Sign in with
Sign up | Sign in

Image Quality And Settings

Crysis 3 Performance, Benchmarked On 16 Graphics Cards
By

Crysis 3 is powered by Crytek's CryEngine 3, based on the same technology employed by Crysis 2. The latest version, however, is heavily enhanced by new anti-aliasing modes, displacement mapping, 3D lens flares, high-quality first-person shadows, volumetric fog (including cloud shadows), volumetric light shafts, dynamic caustics, semi-deferred shading, improved soft shadows, real-time area lights and shadows, glossy reflections, texture streaming support, and a new breakable glass system. 

No doubt, this is a beautiful-looking game with a stunning engine behind it. With that said, it doesn't raise the bar as much as the original Crysis six years ago in 2007.

The game has four detail presets (Low, Medium, High, and Very High), but individual settings are customizable. These are the user-configurable controls:

As you can see below, the difference between the lowest and highest quality settings can be subtle. The most obvious change is the lighting, though you also see more triangles on-screen as the detail level increases. Additionally, there are some effects inherent to higher detail levels that the screen shots don't capture. When you actually play through the game, you get a very tangible sense for the difference between Low and Medium detail presets, and less of an impact as details increase from there. Moreover, higher settings have a significant impact on the frame rate, as you will see in the benchmarks.

Anti-Aliasing Modes

Let's talk about anti-aliasing. Crytek developed a custom morphological AA mode called Enhanced Subpixel Morphological Anti-Aliasing (SMAA). This method combines AMD's post-process Morphological Anti-Aliasing (MLAA) with multi- and super-sampling techniques, with added temporal reprojection to provide a result that Crytek claims delivers "results of better overall quality than previous approaches, while more closely converging to MSAA/SSAA references but maintaining extremely fast execution times."

In layman's terms, SMAA is a combination of post-process (software) and MSAA/SSAA (hardware) anti-aliasing with the goal of achieving better image quality and a smaller performance hit. It's notable that Crysis 3 offfers 1x, 2x, and 4x SMAA modes for single-GPU cards, but only 2x SMAA for CrossFire or SLI configurations.

In some ways, this sounds similar to Nvidia's Temporal Anti-Aliasing technique known as TXAA. This mode also uses a combination of hardware and software anti-aliasing techniques with a custom resolve and temporal filter. TXAA is only available on Kepler-equipped GeForce hardware, however.

Our focus in this article is not anti-aliasing modes, but raw performance. Having said that, we'd like to deliver a quick look at the different modes available:

FXAA looks little better than a blur filter here, but it's probably preferable over no AA at all. Nvidia's TXAA does perhaps the best job to smooth lines, but at the cost of some blurring of textures in the scene. Crytek's SMAA looks like a nice compromise, with anti-aliased lines and good, crisp textures.

With a working knowledge of the different AA modes, let's have a quick look at each one's performance penalty. We'll use the GeForce GTX 670 for this example, since it's capable of running all of the AA modes available:

Based on the AA image quality comparison and our performance data, we think 2x SMAA gives you the best compromise between image quality and performance impact. Since FXAA and 2x SMAA are compatible across all of the cards we're testing, these are the modes we'll focus on in our benchmarks.

Display all 139 comments.
This thread is closed for comments
Top Comments
  • 39 Hide
    s3anister , March 5, 2013 5:14 AM
    Cool article but every time I see an "ultra-mega-(insert specific game here)-gpu-performance-showdown" type article I can't help but feel that they are always lacking in comparison to older cards. It'd be nice if there were at least a few last gen cards tossed in for reference. Not everyone decided to upgrade from their HD 6970s or GTX 580s.
  • 37 Hide
    stickmansam , March 5, 2013 3:24 AM
    Still feel that the game is unduly harsh for what it displays

    Also hope AMD comes out with better drivers soon
  • 34 Hide
    Immoral Medic , March 5, 2013 4:09 AM
    I completed this game in 4.5 hours. I gotta say, having great graphics does NOT make a good game. It's sad when all you have to attract customers is "Best Graphics in a Game Yet". BUYBUYBUY. Don't even get me started on the absolutely terrible multiplayer...
Other Comments
  • 14 Hide
    will1220 , March 5, 2013 3:24 AM
    Why would you include the top of the line amd, middle of the line intel (ivy bridge i5) and not the top of the line ivy bridge i7 3770k?????????
  • 37 Hide
    stickmansam , March 5, 2013 3:24 AM
    Still feel that the game is unduly harsh for what it displays

