1. Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.

SLI GTX560 VS HD6970 benchmarks (30 games tested on 1600x1200 res)

Discussion in 'Graphics Cards' started by Artas1984, Apr 19, 2012.

  1. Artas1984

    Joined:
    Feb 22, 2009
    Messages:
    211 (0.10/day)
    Thanks Received:
    94
    But you understand what i am saying? I could have tested any checkpoint in the game, and you would not know which checkpoint i am testing..

    Is there are inbuilt Crysis 2 benchmark to download?
     
  2. brandonwh64

    brandonwh64 Addicted to Bacon and StarCrunches!!!

    Joined:
    Sep 6, 2009
    Messages:
    18,582 (10.12/day)
    Thanks Received:
    6,090
    Location:
    Chatsworth, GA
    No it does not, BF3 uses a more complex frostbyte 2 engine and is a lot more stressful on cards. If you see it go on sale, you should pick it up and run some benches ;)
     
    Crunching for Team TPU
  3. Melvis

    Melvis

    Joined:
    Mar 18, 2008
    Messages:
    3,577 (1.51/day)
    Thanks Received:
    522
    Location:
    Australia
    Yea i get what ya saying and since ive covered 80% of the game there is a good chance ive gone through where you benchmark it right?

    No there isnt not like number 1, you have to download the benchmark but it never worked for me anyway.
     
  4. ManofGod

    Joined:
    Jul 24, 2007
    Messages:
    108 (0.04/day)
    Thanks Received:
    15
    If your results are not reproducible, then you just wasted yours and our time with these results. Seems to me, you have just picked a spot that favors one card over another and stuck with that.

    Now, if you were to tell us exactly where you tested Crysis 2 so we can see the results in the same place for ourselves, that would be different.
     
    heky says thanks.
  5. Artas1984

    Joined:
    Feb 22, 2009
    Messages:
    211 (0.10/day)
    Thanks Received:
    94
    If i wasted your time, then get the ^**^ lost ok? I do not appreciate this kind of attitude... Who in the blue hell are you to tell me i wasted anyone's time? As far as i am concerned, it's the other way around - i might have saved time for people choosing SLI cards..

    Now as far as reproducible scenes go, here is the the scene where i tested SLI GTX560 VS HD6970

    http://imageshack.us/photo/my-images/6/crysis2z.jpg/

    And the scene where i tested SLI GTX560 DX9 VS DX11? Well it's the very first autosave in front of the closed door!
     
    Last edited: May 2, 2012
  6. ManofGod

    Joined:
    Jul 24, 2007
    Messages:
    108 (0.04/day)
    Thanks Received:
    15
    Cool, thanks for letting us know where the testing took place.
     
  7. caleb

    Joined:
    Sep 15, 2004
    Messages:
    1,540 (0.42/day)
    Thanks Received:
    204
    Location:
    Poland,Slask
    bump for BF3 benchmarks :)
     
  8. Artas1984

    Joined:
    Feb 22, 2009
    Messages:
    211 (0.10/day)
    Thanks Received:
    94
    I already uninstalled BF BC 2, which i did not even own. I ain't psyched for Call Of Duty or Battlefield games anyway...
     
  9. caleb

    Joined:
    Sep 15, 2004
    Messages:
    1,540 (0.42/day)
    Thanks Received:
    204
    Location:
    Poland,Slask
    Dude no offense but I don't care what you get psyched with :) You posted a large VS review and you don't have the top shooter for comparison so expect people to ask for it.
     
  10. EarthDog

    EarthDog

    Joined:
    Dec 31, 2009
    Messages:
    3,222 (1.87/day)
    Thanks Received:
    647
    I respect the amount of time and effort you put in to this very specific (yet unusual) type of comparison.
     
  11. Artas1984

    Joined:
    Feb 22, 2009
    Messages:
    211 (0.10/day)
    Thanks Received:
    94
    I have already included one modern top shooter from the series - BATTLEFIELD BAD COMPANY 2. I think that will be enough. Anyway in other GTX560 reviews i have seen games like Orange Box ones and Call Of Duty 4 included, so you wana say something about modern?:)
     
  12. heky

    heky

    Joined:
    Sep 9, 2009
    Messages:
    886 (0.48/day)
    Thanks Received:
    144
    Location:
    Slovenia, Europe
    Look, i dont want to sound rude, but there is something wrong with at least a couple of your results of the 6970. I get a lot higher framerates on a lot of games you tested, with the same resoution. For example Crysis 2 with high resolution textures, everything on ultra, dx11 mode, i never get such low framerates and my average is almost double. I dont know, its either the proc holding you back, or driver issues.
    Btw i am using 12.4 drivers.
     
  13. Artas1984

    Joined:
    Feb 22, 2009
    Messages:
    211 (0.10/day)
    Thanks Received:
    94
    LMAO, did you run the test with a Core i7 2600K overclocked at 4,6 GHz? And then you compare it with my Phenom II, which, as i said, MHz per MHz is slightly slower than Core 2 Duo E8400? You did read that part did not you?
     
  14. heky

    heky

    Joined:
    Sep 9, 2009
    Messages:
    886 (0.48/day)
    Thanks Received:
    144
    Location:
    Slovenia, Europe
    Yes i did read it, and that is exactly the point. You are bottlenecked. No offence, but a comparison like this is useless if the system is bottlenecked. The 6970 cant really spread its wings.
    Also, crysis 2 at 1080p with high resolution textures, dx11 mode on and all settings on ultra uses more than 1800mb of vram, so the sli setup is worthless on higher resolutions with high detail settings. And this is just one example.
     
