Skip to main content

AMD Releases New Windows 10 x64 Ryzen Chipset Drivers

Those of you rocking a brand new AMD Ryzen build should head on over to the AMD website and download the v17.10 AMD Ryzen chipset drivers.

According to AMD, this release is primarily intended to make good on the company's promise to include the latest AMD Ryzen Balanced power plan in the chipset driver package. The new 17.10 driver is specifically designed for systems with an AMD Ryzen CPU running Windows 10 x64. This update automatically installs and activates the AMD Ryzen Balanced as a fourth power plan.

The company explained that the new AMD Ryzen Balanced power plan prevents your CPU cores from being "parked" in an idle state that could have a negative impact on gaming performance. The plan also allows your AMD Ryzen processor to more quickly raise clock speeds by reducing the timers and thresholds for P-state transitions.

For a primer on the issue at hand, our very own Paul Alcorn explained it this way:

The key issue revolves around latency--there is up to a 30ms delay between when the operating system orders a power transition and the processor acts upon it. The delay varies based upon power plan parameters. Switching to the high performance power profile hands the power management tasks back to the processor. With sole control of its own power states, a modern processor can transition between the various sleep states in 1ms, which reduces the performance problem. Unfortunately, it isn't as power efficient.

The company expressed its appreciation for the feedback provided by end users on installation and usage. Moving forward, AMD promised continued support and improvements in future driver updates.

  • JimmiG
    The drawback of disabling core parking if you're not overclocking, is that you effectively disable XFR, so that's something to keep in mind.

    Core parking is different from letting Windows manage the p-states, which is what introduces most of the delay.
    Reply
  • valeman2012
    It maybe wise to be messing with the Power Plans.

    They should prevented from the start instead of focusing on graph 69%
    Reply
  • pentiuman
    @ Jimmig - I'm a long time reader, long time PC user, but new account and apparently hit the neg. button on your comment. Sorry. Tried to undo too, w/out luck.

    - Can anyone comment on Jimmig's comment to this article re: core parking, and also p-states? At this very moment, I'm building a Ryzen 1600X sys for my son, so I will be downloading the newest chipset shortly. I likely won't OC this system, as my son lives out of state and wouldn't know what to do if the settings needed fixing. Also - Ryzen is so new, that even if I was going to OC, I first want to wait and verify the system is stable before revving it up. But if disabling core parking disables XFR, I'll have to get more research. TIA
    Reply
  • JamesSneed
    19614905 said:
    @ Jimmig - I'm a long time reader, long time PC user, but new account and apparently hit the neg. button on your comment. Sorry. Tried to undo too, w/out luck.

    - Can anyone comment on Jimmig's comment to this article re: core parking, and also p-states? At this very moment, I'm building a Ryzen 1600X sys for my son, so I will be downloading the newest chipset shortly. I likely won't OC this system, as my son lives out of state and wouldn't know what to do if the settings needed fixing. Also - Ryzen is so new, that even if I was going to OC, I first want to wait and verify the system is stable before revving it up. But if disabling core parking disables XFR, I'll have to get more research. TIA

    Some are saying since AGESA 1.0.0.4a firmware update that comes with the latest BIOS from most vendors XFR works with all power plans i.e. core parking doesn't effect XFR anymore. I cant confirm it though.
    Reply
  • photonboy
    Pentiuman,
    AMD's new power profile is mean to get the MOST out of your CPU right now which should really answer your question; WHY would AMD do that if it gave worse performance?

    Not sure WHY some think it disables Turbo... I have an Intel CPU and my Turbo is not affected doing something similar (just changing Power Options CPU minimum to closer to 100%).
    https://www.kotaku.com.au/2017/04/if-youve-got-a-ryzen-cpu-amd-has-a-power-plan-for-you/

    "At the end of the day: if you have a Ryzen desktop or you're buying a Ryzen CPU, use the High Performance plan."

    THAT may have been a previous power plan, so the new one (download link in article) is either the same or better.
    Reply
  • photonboy
    If in DOUBT just use Task Manager to monitor the CPU usage and see if testing under LOAD (Prime95, convert video etc) has the CPU frequency where you would expect (i.e. 4.0GHz).

    *Don't get too hung up on Ryzen, but DO check for UEFI/BIOS updates periodically. Note that applying those probably puts your settings (i.e. overclock etc) back to default but the updates should help fix memory, overclock and various chipset issues.
    Reply
  • navin001
    East or West AMD is the best and it will rock.
    Reply
  • josejones
    Does this also mean that my AMD 955 is finally optimized for Windows 10 too?
    Reply
  • AkuBerhala
    19614905 said:
    @ Jimmig - I'm a long time reader, long time PC user, but new account and apparently hit the neg. button on your comment. Sorry. Tried to undo too, w/out luck.

    - Can anyone comment on Jimmig's comment to this article re: core parking, and also p-states? At this very moment, I'm building a Ryzen 1600X sys for my son, so I will be downloading the newest chipset shortly. I likely won't OC this system, as my son lives out of state and wouldn't know what to do if the settings needed fixing. Also - Ryzen is so new, that even if I was going to OC, I first want to wait and verify the system is stable before revving it up. But if disabling core parking disables XFR, I'll have to get more research. TIA
    I've been using the R5 1500X since launch day with AGESA 1.0.0.4a and can confirm that XFR is working on all power profiles.
    Reply
  • Justin_H
    JIMMIG - disinformation on tomshardware about AMD, why am I not surprised?
    Reply