Crysis Warhead performance WORSE!?

How do you switch from dx9 and 10

how you do it depends whether its frim Steam or from a dvd. if from dvd just go to the games menu, right - click Warhead and click run dx 9 mode. If its from steam, well im still trying to figure that out since i got it from steam...
 
I'm not sure why you guys are having so much trouble. My Vista64/4870 box is doing very well with the Steam version of Warhead @ 1680x1050. Shadows on Mainstream, Objects/Physics on Gamer, everything else on Enthusiast, no AA, and it's running nice and smooth. /shrug

 
I'm running everything on Enthus. except for shadows and motion blur. I'm gettin about 35FPS with my sig rig. This game DOES NOT look like the orginal crysis. The orginal had MUCH better graphics and was a lot better. Everything has a hint of brown/tin like another poster mentioned... why this is I have on idea.
 
Frankly I think Crysis performed better. And one other thing I can say for absolute certain is that it was more stable. Warhead has crashed 3 times for me in about 4 hours playing it. Crysis didn't crash once on me in 3 full playthroughs of the game.
 
how you do it depends whether its frim Steam or from a dvd. if from dvd just go to the games menu, right - click Warhead and click run dx 9 mode. If its from steam, well im still trying to figure that out since i got it from steam...

lol, same here. I just added the -dx9 tot he desktop shortcut. Seemed to get another 5-10fps when I did that.
 
No, its NOT from Crytek. It is from PC gamer. here is the link to Cryteks posted specs from THEIR web site ...for the third time even :rolleyes:
You seem to have a problem reading. PC Gamer got those specs from Crytek, which is very clearly stated in that pic. Do I really need to post the rest of the feature on Warhead where Cevat Yerli goes on and on about the 653$ machine they built which runs the game "silky smooth" "with all the bells and whistles" PC Gamer did not arbitrarily make these specs up, Crytek provided them.

I really cannot understand why you have such a problem understand this.
 
You seem to have a problem reading. PC Gamer got those specs from Crytek, which is very clearly stated in that pic. Do I really need to post the rest of the feature on Warhead where Cevat Yerli goes on and on about the 653$ machine they built which runs the game "silky smooth" "with all the bells and whistles" PC Gamer did not arbitrarily make these specs up, Crytek provided them.

I really cannot understand why you have such a problem understand this.


I have a problem with it because that is not true. I have a problem with it because you are claiming to all that your word is truth when it is obviously not. You are saying PC gamer says this. Im telling you that Crytek says otherwise and I have linked to that on more than one occasion on this thread. You want to continue believing some second hand information even when presented with the ACTUAL information from Crytek themselves. Like I said before, you keep quoting something that did not come from Crytek but from another source claiming thats what Crytek told them. Funny thing is, you fail to provide any type of links to show your information is true. I posted many links from other sites that got their information from Crytek (including Cryteks own site as well as the official site for Warhead) and it is nothing like PC gamer says. Therefore either they are lying, its a typo or the reviewer is an ignorant ass munch that doesnt understand the difference between 8800 and 9800 in print. And to top it all off, you eat it up like ice cream and wonder on your merry way.

So go right on ahead believing PC Gamer even when no other site agrees with them. Keep on spreading their falicy as if it were a biblical belief. Hell, go on and post a true link to PC Gamer. After all, it is the only link you can post that shows that because no one else is saying that is what Crytek told them.

If you want to continue this obsessive belief in second hand information, go right ahead, but Im through arguing with you. If you need another reply, just go back and reread the other replies I already gave that proved those specs you posted are inaccurate. And maybe try clicking on those links and reading them before you do so. You might learn something.
 
how you do it depends whether its frim Steam or from a dvd. if from dvd just go to the games menu, right - click Warhead and click run dx 9 mode. If its from steam, well im still trying to figure that out since i got it from steam...


Found it on Steam's forum.
.
From the Games tab, Crysis properties, Set launch options then -dx9.
 
