Jump to content

SimonOCedition

Member
  • Posts

    3
  • Joined

  • Last visited

Awards

This user doesn't have any awards

SimonOCedition's Achievements

  1. In truth, it never pushes to hit a limit, it OCs to something that could be regarded as ok, good but not perfect, (it ussually misses the cap by at least 200mhz, meaning you can manually push it that much more, on a 1060, numbers are smaller on stronger cards.), and then measures the difference from power/temp limit. As long as this margin gets smaller, it adjusts the clocks downwards, so as not to hit a wall at some point, resulting in major fps drop. So you can think of them as 2 seperate systems, 1 is pre-programmed to kick in as soon as you start your application, and OC the card as much as it can, then a 2nd system kicks in, and tries to actively control clocks to reduce the chance of you hitting the temp limit to 0, under normal circumstances.
  2. Hello, I noticed from the video you posted that many of the cpu cores where reaching 94%, which technically might be 100% since some games cannot fully utilize a core to 100%, rather reach close to 95%. Also, if a core on a major function hits 100%, while not using aSync compute then it slows down all the other cores because they have to wait for it. so then the GPU is normal to stay at low percentages, it can't push more frames because it has to wait for the cpu, I think this is all normal and probably has to do with your ram speed. Anyone, feel free to correct me if i'm wrong with something but a friend of mine had a similar issue, we have almost identical systems, with the same GPU from the same Vendor, and i was getting 30 fps more in games. He then found out that it was incompatibility between his mobo and ram, making ram run in single channel, even tho CPU-Z showed dual. his firestrike score was 2k points behind, I was getting 12k and he was getting 10k. fixed by installing ram compatible with his board.
×