Announcement

Collapse
No announcement yet.

The Gaming Performance Impact From The Intel JCC Erratum Microcode Update

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

  • #11
    Originally posted by atomsymbol View Post
    Notice that all of the tests achieved a framerate higher than 60 Hz. With FreeSync, a game needs only about 50 Hz to appear smooth to most gamers. For the most part, anything higher than 60 Hz is an unnecessary waste of electrical energy and increases fan noise, unless the gamer is using a 100+ Hz monitor and gains an actual advantage from using the 100+ Hz refresh rate. (I do not own a 144 Hz display, so I what I wrote may be completely invalid.)
    Well then the recent Intel CPU's don't have even the gaming excuse to be bought, as that's the only area where they are having some small frame-rate advantage over AMD. Even that's mostly the case only with 9900K + 2080 Ti. Any slower GPU will be the bottleneck in several games even with slower CPUs. And on 2080 Ti, 3900X (even 3700X) is usually within a few percents of the 9900K, so there's no point paying more for the CPU with security flaws, producing more heat and no platform upgrade path.

    Because the human eyes don't see frames, there are things like ghosting and blurring even with higher than 60 FPS. See for example this: https://www.testufo.com/eyetracking - how different the background looks depending on which UFO you are focusing your eyes to. And blurbusters.com has much more interesting content explaining why things matter beyond 60 Hz (and even beyond 144 Hz).

    Also, any benchmarks on current game mean that a future CPU-bound game would suffer this in similar ratio. So it could be a difference between say, 57 and 60 Hz, which would matter especially without FreeSync. Also, this will likely effect CPUs like older i3, i5 etc., where the starting frame-rate would not be this high. It's also worth pointing out that these changes again further decrease the performance that has already taken hit by the numerous earlier patches to flaws in the Intel CPUs.

    Comment


    • #12
      Originally posted by DoMiNeLa10 View Post
      Having a JCC patched version of proprietary software only for affected chips would be a win for everyone. Binaries themselves aren't that big.
      An example: Shadow of the Tomb Raider executables+libraries have size of about 262 MiB (Linux) and 278 MiB (Windows).

      Comment


      • #13
        Originally posted by fintux View Post
        Most of the testing was done at 1440p or 4k and RTX 2080 Super. This means that a lot of the results can be actually GPU-bound. The results would be more informative and reliable if they were done at 1080p and using RTX 2080 Ti to minimize any GPU bottlenecks.
        Yes, the GPU is a bottleneck in a lot of these cases. But despite this, we have still seen performance drops. It's good to see 1440p and 4K results, so people playing games at those resolutions know they don't really have anything to worry about. But I agree seeing 1080p results would be nice, since some people prefer higher frame rates over resolution (I personally am not one of them).

        Comment


        • #14
          Originally posted by schmidtbag View Post
          Yes, the GPU is a bottleneck in a lot of these cases. But despite this, we have still seen performance drops. It's good to see 1440p and 4K results, so people playing games at those resolutions know they don't really have anything to worry about. But I agree seeing 1080p results would be nice, since some people prefer higher frame rates over resolution (I personally am not one of them).
          Just a note: An average/median Linux user would experience the effects of the mitigation somewhat differently than in the article. The article uses Core i9-7960X 4.2-4.4 GHz 16/32 cores/threads, which isn't representative of an average/median Intel Linux user. Most people have slower CPUs than i9-7960X: if for example their CPU is 50% slower than i9-7960X it means that the microcode update might have about 50% more severe impact - but only if some OS thread was already near 100% CPU usage and due to microcode update it crosses the threshold of 100% CPU usage while benchmarking the game. In other words, games in the article that for example do not to experience any slowdown from the microcode update might in certain other hardware setups experience a slowdown of a few percent. It would be of great help if the article was measuring the CPU time consumed by each of the benchmarks.

          Comment

          Working...
          X