    Also hope AMD comes out with better drivers soon
  • 8 Hide
    johnsonjohnson , March 5, 2013 3:26 AM
    Right on time. I kinda suspect the i3-3220 performance from Techspot was unusual..
  • 17 Hide
    rawrrr151 , March 5, 2013 3:33 AM
    I thought i3 3220 was IB, not SB?
  • 5 Hide
    hero1 , March 5, 2013 3:35 AM
    Time to make an i7 rig and pass my current system to wife because Crysis demands. Nice review and the 13.2 driver from AMD has really improved frame variance for their cards. Keep it up red team so green team can do the same. The better the drivers the better our gaming experience. After all, we pay pretty penny looking for better experience. Cheers!
  • 9 Hide
    DryCreamer , March 5, 2013 3:54 AM
    I have a hand ful of benchmarks I ran when I upgraded to from the i3 3220 to the i7 3770K and I DEFINITELY noticed a jump in the minimum frame rates:

    http://www.tomshardware.com/forum/395367-33-crysis-benchmark-560ti

    Dry
  • 34 Hide
    Immoral Medic , March 5, 2013 4:09 AM
    I completed this game in 4.5 hours. I gotta say, having great graphics does NOT make a good game. It's sad when all you have to attract customers is "Best Graphics in a Game Yet". BUYBUYBUY. Don't even get me started on the absolutely terrible multiplayer...
  • 1 Hide
    aussiejunior , March 5, 2013 4:25 AM
    Wheres the gtx 680?
  • 16 Hide
    xpeh , March 5, 2013 4:39 AM
    The only thing this game has going for it are the graphics. I beat the game in under 6 hours. The story was simply tossed in the gutter. They should have stuck with fighting the Koreans instead of introducing Aliens.
  • 22 Hide
    iam2thecrowe , March 5, 2013 4:49 AM
    toms, your method of monitoring frame times must be screwed up, the cards vary wildly and at some point the lowly gtx 650ti was showing an unbelievably good score, even better than the gtx 670. There is something wrong with your testing method. I have also noticed the same thing in previous benchmarks where you measured frame time, not consistent results. Please look into this.
  • 3 Hide
    JJ1217 , March 5, 2013 4:53 AM
    xpehThe only thing this game has going for it are the graphics. I beat the game in under 6 hours. The story was simply tossed in the gutter. They should have stuck with fighting the Koreans instead of introducing Aliens.


    While its no where near to Crysis 1, I don't understand the hate for C2/C3's campaign. I thought it was amazing, good fun, while crysis was just too serious. I loved jumping around in c2, sliding through hallways, spamming my shotgun.

    I do think that C2 and C3 shouldn't be C2 and C3, if you know what I mean, like it should be called something different, not in the same Crysis franchise.
  • 17 Hide
    mouse24 , March 5, 2013 5:02 AM
    Wonder if theres tessellation under the ocean in this one to.
  • -5 Hide
    cleeve , March 5, 2013 5:11 AM
    iam2thecrowetoms, your method of monitoring frame times must be screwed up, the cards vary wildly and at some point the lowly gtx 650ti was showing an unbelievably good score, even better than the gtx 670. There is something wrong with your testing method. I have also noticed the same thing in previous benchmarks where you measured frame time, not consistent results. Please look into this.


    The method is fine, but the graphics load has a lot to do with the results. It's not cut and dry.

    Were writing an article around it this month, it should explain a lot.
  • 5 Hide
    cleeve , March 5, 2013 5:12 AM
    aussiejuniorWheres the gtx 680?


    In the high detail and triple-monitor benchmarks
  • 39 Hide
    s3anister , March 5, 2013 5:14 AM
    Cool article but every time I see an "ultra-mega-(insert specific game here)-gpu-performance-showdown" type article I can't help but feel that they are always lacking in comparison to older cards. It'd be nice if there were at least a few last gen cards tossed in for reference. Not everyone decided to upgrade from their HD 6970s or GTX 580s.
  • 10 Hide
    de5_Roy , March 5, 2013 5:16 AM
    there was something more into fx8350's 'higher' performance after all. proof that average fps don't tell the full story.
    i hope crytek can fix this with an update.
    i can totally see this game becoming a benchmark staple very soon. :D 
  • 11 Hide
    Novuake , March 5, 2013 5:19 AM
    The first graphics benching AA with GTX670???? Whats up with that?

    How can the minimum FPS be 30FPS but the average is 24 FPS?

    And its a little odd that the min FPS is so close across the the board... Explain?
  • 6 Hide
    slomo4sho , March 5, 2013 5:23 AM
    I wanted to see if the hd 7870 or a 2gb 7850 would be able to support 5760x1080 on low settings.
  • 20 Hide
    mouse24 , March 5, 2013 5:47 AM
    JonnyDoughJust like Crysis 1 and 2, I still don't care. I play TF2, Skyrim, and any other game that isn't brand new because I refuse to pay $60 for a game that I can't return to the store and I don't have time to play all the titles out there I want to anyway. Anyone who has to jump on the latest and greatest bandwagon doesn't understand what "good gameplay" is.


    Some people don't understand that peoples opinions/gameplay/genres are different.
Display more comments