Game Performance problems on Radeon HD 7970

Discussion in 'Videocards - AMD Radeon' started by l0stPlayer, Apr 2, 2013.

  1. l0stPlayer

    l0stPlayer Guest

    Messages:
    11
    Likes Received:
    0
    GPU:
    Sapphire Radeon HD 7970
    During the fall of 2012 I completed a first time PC build. Since then, I have not been absolutely certain on whether my system is performing as it should in intensive games including the Crysis series and Battlefield 3. If my system is underperforming I am suspecting a hardware issue with my motherboard. More specifically, the 16x PCI-E slot where my Sapphire Radeon HD 7970 Dual-X graphics card w/ 3GB of VRAM is plugged in.

    Once I started gaming on my new PC, I encountered some unusual performance snags in Half-Life 2 and other source games at max settings, 8x SSAA, 16x AF, and 1920x1200 resolution. The most glaring was at one specific point in Ravenholm where performance significantly decreased. I found this to be unusual given the game’s age and specs of my PC. None of the games I had on my machine at the time were particularly demanding, the most demanding game was probably either Mass Effect 2 or Portal 2. I glossed over these issues until I bought the Crysis Collection during the Steam 2012 Autumn sale. It was once I started playing Crysis it was clear something was wrong. Until this point I did not bother to use FRAPS or other methods to gauge fps. Once I did, I was stunned to find I was getting a mediocre fps mostly in the 30s in Crysis 1 at the maximum graphics settings, 4x SSAA, 1920x1200 resolution, and noVsync. Crysis 2, Crysis Warhead, Unigine Heaven Benchmark, and Battlefield 3 (once I got my hands on a copy). Also displayed disappointing results. After communicating on a different computer enthusiast fourm about my problems, I followed their advice I got a replacement graphics card from Sapphire only to find no improvement. Having up to date drivers and BIOS did not help either. If the graphics card is not bad, the other possibility that was brought up was a bad or failing PCI-E slot on my ASUS P8Z77-V Pro motherboard.

    Thankfully, I found out that I did not have to replace my motherboard, at least not yet. I went into the Catalyst Control Center and switched my Anti-Aliasing method from Supersampling to Multisampling resulting in a massive performance increase. The Battlefield 3 campaign, runs at a brisk and buttery smooth frame rate at maximum settings with 4x MSAA and 16x AF at 1920x1200 with no Vsync. On the level “Operation Swordbreaker,” the frame rate usually hovers in the 60s or 70s, goes as high as the very low 90s and drops in one section into the 50s. . Despite the performance I am getting in BF3 campaign, performance in the Crysis series remains disappointing. Right now Crysis 1 shows fps in the 40s most of the time, will sometimes dip into the 30s, and may go as high as 60 fps during gameplay (60fps does not happen often in Crysis 1). The game is running at max settings with 8x MSAA and 16x AF at 1920x1200, no vsync. The level “Second Chance” in Crysis 2 at very beginning the runs buttery smooth with 60+ fps until I step outdoors where fps then hovers in the 40s most of the time. I did not progress very far into the level given that I have not yet completed Crysis 1 and Crysis Warhead. I do not know about the effect of AA and AF settings in Crysis 2 given the lack of in-game options for them and not knowing if the desired AA and AF settings are applied when forced through Catalyst Control Center. Running the GPU benchmarks for Crysis 2 & Crysis Warhead have made things more confusing and inconclusive. The benchmarks provided by online publications are also adding to my confusion whether I am getting the proper performance. This is all further compounded by youtube videos from users claiming to be getting higher fps in the crysis games than what I am getting. Shouldn’t I be getting more in the ballpark of these users are getting even though I am playing at a slightly higher resolution? (1920x1200 vs. 1080p). Here are the links those videos.

    http://www.youtube.com/watch?v=Df0lxJTVbFc

    http://www.youtube.com/watch?v=B-XGYNaLsUY

    http://www.youtube.com/watch?v=9XHjUWztgzs

    http://www.youtube.com/watch?v=llDUohfnrws

    Turning Down MSAA to 4x did hardly anything to improve performance in the Crysis series. I am not entirely sure of performance of BF3 either, especially multiplayer where in one game on a large outdoor map, the game would run mostly in the 50s below what some other people where claiming to getting in multiplayer games such as in these videos that are linked.

    http://www.youtube.com/watch?v=NWsjQINgXLg

    https://www.youtube.com/watch?v=ty2MC3p_lLo


    In closing I do realize that the Crysis series and Battlefield 3 are highly demanding of even the highest end PC hardware. Crysis 1 has been said to be poorly optimized even by developer Crytek. However, I am still very baffled and disappointed about the performance I am getting given the evidence leaning towards my system underperforming. If my PC is indeed underperforming, I need some help finding the root cause, so I can fix it. If there is hardware problem I am leaning towards the possibility of a problem with my PCI-E 16x slot given that I already had my graphics card replaced (although this was before I disabled supersampling AA). My BIOS and drivers are completely up to date. I apologize for the length of this post and any help is greatly appreciated.

