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

DICE Posts its Own Battlefield 4 DirectX vs. Mantle Performance Numbers

Discussion in 'News' started by btarunr, Jan 30, 2014.

  1. zhangteng242007

    zhangteng242007

    Joined:
    Nov 2, 2011
    Messages:
    5 (0.00/day)
    Thanks Received:
    1
    Location:
    WenZhou
    The Mantle's Visual effects same as DirectX?
    And DirectX 11,11.2 was support?
    I think this technology is not practical.
     
  2. zhangteng242007

    zhangteng242007

    Joined:
    Nov 2, 2011
    Messages:
    5 (0.00/day)
    Thanks Received:
    1
    Location:
    WenZhou
    We need to be optimized, rather than an independent technology, it couldn't survive long!Unless everyone want to the API.
     
  3. the54thvoid

    the54thvoid

    Joined:
    Dec 14, 2009
    Messages:
    5,498 (2.15/day)
    Thanks Received:
    4,377
    Location:
    Glasgow - home of formal profanity
    Dude, stop shouting.....
     
    Nihilus and RCoon say thanks.
  4. TheMailMan78

    TheMailMan78 Big Member

    Joined:
    Jun 3, 2007
    Messages:
    21,851 (6.28/day)
    Thanks Received:
    8,503
    I THINK ITS AWESOME! WE SHOULD ALL TALK WITH BOLD CAPS! I LOVE PIE!
     
    MxPhenom 216 and Lionheart say thanks.
  5. Cheeseball

    Joined:
    Jan 2, 2009
    Messages:
    724 (0.25/day)
    Thanks Received:
    103
    I WISH I WERE A BIRD

    Hopefully I'll be able to see great gains with my current setup.
     
  6. TheHunter

    TheHunter

    Joined:
    Apr 10, 2012
    Messages:
    1,332 (0.78/day)
    Thanks Received:
    639
    Location:
    78°55' N, 11°56' E
    nvidia fanboys need to bitch @ nvidia

    Let NV make a proper mantle driver - its a open API not bound to GCN, end of story.

    But nooo, its AMD fault and AMD sucks and what not.. Like i said this once some nv fanboys are like a plague in gaming industry...

    You should have seen all the commnets at battlelog, Omg:rolleyes:


    I own a NV gpu and I still think Mantle is almost God sent.. Just wait until its more widespread.
     
    Last edited: Jan 30, 2014
    rpsgc and Lionheart say thanks.
  7. tokyoduong

    Joined:
    Jan 22, 2013
    Messages:
    478 (0.34/day)
    Thanks Received:
    119

    English? Are you in China being paid by NV or something?
    Are you using your translator properly?
     
    Nihilus and NeoXF say thanks.
  8. TRWOV

    TRWOV

    Joined:
    Aug 11, 2011
    Messages:
    4,312 (2.21/day)
    Thanks Received:
    3,001
    Location:
    Mexico

    I think that's because of the iGPU's memory bandwidth, think of a DDR3 HD7750 vs the GDDR5 version. The other setups had the benefit of using GDDR5 plus I suppose ROPs have a hand in it too (8 vs 32 vs 64x2)






    [​IMG]
     
    Last edited: Jan 30, 2014
    Crunching for Team TPU
  9. Prima.Vera

    Prima.Vera

    Joined:
    Sep 15, 2011
    Messages:
    3,456 (1.81/day)
    Thanks Received:
    683
    Stop feeding the trolls please.
     
  10. Crap Daddy

    Crap Daddy

    Joined:
    Oct 29, 2010
    Messages:
    2,867 (1.28/day)
    Thanks Received:
    1,238
    Good, in multiplayer they brought the 8350 close to a 4670, so basically if you run a Haswell i5 you'd still be better off that with an FX

    BF-4-1920-x-1080-Ultra-settings-GTX-770-vs-7970.jpg
     
  11. btarunr

    btarunr Editor & Senior Moderator Staff Member

    Joined:
    Oct 9, 2007
    Messages:
    32,791 (9.79/day)
    Thanks Received:
    16,528
    Location:
    Hyderabad, India
    YES IDENTICAL (THEY ARE CODE COMPATIBLE, WITH MANTLE HAVING ADDITIONAL CAPABILITIES THAT ARE EXCLUSIVE TO XBOX DIRECTX.)



