Video of iGPU comparison of 2016 vs 2015 models

Discussion in 'MacBook Pro' started by dark_mark, Nov 19, 2016.

  1. dark_mark Suspended

    Joined:
    Nov 5, 2016
    #1
    Some of you may be concerned about iGPU performance on 2016 Macbook Pro 15".

    iGPU inside 2016 model is Intel HD 530 which has performance of 441.6 GFLOPS
    iGPU inside 2015 model is Intel Iris Pro 5200 which has performance of 832 GFLOPS

    So I made a video comparison between 2015 vs 2016 models

    Btw, iGPU is used for 90% of the daily tasks you will be doing on a laptop. dGPU only kicks in when you are using GPU intensive apps such as Final Cut or Photoshop.

     
  2. BondFancy macrumors 6502

    Joined:
    Jun 26, 2016
    #2
    If this isn't a software issue, I may very well pickup the 13" instead. What are the Flops of the Intel Iris 550?
     
  3. dark_mark thread starter Suspended

    Joined:
    Nov 5, 2016
    #3
    Iris 550 has 844.8 GFLOPS and is twice as powerful as HD 530. I played with 13 TB and it was smooth as butter.
    --- Post Merged, Nov 19, 2016 ---
    Another thing I noticed is that scrolling through messages in Microsoft Outlook for mac stutters like there's no tomorrow on 2016 15"
     
  4. swein1992, Nov 19, 2016
    Last edited: Nov 19, 2016

    swein1992 macrumors member

    Joined:
    Jul 9, 2015
    #4
    Stop TROLLING!
    My 2012 rMBP with the miserable HD4000 iGPU doesn't stutter at all in any of the scenarios you mentioned.
    And if you don't like the new MBP just return it then. What is bashing it over and over again gonna get you? Do you expect Tim Cook to show up in your house and apologize and reinvent the whole MBP line? Return it while you can and stop trolling on the forum.
     
  5. dark_mark thread starter Suspended

    Joined:
    Nov 5, 2016
    #5
    Lol. How can you compare 2012 model to 2016? Did you know that 2016 model uses different rendering system, internally it renders close to 3340 pixels as compared to 2880. It also scales to 1680 by 1050.

    But keep comparing to your 2012 which scales at 1440 by 900. Bye
     
  6. Kudos6612 Suspended

    Joined:
    Aug 10, 2016
    #6
    Lol this is a very unscientific way of igu comparison, and I might have to go to the Apple Store to check this out I just don't believe it. Flops is not really a good measurement in predicting whether there is u.i. Lag> anyone ELSE with the 15" to compare it to because are you telling me that anytime you scroll through a folder there is lag?


    I mean not one review stated there is u.i. Lag and this is such an easy thing to notice.
     
  7. dark_mark thread starter Suspended

    Joined:
    Nov 5, 2016
    #7
    Not in any folder. It stutters inside "Applications" folder, on facebook and on UI animations such as Mission Control. It also stutters when scrolling through messages inside Microsoft Outlook for mac.
    --- Post Merged, Nov 19, 2016 ---
    If Flops is not a good measurement then why did Apple use FLOPS to compare Radeon 460 Performance to Radeon 370X ?
    Don't you think that the best marketing machine in the world would have chosen the best way to tout their new GPUs if Flops wasn't the best measurement?
     
  8. swein1992 macrumors member

    Joined:
    Jul 9, 2015
    #8
    OH WOW so the 16 model renders 1.36x pixels than my 2012 while the HD530 is like 2.5 times faster than my 2012. And any hard evidence on the 16 model rendering 3340? Where did you get that number?
     
  9. dark_mark thread starter Suspended

    Joined:
    Nov 5, 2016
    #9
    Listen, I am not here to defend myself. Whoever finds my video useful will decide for themselves.

    Keep defending Apple or your purchase. I don't care.
     
  10. Cariboue macrumors newbie

    Joined:
    Nov 13, 2016
    #10
    thumbs up dark_mark. I think that is very important to know!
     
  11. swein1992 macrumors member

    Joined:
    Jul 9, 2015
    #11
    LOL when I ask you for concrete evidence you start talking about 'defending Apple'. I am not defending Apple. I am just tired of people trolling on this forum. Anyways, people will begin getting their orders and let's see how other machines are performing.
     
  12. dark_mark thread starter Suspended

    Joined:
    Nov 5, 2016
    #12
  13. dk808 macrumors 6502a

    Joined:
    May 13, 2015
    #13
    Just got back from the apple store, zero lag on the 15 inch. The one i played around on was 2.6 ghz and 450 gpu. Another thing i noticed was that the space grey doesnt look as nice as i thought, i actually prefer silver now. I kind of want to change my order but im scared they'll delay my delivery even more
     
  14. torana355 macrumors 68020

    Joined:
    Dec 8, 2009
    Location:
    Sydney, Australia
    #14
    Were you testing at the default scaling setting? The UI Lag is going to be much more prevalent at the higher scaling settings like the OP was using.
     
  15. dk808 macrumors 6502a

    Joined:
    May 13, 2015
    #15
    op told me he was using the default resolution
     
  16. high3r macrumors member

    Joined:
    Jul 23, 2015
    Location:
    Hungary
    #16
    It was connected to power wasn't it? When connected it uses the dGPU, and the iGPU (HD 530) while on battery and the complaints are about the latter.
     
  17. torana355 macrumors 68020

    Joined:
    Dec 8, 2009
    Location:
    Sydney, Australia
    #17
    I have not tested the 15" model myself, can anyone confirm the default scaling setting on the 15" model?
     
  18. dark_mark thread starter Suspended

    Joined:
    Nov 5, 2016
    #18
    Default scaling on 2016 is 1680 by 1050. On 2015 it's 1440 by 900. I used 1680 by 1050 on both.
    --- Post Merged, Nov 19, 2016 ---
    Possible that automatic graphics switching is turned off on display models to provide best demo experience.
     
  19. dk808 macrumors 6502a

    Joined:
    May 13, 2015
    #19
    Is this true? I think it still uses the igpu while charging
     
  20. torana355 macrumors 68020

    Joined:
    Dec 8, 2009
    Location:
    Sydney, Australia
    #20
    Ok so 1680x1050 is the default setting on the 2016 15" machines, one person with a video as proof says it does lag and another person with no proof say it doesn't. Im going to have to check this out for myself when im at the Apple store next. I know some people think this is trolling but the UI stutter is a real issue for some people.
     
  21. dk808 macrumors 6502a

    Joined:
    May 13, 2015
    #21
    That could be true, i didnt bother to check if the setting was turned off or not
     
  22. Robstevo macrumors 6502

    Joined:
    Jun 7, 2014
    #22
    Wow that is a total lad fest. Hopefully it's a software issue but if it isn't, that's a disgrace for such an expensive machine. People would be laughing their ass off and slamming windows of the shoe was on the other foot.
     
  23. Kudos6612 Suspended

    Joined:
    Aug 10, 2016
    #23
    The hd530 can natively drive 3 4K displays at 60 vs last years one 4K at 60. It also has other improvements in eram 3D graphics etc. you can find this via internet I don't think it should have a problem driving its 2800 x 1800 internal display at whatever resolution you set it to.
     
  24. dark_mark thread starter Suspended

    Joined:
    Nov 5, 2016
    #24
    That's not true. Connecting external display to MacBook always causes the graphics to switch to dedicated Radeon.
     
  25. Kudos6612 Suspended

    Joined:
    Aug 10, 2016
    #25
    Yeah for past models it's true and could be true about the new 15" macbook pro 2016 i don't know. But the 13" doesn't have a dedicated graphics card and can drive that many pixels without one.


    http://www.digitaltrends.com/computing/three-generations-intel-hd-graphics-tested/
    this article talks about the desktop 530 and mobile 540 compared to past versions of intel hd graphics but doesn't compare to iris pro specifically.

    arstechnica has a review of the 2016 13" macbook pro without touch bar and shows that the hd 540 beats the iris pro of the 2015 15" on their gfxbench and cinebench tests you can look at that although it's not the 530. These tests are also not the be all end all like flops performance.

    you won't find too many direct comparisons as there are other factors at play.
     

Share This Page