Official Rf2 graphic card performance list

Discussion in 'General Discussion' started by tsunamibr, Sep 21, 2012.

  1. tsunamibr

    tsunamibr Registered

    Joined:
    Mar 25, 2012
    Messages:
    15
    Likes Received:
    0
    Hi there,

    I'm looking for a new graphic card for the upcoming release of Rf2. I've been trying to gather all the information about several cards and its performance on Rf2 to choose my next one, but all I found was rather sparse and incomplete. So, I´d like to ask you to help me fill this thread with your specific graphic card (system performance), your settings and your results (fps). I suppose this will become a very useful thread for many of us.

    P.S. It's a specific video performance thread, so I´m assuming everybody posting here have a good enough CPU not running as the bottleneck system.

    Video Card:
    FPS (min, max, avg):
    Resolution:
    Anti-aliasing (type, level):
    HDR:
    Anisotropic Filtering:
    V-sync:
    Circuit Detail:
    Player Detail:
    Opponent Detail:
    Special Effects:
    Shadows:
    Shadows Blur:
    Sun Occlusion:
    Road Reflection:
    Environment Reflection:
    AI in field:
    Number of visible vehicle:
     
  2. Gearjammer

    Gearjammer Registered

    Joined:
    Jun 11, 2012
    Messages:
    1,823
    Likes Received:
    24
    The best place I have found to find performance information on graphics cards is tomshardware.com. Every quarter they release an article called best graphics card for the money for what ever month it is released. At the end of the article is a list of graphics cards from both major manufacturers and the list is top down based on performance, so if you look at the top one, that will be the most powerful GPU on the market as of the date of the article.

    Using that list it is easy to compare cards for performance.

    One other thing that site does well is test cards against each other, separately and more. They provide complete system specs and testing specs so that anyone could duplicate their tests if desired. They are also one of few sites that are not biased that I can tell.
     
  3. MaXyM

    MaXyM Registered

    Joined:
    Oct 5, 2010
    Messages:
    1,774
    Likes Received:
    29
    problem is that cpu is actually the bottleneck (due to gfx engine design)
     
  4. Gearjammer

    Gearjammer Registered

    Joined:
    Jun 11, 2012
    Messages:
    1,823
    Likes Received:
    24
    If you are referring to the lack of use of all cores, then yes, you are correct MaXyM.
     
  5. Ricknau

    Ricknau Registered

    Joined:
    Nov 12, 2011
    Messages:
    778
    Likes Received:
    39
    If you don't specify the conditions you will end up with a pile of non-comparable results. Seems i remember someone else already did this in another thread.
     
  6. MarcG

    MarcG Registered

    Joined:
    Oct 4, 2010
    Messages:
    6,854
    Likes Received:
    2,234
    another vote for Tomshardware comparison pages, used them for years. Dont forgoet that RF2 is still getting optimizations so results will vary from Build to Build, best bet is just to use Compairson Guides, review sites and google, find whats best in your price range and go for that.
     
  7. martymoose

    martymoose Registered

    Joined:
    Aug 10, 2012
    Messages:
    461
    Likes Received:
    13
    If someone posts a replay file that could be useful as a benchmark tool along with fraps.
     
  8. jtbo

    jtbo Registered

    Joined:
    Jan 12, 2012
    Messages:
    1,668
    Likes Received:
    48
    Monaco with GTR car has been heaviest for me, all my replays seem to be from older versions so not working, but I can drive another slow lap there to be benchmark, my computer is bit slow by today's standards so I haven't had luck to get AI cars to work on that track without dropping under 60fps. Even driving alone I need to put track to low, car to med-high and textures to high instead of full, otherwise I run into problems.
     
  9. jtbo

    jtbo Registered

    Joined:
    Jan 12, 2012
    Messages:
    1,668
    Likes Received:
    48
    1:24.x slow lap (I think it was 2nd lap after exiting pits for that stint), but should be usable for performance testing, running this at full detail settings is something my computer can't handle, add 20AI cars and things grind to halt. So 8800GTS 512 simply is not enough to get decent performance from rF2, at least not with E6750 Dual Core.

    I use very low detail, no HDR and 1920x1080 resolution, when attempting to drive even bit seriously.
     
  10. MaXyM

    MaXyM Registered

    Joined:
    Oct 5, 2010
    Messages:
    1,774
    Likes Received:
    29
    Not exactly. I'm referring to flaws in gfx engine, which loads CPU to extent making impossible to utilize power of gfx card in full range. In other words you may have powerful gfx card but still struggle with performance (see SLI issues) because of CPU usage by gfx related tasks. Read about draw calls in posts of KSzczech for details.
     
    Last edited by a moderator: Sep 21, 2012
  11. martymoose

    martymoose Registered

    Joined:
    Aug 10, 2012
    Messages:
    461
    Likes Received:
    13
    I will run your replay file on a few bench settings, my earlier tests showed the engine being cpu limited on my 5ghz i7 as changing graphic settings has very little effect on performance for me until I run out of gpu ram that is. Using SLI gtx 480's there are times it could do with more then 1.5GB of video ram with a full field and some aa settings.

    It doesnt really matter about the driving for the replay file just so others can use the same and give a more usefull comparison between setups.
     
  12. 88mphTim

    88mphTim racesimcentral.net

    Joined:
    Sep 23, 2010
    Messages:
    10,840
    Likes Received:
    314
    At this point, I am not sure how valid benchmarking using rF2 is for comparative purposes. If you are planning to use that to make a purchase? Don't. All you need to do is go to the tomshardware.com comparison pages, and look at the graphics card (desktop) charts with performance, and if you want to, specify the results for DX9 (although we haven't said we will always be DX9 either). And as much as I would like to not say this, being an ATI user myself at the moment, things are looking better for nVidia with us, so perhaps look for something well up the charts from that manufacturer.
     