Found it on Steam's forum.
.
From the Games tab, Crysis properties, Set launch options then -dx9.

Cool, just did that.

What a performance difference; double that of dx10. dx10on gamer I was getting 15, noe dx9 on gamer over 30fps, and mainstream went from 25 to over 50fps.
 
Cool, just did that.

What a performance difference; double that of dx10. dx10on gamer I was getting 15, noe dx9 on gamer over 30fps, and mainstream went from 25 to over 50fps.

yes here is a preformance increase, but for me its more a less more stable frame rate since i dont got the ram to feed the card with a steady amount of data with Warhead dx10 :( i run enthusiast dx9 and get around 30-40 frames stable, sometimes 20 frames, but you cant notice it from gameplay, only from the counter :)

But with vista 32 bit and forcing dx9 through steam every once in a while, usually when a new map is loading or anything big, the game crashes, while in dx10 it doesnt, weird. Anyone know what may be the cause?
 
Will be getting the game this afternoon, will do some tests and post results, would be curious to see the difference between the system in my sig and a similar one with a GTX260 as in the original the 4870 comes out just above the GTX260. I really hope all that TWIMTBP bullshit isn't true this time. :rolleyes:

Also to the guy going on about warhead being able to run on a 9800gt but not an 8800gt you my friend need to get your facts right 9800GT = 8800GT with 3 way SLI read some reviews before harassing people.
 
yes here is a preformance increase, but for me its more a less more stable frame rate since i dont got the ram to feed the card with a steady amount of data with Warhead dx10 :( i run enthusiast dx9 and get around 30-40 frames stable, sometimes 20 frames, but you cant notice it from gameplay, only from the counter :)

But with vista 32 bit and forcing dx9 through steam every once in a while, usually when a new map is loading or anything big, the game crashes, while in dx10 it doesnt, weird. Anyone know what may be the cause?
for some reason DX9 uses MORE ram than DX10.

DX9 is definitely smoother with higher framerates though.
 
Also to the guy going on about warhead being able to run on a 9800gt but not an 8800gt you my friend need to get your facts right 9800GT = 8800GT with 3 way SLI read some reviews before harassing people.

Sounds like you are reffering to me there....

My facts ARE straight. Read the links I posted back in the thread for what Crytek said. And who said anything about a 9800GT=8800GT's in SLI? Way to come in here on day one and mouth off without reading the whole thread.:rolleyes: <---look, I can roll my eyes too.
 
Does anyone happen to know if the modded config files for DX10 effects under DX9 from the original work with warhead?
 
Sounds like you are reffering to me there....

My facts ARE straight. Read the links I posted back in the thread for what Crytek said. And who said anything about a 9800GT=8800GT's in SLI? Way to come in here on day one and mouth off without reading the whole thread.:rolleyes: <---look, I can roll my eyes too.

first i did rad the wole thread, i have been a long time reader of Hard Forum, just never felt like signing up. i did not mean that a 9800GT is the same as 8800GT's is SLI.
i mean that the 9800GT is EXACTLY the same card as the 8800GT rebadged as a 9800GT and given the capability to utilize 3-way SLI. They have exactly the same architecture and memory sub system and even have the same clock speeds. so explain why a 9800GT would run any better than a 8800GT? *sigh*
 
for some reason DX9 uses MORE ram than DX10.

DX9 is definitely smoother with higher framerates though.

well not for me, well atleast not anymore. i ran the in game info display and dx10 uses about 300mb of more ram maxing at around 1800mb with an average of around 1600 whiile dx9 is an average 1400 maxing at arounf 1600 only when loading new maps, after that it drops.

