Testing the Impact of Windows Game Mode and Hardware-Accelerated GPU Scheduling on Microsoft Flight Simulator 2020

This article delves into a user-conducted test exploring the effects of two Windows features, “Game Mode” and “Hardware-Accelerated GPU Scheduling” (HAGS), on the performance of Microsoft Flight Simulator 2020 (MSFS 2020). While definitive, professional benchmarks are complex to achieve, this personal experiment aims to provide insights for fellow sim enthusiasts.

The Testing Methodology

The test was conducted on a high-end PC featuring an AMD 5900x CPU, MSI Suprim X RTX 3090 GPU, 64GB RAM, and a 2TB M.2 SSD. Settings in MSFS 2020 were predominantly set to Ultra, with Level of Detail (LOD) reduced to approximately 150. The testing environment included a 4K 60Hz Dell monitor and the latest NVIDIA drivers available at the time of the test (465.89).

The procedure involved loading into London Heathrow (EGLL) at Runway 27 in a Cessna 172 G1000, with clear skies and specific time/date settings. Multiplayer traffic was disabled, while Live AI traffic was enabled. After initiating the flight and waiting for FPS to stabilize, a one-minute capture using Riva Statistics was performed while idling on the runway. Crucially, the system was fully restarted between each test variation of Game Mode on/off and HAGS on/off.

The test was further broken down into three background task scenarios:

  • BG: Standard background tasks running, including RGB software (Razer Synapse, Corsair iCue), Dell monitor management, and other utilities.
  • No-BG: All non-essential background applications were closed. The simulation was reloaded after closing these applications.
  • Sim-BG: Additional software crucial for peripherals like the motion platform, FIPs (flight information panels) via SPAD.next, and wind simulation via SRS were running alongside the sim.

Performance was reported as average FPS and 1% lows in the format “Avg/1% lows.”

Performance Results and Analysis

The collected data presented some surprising findings, with minimal performance differences observed across the various configurations.

  • Game Mode On + HAGS On: BG: 50/43, No-BG: 50/42, Sim-BG: 48/41
  • Game Mode Off + HAGS On: BG: 50/42, No-BG: 49/39, Sim-BG: 48/38
  • Game Mode On + HAGS Off: BG: 51/46, No-BG: 51/45, Sim-BG: 51/38
  • Game Mode Off + HAGS Off: BG: 52/49, No-BG: 51/48, Sim-BG: 50/40

The analysis revealed that the impact of these features was largely insignificant. The tester noted a tendency for performance to slightly drop when closing background tasks, which could indicate a flaw in the testing methodology.

Hardware-Accelerated GPU Scheduling (HAGS): In all tested scenarios, leaving HAGS off appeared to provide a marginal increase of 1-2 FPS. This led the tester to conclude that disabling HAGS would be the preferred setting.

Game Mode: The impact of Game Mode was less clear. While it might have offered a very small positive impact when HAGS was enabled, it seemed to have no effect on average FPS when HAGS was off, and potentially even a slight negative impact on 1% lows. The tester acknowledged the potential benefit of Game Mode in preventing Windows from performing background updates during gameplay, but the current data suggested otherwise.

Conclusion and Recommendations

For the majority of users, the impact of Windows Game Mode and Hardware-Accelerated GPU Scheduling on MSFS 2020 appears to be negligible. The tester’s personal decision is to leave HAGS disabled due to the slight performance benefit observed. The stance on Game Mode remains on the fence, with the potential for background update prevention being weighed against the data suggesting a minor negative impact.

It is suggested that longer testing periods, particularly during extensive flight sessions where new data is streamed, might reveal different results, especially if Windows happens to initiate an update concurrently. For now, the recommendation leans towards leaving Game Mode off.

Users are encouraged to share their own experiences and interpretations of these results, or any contrary findings from similar tests. While the tester is open to further suggestions, the preference is to spend time flying rather than conducting extensive performance analysis.

Dave.