EVGA GTX 680 lacks performace?

Wondering if anyone can help me figure out what's wrong with my 680. If I run 3DMark 11 with stock CPU and stock GPU my scores are lower than expected:
http://img688.imageshack.us/img688/2940/3dmarkw.png

Now if I run it with my i5-2500k OC to 4.5ghz:
http://img210.imageshack.us/img210/3158/3dmarkcpuoc.png

With my CPU at 4.5ghz and GPU OC to +125core/+400mem:
http://img585.imageshack.us/img585/6200/c125m400.png

I've seen people with scores at stock at least pass the graphics with the GTX 680 and I've seen others that pass everything with stock CPU & GPU. What might be the problem?
62 answers Last reply
More about evga lacks performace
  1. Specs:

    Case: Corsair Carbide 500r White
    MOBO: Asus P8Z68-V PRO
    CPU: i5 2500K @ 4.5ghz
    GPU: GTX 680
    RAM: 2x4GB 1600 G.Skill RipJawsX 8-8-8-24
    PSU: Corsair HX750
    SSD: Crucial M4 128GB
    HDD: Western Digital Caviar Black 1TB
    Optical Drive: Asus
    Keyboard: Microsoft Sidewider X4
    Mouse: Logitech G9x
    Speakers: 5.1 Logitech X-540
    CPU Cooler: Cooler Master Hyper 212 Evo
    Monitor: Asus VS248
    Wireless Adapter: Rosewill RNX-N250PCe
  2. Your 3Dmark11 screens look different than mine. However, you are right to be concerned about the score. However, one thing that looks very odd to me is the target score it gives you.

    I had to get a score of P9500 to be considered comparable to other systems and that took some overclocking.

    Your score is saying it only needs P8600, which leads me to believe you have a slightly different version of the software.
  3. I believe that changes with the hardware. Considering that you have an i7 which supports hyper threading you require a higher score. I've seen people with the same processor as you and your score is similar to theirs.

    With your GPU at stock does your graphics score pass?
  4. Also if your processor has more than 4 cores then it requires a much higher score than mine being that the i5 only has 4 cores and doesn't support HT
  5. My graphics score was much higher as well. Are you running at stock, or do you have an OC? I currently have HT turned off, and I believe I scored 10k without HT on, which means I had a slower CPU than you. I have the original i7 that had a stock clock of 2.66ghz.
  6. The first and second SS above is with my GPU at stock settings the third has an OC of 125/400

    Even if you disable HT 3DMark 11 expects you to score a certain amount based on your hardware.
  7. You think you can post a SS like the ones I have above with your GPU at stock? using 3DMark 11 Performance.
  8. I did notice that with HT turned off, they score needed has lowered quite a bit. Here is a score with HT off and my i7 at 3.8ghz (you can see in picture). The GTX 680 is stock, which is a higher graphics score than yours.

    http://img37.imageshack.us/img37/7326/3dmark11stock680.png

    Uploaded with ImageShack.us

    Out of curiosity do you have 3d vision enabled in the drivers (still not on in game)? I've found with it simply enabled in the Nvidia control panel hurts your score.
  9. It isn't a large difference when at stock, but with HT off and a lower CPU speed, I'd think your i5 would score higher with that OC you have.
  10. I have no idea if 3D vision is enabled or not. How can I disable this?
  11. Open up the Nvidia Control Panel, and look to the left for the sub section "Stereoscopic 3d" and select "Set up stereoscopic 3d" under that, and you just need to make sure nothing is enabled on the screen to the right.
  12. Nope. It's off.
  13. youll be chasing scores till you die and what do they give you? nothing so be happy maxing out most games you play and dont chase the benchmarks.
  14. I think the point was that he felt the score was lower than it should be. It does seem a little low, but not a huge amount. He likely has something in the background slowing things down.
  15. I have two 6970s in crossfire which is somewhat stronger than a single 680 (but not by much), but you whip my butt on 3D mark 11. All of you guys with 680s do. I wonder why...
  16. cbrunnem said:
    youll be chasing scores till you die and what do they give you? nothing so be happy maxing out most games you play and dont chase the benchmarks.