  15. EarthDog

    EarthDog

    Joined:
    Dec 31, 2009
    Messages:
    3,222 (1.87/day)
    Thanks Received:
    647
    +1. Plenty of commercial reviews showing the difference between AMD and Intel, especially with SLI/Crossfire configs and games like CivV (the RTS stuff) that is more or less CPU bound anyway. That said it does vary significantly by GAME.

    http://www.techpowerup.com/reviews/AMD/HD_7970_CPU_Scaling/20.html

    Obviously what you have is slower than BD and Nehalem too so the point his is trying to make, Im afraid, is valid.

    EDIT: And please note, Im not trying to be a douche here supporting his statement. You know me from OCF that I am also a reviewer so I understand the time and effort it takes to complete this, especially with 30 games. I like what you do for the most part (the reso used is pretty low for high end cards and especially SLI/Crossfire leaving the CPU to shoulder more of the load) and with some tweaks this testing can become even more worthwhile. :)
     
    Last edited: May 3, 2012
    heky says thanks.
  16. Artas1984

    Joined:
    Feb 22, 2009
    Messages:
    211 (0.10/day)
    Thanks Received:
    94
    That is why i also made the 1920X1200 benchmarks.

    It's strange that the main discussion is even talked here rather than in the higher-res thread..

    So it seems i was indeed bottlenecked. Well next time when i will do something like this, i will certainly do a review on an Ivy Bridge CPU.

    That is the biggest BS said here - or did you miss the part where SLI GTX560 DOUBLED the frame rates from single card? So it is not vram.
     
    Last edited: May 4, 2012
  17. heky

    heky

    Joined:
    Sep 9, 2009
    Messages:
    886 (0.48/day)
    Thanks Received:
    144
    Location:
    Slovenia, Europe
    You clearly dont understand. I can proove to you, that Crysis 2 with High resolution textures, dx11 mode on and everything on ultra uses more than 1800mb of vram. That means you are not running the same IQ on the gtx560sli becouse it can not be run!!!It doesnt use the same IQ becouse of the lack of vram.
     
  18. lyndonguitar

    lyndonguitar I play games

    Joined:
    Apr 1, 2010
    Messages:
    1,629 (1.00/day)
    Thanks Received:
    338
    Location:
    Philippines
    wow, 1 video card all day
     
  19. Artas1984

    Joined:
    Feb 22, 2009
    Messages:
    211 (0.10/day)
    Thanks Received:
    94
    Damn, this is getting annoying. What is your point in all this? Are you saying that SLI GTX560 should have benchmarked the same FPS as a single GTX560 - is that your point or not?

    What kind of ABSURD talk is this? Of course i was running the same IQ settings with SLI GTX560and single GTX560!!! Do i have to film this on camera to you to prove it?
     
    Last edited: May 4, 2012
  20. heky

    heky

    Joined:
    Sep 9, 2009
    Messages:
    886 (0.48/day)
    Thanks Received:
    144
    Location:
    Slovenia, Europe
    Do you undestand the use of vram? Judgeing by your replys you dont. You can set the ingame settings to the max, but it doesnt use the same IQ if there is not enough vram availible!!! And since the game uses more than 1800mb vram on ultra settings, your sli setup cant run the same IQ as the 6970! Get it now?
     
  21. Artas1984

    Joined:
    Feb 22, 2009
    Messages:
    211 (0.10/day)
    Thanks Received:
    94
    Ye i get the use or vram - each card uses it's own RAM and renders half of the image, not summing the vram..

    Lowering IQ settings when a bottleneck applies - is this common in all games or just in Crysis 2? And in what exact resolution and settings does the 1800 Mb limit for Crysis 2 apply?
     
  22. heky

    heky

    Joined:
    Sep 9, 2009
    Messages:
    886 (0.48/day)
    Thanks Received:
    144
    Location:
    Slovenia, Europe
    I dont know if it is common in all games, i do know that in some games you cant even select the highest settings if there is not enough vram available.
    The 1800mb usage is at 1080p with everything on ultra, dx11 enabled and high resolution textures.
    And it is not a 1800mb limit, you can see the vram usage in gpu-z, and i can see it using more than 1800mb with the settings i mentioned.
     
  23. Artas1984

    Joined:
    Feb 22, 2009
    Messages:
    211 (0.10/day)
    Thanks Received:
    94
    OMG, i overclocked my Phenom II X6 1075T to 4 GHz and the FPS in Crysis Warhead jumped from from 22/25/26 to 37/40/42.

    EDIT: overclocked my Phenom II X6 1075T to 4,2 Ghz and the FPS in Crysis Warhead jumped frpm 37/40/42 to 37/41/44.

    CPU won't overclock more. Also it seems there is no reason to go beyond 4 GHz to 4,2 GHz having in mind how much voltage i have applied just for 200 MHz increase.
     
    Last edited: May 4, 2012
  24. ChristTheGreat

    ChristTheGreat

    Joined:
    Jun 29, 2007
    Messages:
    922 (0.35/day)
    Thanks Received:
    364
    some game will run better wit hfaster CPU. Just For Crysis 2, on a GTX 590 (I know it's dual GPU card..)

    http://www.techspot.com/review/379-crysis-2-performance/page8.html

    Anyway, thanks for your time about this, it's a good work ;) Just my 2 cent, you should try to test a game if it's possible, inside a building, and outside. Why? Some card could be faster in a building, but when you get alot of graphics, Outside, it can drop fps.

    good work!
     
    Crunching for Team TPU

Currently Active Users Viewing This Thread: 1 (0 members and 1 guest)

Share This Page