And i agree, frames are better in dx9, but dx10 enthusiast does look beter than dx9 enthusiast, so i cant wait for more ram to run dx10 smoothly :)
 
DX10 1280x1024 Gamer - 25 fps avg
DX9 1920x1080 Gamer - 35 fps avg

Something is not happy in Warhead land.
 
well not for me, well atleast not anymore. i ran the in game info display and dx10 uses about 300mb of more ram maxing at around 1800mb with an average of around 1600 whiile XP is an average 1400 maxing at arounf 1600 only when loading new maps, after that it drops.

And i agree, frames are better in dx9, but dx10 enthusiast does look beter than dx9 enthusiast, so i cant wait for more ram to run dx10 smoothly :)

i just checked using the r_displayinfo=1 in the game console.

DX9 1090 MB
DX10 860 MB

also your post is confusing... so you checked DX10 in Vista but DX9 in XP? thats two different OS and of course XP would probably use less. you need to compare both with Vista.
 
kinda strange that DX9 uses more ram than DX10 can anyone test and see if this was also the case on the original?
 
i just checked using the r_displayinfo=1 in the game console.

DX9 1090 MB
DX10 860 MB

also your post is confusing... so you checked DX10 in Vista but DX9 in XP? thats two different OS and of course XP would probably use less. you need to compare both with Vista.

i did, i used dx10 and dx9 in vista. dx10 used more ram for me, and i can understand this since dx10 did look better, while dx9 didnt look as good but had better preformance by using less ram.
 
Going back to my original post, I finally got the catalyst 8..9 to work. Dunno how. But no improvement. at 1280X1024, all minimum settings, it's unplayable (I'm at the point right after you navigate through the frozen ship.
See sig for system specs, any ideas?
 
i did, i used dx10 and dx9 in vista. dx10 used more ram for me, and i can understand this since dx10 did look better, while dx9 didnt look as good but had better preformance by using less ram.
if you look at your post you said XP though.

so if you are comparing both under Vista then something is WRONG with YOUR results as Crysis in DX10 has ALWAYS used less memory than DX9.


"DX10 also has the advantage of reduced system memory usage, which could be more important on machines with lower amounts of physical RAM."

http://enthusiast.hardocp.com/article.html?art=MTQ4MCw4LCxoZW50aHVzaWFzdA==


"As we see in the NVIDIA screenshots in DX10 there is about 1GB of system memory being used but when we look at that same screenshot in DX9 there is about 1.5GB of system memory used, and sometimes it was even higher. That means DX9 is so inefficient that it uses about a half a GB of RAM or more with the exact same image quality settings."

http://enthusiast.hardocp.com/article.html?art=MTQ4MCw3LCxoZW50aHVzaWFzdA==
 
if you look at your post you said XP though.

so if you are comparing both under Vista then something is WRONG with YOUR results as Crysis in DX10 has always used less memory than DX9

"DX10 also has the advantage of reduced system memory usage, which could be more important on machines with lower amounts of physical RAM."

http://enthusiast.hardocp.com/article.html?art=MTQ4MCw4LCxoZW50aHVzaWFzdA==

my bad, i meant to say dx9 and not XP. Also, i've tried multiple times and dx10 is using more memory. i dont know why but maybe warhead manages memory different? i'm also on a clean vista install with all the my hardware drivers at their newest as well as windows updates, so i dont think anything would be affecting the results.

More things can come into play such as ati hardware runs dx10.1 and vista sp1 tweaked dx10 some more, so since april, the actual results may have changed from what hardocp got.
 
my bad, i meant to say dx9 and not XP. Also, i've tried multiple times and dx10 is using more memory. i dont know why but maybe warhead manages memory different? i'm also on a clean vista install with all the my hardware drivers at their newest as well as windows updates, so i dont think anything would be affecting the results.

More things can come into play such as ati hardware runs dx10.1 and vista sp1 tweaked dx10 some more, so since april, the actual results may have changed from what hardocp got.
well unless its an ATI thing then it still stands because I have SP1 and you see my results. also if you search around it clearly looks like DX10 still uses less memory in Warhead just like in Crysis. unfortunately none of the other people saying the same thing have ATI cards.
 
