Recent content by Vitaly Vidmirov

  1. V

    Apple is an existential threat to the PC

    These values are shown in system info and they are wrong. System info in GB is typically broken. Many consider the whole GB as totally broken benchmark, but I'm optimistic. The point is that CPUs works at exactly the same measured frequency. My Windows (on ARM) laptop has touch screen, but I...
  2. V

    Apple is an existential threat to the PC

    No, it is not true. https://browser.geekbench.com/v5/cpu/2748062.gb5 "processor_frequency": { "frequencies": [ 2493, 2493, 2493, 2493, 2493, 2493, 2493, iPad: https://browser.geekbench.com/v5/cpu/2757729.gb5 "processor_frequency": { "frequencies": [...
  3. V

    Apple A12, A12X, and A12Z SoCs

    In theory, but we have frequency range/steps defined by Apple. BTW do you have a plan to dig into core microarchitecture? I'm trying to code some uarch tests as well. You did a good job with review, a lot of Intel fanboys were seriously butthurt :D
  4. V

    Apple A12, A12X, and A12Z SoCs

    Andrei, how did you run SPEC on little cores? Seems iOS11 didn't paid attention to thread affinity, at least last time I tried. https://developer.apple.com/library/archive/releasenotes/Performance/RN-AffinityAPI/#//apple_ref/doc/uid/TP40006635-CH1-DontLinkElementID_2 BTW, your frequency...
  5. V

    Samsung Exynos 9810

    Boring. This is why Anandtech is interesting in a sea of poor tech journalism. Skip this page. Don't read it. The people who are interested in tech will enjoy reading it.
  6. V

    Apple A11 SoC

    Seems like this is bullshit benchmark. For example they show 297 GFlops for A11. With NEON, each A11 monsoon core can theoretically achieve 57 GFlops. It think they just multiplied single-core measured result by 6. The only adequate peak FLOPS benchmarking utility in App Store is vfpbench...
  7. V

    Apple A11 SoC

    I finally uploaded iOS Spectre attack proof-of-concept to github, after all the hype has died down. Actually it was done a couple of weeks ago, but I was too lazy to prepare and submit it :-/ https://github.com/vvid/ios-spectre-poc There is also disabled code to check for meltdown, but it...
  8. V

    Apple A11 SoC

    Finally I've bought iPhone8, so benchmark iteration time will be shorter. I'm still struggling with stable autodetect. Threads are jumping between cores like crazy. Repo is updated with WIP commit, just in case. ~6 seconds measurement time should be enough even with high thread count (otherwise...
  9. V

    Apple A11 SoC

    I'm pretty certain 2064MHz is a frequency of low-power cores. In a period of 30 secs we see that single thread was relocated to fastest core (2063->2376), while the rest of threads are freely migrating between cores 2063 <-> 2304. The instantaneous frequency is broken - we have 2 cores @...
  10. V

    Apple A11 SoC

    I did some changes in the app. It uses threads now. I tried to set affinity, but it is not guaranteed in xnu. Result is shown after the Stop button is pressed. My results with 15 second processing time: model 1-core / 2-cores A7(Air)..1390/1390 (1400? expected - the only figure is from...
  11. V

    Apple A11 SoC

    Thanks. But iPhone7 can't run 4 threads simultaneously. Two DispatchQueue jobs were just executed sequentially. I'll fix it tomorrow to use threads. Tried to keep things simple and it worked pretty well for my devices. Probably it is too small workload to launch big cores. It could be...
  12. V

    Apple A11 SoC

    I hacked up a little app to get actual CPU frequency for all cores https://github.com/vvid/ios-cpu-tool I hope someone with iPhone8 could compile and run it. Just set 1..6 to "Threads" field and tap "Get CPU frequency" several times. In theory there should be 2 different frequencies of...
  13. V

    Samsung Exynos 8890

    Seems there are no significant (if at all) differences from A72. http://www.realworldtech.com/forum/?threadid=155276&curpostid=155276
  14. V

    Would GTA V be possible on Wii U?

    Of course and can go straight from PPU cache, but it is considered as bad practice for large blocks. according to Cell Broadband Engine Programming Handbook 19.5 Performance Guidelines for MFC Commands DMA transfers from main storage have high bandwidth with moderate latency, whereas...
  15. V

    Intel's smartphone platforms

    Isn't they stated "emerging markets" explicitly? I don't see how Lexington could be competitive with ARM unless Intel ask a very low price or "sponsor" manufacturers in some ways. PS: Sorry for offtopic :cool:
Back
Top