Crysis 1.1 patch is OUT with Multi GPU updates!

I'm hoping for some ATI love from crytec. I don't think x1800 series is running the game as well as a 7800's. I just noticed this from a friend who has the card and has been playing the game too, shader seem to take less of a hit on his card. Anyone know of differences between ati and nvidia's older cards and crysis? I could be completely off he has a better cpu : p
 
imonthemotherfucker.jpg
 
Installing now. WIll report back.

Keep it real Jewels!

Direct Links...

Information: http://www.crymod.com/filebase.php?fileid=636

Download: http://www.crymod.com/filebase_redirect.php?fid=2137

Crytek GmbH Copyright
======================================
Software and documentation © 2007 Crytek GmbH.
ALL RIGHTS RESERVED.

==================================================================
Crysis(R)
Patch 1.1 Change List - November, 2007
==================================================================
==================================================================
Fixes:
==================================================================
*Fixed: Potential crash in D3D10
*Fixed: Orange boxes apearing when hispec savegame loaded into lowspec game.
*Fixed: Inconsistent damage dealt to vehicles when shot by LAW.
*Fixed: Reflection resolution on D3D10, MultiGPU reflection update fix
*Fixed: Memory leak with FSAA modes
*Fixed: Infinite ammo hacks.
*Fixed: Memory leak in D3D10 when switching screen modes
*Fixed: Muti optimizations
*Fixed: When player melees during gun raise animation, their gun will be in a permanantly raised position.
*Fixed: Crash when loading savegame with level exported recently by editor

*Fixed: Virtual keyboard does not function properly when a game pad is connected
*Fixed: Users can lose the ability to look around with the Right Stick
*Fixed: Setting screen resolution to "default" stops user from selecting last resolution
*Fixed: Bug when changing resolution in D3D10
*Fixed: Issues with Depth of field and water droplets in D3D10
*Fixed: Crash on NaN warning
==================================================================
Updates:
==================================================================
*Added: Motion Blur UI and V.SYNC UI options
*Optimized: Motion blur
*Optimized: FSAA (Full Scene Anti-Aliasing)
*Optimized sound id implementation
*Enabled VSync functionality in D3D10
*New benchmarking files for ice CPU benchmark.
*http/xmlrpc password protected remote control session
*Marked debug cvars as cheat
*F12 (screenshot) now works in restricted mode as well
==================================================================
Tweaks:
==================================================================
*Reduced LAW splash damage vs. infantry in PowerStruggle mode
*Slowed Rocket projectile speed down in MP slightly
*Disabled automatic turret bounding boxes on vehicles to prevent issues with LAW hit detection
*Reduced grenade explosion radius in multiplayer
*Clamped water tesselation to avoid cheating in MP
*MultiGPU improvements with depth map updates
 
great. I will test it as well to see what, if any, performance gains there are for single cards
 
Hartz said:
I'm hoping for some ATI love from crytec. I don't think x1800 series is running the game as well as a 7800's. I just noticed this from a friend who has the card and has been playing the game too, shader seem to take less of a hit on his card. Anyone know of differences between ati and nvidia's older cards and crysis? I could be completely off he has a better cpu : p
Click to expand...

which x1800 and which 7800
 
No performance increase for me.

dx10 all high, 1280x1024, 2x AA.

Kind of odd though, with Vsync enabled, my framerate never goes below 20, or above 30 while in game.
 
Performance is better for me.
The game is now very playable at high settings (shadows and motion blur - medium) at 1024x768 on my system.
 
performance went down in the gpu benchmark for me. running at 1024x768 all medium settings DX10. :(

before patch....31.02 fps
after patch ......29.80 fps
 
Note: If you get an error about missing files in the CAB or somesuch while installing the patch then you need to put the intro movies back before installing. :rolleyes:

I tested at 1600x1200 2x AA all "High" settings using GPU benchmark - DX9 on XP - 8800GTS 512.

Before patch (@ 670/970):
Code: 
Min FPS: 11.79 at frame 148, Max FPS: 30.22 at frame 997
Run #1- DX9 1600x1200 AA=2x, 32 bit test, Quality: High ~ Average FPS: 26.00

After 1.1 patch (@ 670/970):
Code: 
Min FPS: 15.90 at frame 153, Max FPS: 31.47 at frame 989
Run #1- DX9 1600x1200 AA=2x, 32 bit test, Quality: High ~ Average FPS: 27.14

After 1.1 patch (@ 750/1100):
Code: 
Min FPS: 15.31 at frame 142, Max FPS: 35.49 at frame 1004
Run #1- DX9 1600x1200 AA=2x, 32 bit test, Quality: High ~ Average FPS: 29.15

Also note I only ran this test group once so it could vary a little here or there.
 
For me, the performance change was negligible.

1.0 - Run #1- DX9 1280x1024 AA=No AA, 32 bit test, Quality: Medium ~~ Last Average FPS: 43.48
1.1 - Run #1- DX9 1280x1024 AA=No AA, 32 bit test, Quality: Medium ~~ Last Average FPS: 44.32


1.0 - Run #2- DX9 1280x1024 AA=No AA, 32 bit test, Quality: High ~~ Last Average FPS: 28.90
1.1 - Run #2- DX9 1280x1024 AA=No AA, 32 bit test, Quality: High ~~ Last Average FPS: 29.34

Both tests in XP x64, 2.7GHZ cpu, ram @ 719MHz. No driver changes or other tweaks between the 2

As a matter of fact, at 2.5GHz, ram at 833MHz, 4-4-4-15-2T, I get slightly better (compare to the first set of tests)..

Run #1- DX9 1280x1024 AA=No AA, 32 bit test, Quality: Medium ~~ Last Average FPS: 45.00
 
Good news, bad news.

Good news is, Crossfire is without a doubt working in DX10 mode, and it's helping, a LOT.

Bad news is, DX9 Crossfire performance seemed to have dropped a bit.

Here's 1.0 DX9 Crossfire, 1680x1050, no AA/AF, custom cfg file.

Code: 
==============================================================
TimeDemo Play Started , (Total Frames: 1500, Recorded Time: 44.62s)
!TimeDemo Run 0 Finished.
    Play Time: 26.24s, Average FPS: 57.16
    Min FPS: 40.31 at frame 831, Max FPS: 83.45 at frame 351
    Average Tri/Sec: 41248536, Tri/Frame: 721656
    Recorded/Played Tris ratio: 1.00
!TimeDemo Run 1 Finished.
    Play Time: 23.57s, Average FPS: 63.64
    Min FPS: 40.31 at frame 831, Max FPS: 83.45 at frame 351
    Average Tri/Sec: 45854524, Tri/Frame: 720526
    Recorded/Played Tris ratio: 1.00
!TimeDemo Run 2 Finished.
    Play Time: 23.66s, Average FPS: 63.39
    Min FPS: 40.31 at frame 831, Max FPS: 83.45 at frame 351
    Average Tri/Sec: 46277360, Tri/Frame: 730046
    Recorded/Played Tris ratio: 0.99
!TimeDemo Run 3 Finished.
    Play Time: 23.12s, Average FPS: 64.87
    Min FPS: 40.31 at frame 831, Max FPS: 83.45 at frame 351
    Average Tri/Sec: 46674764, Tri/Frame: 719542
    Recorded/Played Tris ratio: 1.00
TimeDemo Play Ended, (4 Runs Performed)

Here's patch 1.1 Crossfire, same config:

Code: 
==============================================================
TimeDemo Play Started , (Total Frames: 1500, Recorded Time: 44.62s)
!TimeDemo Run 0 Finished.
    Play Time: 29.01s, Average FPS: 51.70
    Min FPS: 34.35 at frame 843, Max FPS: 74.37 at frame 127
    Average Tri/Sec: 48859852, Tri/Frame: 945042
    Recorded/Played Tris ratio: 0.76
!TimeDemo Run 1 Finished.
    Play Time: 25.05s, Average FPS: 59.87
    Min FPS: 34.35 at frame 843, Max FPS: 75.84 at frame 363
    Average Tri/Sec: 56916268, Tri/Frame: 950670
    Recorded/Played Tris ratio: 0.76
!TimeDemo Run 2 Finished.
    Play Time: 24.73s, Average FPS: 60.66
    Min FPS: 34.35 at frame 843, Max FPS: 75.84 at frame 363
    Average Tri/Sec: 57772560, Tri/Frame: 952423
    Recorded/Played Tris ratio: 0.76
!TimeDemo Run 3 Finished.
    Play Time: 24.86s, Average FPS: 60.33
    Min FPS: 34.35 at frame 843, Max FPS: 77.79 at frame 367
    Average Tri/Sec: 57205516, Tri/Frame: 948136
    Recorded/Played Tris ratio: 0.76
TimeDemo Play Ended, (4 Runs Performed)
==============================================================

Now for the good news:


Here's DX10, same cfg, Crossfire disabled:

Code: 
==============================================================
TimeDemo Play Started , (Total Frames: 1500, Recorded Time: 44.62s)
!TimeDemo Run 0 Finished.
    Play Time: 52.06s, Average FPS: 28.81
    Min FPS: 8.90 at frame 199, Max FPS: 42.49 at frame 119
    Average Tri/Sec: 25240396, Tri/Frame: 875956
    Recorded/Played Tris ratio: 0.82
!TimeDemo Run 1 Finished.
    Play Time: 47.46s, Average FPS: 31.61
    Min FPS: 8.90 at frame 199, Max FPS: 42.49 at frame 119
    Average Tri/Sec: 27811818, Tri/Frame: 879892
    Recorded/Played Tris ratio: 0.82
!TimeDemo Run 2 Finished.
    Play Time: 47.39s, Average FPS: 31.65
    Min FPS: 8.90 at frame 199, Max FPS: 42.49 at frame 119
    Average Tri/Sec: 27835384, Tri/Frame: 879347
    Recorded/Played Tris ratio: 0.82
!TimeDemo Run 3 Finished.
    Play Time: 46.99s, Average FPS: 31.92
    Min FPS: 8.90 at frame 199, Max FPS: 42.49 at frame 119
    Average Tri/Sec: 28013810, Tri/Frame: 877569
    Recorded/Played Tris ratio: 0.82
TimeDemo Play Ended, (4 Runs Performed)
==============================================================

Here's DX10 Crossfire:

Code: 
==============================================================
TimeDemo Play Started , (Total Frames: 1500, Recorded Time: 44.62s)
!TimeDemo Run 0 Finished.
    Play Time: 41.34s, Average FPS: 36.28
    Min FPS: 17.15 at frame 196, Max FPS: 57.20 at frame 358
    Average Tri/Sec: 34260396, Tri/Frame: 944279
    Recorded/Played Tris ratio: 0.76
!TimeDemo Run 1 Finished.
    Play Time: 34.70s, Average FPS: 43.23
    Min FPS: 17.15 at frame 196, Max FPS: 57.20 at frame 358
    Average Tri/Sec: 40983672, Tri/Frame: 948042
    Recorded/Played Tris ratio: 0.76
!TimeDemo Run 2 Finished.
    Play Time: 34.39s, Average FPS: 43.62
    Min FPS: 17.15 at frame 196, Max FPS: 57.20 at frame 358
    Average Tri/Sec: 41279668, Tri/Frame: 946418
    Recorded/Played Tris ratio: 0.76
!TimeDemo Run 3 Finished.
    Play Time: 34.34s, Average FPS: 43.68
    Min FPS: 17.15 at frame 196, Max FPS: 57.20 at frame 358
    Average Tri/Sec: 41630512, Tri/Frame: 953078
    Recorded/Played Tris ratio: 0.75
TimeDemo Play Ended, (4 Runs Performed)
==============================================================
 
cannondale06 said:
performance went down in the gpu benchmark for me. running at 1024x768 all medium settings DX10. :(

before patch....31.02 fps
after patch ......29.80 fps
Click to expand...

