Skip to content

ThreadPilot Core Parking toggle vs Windows power settings #60

Closed

Description

@ende-es

Version

1.7.4

Windows Version

Windows 11

Reproduction Steps

I’m testing ThreadPilot’s Core Parking feature on Windows 11 and found what appears to be a mismatch between the ThreadPilot UI and the actual Windows power-plan settings.

System:

  • Windows 11 build 26200
  • Intel Core i7-10510U (4 cores / 8 threads)

On my Power Saver plan, I checked these settings with powercfg:

  • CPMINCORES: AC/DC 0/0
  • CPMAXCORES: AC/DC 100/100
  • CPINCREASEPOL: AC/DC 0/0
  • CPINCREASETIME: AC/DC 7/7
  • CPOVERUTIL: AC/DC 60/60
  • CPPERF: AC/DC 0/0

I then toggled ThreadPilot’s Core Parking switch and immediately checked all six settings again.

None of the six Windows settings changed, but ThreadPilot’s displayed Core Parking status changed.

In an earlier test, I also observed ThreadPilot changing CPMINCORES on AC while DC remained at 0, so the behavior appears inconsistent between tests.

Could you clarify:

  1. Which Windows setting/API does ThreadPilot use to determine the Core Parking status?
  2. What exactly does the Core Parking toggle modify?
  3. Why can ThreadPilot show different states when the Windows CPMINCORES/CPMAXCORES values are identical?

I can provide the exact before/after powercfg output if needed.

Expected Behavior

I’m testing ThreadPilot’s Core Parking feature on Windows 11 and found what appears to be a mismatch between the ThreadPilot UI and the actual Windows power-plan settings.

System:

  • Windows 11 build 26200
  • Intel Core i7-10510U (4 cores / 8 threads)

On my Power Saver plan, I checked these settings with powercfg:

  • CPMINCORES: AC/DC 0/0
  • CPMAXCORES: AC/DC 100/100
  • CPINCREASEPOL: AC/DC 0/0
  • CPINCREASETIME: AC/DC 7/7
  • CPOVERUTIL: AC/DC 60/60
  • CPPERF: AC/DC 0/0

I then toggled ThreadPilot’s Core Parking switch and immediately checked all six settings again.

None of the six Windows settings changed, but ThreadPilot’s displayed Core Parking status changed.

In an earlier test, I also observed ThreadPilot changing CPMINCORES on AC while DC remained at 0, so the behavior appears inconsistent between tests.

Could you clarify:

  1. Which Windows setting/API does ThreadPilot use to determine the Core Parking status?
  2. What exactly does the Core Parking toggle modify?
  3. Why can ThreadPilot show different states when the Windows CPMINCORES/CPMAXCORES values are identical?

I can provide the exact before/after powercfg output if needed.

Actual Behavior

I’m testing ThreadPilot’s Core Parking feature on Windows 11 and found what appears to be a mismatch between the ThreadPilot UI and the actual Windows power-plan settings.

System:

  • Windows 11 build 26200
  • Intel Core i7-10510U (4 cores / 8 threads)

On my Power Saver plan, I checked these settings with powercfg:

  • CPMINCORES: AC/DC 0/0
  • CPMAXCORES: AC/DC 100/100
  • CPINCREASEPOL: AC/DC 0/0
  • CPINCREASETIME: AC/DC 7/7
  • CPOVERUTIL: AC/DC 60/60
  • CPPERF: AC/DC 0/0

I then toggled ThreadPilot’s Core Parking switch and immediately checked all six settings again.

None of the six Windows settings changed, but ThreadPilot’s displayed Core Parking status changed.

In an earlier test, I also observed ThreadPilot changing CPMINCORES on AC while DC remained at 0, so the behavior appears inconsistent between tests.

Could you clarify:

  1. Which Windows setting/API does ThreadPilot use to determine the Core Parking status?
  2. What exactly does the Core Parking toggle modify?
  3. Why can ThreadPilot show different states when the Windows CPMINCORES/CPMAXCORES values are identical?

I can provide the exact before/after powercfg output if needed.

Logs and Diagnostics

I’m testing ThreadPilot’s Core Parking feature on Windows 11 and found what appears to be a mismatch between the ThreadPilot UI and the actual Windows power-plan settings.

System:

  • Windows 11 build 26200
  • Intel Core i7-10510U (4 cores / 8 threads)

On my Power Saver plan, I checked these settings with powercfg:

  • CPMINCORES: AC/DC 0/0
  • CPMAXCORES: AC/DC 100/100
  • CPINCREASEPOL: AC/DC 0/0
  • CPINCREASETIME: AC/DC 7/7
  • CPOVERUTIL: AC/DC 60/60
  • CPPERF: AC/DC 0/0

I then toggled ThreadPilot’s Core Parking switch and immediately checked all six settings again.