    My PC specifications

    Intel Core i7 3770k
    Asus P8Z77-V Pro
    Sapphire Radeon HD 7970 OC Dual X 3GB
    Cooler Master Hyper 612 CPU cooler
    Corsair Vengeance 16GB LP
    SeaSonic Platinum 860 PSU
    Corsair Carbide 500r Case
     
  2. teleguy

    teleguy Maha Guru

    Messages:
    1,363
    Likes Received:
    221
    GPU:
    GTX 1070/Vega 56
    What are your Heaven results?
     
  3. Mraz

    Mraz Master Guru

    Messages:
    664
    Likes Received:
    12
    GPU:
    /
    MSAA x2 eats up you performance like crazy, x4 MSAA I would recommend only for CFX or SLI configurations because it is very demanding.

    Turn off MSAA completely in any game and play with AA x16 or similar that is more then enough, and do not expect anything special really.

    I never use MSAA because I just dont see any reason why I should use a performance eater for eye candy when you can set everything else max in any game, and just turn off MSAA and have amazing performance.

    Just for the record I play BF3 everything maxed out, without MSAA and without blur which I find annoying, and I have well over 100FPS, I know some will say why would you do that, you cant even see it. But trust me I can feel the difference when playing, it is much smoother.

    I would recommend you turning off MSAA completely in any game and just enjoy it as it is with all other settings maxed out.
     
  4. Fox2232

    Fox2232 Guest

    Messages:
    11,808
    Likes Received:
    3,371
    GPU:
    6900XT+AW@240Hz
    50% oversampling > SMAA > precise downsampling to native = best IQ possible with lowes performance impact on most of games.

    MSAA/SSAA are dinosaurs, AMD & nV should give players better options.
    (Dinosaurs are slow and eat a lot.)

    @Mraz, they may not see difference between 60Hz and 100Hz. Their incapability, their death in fps games.
     

  5. l0stPlayer

    l0stPlayer Guest

    Messages:
    11
    Likes Received:
    0
    GPU:
    Sapphire Radeon HD 7970
    Unigine Heaven Benchmark 4.0 results:

    Avg. FPS: 30.3
    Score: 764
    Min FPS: 7.2
    Max FPS: 67.4

    Settings for Heaven Benchmark:

    Render: Direct3D11
    Mode: 1920x1200 8xMSAA fullscreen
    Preset: Custom
    Quality: Ultra
    Tessellation: Extreme
     
  6. Lowki

    Lowki Master Guru

    Messages:
    631
    Likes Received:
    14
    GPU:
    RX 7800 Xt
    Yeah I was going to say man 4x ssaa in crysis lol you are over using aa. I only use ssaa in old games like doom 3 or weak ones like skyrim. I actually think no aa looks great if a game supports diffusion post. That's direct x 11 so hopefully seeing a lot more of it.
     
  7. l0stPlayer

    l0stPlayer Guest

    Messages:
    11
    Likes Received:
    0
    GPU:
    Sapphire Radeon HD 7970
    As I said in my original post, I switched to SSAA to MSAA resulting in a significant performance boost. However, it still seems that games are underperforming especially when compared what other people have been claiming in the videos I have linked. This benchmark for Crysis 1 on the Radeon HD 7970 seems to point to my system underperforming.

    http://www.techpowerup.com/reviews/HIS/HD_7970_X_Turbo/10.html
     
  8. teleguy

    teleguy Maha Guru

    Messages:
    1,363
    Likes Received:
    221
    GPU:
    GTX 1070/Vega 56
    Could you download this tool, run each map once and post a screenshot of the results here?

    PS: Try Adaptive AA instead of SSAA.
     
  9. l0stPlayer

    l0stPlayer Guest

    Messages:
    11
    Likes Received:
    0
    GPU:
    Sapphire Radeon HD 7970
    [​IMG]
     
  10. teleguy

    teleguy Maha Guru

    Messages:
    1,363
    Likes Received:
    221
    GPU:
    GTX 1070/Vega 56
    Your results look alright to me. They're not much different from mine.

    [​IMG]

    [​IMG]
     
    Last edited: Apr 3, 2013

  11. Fox2232

    Fox2232 Guest

    Messages:
    11,808
    Likes Received:
    3,371
    GPU:
    6900XT+AW@240Hz
    There are people who turn Mipmap/Textures to High Performance instead of Quality. Then They may have higher fps at same settings in game, but have much worse image quality.

    Considering how much you can loose by reducing mip/tex quality and how small improvement 8xAA makes. Not really worth running around telling people you have 10% higher fps on same hardware.

    They are just ones who don't know what they are doing.

    And 1200p vs 1080p has 11% more pixels in such shader heavy game it makes quite a difference.
     
  12. l0stPlayer

    l0stPlayer Guest

    Messages:
    11
    Likes Received:
    0
    GPU:
    Sapphire Radeon HD 7970
    This is where I am getting quite confused and conflicted. Performance may seem right in both benchmarks but those videos I have linked in addition to a benchmark I found are indicating that I should be getting 60+ fps in Crysis 1 (at nearly the same graphics settings that I am using) while I am not and a large outdoor map hovers in the 50s in BF3 when another video claims to be getting better performance on a large outdoor map. Granted, I do realize that the number of players does affect performance and I don't remember how many players there were in my last BF3 multiplayer match. In regards to BF3, I have not found many benchmarks for multiplayer (almost all of the benchmarks I have found are single player). I will also admit that I am being reintroduced to PC gaming as I wound up taking a long hiatus years ago and my hardware knowledge is not in the best shape.
     
  13. flow

    flow Maha Guru

    Messages:
    1,023
    Likes Received:
    17
    GPU:
    Asus TUF RTX3080 oc
    Op, you should use default settings in CCC, or leave everything as it is after a clean install. Besides overdrive ofcourse if you use that. If you use other software to manipulate your videocard, then it's best to leave things like AA, vsync and so forth at game managed.
    That way you can decide per game if you want to enable 4x AA and so on.
    Also, a benchmark will be better that way, or at least accurate.

    With all these things enabled in your driver it could lead to performance issues in games.
     
  14. TechnikL

    TechnikL Master Guru

    Messages:
    205
    Likes Received:
    0
    GPU:
    AMD R9 290X 1130/1300
    Nothing wrong with the performance and all your descriptions...

    the only thing slowing you down is over use of AA, (any type).

    with your res 1920X1200 I would run almost all games without AA, because it will be hardly noticeable and send your FPS way up.

    reserve AA for titles that run well over 60 already and then apply AA if you really want it, but it can and will significantly lower your FPS.

    thats my advice, Good Luck :)
     
  15. eclap

    eclap Banned

    Messages:
    31,468
    Likes Received:
    4
    GPU:
    Palit GR 1080 2000/11000
    you're running very high amounts of AA. performance seems ok otherwise. what's your 3dmark11 score with default settings?
     

  16. l0stPlayer

    l0stPlayer Guest

    Messages:
    11
    Likes Received:
    0
    GPU:
    Sapphire Radeon HD 7970
    I have not used 3DMark 11 due to the fact that I cannot use the "Extreme" preset unless I step up and pay for the advanced edition of the software. I am not sure how indicative the "performance" preset would be of my PC's performance.
     
  17. eclap

    eclap Banned

    Messages:
    31,468
    Likes Received:
    4
    GPU:
    Palit GR 1080 2000/11000
    just run the free version default settings, that's what everyone is using and post back with your score.
     
  18. l0stPlayer

    l0stPlayer Guest

    Messages:
    11
    Likes Received:
    0
    GPU:
    Sapphire Radeon HD 7970
    3DMark 11 results, "performance" preset:

    Score: P9355

    Graphics Score: 9403

    Physics Score: 9688

    Combined Score: 8591
     
  19. eclap

    eclap Banned

    Messages:
    31,468
    Likes Received:
    4
    GPU:
    Palit GR 1080 2000/11000
    hmmm, that's around normal for a stock hd 7970 I think. is that with stock clocks?
     
  20. l0stPlayer

    l0stPlayer Guest

    Messages:
    11
    Likes Received:
    0
    GPU:
    Sapphire Radeon HD 7970
    It might go higher as I might have forgot to disable 8x MSAA via Catalyst Control Center which I was using in Mass Effect 2 right before I did the benchmark. As far as my hd 7970 is concerned, it is a factory overclocked version running with a 950mhz GPU clock and 1425mhz memory.
     

Share This Page