    DIRECTX 11.1 IS SUPPORTED.

    THE GREEDIEST GAME STUDIO ON THE PLANET (EA) DISAGREES WITH YOU.

    Alright guys, enough screaming.
     
    Nihilus says thanks.
  12. zhangteng242007

    zhangteng242007

    Joined:
    Nov 2, 2011
    Messages:
    5 (0.00/day)
    Thanks Received:
    1
    Location:
    WenZhou
    Yes,China's English!Forgive me,My English was bad,I will to learn more in free time.
    Actually,I really like english,but when I was a child did not study hard
     
  13. zhangteng242007

    zhangteng242007

    Joined:
    Nov 2, 2011
    Messages:
    5 (0.00/day)
    Thanks Received:
    1
    Location:
    WenZhou
    Good reply!God reply!Thanks
     
  14. Prima.Vera

    Prima.Vera

    Joined:
    Sep 15, 2011
    Messages:
    3,456 (1.81/day)
    Thanks Received:
    683
    with all this huge bold text all over I thought my Windows 8'd DPI are fracked up again. ;))) Actually I was also trying to zoom out like an idiot...
     
    Lionheart says thanks.
  15. FX-GMC

    FX-GMC

    Joined:
    Aug 26, 2013
    Messages:
    954 (0.79/day)
    Thanks Received:
    402
    Location:
    Cumberland Plateau
    I'm curious to know why the FX6300 has a higher fps than the FX8350 in your picture there. This is contrary to logic, all other tests i've seen, to all the people whining that their FX six cores are struggling (From battlelog forums) and my own experience. Unless the combination of a FX8320 (4.6) + GTX760 is faster than a FX8350 (stock) + 7970Ghz or GTX770.
     
  16. Crap Daddy

    Crap Daddy

    Joined:
    Oct 29, 2010
    Messages:
    2,867 (1.28/day)
    Thanks Received:
    1,238
    Well, hardwarepal, where I find these benchmarks says:

    I just pass the info. We mostly rely on reviews to make decisions regarding what hardware we buy. This is one of them and it tests the CPUs in multiplayer conditions, the only thing worth playing in BF4. Whether it's accurate, I realy can't say.
     
  17. FX-GMC

    FX-GMC

    Joined:
    Aug 26, 2013
    Messages:
    954 (0.79/day)
    Thanks Received:
    402
    Location:
    Cumberland Plateau
    That weird situation where the i7 and 8350 are performing worse that slower cpu's means something with their test system is not optimized.

    These are more in line with the real world (or at least what i've seen/read):

    [​IMG]

    [​IMG]

    To give reference, my setup (8350 @4.6 + GTX760 @ 1267MHz) gets an average of 72fps with a minimum of around 58-60fps.

    1080p Ultra settings 4xmsaa high post, etc.
     
    Last edited: Jan 30, 2014
    NeoXF says thanks.
  18. NeoXF

    Joined:
    Sep 19, 2012
    Messages:
    615 (0.40/day)
    Thanks Received:
    80
    They sure don't look accurate. Most reviews on CPU performance, including the early Alpha/Beta tests, show it being close to, or even faster than the i5 4670K, and FX 9590 being at or slightly under i7 4770K levels.

    Edit: Pretty much what FX-GMC just said/posted before I pressed post (thanks!).
    Edit2: So what now again... is a FX 8350 build ~15% faster than i5 4670K one in BF4? AMD are witches! We all now Bulldozer sucks! Booooo! etc etc...