I just stuck the autoexec.cfg file from the Crysis CCC 2.21 straight into the Warhead folder and I found that it helped a lot. At least some of the commands present do work in Warhead, and they help to smooth out the game in the same way they did for Crysis.
 
well unless its an ATI thing then it still stands because I have SP1 and you see my results. also if you search around it clearly looks like DX10 still uses less memory in Warhead just like in Crysis. unfortunately none of the other people saying the same thing have ATI cards.

DX9

crysis-2008-09-19-00-21-36-80.bmp


DX10

crysis-2008-09-19-00-31-18-42.bmp


DX9

crysis-2008-09-19-00-23-26-44.bmp


DX10

crysis-2008-09-19-00-33-58-68.bmp


DX9

crysis-2008-09-19-00-24-43-09.bmp


DX10

crysis-2008-09-19-00-35-01-86.bmp


DX9

crysis-2008-09-19-00-25-11-76.bmp


DX10

crysis-2008-09-19-00-35-38-20.bmp


So by comparing both at the same settings, same areas, only difference is dx9 vs dx10, i can say for around 50mb more you get almost double the frames,

DX10 also has texture problems as shown here ( All Screens Below Are DX10)

crysis-2008-09-17-20-35-00-54.bmp


crysis-2008-09-17-20-36-38-11.bmp


crysis-2008-09-17-20-39-52-70.bmp


crysis-2008-09-17-20-42-48-94.bmp


crysis-2008-09-19-00-33-19-03.bmp


All of these only occur in DX10 and arent due to memory artifacting or gpu artifacting.
 
so are you drunk because even your screenshots show you using MORE memory in DX9 than DX10 just like I said? :confused:

your difference isnt quite as big as mine but maybe the higher settings you are using are more efficient in DX9 then they are in the lower settings I am using.

also those screens look like poo and really arent better looking than mainstream settings.
 
I still do not understand how anyone can defend Crytek and their games... A history of shitty, buggy beautiful looking games, in a couple years however... I like to "game" whereas do stuff, not stare at leaves and edit my cfg 14 million fucking times just to make that leaf extra pretty... I kinda like gameplay over graphics... I'm a gamer, not a desktop publisher. All you guys defending this shitty mess, end up doing it in massive threads where everyone complains about performance... Meanwhile you could actually be playing a ton of games that don't require all this bs and experience GAMEPLAY better than anything offered in Crysis... Then again this is my opinion... a gamers opinion... is this the gaming forum or video card forum? Oh well.
 
so are you drunk because even your screenshots show you using MORE memory in DX9 than DX10 just like I said? :confused:

your difference isnt quite as big as mine but maybe the higher settings you are using are more efficient in DX9 then they are in the lower settings I am using.

also those screens look like poo and really arent better looking than mainstream settings.

no im just showing you how closely they are, and these screens are from the beginning. Later in the game when the graphics get more and more stressed the difference becomes in the 100s of megs between dx9 and dx10, usually lower in dx9.

I just took some quick shots from the first level because you said dx10 always uses less memory and is more efficient than dx9, when it isnt all the time. Later ill take some screens of levels deeper into the game where i started to hit around 1800mb use in dx10.
 
I found the problem. This may or may not be the solution for other people having issues, but for me, Cool n' Quiet was interfering with this game. I turned up the minimum processor state to 100% and my FPS is much more stable now. With all settings at Gamer (1280x960) I average around 25 FPS on the ice levels... no more dips down to 10FPS. I brought some settings like shadow quality down to Mainsteam to stay around 30 though.
DX10 or 9?
 
I'm running the game @ 1680x1050 -dx9 on gamer + a few other custom options (edgeAA=2, triplebuffering, vsync, view distance 4096) and everything looks fantastic and runs beautifully. I'm getting 25-60 frames w/ avg around 30-35 outdoors. Indoors it's a constant 60, looks so fluid, really really nice. They most certainly did optimize this engine, I can see it and feel it - when compared to crysis. I think most of the people with problems need to sort out their operating system or drivers. I'm about 3 or so hours in and enjoying the experience.

e6850 @ 3.2ghz (8x400)
Abit IP35-Pro
4GB DDR2 800 (1:1 ratio)
HD4870 (800/1050 - fan speed 40%) - catalyst 8.9
Vista x64 SP1
 
I'm running the game @ 1680x1050 -dx9 on gamer + a few other custom options (edgeAA=2, triplebuffering, vsync, view distance 4096) and everything looks fantastic and runs beautifully. I'm getting 25-60 frames w/ avg around 30-35 outdoors. Indoors it's a constant 60, looks so fluid, really really nice. They most certainly did optimize this engine, I can see it and feel it - when compared to crysis. I think most of the people with problems need to sort out their operating system or drivers. I'm about 3 or so hours in and enjoying the experience.

e6850 @ 3.2ghz (8x400)
Abit IP35-Pro
4GB DDR2 800 (1:1 ratio)
HD4870 (800/1050 - fan speed 40%) - catalyst 8.9
Vista x64 SP1
unless you are using some third party app to force triple buffering for DX games then you dont actually have it on. triple buffering doesnt work on DX games, only openGL, from the ATI and Nvidia control panels.

the game is WAY slower in DX10 than the original for many people including myself. DX9 performance does seem a little better(faster) than the original though. also on the original game the difference in performance between DX9 and 10 was very minimal. on Warhead the difference is about 30% slower in DX10 as opposed to DX9.
 
SO I opted for dx9, and the performance is WAY better. EVen then I have to keep most on minimum, but at least the framerates are playable.

However, it only took me 2.8 hours (via steam) to beat the whole thing. That's less than Portal, which cost ALOT less. That freakin sucks. Another poorly coded game from Crytek that I can barely play with decent settings, and not at all under DX10 (and I have 2 4850's in crossfire), and all I get is 3 hours of gameplay? $10/hr!? really?
 
I have the game running on DX10 with the Enthusiast setting @ 1920x1200 with no AA. Looks amazing & its very playable.
 
SO I opted for dx9, and the performance is WAY better. EVen then I have to keep most on minimum, but at least the framerates are playable.

However, it only took me 2.8 hours (via steam) to beat the whole thing. That's less than Portal, which cost ALOT less. That freakin sucks. Another poorly coded game from Crytek that I can barely play with decent settings, and not at all under DX10 (and I have 2 4850's in crossfire), and all I get is 3 hours of gameplay? $10/hr!? really?

How long the game is depends highly on gaming style. I've been playing for about 5 hours and I've just finished the first level.
 
at least you guys can start to play it. my notebook in my sig it plays fine at1440x900 on medium settings. my desktop which i built less than a week ago has the following specs
be-2300 @ 2.4,
2gb gskill ddr 2
7200.10 320gb sata
visiontek 4850
vista basic

locks up and says not responding as soon as the game starts after the loading the first level. total bs. tried reinstalling 2x same result. running the 8.9 drivers.
 
I have the game running on DX10 with the Enthusiast setting @ 1920x1200 with no AA. Looks amazing & its very playable.
the higher two settings are more playable in Warhead because they are uglier than the original game. there isnt as big of a gap in visuals between medium/mainstream and very high/enthusiast like there was in Crysis.
 
at least you guys can start to play it. my notebook in my sig it plays fine at1440x900 on medium settings. my desktop which i built less than a week ago has the following specs
be-2300 @ 2.4,
2gb gskill ddr 2
7200.10 320gb sata
visiontek 4850
vista basic

locks up and says not responding as soon as the game starts after the loading the first level. total bs. tried reinstalling 2x same result. running the 8.9 drivers.

Turn off the CPU power saving in BIOS (cool and quiet?) seen it help on a number of games.
 
Back
Top