Um, you do realize you're testing at settings that are CPU-bound right? Of course the GPU performance can't improve there lol
 
============================================================ ======
Tweaks:
============================================================ ======
*Reduced LAW splash damage vs. infantry in PowerStruggle mode
*Slowed Rocket projectile speed down in MP slightly
*Disabled automatic turret bounding boxes on vehicles to prevent issues with LAW hit detection
*Reduced grenade explosion radius in multiplayer
*Clamped water tesselation to avoid cheating in MP
*MultiGPU improvements with depth map updates
Click to expand...

Thumbs up to all of these but the bolded. Basically this comes down to "Hey, I have an idea, lets make flying vehicles even harder to destroy!"
 
All the people here showing that it made little difference in your bench marks are fucking killing me.

They arent patching it so you can benchmark better you terds. They are patching it for those of us who play the game not buy it so we can run a standard bench mark and say i got 31.9847283464829394 FPS!! WOOT

Step away from your computers and give em to someone who will use them for what they were made for
 
MrMike said:
Thumbs up to all of these but the bolded. Basically this comes down to "Hey, I have an idea, lets make flying vehicles even harder to destroy!"
Click to expand...

Can anyone confirm that AA bug is fixed if so there ya go.
 
Remember guys, the Timedemo is NOT a good indicator of performance, and will not show the advantages of performance gains properly.
 
Lamont Cranston said:
*snip* you terds *snip*
Click to expand...

As most of the tests here have shown here, the game is 'unplayable' at the quality some of us prefer, despite running the best cpu with the best gpu's. You can ask us to hand the game over, but that next person will gripe that the game is unplayable as well, and end up posting benchmarks as well wondering why his shiny new 8800GT isn't that much better than his old 8600GT SLI setup.

One of the biggest problems people have with this game is the performance vs the visual quality. Otherwise, people either love the gameplay, or hate it. The fact that this game is so demanding on even the best systems out there gives us a reason to do such testing - in a quest to make it look good, and run well, so we can enjoy the game in its entirety, not with its blocky palm tree leafs.

Some of the mentioned fixes or tweaks show possible improvements in graphical or performance capability of the game. If that's an issue for the people here who have benched after the patch, so be it. We want a game that's playable at decent frames in a decent quality.

I take pleasure in being able to make you chuckle, but let us be in our ways as we try to validate why we bought newer/better stuff in order to play one of the better games available. :D
 
I'll install it later and see if it fixes my ATI driver crashing every 15 minutes on 64-bit Vista.
 
Installed the patch and came to the conclusion that both GOW and COD4 are alot better games still :D.

Too bad 1.1 doesn't remove the suck from the performance or the whole lame ass alien sequence that almost made me quit and uninstall the POS altogether.

Crysis is still 'disappointment of the year' for me.
 
Brent_Justice said:
Remember guys, the Timedemo is NOT a good indicator of performance, and will not show the advantages of performance gains properly.
Click to expand...

But it is the only standardized test available to everyone that owns the game. What do you suggest that is objective and quantifiable?
 
Clockz said:
But it is the only standardized test available to everyone that owns the game. What do you suggest that is objective and quantifiable?
Click to expand...

Looking at the overall performance while playing some levels. I got close to 40fps in the time demo, when I get to the third level, it drops to teens constantly. IMO, bad coding, but I think that the best indicator of any improvements would be to play those levels again. I have tried the patch, no difference. Buddy of mine tried it on 2x 8800GTX, still no difference. I am very disappointed, will think twice prior to buying another game from these guys in the future.
 
is it true that when you get to the icy areas in the game youre FPS drop even more?
 
Lamont Cranston said:
All the people here showing that it made little difference in your bench marks are fucking killing me.

