Comparing Benchmarks MBP 15 2017 vs Late 2014 is 30% SLOWER

Discussion in 'MacBook Pro' started by gadgetfreaky, Jun 17, 2017.

  1. gadgetfreaky macrumors 65816

    Oct 28, 2007
    So I talked about this in the other thread but I've done a few more tests and I get the same result each time. Both machines are high end Spec- the AMD 560 Card MBP vs the 750m Nvidia.

    What's tested--

    Well I call this the Civ VI test. Play Civ VI and set up a map on Huge. Then time each turns processing time.

    I sit both computers side by side. It's not that the results are the same-- or even close.

    The older MBP - 3 years mid/late 2014 is FASTER by about 30%.

    What are possible reasons for this? The AMD 560 should blow away the 3 year old Nvidia. They are both the same memory- 16 gig. Iris pro in the older machine (maybe the GPU is not being used)....

    Any ideas?
  2. fokmik macrumors 68040

    Oct 28, 2016
  3. UBS28 macrumors 6502a

    Oct 2, 2012
    Can you do more tests under OS X such as Starcraft 2 for example?

    Would be funny if the old 750M outperforms the 560M despite NVIDIA or Apple not putting alot of effort for the MAC drivers.
  4. HiRez macrumors 603


    Jan 6, 2004
    Western US
    Possible the Mac GPU drivers on the new machines are still immature and not fully optimized yet?

    Does running games in Windows (bootcamp) exhibit a similar difference?
  5. Poki macrumors 65816


    Mar 21, 2012
    Comparing benchmarks, the Radeon Pro 560 should be about three times faster than the GT 750M. Here are a few benchmarks (using the Radeon Pro 460, but the 560 should be just slightly faster):

    NVidia GeForce GT 750M:

    AMD Radeon Pro 460:

    If you compare XCOM 2, for example, on medium settings at 1080p the Radeon Pro blows away the GeForce at 31.9 fps vs. 10.4.

    The Civ benchmark seems a little unscientific, since not all worlds are created equal. Also, some games just heavily favour NVidia chips over AMD chips.
  6. UBS28 macrumors 6502a

    Oct 2, 2012
    I am not interested in Windows benchmark. I want to know how the performance is under OS X.
  7. Lennyvalentin macrumors 65816


    Apr 25, 2011
    It's not that simple. You're comparing different hardware architectures in a single benchmark. Due to the different ways games software can be written, it may perform (hugely) differently on another manufacturer's graphics architecture. It's easier to unintentionally make a game run fast on one graphics accelerator and quite badly on another. Making a game perform well on every system takes quite a bit of wizardry.

    It's unrealistic to expect one system to consistently outperform another (perhaps somewhat older) system in absolutely everything. It's more likely there will be statistical outliers than that there won't be. :)
  8. leman macrumors G3

    Oct 14, 2008
    Was it the same map? With same civs and everything else same? You are not comparing ardware here, you are measuring the AI performance. Use the Civ's internal benchmark tool.

Share This Page