Advertisement
If you have a new account but are having problems posting or verifying your account, please email us on hello@boards.ie for help. Thanks :)
Hello all! Please ensure that you are posting a new thread or question in the appropriate forum. The Feedback forum is overwhelmed with questions that are having to be moved elsewhere. If you need help to verify your account contact hello@boards.ie

AnandTech Ryzen Benchmarking Audit

Options
  • 26-04-2018 9:48am
    #1
    Xenoronin
    Registered Users Posts: 1,307 ✭✭✭


    https://www.anandtech.com/show/12678/a-timely-discovery-examining-amd-2nd-gen-ryzen-results

    Super interesting stuff which has the comments abuzz with speculation. Still sounds a little weird that it has such a massive effect on Intel, even if it isn't a default option. It could be forced on by certain applications (AnandTech couldn't compile a list) which leads to a massive bottleneck. Ryzen 2 is barely affected in comparison, though they didn't have Meltdown to deal with.

    TL;DR: AnandTech forced a timer called HPET on during benchmarking (it is not on by default in the OS, even if enabled in the BIOS), which created a perfect storm for Intel Coffee Lake CPUs in a post Meltdown/Spectre world. Other publications don't seem to have done this, explaining the gaming benchmark discrepancy.


Comments

  • Options
    #2
    TerrorFirmer
    Closed Accounts Posts: 29,930 ✭✭✭✭TerrorFirmer


    Interesting.

    What I found particularly interesting too prior to this update was the amount of people claiming in the comments that Ryzen 2 was just clearly better than Intel for gaming and people just couldn't accept it by questioning the initial article!


  • Options
    #3
    cherryghost
    Moderators, Technology & Internet Moderators Posts: 17,133 Mod ✭✭✭✭cherryghost


    Nice little ****storm for anyone wanting to overclock so.


  • Options
    #4
    Cuddlesworth
    Registered Users Posts: 13,986 ✭✭✭✭Cuddlesworth


    Doesn't make sense. If its the recent patches and the IO calls doing it, why were their 8700K review results so on par.

    On a side note, my 5820k is getting hit hard. I run a lot of small active VM's and its gotten very slow in the last few months.


  • Options
    #5
    Xenoronin
    Registered Users Posts: 1,307 ✭✭✭Xenoronin


    More information added on the final page of the AnandTech article.
    Ian Cutress (Senior Editor and reviewer in question) tweeted this article too: https://www.overclockers.at/articles/the-hpet-bug-what-it-is-and-what-it-isnt

    Overclockers.at also released a windows timer benchmark, which could be useful in checking to see if you have this issue.
    Because the HPET bug can be difficult to spot, we have implemented a timer benchmark for windows that sheds some light on your timer configuration and its performance. It's called TimerBench and mainly focuses on QPC because it's the defacto standard in Windows. There is a synthetic test to show the maximum number of possible timer calls and a game test to analyze the impact of your configured timer in 3D applications. It uses Unreal Engine 4 and DirectX 11, a famous combination for games.


  • Options
    #6
    Lu Tze
    Registered Users Posts: 3,495 ✭✭✭Lu Tze


    Surprised there aren't articles (or maybe I haven't seen them) on the performance drop on Intel processors. Tom's hardware have a review up of the 2600x and it has pre and post patch results for Intel, but only have a limited number of productivity tests. Would be interesting to see somebody do a full suite of tests to see which workloads would be impacted the most, and by how much.


  • Advertisement
Advertisement