Battlefield 3 Benchmarked, GPU and CPU Performance Tested

By Julio Franco
Oct 28, 2011
Post New Reply
  1. red1776

    red1776 Omnipotent Ruler of the Universe Posts: 5,867   +74

    no you didn't
    .
    yeah ..C&P'ing that was really labor intensive.

    why would you expect that every two or three old gen CPU under the sun would be/ should be on the list anyway? If you think that (like you inferred) he is an AMD shill, you should read, well...the review for one.
  2. dividebyzero

    dividebyzero trainee n00b Posts: 4,783   +639

    Toto, I've a feeling we're not in 2007 any more.
    ....but it's too taxing ?

    It's a competition right? what I win ?

    Troll = picture gazing only. Troll reading = A bridge too far.

    Hopefully the troll will scour the interwebs looking fruitlessly for a site showing yesteryears CPU performance in BF3 in order to make a glorious reappearance in the forum......or not
  3. Steve

    Steve TechSpot Staff Posts: 1,281   +395 Staff Member

    I was going to include the old Core 2 Duo processors but then Athlon64 X2 owners would want their processors included and if I was going to test them I should include the Pentium 4 range as well and before you know it I would be testing BF3 with Pentium III processors.

    In all seriousness though we understand everyone wants their configuration tested but let's be realistic that's just not possible, we do our best to fully test the current generation of hardware as well as the previous and for Intel that does not include the Core 2 range.
  4. fpsgamerJR62

    fpsgamerJR62 Newcomer, in training Posts: 489

    Great article, Steve. I still have one PC running a Core 2 Duo E7500 CPU along with a 4-year old 512 MB 8800GT video card plus another GTX 275 card inside my Windows 7 box and I don't think I need benchmarks to show me that all of the above hardware are seriously past their prime and no longer suited for current generation games.
  5. I think it would be in your best interest to include a GTX 570 SLI benchmark when you write these types of reviews. If it's listed in the Desktop PC Buying Guide in your Luxury System configuration, wouldn't you want to prove to your readers how good the recommendations you make actually are?

    Just a thought -- not trying to be rude.
  6. Ultraman1966

    Ultraman1966 Newcomer, in training Posts: 69

    I only upgraded my CPU to a q6700 just over 2 years ago (due to mobo compatibility). Let's be honest, the majority of people out there are probably still using Conroe processors. However, it's understandable that for the hardcores such technology is archaic and not worth a second look!

    At any rate, thanks for the fairly comprehensive report!
  7. can we turn off msaa in ultra settings or it forces msaa to 4x?????
  8. Per Hansson

    Per Hansson TS Server Guru Posts: 1,917   +119 Staff Member

    No, but you can select "Custom" and make any graphical setup you would like.
    So for example a "Ultra" without MSAA is possible to setup manually...
  9. Nah... I'll just get a PS3 instead.
  10. Steve

    Steve TechSpot Staff Posts: 1,281   +395 Staff Member

    If you want to be really cool you can play BF3 on PC in the PS3 mode. Just lower the resolution to 1280x800 and use the low quality preset. Then buy a $50 graphics card and you are good to go, ohh once you ditch the keyboard and mouse for a sweet game controller.
  11. dividebyzero

    dividebyzero trainee n00b Posts: 4,783   +639

     
  12. Per Hansson

    Per Hansson TS Server Guru Posts: 1,917   +119 Staff Member

    Referring to my first post in this thread HardOCP now updated their BF3 article with IQ and Performance analysis: http://www.hardocp.com/article/2011/11/02/battlefield_3_single_player_performance_iq_review
    I found their testing highly interesting, and especially the part about the MSAA vs FXAA:

    "We found that 2X AA was playable at this resolution, but the performance hit was too severe and there was no IQ advantage compares to High FXAA. In the case of Battlefield 3, there is no practical reason to choose MSAA over FXAA. In these tests, the Radeon HD 6950 was about 4.9% slower than the GeForce GTX 570."
    -----------------------
    "Let's not forget that MSAA and FXAA can be used together uniquely in this game. DICE talked up that possibility, and even mentioned that they "complement each other," but we feel that the reality of the situation does not warrant much excitement. Yes, they work together, but there is no immediately and persistently noticeable reason to do it. If you take still screenshots and zoom in a few hundred percent, it is easy to find differences side by side. But if you play the game, chances are you'll never actually see what is different with MSAA and FXAA as opposed to just FXAA."

    "In AMD's own review guide for this game AMD recommends to use FXAA in this game instead of MSAA. This is a bold statement from AMD since FXAA is the competitors technology. This leads credence to the benefit and positive effect that today's shader based AA technology provides. It is very easy to see that FXAA is more effective than MSAA in this particular game title."
  13. DanUK

    DanUK TechSpot Enthusiast Posts: 189   +7

    +1 :'D
  14. lipe123

    lipe123 TechSpot Guru Posts: 408   +73

    Hey I have a request, instead of testing the biggest baddest cards in SLI/X-fire how about testing a few low end popular cards in SLI like the 460/560 ti/and whatever ATI has in that range.