  13. Bart S

    Bart S Member

    Joined:
    Oct 5, 2010
    Messages:
    851
    Likes Received:
    104
    Tim your such a nice chap, this is a real heads up from the inside to make a favourable decision.
    Personally from my experience on ATI and nVidia, nVidia is far better on comparative cards for rfactor 2.
    As a GTX680 owner and triple screen user and trying many configurations I've just about found a satisfactory nVidia and rfactor myself and it gives me confidence my 680 will be sufficient for the future of rf2.
     
  14. jtbo

    jtbo Registered

    Joined:
    Jan 12, 2012
    Messages:
    1,668
    Likes Received:
    48
    We can always start keeping track between builds how performance improves ;)

    Only result I got from testing was what I knew, my machine is barely meeting minimum usable in rF2.

    When I set every bell and whistle to their maximum glory, I get between 6-10fps mostly, with one place before tunnel peaking to 15-17fps and despite setting fxaa and anti aliasing on, they did go to off for some reason, maybe machine has some self preservation routine or maybe I forgot to press apply, anyway it just shows how I run out of gfx memory, power and cpu power with this old system of mine, 512MB gfx memory just is not enough today, I don't except this system to be able to perform much better when putting settings so that it overtaxes available resources.

    My gaming settings are 2x AF, at Monaco track detail is set to LOW, car details high and texture details high, special effects are low and rest are off, no HDR either as in Monaco my machine struggles to keep 60fps with it.

    With those settings I get 70fps to 225fps with this replay. MAX 225.394, MIN 69.882, AVG 123.4405952 as I usually lock fps to 60, it is crucial that fps never drops below 60fps, which I can manage with those settings, but I really need to upgrade to be able to run something else than hotlapping, I think that it was 2007 when I build this machine, upgraded 2008(or 2009?) with new gfx card and even then 8800GTS 512MB version was end of life product and shops were selling remaining stores at silly price, I think that I did pay 130 euros from the card. So it is no wonder that my machine has difficulties, I'm more surprised that I don't necessarily need new hardware yet.

    But anyone asking, I would say my setup is bare minimum, to be able to race on big grids, or use high details stronger machine is needed. 19" 4:3 monitor of course would save a lot, with low resolution performance would be decent, but nobody probably uses such anymore.

    Anyway for those looking just fps data some facts:
    Video Card: Nvidia 8800GTS 512MB
    FPS (min, max, avg): 69.882, 225.394, 123.4405952
    Resolution: 1920 x 1080
    Anti-aliasing (type, level): OFF
    HDR: OFF
    Anisotropic Filtering: 2X
    V-sync: OFF
    Circuit Detail: LOW
    Player Detail: HIGH
    Opponent Detail: HIGH
    Texture Detail: HIGH
    Special Effects: LOW
    Shadows: OFF
    Shadows Blur: OFF
    Sun Occlusion: OFF
    Road Reflection: OFF
    Environment Reflection: OFF
    AI in field: 0
    Number of visible vehicle: NA
     
  15. coops

    coops Banned

    Joined:
    Feb 1, 2011
    Messages:
    1,680
    Likes Received:
    9
    i think it would be in every1s best interest to wait for gold and hopefully ISI will give us a benchmark
     
  16. martymoose

    martymoose Registered

    Joined:
    Aug 10, 2012
    Messages:
    461
    Likes Received:
    13
    Problem with a replay file to compare builds is the replays ive tried have never worked from the older builds. I rarely go for max fps anyway as I only want to ensure I never go below v-sync. I dont see any benefit from 200+ fps other then generating more heat causing my gtx 480's to up the fan rates and sounding like a helicopter just landed in my room and it also causes terrible screen tearing.

    My battle with gfx in RF2 has always been to get the smoothest lag free v-sync setting, getting this right looks much better then a 400fps tear fest.
     
  17. jtbo

    jtbo Registered

    Joined:
    Jan 12, 2012
    Messages:
    1,668
    Likes Received:
    48
    Have you checked your GPU usage with MSI Aferburner (works with any card) ?

    For me that is 100% GPU usage, so even my slow CPU could do some more with faster card, I believe.

    Anything below 60 (vsync if used, I use MSI Afterburner fps limiter instead of vsync) is making rF2 to feel completely broken and wrong, so for me it does not matter how it looks as long as it is smooth and not changing from that 60.

    At Monaco with my gaming settings I have under 10fps room for slowdowns, which is not all that much. I would need to reduce settings further to be able to get AI cars on track also.
    I'm not sure if there is much to be gained with game optimizations, I think that my hardware is just too slow and I need to look for upgrade, well I would look if it would not cost so insanely lot to get 2-3 times performance that I have now, anything less is not worth of upgrading.
     
  18. MarcG

    MarcG Registered

    Joined:
    Oct 4, 2010
    Messages:
    6,854
    Likes Received:
    2,234
    So the recent interaction with ATI has led you to this conclusion? or is you opionion based on before that?
    I've gathered that since beta began nVidia have been marginally better overall from the forum posts I've read, was hoping the recent converse between you and ATI would further improve performance for us ATI users though.

    Still I'm happy with my HD7850 as it runs other games just fine and RF2 is steadily getting better each update, so somethings alright!
     

Share This Page