Read from the PC's own files. Controls are the devices the sims hold settings for, plugged in or not.
The headline is graphics: Windows has logged 4,401 driver fault records in the last 90 days, 4,083 of them graphics timeouts, and the most recent was about two hours before this snapshot was taken. Everything else on this PC is secondary to that.
The crash history lines up with it — the PC restarted from an unclean shutdown two days before capture, X-Plane 12's last session ended in a crash, seven X-Plane 12 crash reports have been uploaded to Laminar, and both MSFS installs are holding crash reports and NVIDIA GPU dump files. Your RTX 2060 looks physically healthy at idle (39 °C, no throttling), so the graphics driver is the first suspect and the card the second, but these files can't separate the two on their own.
On the setup side there are a few things quietly working against you: your C: drive is down to 30 GB free of 931 GB with all four sims on it, the master sound switch is off in both X-Plane 11 and X-Plane 12, X-Plane 11's rudder is bound only to a device the sim couldn't see, and X-Plane 11 is still overriding its nav data with a cycle that expired in 2020.
Worth knowing: this snapshot was taken while you were connected to the PC remotely, so the single generic 32 Hz "monitor" and the sim windows saved "on no active panel" are artefacts of that session rather than a missing screen.
The thousands of graphics-driver timeout faults Windows has recorded, still occurring on the day of capture — this is the single biggest thing on the machine.
X-Plane 12's last session ended in a crash, with seven crash reports already uploaded to Laminar, which may well share a cause with those graphics faults.
In X-Plane 11 the rudder axis is tied only to a device the sim could not see, so it is worth checking what your rudder is actually bound to before the next XP11 flight.
13 items
Windows recorded 4,401 live dumps in the last 90 days — a live dump is Windows quietly saving a record of a driver fault without going to a blue screen. 4,083 of them are graphics faults: the GPU engine hung, or the display driver stopped responding and had to be restarted. They start on 7 July and run right up to 2026-08-27 14:41, roughly two hours before this snapshot. Thirty-six of them landed in the ten minutes after the PC came back from its unclean restart on 25 August.
In the cockpit this is what a momentary freeze, a black flash, a stutter that lasts a second, or an outright crash to desktop looks like. At this volume it is not background noise — it is the most likely single explanation for the sim crashes further down this report.
What these files cannot tell us is whether the fault is in the driver software or in the card itself. The card reads healthy at idle: 39 °C, no thermal or power throttling ever recorded, PCIe link at full gen-3 x16, no memory errors. The driver is version 591.86, dated 20 January 2026, so about seven months old.
Given the volume and how current it is, getting the graphics driver reinstalled cleanly — or moving to a different driver version — is a reasonable next step, following NVIDIA's own guidance rather than any procedure from here. Be aware that NVIDIA's "clean installation" option wipes every custom Control Panel 3D setting, and this PC has four that differ from default (including a Threaded optimization value on the X-Plane profile), so they would need setting again afterwards. MSI Afterburner is also installed; these files cannot show whether a GPU overclock or undervolt profile is being applied at startup, but if you use one, flying a few sessions without it is a cheap way to rule it out. If the faults continue on a freshly installed driver with no tuning profile, the card itself becomes the prime suspect.
X-Plane 12's log for the session that started at 09:52 on the morning of capture carries the sim's own crash footer — that run did not shut down cleanly. The sim's crash folder is empty because seven dumps have already been auto-uploaded to Laminar, with the upload index last touched on 19 August. X-Plane 11 has five crash reports on disk from the last 90 days, MSFS 2024 is holding five older crash reports plus five NVIDIA GPU dump files, and MSFS 2020 has one.
What is notable is what X-Plane 12 did *not* log: zero errors, zero warnings, no plugin load failures. A sim that dies with a clean log is usually being taken down from underneath, which fits the graphics-driver faults above rather than any add-on misbehaving. The NVIDIA GPU dump files sitting in both MSFS folders point the same way. That is a consistent pattern, not proof — nothing here timestamps a specific crash against a specific driver fault.
One other thing shows up in the X-Plane 12 log: six pieces of software inject themselves into the sim's rendering — OBS, the Steam overlay and its shader cache layer, the Epic (EOS) overlay, and two NVIDIA layers. Overlays hooking the renderer are a well-known source of stutters and crashes in Vulkan, and X-Plane 12 runs on Vulkan.
If the crashes continue after the graphics driver has been dealt with, shutting those overlay apps down before launching — one at a time so you can tell which matters — is the cheapest test available, and it changes nothing permanently.
In X-Plane 11, pitch, roll and throttle are all on a device that is plugged in right now — your Thrustmaster stick. Yaw (rudder) is not. It sits on a device slot the sim last saw on 17 August and which was not present at capture. That slot is the virtual joystick (vJoy) that carries your one-G PC-12 panel: the same slot holds 87 button assignments for the PC-12's battery, avionics, autopilot and de-ice switches.