They arent patching it so you can benchmark better you terds. They are patching it for those of us who play the game not buy it so we can run a standard bench mark and say i got 31.9847283464829394 FPS!! WOOT

Step away from your computers and give em to someone who will use them for what they were made for
Click to expand...


i havent benchmarked the game yet but guys who have paid for the game and paid for their machines can do whatever the hell they want to with them, imho. so if it is fucking killing you, i guess you can go up to game faqs with the real gamers.
 
meatfestival said:
No SLI performance difference here. And I still have flickering water.
Click to expand...

I have pretty decent gains with SLI. The difference doesn't seem like it is as much as it should be. The performance is odd. I basically still have to leave post processing and a few other settings on high instead of very high but now I basically get the same performance I always did at 1680x1050 (actually slightly better, especially with V-Sync on) but at 1920x1200 the difference is massive. I'm not sure what the exact FPS numbers are but the game now feels playable and it wasn't before at that resolution. Also the tearing isn't as bad as it was before even with V-Sync off. BTW V-Sync seems to work properly.

Brent_Justice said:
Remember guys, the Timedemo is NOT a good indicator of performance, and will not show the advantages of performance gains properly.
Click to expand...

I agree. I have looked at FRAPS and while the numbers aren't much different than they were before, the game feels worlds better to play.
 
And if you're going to run one of the canned benchmarks, I recommend the CPU test, as it's closer to "real world" gameplay than the GPU flyover.
 
AMD_Gamer said:
is it true that when you get to the icy areas in the game youre FPS drop even more?
Click to expand...

Yes, It drops right after you beat the demo level. It goes from.
Lvl1: Decent
lvl2: - 5fps
lvl3: - 15fps
Ice - 20fps

Not the exact numbers, that's how it felt to me. I had to decrease my resolution by a mile to get playable frame rates.
 
aFive said:
Yes, It drops right after you beat the demo level. It goes from.
Lvl1: Decent
lvl2: - 5fps
lvl3: - 15fps
Ice - 20fps

Not the exact numbers, that's how it felt to me. I had to decrease my resolution by a mile to get playable frame rates.
Click to expand...

For me (before the patch) the snow levels were smoother than the rest of the game. I chalked that up to less detailed texturing due to the snow.
 
After installing the patch, I saw very noticeable fps gains. I play at a combo of very high (cvar change), high, and medium (post-processing, shaders, textures, and water at very high), and it runs about 10-15 fps faster.

I'm running windows xp sp2, 2x8800GT, AMD X2 5000 BE, and 2 gigs of ram. Very pleased with this patch. Performance was tolerable before, now it's very playable.
 
deliciouspork said:
After installing the patch, I saw very noticeable fps gains. I play at a combo of very high (cvar change), high, and medium (post-processing, shaders, textures, and water at very high), and it runs about 10-15 fps faster.
Click to expand...

LOL at what resolution 800x600? Most people running high res are lucky to get 10 - 15 FPS at all let alone another 10-15 haha. I really don't believe you. I gained .051 FPS. Wow. Still sux.
 
what the ka-gigger!!!!!!!! dx9 cf even wonkier than before. loaded up an old save point and min/max frames swing from one extreme to the other. mongo angry!!!!!
 
dnottis said:
LOL at what resolution 800x600? Most people running high res are lucky to get 10 - 15 FPS at all let alone another 10-15 haha. I really don't believe you. I gained .051 FPS. Wow. Still sux.
Click to expand...
Are you running Crossfire on DX10 with your P5K? Don't expect "magic" performance boosts for everyone out of this patch. It's clear this patch was meant to alleviate the poor mutli-gpu scaling.

If Dan-D says that it's really improving his 3-way SLI performance, and others seem to be getting a little out of it too, then obviously the patch is better than nothing at all.
 
dnottis said:
LOL at what resolution 800x600? Most people running high res are lucky to get 10 - 15 FPS at all let alone another 10-15 haha. I really don't believe you. I gained .051 FPS. Wow. Still sux.
Click to expand...

Hes running SLI, are you? Some folks running SLI are reporting considerable gains. Considering SLI was not really working at all before the patch and seems to be halfass now that would gain SLI folks a good chunk of performance.
 
You must log in or register to reply here.
Back
Top