    I'm not aiming for a certain score but it seems to me that my score isn't on par at where it should be for stock settings. I'm not worried about having a better score. I do know that not all GPU are the same and some OC better than others but at stock I should be able to pass the score for my hardware.
  17. PCgamer81 said:
    I have two 6970s in crossfire which is somewhat stronger than a single 680 (but not by much), but you whip my butt on 3D mark 11. All of you guys with 680s do. I wonder why...


    I'm trying to figure out if 3DMark 11 keeps tracks of overclocks.

    You think you can run 3DMark 11 Performance with your CPU at stock and your cards as well and post a SS.
  18. BTW to get to the screen like mines with the clouds and the sun you need to hit the how to make your system better button.
  19. Ya, I did find that one.
  20. aRodr1guez said:
    I'm trying to figure out if 3DMark 11 keeps tracks of overclocks.

    You think you can run 3DMark 11 Performance with your CPU at stock and your cards as well and post a SS.


    I did the GPU at stock, but changing my CPU OC is much more complicated, and I won't be doing that.
  21. bystander said:
    I did the GPU at stock, but changing my CPU OC is much more complicated, and I won't be doing that.


    I was referring to PCgamer81
  22. I will give it a whirl. It has been a while since I last done it, but my score isn't great.

    Just let me know how you want me to run the tests, and I will have the results of each test posted by noon tomorrow. I hope I will be of some help.
  23. PCgamer81 said:
    I have two 6970s in crossfire which is somewhat stronger than a single 680 (but not by much), but you whip my butt on 3D mark 11. All of you guys with 680s do. I wonder why...


    There are a couple benchmark threads around here. One thing I noticed from one 7970 owner clocked to 1250 core is that my GTX 680 scored higher with 3dmark11 even with his i7 3950 clocked to 4.5+. However, his score was quite a bit higher in 3dmark Vantage. The two cards have different strengths.
  24. PCgamer81 said:
    I will give it a whirl. It has been a while since I last done it, but my score isn't great.

    Just let me know how you want me to run the tests, and I will have the results of each test posted by noon tomorrow. I hope I will be of some help.


    Just run 3DMark 11 on Performance with your CPU & GPUs at stock settings. Thanks.
  25. Sorry I have been busy.

    I will run 3D mark right away...
  26. So far my passmark score is 2675.

    I will try to figure out how to do the 3D mark...
  27. P8864
  28. That's nearly identical to my system with HT off, and OC'ed to 3.8ghz, with the stock gtx 680.

    http://3dmark.com/3dm11/3166558
  29. PCgamer81 said:
    P8864


    Do you have a break down of the Graphics score? The physics scores will play a part in the total score, but the graphics score is more meaningful in this case.
  30. bystander said:
    Do you have a break down of the Graphics score? The physics scores will play a part in the total score, but the graphics score is more meaningful in this case.


    That's exactly how I feel about it. I need to see someone else run the same test as me with an i5 and gtx 680 I'm starting to think the i5 is being a bottleneck but that really shouldn't be the case. It's weird.
  31. Some people get unlucky with the dynamic OC is what I'd assume.
  32. bystander said:
    Do you have a break down of the Graphics score? The physics scores will play a part in the total score, but the graphics score is more meaningful in this case.

    I am trying to post the screenshot but am having trouble getting back there...hang with me...
  33. Okay, so I couldn't figure out how to post a screenshot that covers the entire page so I will just post the link...

    http://3dmark.com/3dm11/3180613;jsessionid=4lriwubf7t8n1ij3fkn3t8dbw

    As you can see, I gain a significant jump from overclocking, but it's still significantly less than what you get overclocked.
  34. Your graphics score is pretty damn amazing. Do you use one of your GPU for physX? or do you have it set to auto?
  35. aRodr1guez said:
    Your graphics score is pretty damn amazing. Do you use one of your GPU for physX? or do you have it set to auto?

    I wish. As for as I know AMD GPU's can't be used for PhysX.