So the practical picture is: when the one-G rig and its bridge software are running, you have rudder. When you launch X-Plane 11 with just the Thrustmaster stick, that stick has no yaw axis assigned in XP11 at all, and you will be flying with no rudder input beyond the keyboard. X-Plane 12 is not affected — there the same Thrustmaster device has yaw on axis 3.
The uncertainty here is only about intent, not about the reading: the binding genuinely points at a device the sim did not have. If you always fly XP11 with the one-G panel powered up, this is normal and nothing needs doing. If you sometimes fly it with the stick alone, assigning yaw on the Thrustmaster in XP11's joystick screen is worth doing before the next flight.
MSFS 2024 is holding 25 saved control profiles, including eleven for the T.A320 Pilot stick alone and seven for the Logitech Flight Yoke, with names like "Test", "Blank test", "My new general profile_1/_2/_3". Three of them contain no bindings at all: the keyboard profile "Props", the T-Rudder profile "My new general profile", and the vJoy "Tobii" profile.
These records carry no marker for which preset is actually selected, and deleted profiles can linger in the data, so nothing here says a blank one is in use — and you have not reported a dead device. The risk with this many near-identical names is simply picking the wrong one and then wondering why the stick behaves differently from last time.
Worth a glance at the Controls screen to confirm which preset each device is currently on. Do not delete anything unless it still shows up in the sim and you know you do not want it.
X-Plane 12 has four keys carrying two commands each: Ctrl+F is bound to both "fail system" and "rotate view down slowly", Ctrl+comma to both a dataref test and "backward slow", and Ctrl+minus / Ctrl+equals each to both a radio standby flip and a zoom command.
The practical effect is that one of the two fires and the other does not, so a shortcut occasionally does the wrong thing — and Ctrl+F triggering a system failure when you meant to move the view is the one that could actually spoil a flight.
This is minor and easy to leave alone; if any of those four shortcuts have ever behaved oddly for you, the keyboard page in X-Plane 12's settings is where to untangle them. X-Plane 11's keyboard map has no duplicates.
Your only working drive, C:, has 30 GB free out of 931 GB. On it sit MSFS 2020 (Store), MSFS 2024 (Store), X-Plane 11 and X-Plane 12, plus Windows itself and a full Visual Studio installation.
That is enough to fly today, but it is not enough room for a single MSFS update, which routinely wants tens of gigabytes of temporary space on top of what it finally installs. A drive this close to full also tends to show up as longer scenery loads and the occasional stutter, because Windows has less room to work with when it swaps memory to disk.
There is nothing here that says the sims are currently failing for want of space. Clearing sim leftovers would barely dent it — X-Plane 11's replay folder is 164 MB and X-Plane 12's shader cache 137 MB, and the shader cache simply gets rebuilt (making the first flight after slower). Getting back to comfortable means moving or removing something substantial, or adding a drive; worth planning before the next big sim update lands rather than during it.
Both X-Plane installs have the master sound switch set to off in their saved preferences. This is the single switch in Settings > Sound that mutes everything the sim produces — engines, warnings, ATC, the lot.
If you have been flying X-Plane in silence, or wondering why SayIntentions ATC audio never arrives in the sim, this is the first thing to look at. It is one toggle in each sim's own sound page.
One caveat worth stating: it is possible this was set deliberately, and it is also possible it reads that way because you sometimes use this PC remotely (the only audio endpoint Windows reported at capture was the remote-session one). Either way, checking the master slider on X-Plane's sound page costs nothing.
MSFS 2024's settings define three sim windows totalling 8.2 million pixels a frame, and MSFS 2020 defines two totalling 7.2 MP. Pixels per frame is the single biggest lever on frame rate, and on a 6 GB RTX 2060 an 8 MP frame is a heavy ask — that is the same pixel count as a 4K screen, drawn across several windows.
Two caveats matter before you act on that. First, this counts every window the settings file *defines*, not the ones actually open on your last flight — MSFS keeps a definition for every extra window you have ever set up. Second, two of the MSFS 2024 windows are saved at positions on no currently active panel, and this snapshot was taken over a remote connection, so that may simply reflect your normal monitors being absent from the remote session rather than a screen you no longer own.
Your settings already show you compensating: MSFS 2024 is on the Low preset with a 30 fps target, dynamic settings on, DLSS anti-aliasing and FSR frame generation enabled. MSFS 2020 is on High with vsync on. That is a sensible read of the hardware.
If MSFS 2024 feels heavy with several windows open, the specific lever for you is SecondaryScaling, currently 1.000000 — that is the render resolution of the extra windows, and lowering it keeps outboard panels usable without paying full price for them. It has no effect on the main view.
