Deus Ex: Human Revolution Performance Analysis

Benchmark Results: DirectX 11, Maximum Detail With Anti-Aliasing

In our next set of benchmarks, we're turning on soft shadows, increasing anisotropic filtering to 16x, and turning on anti-aliasing. We’ll compare both FXAA and MLAA to find out what kind of performance hit each method causes.

In general, it looks like MLAA incurs a slight performance hit at low resolutions. But by 1680x1050, there is no notable difference between MLAA and the FXAA High setting. In fact, MLAA appears to be easier for the GeForce cards to handle at 1920x1080.

What’s really impressive here is that, of all of the cards tested, only the Radeon HD 6570 struggles to provide a minimum of 30 FPS above 1680x1050. All of the other cards handle these maximum settings without a problem at 1920x1080.

  • Soma42
    Man as much as I prefer PC gaming I can't wait for the PS4 and Xbox 720 to come out so games will look noticeably better. Might actually want to upgrade my computer by then...

    Anywho, I didn't play the first two am I missing anything if I wanted to pick this up?
    Reply
  • festerovic
    @soma - I personally thought they were average - good, for the time. Not sure if they would stand up to time...

    Interesting to read the dual core HT chips outperformed real cores. Can we look forward to the 2600's HT being utilized in games before the next generation of CPUs comes out?
    Reply
  • haplo602
    nice review, finaly a new DeusEx game for me :-)
    Reply
  • Was AMD dual-core optimizer installed?
    Reply
  • gerchokas
    Good article - i still remember when i first saw the 2000' Deus Ex graphics on my friend's then-brand-new pc, i thought 'maaan... this looks *friggin* REAL!' I instantly knew my old Pentium cpu needed replacing ASAP...
    11 years later, i praise again the great graphics.. but this time they havent cought me off-guard!
    Reply
  • tacoslave
    i really wonder how far developers can take the graphics in 3 years.
    Reply
  • aznshinobi
    Hmm... The Nvidia cards perform better than the AMD cards of equivalent rank. I'm not playing fanboy but didn't AMD fund the studio? Afterall Eyefinity was made use of.
    Reply
  • th3loonatic
    Are there any typos? Coz I see a GTX560 Ti listed as a card used to test, but it doesn't appear in the results.
    Reply
  • fyasko
    festerovic@soma - I personally thought they were average - good, for the time. Not sure if they would stand up to time...Interesting to read the dual core HT chips outperformed real cores. Can we look forward to the 2600's HT being utilized in games before the next generation of CPUs comes out?
    HT isn't the reason dual core SB CPU's beat 6 core thubans. SB is a better architecture. Hurry up Bulldozer!
    Reply
  • mayankleoboy1
    @ Don Woligroski
    i want the CPU benchmarks at 1080p with highest settings.
    benches at 1024x768 are irrelevant. the gamer of today is atleast 1680, preferable 1080.
    so please add to the benches. also, this would show the real impact of CPU on FPS.
    Reply