    I mean two 560 Ti's is cheaper than 1 gtx 580 and I've been looking into other tests done and it seems like the are faster than a 580.

    Just something to think about.
  15. dividebyzero

    dividebyzero trainee n00b Posts: 4,783   +639

    @lipe123
    You may want to PM Steve to organise sending your card(s) to him, as I'm pretty certain that he doesn't have every graphics combination to hand...I mean to say, if you're going to benchmark SLI 560Ti, then you pretty much have to bench with CFX HD 6870 and CFX HD 6950....then of course you would also need to bench SLI'ed GTX 460 1GB to satisfy the "bang-for-buck" crowd (see previous game perf ormance review comments)...which would necessitate CFX HD 6850's and/or HD 5850's...then of course three HD 6770's are barely more expensive than a HD 6970.

    So, aside from the obvious investment required in obtaining these extra cards ( I doubt any vendor would give out a sizeable quantity of -already released- freebee cards that won't benefit from the publicity in a standalone review), you also run into problems with swap outs and testing ( i.e. time) and storage. Look at poor old Tom's Hardware...they had so many cards they had to put a few grands worth directly in contact with a carpet!
    [​IMG]

    [click to enbiggen the pic]
  16. Steve

    Steve TechSpot Staff Posts: 1,281   +395 Staff Member

    That's exactly right it is not possible nor is it realistic to start including random SLI and Crossfire setups. The graphics card collection that I have sitting on my office desk is about twice as large as the one pictured and its still not possible.

    Generally those that have gone for setups such as a Radeon HD 5770 Crossfire config know why they have selected such a setup and how it performs when compared to a single Radeon HD 5870 or 6950 for example. So really given the measuring stick that we provide with our articles you can work a lot out for yourselves ;)
  17. This report is not an accurate representation of CPU performance on large 64 player maps such as Caspian Border.

    These maps and MP modes stress the CPU a lot more than some pre-scripted SP portion of the game.

    I'd advise you re-do your CPU test on Caspian Border when its full of players, yes it may not be easy to test as there is no timedemo or exact sequence but all you need to do is run an average with FRAPS and see what results you get.

    I can tell you that dual core CPU's struggle to get over 30fps, and older Quad often dip into the low 30's.... so the conclusion in your CPU page is wrong for BF3 multiplayer.
  18. I currently running a

    Amd phemon II x 4 965 BE (3.4Ghz) (at stock speeds with a stock cooler "yes it is loud but just turn the volume up")

    8 Gigs DDR3 1333mhz kingston ram

    500 gig caviar drive

    EVGA GTX 275 SC (Overclocked to FTW settings)

    and im currently playing BF3 with custom settings

    geforce 275.33 driver (as the newest driver has major issues still)

    All on Ultra, no AA, res at 1600x900 as that is what my 20" monitor will take

    AF at 16x

    Terran at medium,

    turned of HBAO and SSAO

    and im getting 70 fps+ on average and at the lowest 58fps thus far.

    yes i dont have DX11 but who needs it for now.

    cheers
     
  19. Alpha Gamer

    Alpha Gamer TechSpot Booster Posts: 242   +42

    Well, with my setup (Phenom II X6, 1055T, GTX 560 Ti, 4Gb ram), I can play single player on ultra with MSAA OFF and have very decent framerate on 1920x1080. I mean, most of the time it is above 50 and very rarely below 40. But if you value performance over eye candy, you might want to stick to the high preset on a gtx 560 ti in full hd.
  20. which monitor are you using for testing at 2560*1600??
  21. Steve

    Steve TechSpot Staff Posts: 1,281   +395 Staff Member

  22. I run all my games at 1366x768 this review is useless for me, still I enjoy gaming and have fun as much as any of you guys, You may say thats obsolete, to you yes, for me no. Hey I still get to shoot and kill the bad guys, come on, in 20 years 2560X1600 will be laughed at. there will be a time when 2000000 X1600000 resolutions come. Look back and you'll see how foolish you all had been for wasting that much money for so ultra x ultra x ultra low a resolution. You may say see what 2560 X 1600 looks like, no I will not look at it. All I have seen I thoroughly admire it anyway. Days back when 1024X768 was highest possible did you miss 2560X1600? no well. you don't know what you miss until you've seen it.
  23. red1776

    red1776 Omnipotent Ruler of the Universe Posts: 5,867   +74

    Hey look! here comes the nurse with the medication cart.
  24. LNCPapa

    LNCPapa TS Special Forces Posts: 4,272   +258

    ROFL - I was starting to think along those same lines as I was reading that last post, red.
  25. Steve

    Steve TechSpot Staff Posts: 1,281   +395 Staff Member

    I don’t think you were making the point you were trying to.


Add New Comment

TechSpot Members
Login or sign up for free,
it takes about 30 seconds.
You may also...
Get complete access to the TechSpot community. Join thousands of technology enthusiasts that contribute and share knowledge in our forum. Get a private inbox, upload your own photo gallery and more.