rMBP - Xbench - User Interface Test

Discussion in 'MacBook Pro' started by ppone, Feb 27, 2013.

  1. ppone macrumors regular

    Joined:
    Sep 1, 2011
    #1
    2.3 Ghz, 8Gb, base rMBP

    I get a score of 371.23 on the User Interface test

    what is your score for the User Interface Test in Xbench, for the retina.


    http://imgur.com/cvVEN6D
     
  2. andy318 macrumors member

    Joined:
    Jun 18, 2009
    #2
    Ugh. I'm getting a much lower score. I wonder why. I even used gfxCardStatus to switch to discrete graphics and when i do that, the score is even lower (128.5)

    Results 158.97
    System Info
    Xbench Version 1.3
    System Version 10.8.4 (12E55)
    Physical RAM 8192 MB
    Model MacBookPro10,1
    Drive Type APPLE SSD SM256E
    User Interface Test 158.97
    Elements 158.97 729.58 refresh/sec
     
  3. dusk007 macrumors 68040

    dusk007

    Joined:
    Dec 5, 2009
    #3
    Try to switch graphics before you launch the application.
    Switch with a running application usually doesn't work too well. It isn't Optimus.
     
  4. Jesla macrumors 6502

    Jesla

    Joined:
    Jan 7, 2013
    Location:
    Tennessee USA
    #4
    338.92…..With pandora, webkit and Mail open and running….
     
  5. Yahooligan macrumors 6502a

    Yahooligan

    Joined:
    Aug 7, 2011
    Location:
    Illinois
    #5
    141.18 with nothing running, graphics switching disabled, and a scaled UI to "more space".

    144.65 with "best for retina"

    Compare that to my cMBP connected to an ATD which scored 426.98.
     
  6. Jesla macrumors 6502

    Jesla

    Joined:
    Jan 7, 2013
    Location:
    Tennessee USA
    #6
    362.91

    2.5 i5 8Gb. Ram 120 SSd on best for Retina resolution.
    14 tabs open in webkit and Pandora Muse running.

    [​IMG]
     
  7. Yahooligan macrumors 6502a

    Yahooligan

    Joined:
    Aug 7, 2011
    Location:
    Illinois
    #7
    That's not the UI test. You need to uncheck all the boxes except for "User Interface Test" and then run the test, post up the results of that.
     
  8. nontroppo macrumors 6502

    nontroppo

    Joined:
    Mar 11, 2009
    #9
    Um, wasn't the last version of XBench made in 2006 -- surely this is not going to accurately benchmark current generation OSes and machines?
     
  9. Yahooligan macrumors 6502a

    Yahooligan

    Joined:
    Aug 7, 2011
    Location:
    Illinois
    #10
    Doesn't seem to have any problem accurately showing what we already know, the UI performance on rMBPs is much slower than non-Retina models. :D Beyond that, I don't know what anyone else is expecting from this thread.
     
  10. nontroppo macrumors 6502

    nontroppo

    Joined:
    Mar 11, 2009
    #11
    We don't "know" that (i.e. it has been heavily debated and differs substantially between users), and a very outdated program using old APIs is certainly not going to prove anything either way...
     
  11. Yahooligan macrumors 6502a

    Yahooligan

    Joined:
    Aug 7, 2011
    Location:
    Illinois
    #12
    It's just logical that the rMBP would have slower UI performance when you consider that the same GPU is used in the non-Retina models and the rMBP GPU is having to drive a lot more pixels. Nobody should be surprised that the rMBP UI performance suffers as a result. More pixels is more pixels.

    What would be interesting is if the rMBP UI performance test differs greatly when also connected to an external display, but I suspect it will perform the same and I just might test that this evening.
     
  12. nontroppo macrumors 6502

    nontroppo

    Joined:
    Mar 11, 2009
    #13
    If the bandwidth is higher than than the data demand, your assertion is not logical IMO. 10 or 100 people can walk through a gate at the same time if it was designed for 500 people to walk through!

    The bandwidth of both the discrete and integrated chips are significantly higher than the highest required resolution (3840x2400). I calculated it and the values are in a thread on this forum somewhere...

    Several people have already tested using an external monitor, claiming that the rMBP is less "laggy" on the external, even when logically it is driving more pixels overall!