None of the six Windows settings changed, but ThreadPilot’s displayed Core Parking status changed.

In an earlier test, I also observed ThreadPilot changing CPMINCORES on AC while DC remained at 0, so the behavior appears inconsistent between tests.

Could you clarify:

  1. Which Windows setting/API does ThreadPilot use to determine the Core Parking status?
  2. What exactly does the Core Parking toggle modify?
  3. Why can ThreadPilot show different states when the Windows CPMINCORES/CPMAXCORES values are identical?

I can provide the exact before/after powercfg output if needed.

Activity

  1. PrimeBuild-pc commented on Sep 21, 2026

    @PrimeBuild-pc
    Owner

    Hi, thanks for the detailed report. You have identified both a confusing UI convention and behavior that we need to investigate.

    In ThreadPilot 1.7.4, the Core Parking control gets the active power scheme through PowerGetActiveScheme and reads the AC value of CPMINCORES through PowerReadACValueIndex.

    The toggle currently changes only CPMINCORES for AC power on the active scheme:

    • ON writes 100, which disables core parking by keeping all cores unparked.
    • OFF writes 10.

    It does not modify the DC value, CPMAXCORES, CPINCREASEPOL, CPINCREASETIME, CPOVERUTIL, or CPPERF. This means the current Core Parking label is ambiguous and should be improved.

    An AC change with DC remaining unchanged is expected. However, if CPMINCORES for the same active scheme remained unchanged while ThreadPilot's displayed state changed, that is not expected.

    Could you please post the exact before and after output of:

    powercfg /getactivescheme
    powercfg /query SCHEME_CURRENT SUB_PROCESSOR CPMINCORES
    

    Please also mention whether the laptop was plugged in. We will keep this issue open while we investigate.

  2. PrimeBuild-pc commented on Sep 22, 2026

    @PrimeBuild-pc
    Owner

    Thanks, Navin. Your email follow-up confirms that ThreadPilot changed the active plan's AC CPMINCORES value from 0 to 100 while DC remained at 0, as intended.

    We found a UI synchronization gap: after a system tweak write fails or is not confirmed, the one-way-bound switch could keep the visual state produced by the click even though the model still held the previous state. PR #61 refreshes the displayed state after every attempt. It also renames the control to "Prevent Core Parking" and explains that it affects the active power plan while plugged in.

    A non-release test build is available here:
    https://github.com/PrimeBuild-pc/ThreadPilot/releases/tag/issue-60-test.1

    For the read-only confirmation you offered:

    1. Keep the laptop plugged in and select Balanced.
    2. Run powercfg /getactivescheme.
    3. Run powercfg /query SCHEME_CURRENT SUB_PROCESSOR CPMINCORES.
    4. Open the test build and compare the Prevent Core Parking switch with the current AC power setting index.

    Expected result: the switch is on only when the AC index is 100 (0x64), and off for lower values. You do not need to toggle the switch or change any setting.

    This is an unsigned test build, so Windows may show a SmartScreen warning. Its SHA-256 is D76C6E14E3699C2DAD04253A89E82616F464ED51162A32F8399F78CEFE859295.

    We will keep the issue and PR open until you confirm the result.

  3. ende-es commented on Sep 23, 2026

    @ende-es
    Author

    Hi Lorenzo,

    I tested the issue #60 test build using the requested read-only procedure.

    • Laptop plugged into AC power
    • Balanced power plan selected
    • Balanced AC CPMINCORES: 0x00000000 (0%)
    • ThreadPilot “Prevent Core Parking” switch: OFF
    • I also checked High Performance, where the switch shows ON, consistent with the AC CPMINCORES setting of 100% (0x64).

    The switch correctly reflects the current AC CPMINCORES value in both cases.

    I also noticed that the switch is now labeled “Prevent Core Parking” rather than simply “Core Parking.” I think this makes the function much clearer, since ON means preventing Windows from parking cores.

    I did not toggle the switch or change any power settings during the test.

    Thanks for preparing the test build.

  4. PrimeBuild-pc commented on Sep 23, 2026

    @PrimeBuild-pc
    Owner

    Thanks again, Navin. Your read-only confirmation matched the acceptance criteria and helped us validate the fix on real hardware. The change from PR #61 has now been merged and released in ThreadPilot v1.7.5: https://github.com/PrimeBuild-pc/ThreadPilot/releases/tag/v1.7.5

    The Prevent Core Parking switch now resynchronizes with the active power plan's AC CPMINCORES value after every write attempt, and the clearer label is included in the release.

  5. added a commit that references this issue on Sep 29, 2026
    2a2b7ca
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    bugSomething isn't working

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions