01
Default is 0.01. 0 means update every frame

e_SkyUpdateRate=0.2 (low 0.12, medium 0.5)
Percentage of a full dynamic sky update calculated per frame (0..100]

e_ViewDistRatio=25 (40)
View distance ratio for objects

e_ViewDistRatioVegetation=12 (15)
View distance ratio for vegetation

r_beams = 0 (4, think it is single layer beams)
Toggles light beams.
Usage: r_Beams [0/1/2/3]
Default is 3 (optimized beams with glow support). Set to 0 to disable beams or 2 to
use fake beams. Set 1 for real beams, full resolution (slower). Set to 3 to use
optimized and with glow support beams.


r_coronas = 0 (1, might be moot in CE3 with deffered renderer, dunno)
Toggles light coronas around light sources.
Usage: r_Coronas [0/1]Default is 1 (on).

e_DecalsNeighborMaxLifeTime=1 (4)
If not zero - new decals will force old decals to fade in X seconds

g_breakage_particles_limit = 40 (80)
Imposes a limit on particles generated during 2d surfaces breaking

p_max_mc_iters = 2000 (4000, physics)
Specifies the maximum number of microcontact solver iterations

e_GsmCacheLodOffset=2 (no shadows in low but medium has a value of 3. Controls which shadow LODs have same updaterate as global framerate, 2 means LOD 0+1+2 have full and the rest after that probably 15fps.)
Makes first X GSM lods not cached

e_TerrainLodRatio = 2 (1, lower is better, 0 is disabled terrain LOD)
Set heightmap LOD
 
Can we really compare CVAR values between two different engines though??

Thx for all that work, but I was thinking just in more general terms not specific cvars, like what general aspects look like low Crysis PC quality to you? But that CVAR table is still interesting.

e_water_tesselation_amount=1 (vs 360, PS3 has 6 like medium going by config) (1)

Wouldn't that be odd since I thought Xenos had tesselation and RSX didn't?

I should also say this game sounds really good through 5.1, with really punchy bass. As expected for a audiovisual showcase title from crytek. I've been dreaming for a while now of getting that Crysis 2 disc, popping it in, turning the lights down and the HTIB up :)
 
Last edited by a moderator:
Can we really compare CVAR values between two different engines though??

Thx for all that work, but I was thinking just in more general terms not specific cvars, like what general aspects look like low Crysis PC quality to you? But that CVAR table is still interesting.

Several of them yes but some are trickier. multipliers are just that, multipliers and multiplies a set value which can differ greatly. GSM cache settings, updates are of same metric.

Then one can speculate about complexity like for example the terrain LOD. However being in a city I doubt terrain geometry would be greatly affected visually like it would in Crysis since Crysis has far more deformed terrain (exclude debris as that is not part of terrain). In Crysis disabling terrain LOD would make terrain and mountains have full LOD upping geometry per frame with 4-6m or more. In Crysis 2 it probably wont even hit half a million with LOD off.


Wouldn't that be odd since I thought Xenos had tesselation and RSX didn't?

See since it has a value of 1 it might be becouse that tesselation method is not used and instead the Xenos tesselator is used thus a value of 1. I doubt tesselation on loaded tesselation 'blocks' would end up being correct. I assume PS3 relies then on same method as Crysis game did for water tesselation and benefits from higher value but might still have less complex water.

I should also say this game sounds really good through 5.1, with really punchy bass. As expected for a audiovisual showcase title from crytek.

Yeah. :smile:
 
Back
Top