    Nope, BF3 was quite CPU agnostic... BF4 is a pretty different kind of beast. I guess you haven't followed many CPU benches on BF4, have you. :)
     
  19. tokyoduong

    Joined:
    Jan 22, 2013
    Messages:
    478 (0.34/day)
    Thanks Received:
    119
    Your system gets that frame rates at what map? how many players? what is happening? Does your settings match theirs? Does your set up match theirs?

    Your FPS can vary by just about anything. This is why benchmarks are made! Same exact repeatable scenario and expected results!
     
  20. FX-GMC

    FX-GMC

    Joined:
    Aug 26, 2013
    Messages:
    954 (0.79/day)
    Thanks Received:
    402
    Location:
    Cumberland Plateau
    Except by nature, multiplayer benchmarks aren't repeatable.

    Here we go:

    Your system gets that frame rates at what map? No specific map (do note I am talking about the vanilla maps). When I play I notice fps in the low 70's most of the time. Lowest I've seen is around 55fps but usually min frames are around 60. Max frames can go up to 90.

    How many players? 64......duh.

    What is happening? Ever played battlefield? Shooting, explosions, people raging, flying, driving. That's what happens in battlefield.

    Does (I think you mean do there) your settings match theirs? Either matches their settings or they are skewing the numbers. You must of missed the part where I said MAX settings 1080p. (Well i didn't say MAX, but Ultra, 4xmsaa, and high post should cover max settings.)

    Does your setup match theirs? No it doesn't. Mine should be weaker, but my performance is better. If you would have followed along this has all been posted already.




    Got anymore questions?
     
  21. Crap Daddy

    Crap Daddy

    Joined:
    Oct 29, 2010
    Messages:
    2,867 (1.28/day)
    Thanks Received:
    1,238
    I'll post another one from sweclockers where the 4670k is considerably faster than an 8350, both stock with a high-end card. 4972.png
     
  22. FX-GMC

    FX-GMC

    Joined:
    Aug 26, 2013
    Messages:
    954 (0.79/day)
    Thanks Received:
    402
    Location:
    Cumberland Plateau
    What do you mean by "both stock and with a high-end card"?

    Also, why are we talking about the i5 4670k now (I'm not taking the bait)? This was strictly about the FX6300 being faster in their "test" when it's not in the real world.
     
    Last edited: Jan 30, 2014
  23. Deadlyraver

    Joined:
    Nov 17, 2011
    Messages:
    286 (0.15/day)
    Thanks Received:
    23
    TBH Frame rate consistency is more my concern than frame rates. Whats the point drooling on high frame rates when they don't even stay up for long? See this kind of comparison brings me assurance that there will be less slowdowns in the midst of air strikes.
     
  24. Crap Daddy

    Crap Daddy

    Joined:
    Oct 29, 2010
    Messages:
    2,867 (1.28/day)
    Thanks Received:
    1,238
    It wasn't about the 6300. It was about the Dice results with Mantle and the 8350 vs. the 4670 paired with the same AMD card. Remember this is a thread about Mantle. Go back to my first post and you'll see that my conclusion regarding mantle is that, with high end set-ups it improves framerates for AMD CPUs better than with Intel CPUs, thus bringing the FX closer if not equal to Haswell i5. I showed you a couple of benchmarks to prove this.
     
  25. FX-GMC

    FX-GMC

    Joined:
    Aug 26, 2013
    Messages:
    954 (0.79/day)
    Thanks Received:
    402
    Location:
    Cumberland Plateau
    The only part of your post that I had an issue with was the first benchmark you posted because it doesn't conform with real world usage. Remember this:
    I wholeheartedly agree that Mantle will benefit AMD procs more than Intel's. I just didn't get why you posted a benchmark that was 4670k and 8350 when I wasn't talking about any intel processors. Then again, I did assume you posted it in response to me. If you weren't posting it in response to me than my apologies and carry on.
     

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