    P9783 isn't any better than yours I don't think. 120 more is not a lot. Especially considering it's two against one.

    Hey, remember my GPU is overclocked on this one.
  36. Here is with a reasonable high OC, you can see the graphics score is pretty much the same as yours. HT is likely why my total score is higher.

    http://3dmark.com/3dm11/3137017

    As far as I know, PhysX is not used in 3dmark11, and from my more recent runs in Vantage, it is no longer used on the GPU in Vantage either.
  37. bystander said:
    Here is with a reasonable high OC, you can see the graphics score is pretty much the same as yours. HT is likely why my total score is higher.

    http://3dmark.com/3dm11/3137017

    As far as I know, PhysX is not used in 3dmark11, and from my more recent runs in Vantage, it is no longer used on the GPU in Vantage either.


    What's your overclock on your GPU?
  38. It was +170 core and +300 on VRAM. I know the Vram will go higher easily, but I afraid to push the VRAM for fear of causing permanent damage.
  39. bystander said:
    It was +170 core and +300 on VRAM. I know the Vram will go higher easily, but I afraid to push the VRAM for fear of causing permanent damage.


    Lucky you. I can't push my core pass +125 and as for my mem I can't go over +425

    What your score look like in Heaven DX11 with everything maxed out at 1080p? With an overclock of +125/+300
  40. By chance, do you have FXAA as on in your Nvidia Control Panel defaults? That would certainly drop your scores a little.
  41. aRodr1guez said:
    Lucky you. I can't push my core pass +125 and as for my mem I can't go over +425

    What your score look like in Heaven DX11 with everything maxed out at 1080p? With an overclock of +125/+300


    This is with everything maxed including x8 AA.
    http://img215.imageshack.us/img215/8081/unigineextreme30125300.png

    Uploaded with ImageShack.us
  42. Here's mine:
    http://img513.imageshack.us/img513/6489/heaven125300.png

    Yours was clocked at 125/300?
  43. What I do notice is that your CPU flags 3800 while mines is at 3300mhz
    Wonder why mines isn't showing the overclock
  44. http://3dmark.com/3dm11/3181039;jsessionid=1gzebvyl8kpzg1ge4gt9zwy2b8

    I overclocked a tad higher.

    4.50GHz on my 2500k and 950/1450 on my two 6970s.
  45. aRodr1guez said:
    What I do notice is that your CPU flags 3800 while mines is at 3300mhz
    Wonder why mines isn't showing the overclock


    It is curious, but from my experience, the CPU plays pretty much no part in the score. Even at 2.66Ghz, I get the same scores.

    But your score is not that different. Only 1.8 DPS different.
  46. Why is your name on my test?

    Weird.

    Also, I had my fan set too low because I forgot to set back to auto after disengaging MSI Afterburner - I think I caused my GPU score to be sabotaged because of heat.

    Let me run the Heaven bench...
  47. PCgamer81 said:
    http://3dmark.com/3dm11/3181039;jsessionid=1gzebvyl8kpzg1ge4gt9zwy2b8

    I overclocked a tad higher.

    4.50GHz on my 2500k and 950/1450 on my two 6970s.


    Your graphic score dropped a hell of allot compared to the last test you did?
  48. PCgamer81 said:
    Why is your name on my test?

    Weird.

    Also, I had my fan set too low because I forgot to set back to auto after disengaging MSI Afterburner - I think I caused my GPU score to be sabotaged because of heat.

    Let me run the Heaven bench...


    That is weird about my name being on your test.

    As far as the fan, the only way that would alter the score is if throttled the clocks due to over heating. That said, Heaven benchmark is a pure GPU test, so it will be more about the cards. I have also noticed 3Dmark11 favors Nvidia while Vantage favors AMD.
  49. bystander said:
    It is curious, but from my experience, the CPU plays pretty much no part in the score. Even at 2.66Ghz, I get the same scores.

    But your score is not that different. Only 1.8 DPS different.


    Yea I can see. Do you have physX on auto and have you changed any of the settings in nvidia's control panel?
Ask a new question

Read More

Graphics Cards Gtx GPUs CPUs Graphics