R9-290 or GTX-770 4GB ??

Discussion in 'General Discussion' started by ViSo, Nov 6, 2013.

  1. Adrian

    Adrian Registered

    Joined:
    Dec 3, 2012
    Messages:
    383
    Likes Received:
    7
    I gave it a go here's mine;

    2013-11-16 14:02:17 - rFactor2
    Frames: 9151 - Time: 200103ms - Avg: 45.731 - Min: 28 - Max: 87
    --3820@4.4GHz crossfire 7970 @955 @1375--

    2013-11-16 14:08:06 - rFactor2
    Frames: 9679 - Time: 199588ms - Avg: 48.495 - Min: 27 - Max: 91
    --3820@4.4GHz crossfire 7970 @1125 @1575--


    As low as it looks it seems to scale quite well at 5760x1080 which I run... But still it's pretty poor compared to most other games.
     
  2. DrR1pper

    DrR1pper Registered

    Joined:
    Apr 29, 2012
    Messages:
    3,294
    Likes Received:
    36
    Thanks for redoing yours and amazing overclock on your 780, you increased performance by 18-19% surpassing the Ti by 5-6%. Gobsmacking overclock!

    All I managed was 7% increase with my o/c. :(
     
    Last edited by a moderator: Nov 16, 2013
  3. Timpie Claessens

    Timpie Claessens Registered

    Joined:
    Jul 26, 2011
    Messages:
    491
    Likes Received:
    1
    This with the latest drivers or the old 12.11's?
     
  4. Adrian

    Adrian Registered

    Joined:
    Dec 3, 2012
    Messages:
    383
    Likes Received:
    7
    latest 13.11 beta 9.2 drivers
     
  5. Shamrock

    Shamrock Registered

    Joined:
    Dec 6, 2012
    Messages:
    287
    Likes Received:
    36
    If you feel there is a problem, anyone can use this form to fill out.

    http://www.amdsurveys.com/se.ashx?s=5A1E27D20B2F3EBB

    But I have a question. Why Lock Drummond and the Clios? I get much better fps on other tracks, than LD, with other cars. Personally I don't like either :p

    Also here is a tip. Use Radeon Pro, it's free graphics. Don't use the included CCC.
     
    Last edited by a moderator: Nov 16, 2013
  6. paarma

    paarma Registered

    Joined:
    Mar 20, 2012
    Messages:
    24
    Likes Received:
    0
    I did it my way. :)
     
  7. DrR1pper

    DrR1pper Registered

    Joined:
    Apr 29, 2012
    Messages:
    3,294
    Likes Received:
    36
    That's precisely the reason why we would want to use a track like Loch Drummond, to test the cards at there limits by ensuring they are the bottleneck in the system when running rf2. This way we can be more sure the fps performance are a true and accurate reflection of your cards performance in rf2 for comparison.
     
  8. Nuno Lourenço

    Nuno Lourenço Registered

    Joined:
    Oct 19, 2010
    Messages:
    593
    Likes Received:
    65
    The graph has been actualized with minimu values also.

    [​IMG]

    crossfire actually scales "well" but the minimum fps are very poor compared to single GPU..
    :)
     
    Last edited by a moderator: Nov 16, 2013
  9. Marek Lesniak

    Marek Lesniak Car Team Staff Member

    Joined:
    Oct 6, 2010
    Messages:
    1,585
    Likes Received:
    101
    Wonderful Nuno! Thanks!
    Assuming SLI gives ~70% boost, two 780's after OC should give ~120 minimum and over 150 average fps. More than enough for one screen racing and probably enough for triple screen.
     
  10. MoodyB

    MoodyB Registered

    Joined:
    Apr 23, 2013
    Messages:
    72
    Likes Received:
    0
    I've been fairly lucky with my 780, although it did boost to 1106mhz out of the box with the factory overclock. Most 3rd party 780's seem to overclock to 1150-1250mhz boost in-game.

    There's a few of the more expensive 780's like the MSI Lightning & EVGA Classified that will do 1300+mhz.

    And if John clocked his 780Ti it would blow right past my 780... giggling as it went :cool:


    7% is still an increase. Plus it was free too :)
     
  11. Golanv

    Golanv Registered

    Joined:
    Nov 7, 2012
    Messages:
    1,041
    Likes Received:
    9
  12. Nuno Lourenço

    Nuno Lourenço Registered

    Joined:
    Oct 19, 2010
    Messages:
    593
    Likes Received:
    65
    Assuming the following test that I've done,


    Resolution: 1920x1080
    Drivers: 331.65

    CPU: Intel Q6600 @ 3.6GHz
    GPU: MSI GTX660 Ti PE (Defaults)
    RAM: 8GB @ 800MHz

    Frames, Time (ms), Min, Max, Avg
    10285, 199978, 39, 66, 51.431

    -----------------------------------------

    Resolution: 5040x1050
    Drivers: 331.65

    CPU: Intel Q6600 @ 3.6GHz
    GPU: MSI GTX660 Ti PE (Defaults)
    RAM: 8GB @ 800MHz

    Frames, Time (ms), Min, Max, Avg
    6772, 200134, 24, 42, 33.837

    ------------------------------------------

    Resolution: 5040x1050 Reflections OFF (i can live with that)
    Drivers: 331.65

    CPU: Intel Q6600 @ 3.6GHz
    GPU: MSI GTX660 Ti PE (Defaults)
    RAM: 8GB @ 800MHz

    Frames, Time (ms), Min, Max, Avg
    8728, 199385, 34, 54, 43.775

    --------------------------------------------------

    With 3 displays I have 65.7% performance compared with only one monitor
    If i disable reflections i have 85%.