    "Lagginess" is not equivalent to UI performance, because the bottlenecks are not being limited by the UI, but by other parts of a program. For example, the problems Chrome had (and probably Safari) had to do with bugs in how images were decoded by the cache mechanism. This caused a perceptual "lag" but it was not rate limited by any graphical drawing or pixel throughputs, but by suboptimal background code that had nothing to do with copying pixels to the GPU buffer... ;)
     
  13. Yahooligan macrumors 6502a

    Yahooligan

    Joined:
    Aug 7, 2011
    Location:
    Illinois
    #14
    You're really grasping on this one, it's not about bandwidth here. It's about workload. Just because the same GPU is driving more pixels with one display than another doesn't mean it can do so with the same speed.

    It's not about the bandwidth.

    Actually, logically, when there's no redrawing going on for a given display the GPU isn't "driving" anything. Hooking a 10,000lb trailer up to a truck but not moving it doesn't put any stress on the engine. The external display isn't a mirror of the first, meaning the GPU isn't logically seeing the external as physical extension of the first and as a result sending all graphics I/O to both displays.

    The GPU will perform the same with an external display whether the lid is open or closed, what will affect performance on the external is activity on the internal display because THEN the GPU is having to do processing for both displays.

    That's all beside the point, IMO. The UI performance test is doing the same test on two different displays. It's the same GPU and the same software.

    What you're basically saying is that a game that runs at 60fps @ 1440x900 should run at 60fps @ 2560x1440 because the GPU has the "bandwidth" to drive all those pixels at the same time.

    Obviously that doesn't happen due to the workload. The GPU simply can't perform the work at the same speed when it has to process more pixels.

    Similarly, it makes sense that for the SAME GPU the UI on rMBPs won't respond/perform just like the non-Retina models simply because of the additional workload that's a direct result of more pixels.
     
  14. nontroppo macrumors 6502

    nontroppo

    Joined:
    Mar 11, 2009
    #15
    Perhaps we have a semantic difference here (my bandwidth = your workload, sorry if I was over-broad in my use of the term bandwidth). All I wanted to say is that if a GPU has much more capacity than the bandwidth/workload requirements, a simplistic "more pixels has to be slower" is really not valid/logical. I program OpenGL based graphical stimuli for psychophysics research, and the retina can push several hundred frames/second (disable VSync) on a 1280x1024 CRT @ 85Hz while updating its main "best for retina" display without breaking a sweat. This bandwidth/workload exceeds the UI compositing requirement for moving one window over another or resizing a window etc. The fact that for e.g. UI lag as described when the rMBP first came out in e.g. Chrome was fixed with software updates, and slowdowns in e.g. mission control disappeared with OS updates for the same animations should suggest these problems are software, not hardware limited...

    The GPU on the rMBP is actually overclocked compared to the cMBP, or was when it was released AFAIK, so it is not really the "same" GPU.
     
  15. Yahooligan macrumors 6502a

    Yahooligan

    Joined:
    Aug 7, 2011
    Location:
    Illinois
    #16
    Unfortunately, reality disagrees with you and it's not just the Xbench test.

    http://www.anandtech.com/show/6023/the-nextgen-macbook-pro-with-retina-display-review/8

    Using ML and Safari, both optimized for Retina, is still slower doing the "same" work as a non-Retina MBP.

    The article may not be super new, but it's still valid in this discussion. More pixels is more work. The GPU is the same. It's simply not logical to expect the same performance with more pixels but that is exactly what you're thinking should happen.
     
  16. andeify macrumors 6502

    andeify

    Joined:
    Jun 10, 2012
    Location:
    UK
    #17
  17. TonyHoyle macrumors 6502a

    Joined:
    Sep 14, 2007
    Location:
    Manchester, UK
    #18
    My god.. 177.14. I reckon my old 2008 model would have scored higher..
     
  18. nontroppo macrumors 6502

    nontroppo

    Joined:
    Mar 11, 2009
    #19
    It was well known that neither ML or Safari were really optimised *at all* for retina, there were glaring bugs in cached image performance that took several iterations of ML and webkit development to fix. Your link actually makes my point perfectly; Safari performance was very poor at launch, same for Chrome, as bugs in the software stack that fed the display (not simply "more" pixels) caused the issues, not some "pixel" workload limit...
     

Share This Page