Announcement

Collapse
No announcement yet.

Core i7 7700K vs. Ryzen 7 1800X With Ubuntu 17.04 + Linux 4.12

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • caligula
    replied
    Originally posted by mlau View Post

    Performance per core is more or less identical on Zen and corei7; corei7 just clocks much much higher when only one or two cores are taxed. And that's what you see with almost all games: intels clockspeed advantage of almost 1ghz. I've never seen my 1800x go beyond 3.6ghz in linux, even when only one core is taxed, so the 4.4/4.5 ghz on kaby make it look better.
    You can overclock your 1800x to 3.9 Ghz if you have a decent motherboard. 3.9 GHz should be doable with 1700/1700x/1800x, maybe even 4.0 Ghz.

    Leave a comment:


  • boxie
    replied
    Michael Can you please do another bench on the 1800X with the Performance governor rather than OnDemand - It might yield some rather interesting results - might not. Either way, it'll be good to know

    Leave a comment:


  • carewolf
    replied
    Originally posted by efikkan View Post
    AMD needs a better prefetcher.
    Well, and a proper 256bit-wide AVX unit. I am guessing that is also a pain point for games that are likely to try to use any vectorization they can get.

    Leave a comment:


  • wizard69
    replied
    Originally posted by macemoneta View Post

    Exactly. Even multi-threaded applications will have their threads limited to the performance of a single core. The I7-7700K and I7-4790K still rule in single thread performance. It doesn't fit well into 'more is better' marketing though.
    More often is better. You can easily see this where AMD wins, sometimes by a large margin. this doesn't even consider users with a lot of processes running at the same time.

    Leave a comment:


  • wizard69
    replied
    Originally posted by bug77 View Post
    Performance per core still matters as some of these benchmarks show.
    Only some of them. Considering how threaded software is becoming more common and my needs Ryzen still looks like the ideal solution.

    Leave a comment:


  • dungeon
    replied
    Just rebench Tomb Raider on nvidia again, AFAIR on nvidia hardware and driver Ryzen was much faster, while on amdgpu driver Kaby is much faster.

    And it seems both times we are famous 60% alcoholic at FullHD and on bigger res diff is even worse
    Last edited by dungeon; 18 May 2017, 08:38 PM.

    Leave a comment:


  • pal666
    replied
    Originally posted by Sidicas View Post
    I bought a 1090T with the expectation that all games would fully utilize 6 or more cores in the future and I was wrong. Now I am looking at getting a 4 core CPU to replace it
    lol, you want to be wrong two times in a row

    Leave a comment:


  • pal666
    replied
    pointless comparison. 1800x's competition is $1100 6950
    if you feel urge to compare 7700, compare it against 6950, not against two times cheaper ryzen

    Leave a comment:


  • lucasbekker
    replied
    Originally posted by monte84 View Post

    I notice the exact same thing. ondemand kills FPS. I asked about this earlier in the thread but have not received a response from Michael about how he tested.
    It's acpi-cpufreq ondemand for AMD according to the first page of the article.

    Leave a comment:


  • monte84
    replied
    Originally posted by aufkrawall View Post
    acpi-cpufreq ondemand butchered fps in some 3D applications like Unigine Valley compared to intel pstate powersave on my 2500k. So it's very likely that the Ryzen CPU runs very disadvantaged in games.
    If you want a fair test, please set every CPU to acpi-cpufreq performance instead.
    (Intel likely would still be quite faster in those tested games though, they are no multithread jewels.)
    I notice the exact same thing. ondemand kills FPS. I asked about this earlier in the thread but have not received a response from Michael about how he tested.

    Leave a comment:

Working...
X