    Doing a fake calculation but practical with 780TI losing the same performance we still get 56fps average but if we overclock it we can easily extract more 20% performance and we can achive 67fps average.
    Doing the same calculation with reflections off we achieve 73fps and 87 with overclock so for me one card is more that enough to play in 3 monitors

    Off course that this calculation ins't exact but we still get an idea from it.
     
  13. Nuno Lourenço

    Nuno Lourenço Registered

    Joined:
    Oct 19, 2010
    Messages:
    593
    Likes Received:
    65
    By the way Lesiu i think that you could change topic name to rF2 benchmarks or something like that :p
     
  14. Marek Lesniak

    Marek Lesniak Car Team Staff Member

    Joined:
    Oct 6, 2010
    Messages:
    1,585
    Likes Received:
    101
    I am not an admin here. Ask Tim :)


    Could you do those tests but on 12.11 drivers?
     
    Last edited by a moderator: Nov 16, 2013
  15. seancleric

    seancleric Registered

    Joined:
    Dec 23, 2011
    Messages:
    54
    Likes Received:
    0
    This is a great thread really helpful for deciding on new cards etc are you guys planning on all doing similar benchs for triples id be very interested to see results to, also Nuno you tabled results make it very easy to see thanks :cool:.

    I did the bench with mines,

    CPU: I7 920 @ 3.8ghz
    GPU: 7970oc 1000mhz
    RAM: 12GB @ 1600MHz

    my average was 27.8 so you guys results are accurate even though slightly depressing for ati/amd lol

    I can achieve quite good fps however with lower settings etc. My on grid fps at 5900x1080 ish my fps is higher than my average for this 1080p test.
    I mostly run high with most others off and still 2/4aa and 16af, if you guys doing at test where people can at higher res for triples, will adjusting the base settings for the test to be helpful to see performances across the cards at the 5760x1080 resolutions ?
     
  16. paarma

    paarma Registered

    Joined:
    Mar 20, 2012
    Messages:
    24
    Likes Received:
    0
    Overclocking my CPU dosent give me any better perfomance so its all about the video card.

    CPU: i5-2500k
    GPU:Gigabyte GTX 670 with 331.65 drivers

    Resolution: 1920x1080
    2013-11-16 18:41:37 - rFactor2 // CPU 3.3GHz, Nvidia default
    Frames: 11380 - Time: 200125ms - Avg: 56.864 - Min: 45 - Max: 77

    2013-11-16 18:47:40 - rFactor2 // CPU 3.3GHz, Overclocked GPU Mem +100 Core +300
    Frames: 12379 - Time: 200172ms - Avg: 61.842 - Min: 49 - Max: 84

    2013-11-16 18:55:56 - rFactor2 // CPU 4.0GHz, Nvidia default
    Frames: 11356 - Time: 200156ms - Avg: 56.736 - Min: 44 - Max: 77

    2013-11-16 19:01:53 - rFactor2 // CPU 4.0GHz, Overclocked GPU Mem +100 Core +300
    Frames: 12351 - Time: 200187ms - Avg: 61.697 - Min: 48 - Max: 84


    Resolution: 5760x1080
    2013-11-16 19:24:18 - rFactor2 // CPU 4.0GHz, Nvidia default
    Frames: 7332 - Time: 200094ms - Avg: 36.643 - Min: 27 - Max: 46

    2013-11-16 19:10:40 - rFactor2 // CPU 4.0GHz, Overclocked GPU Mem +100 Core +300
    Frames: 8050 - Time: 200063ms - Avg: 40.237 - Min: 31 - Max: 51
     
  17. Adrian

    Adrian Registered

    Joined:
    Dec 3, 2012
    Messages:
    383
    Likes Received:
    7
    which version I think it went up to beta 6, there was no WHQL version was there?
     
  18. mospider

    mospider Registered

    Joined:
    May 10, 2013
    Messages:
    16
    Likes Received:
    0
    The result of this threads is: Do not buy AMD graphics card if you want to play rFactor2. I would have been happy if I had known that before I bought my 7970.

    If you can not open an extra thread to point to it?
     
  19. seancleric

    seancleric Registered

    Joined:
    Dec 23, 2011
    Messages:
    54
    Likes Received:
    0
    Ye the difference is quite clear to see is such a shame considering how they perform in other games, we can hope Tim gets somewhere with the contact or next time I upgrade its going to have to be nvdia.
    Maybe we can petition amd to atleast have a look and see how many customers they will lose lol :)
     
  20. Marek Lesniak

    Marek Lesniak Car Team Staff Member

    Joined:
    Oct 6, 2010
    Messages:
    1,585
    Likes Received:
    101
    It was known and obvious in rFactor 1 and since the day 1 of rFactor 2 AMD owners check all the time if there are any improvements from build to build. Nothing new so you must have been asleep until today ;)
     

Share This Page