X-Plane 11's Custom Data folder contains an AIRAC cycle 2009 — valid 13 Aug to 10 Sep 2020, so about six years out of date — and anything in Custom Data overrides the data the sim ships with. Alongside it are 13,769 CIFP procedure files from that same old cycle. The sim itself grumbles about it in its log, warning that the provider data has an outdated format and asking for data built for X-Plane 11.50 or later.
What you will notice is approaches, departures and frequencies that no longer match your charts, or procedures that simply are not there. X-Plane 12 is clean on this front — it has no stale override.
You already have current US terminal data in place: the FAA CIFP cycle 2608, effective 6 August 2026, installed into earth_424.dat by the Hub's Navdata Updater. The old files just sit alongside it. Three ways to resolve the rest, and which is right depends on where you fly: refresh the cycle through Navigraph FMS Data Manager, which is already installed on this PC and covers the whole world including enroute data; keep using the Hub's Navdata Updater, remembering it covers US terminal procedures only — approaches, SIDs, STARs and airport fixes in FAA airspace, with everything enroute and everything outside the US staying on X-Plane's own database; or remove the old Custom Data files entirely and fall back to the cycle X-Plane 11 shipped with (cycle 1802 — older still, but it never expires and never warns).
In the MSFS 2020 Community folder, the package scenery-ma-flyingscool-kbedv0.2.1_VMSQ2 contains another folder of the same name inside it. MSFS only looks one level down, so it sees a folder with nothing recognisable in it and skips the add-on entirely.
The symptom is simply that the KBED scenery is not there when you fly to it, with no error message anywhere to tell you why. This is the most common reason a downloaded add-on "doesn't work".
The fix is to move the inner folder up one level so it sits directly in Community, replacing the outer wrapper — a file-manager job rather than anything in the sim, and worth doing with the sim closed. Everything else in both Community folders is valid; the SayIntentions entries that look unusual are shortcuts created by the SayIntentions installer and are working as intended.
In X-Plane 11's scenery order, KSEA Demo Area sits at position 12, below Global Airports at position 11. X-Plane reads that list top-down and the first pack to claim an airport wins, so the default Global Airports version of Seattle hides the custom layout underneath it.
You would only notice this if you fly to KSEA and wonder why the demo area's layout is not showing. Everything else in the list is above Global Airports and is loading fine.
Two entries in the same list — "FSC Training" and "Aerosoft - EGLL Heathrow" — are empty folders with no scenery inside. They contribute nothing and are not causing any harm; the Heathrow one is worth noting only in case you thought that add-on was installed. X-Plane 12's list is ordered correctly (the only entry below Global Airports there is a Salzburg library, which is exactly where a library belongs).
With no simulator running, the PC was already using 23.2 GB of its 31.9 GB of RAM, with total commitment (memory plus pagefile) at 36.4 GB. The load came from Visual Studio, several Chrome windows, two Claude processes and assorted background apps — CPU was only 8%, so nothing is thrashing, but the memory is spoken for.
MSFS 2024 in particular will happily use 12–16 GB on its own. Starting it on top of a working development session leaves Windows leaning on the pagefile (the space on the drive Windows borrows when RAM runs out), which on a drive with 30 GB free is not where you want to be. In the cockpit that shows up as long scenery loads and stutters when you pan or change view.
Nothing here is broken — the pagefile is system-managed and has never been stressed (peak use 517 MB). It is simply worth closing the development tools and browsers before a long flight, which is free and reversible.
At the moment of capture the PC was being used remotely rather than sat in front of. Windows listed a "Microsoft Remote Display Adapter" alongside your RTX 2060, reported a single generic 3840x2160 panel running at 32 Hz, and offered only a remote audio device.
That matters for how you read the rest of this report: the 32 Hz refresh rate is a remote-session artefact and not a monitor problem, and the sim windows described as saved "on no panel active right now" most likely refer to your real monitors simply not being part of that session. Your actual screens — two LG QHD panels on the RTX 2060 — do show as attached.
Nothing needs doing about this. It is flagged so you do not go chasing a display fault that does not exist, and so you know that any future check of frame rate or refresh behaviour should be done sitting at the machine.
Need help? Our sim techs can remotely connect to your computer to assist with optimizing and troubleshooting your system. Book a session.
This is the data read from your PC and used for the AI's analysis. Captured August 27, 2026 at 4:34 PM.
| Machine | DESKTOP-1KNRQOD |
|---|---|
| Windows | Microsoft Windows 11 Pro build 26200 |
| CPU | Intel(R) Core(TM) i9-9900KF CPU @ 3.60GHz — 8 cores / 16 threads |
| Memory | 32 GB |
| Graphics | Microsoft Remote Display Adapter driver 10.0.26100.8972 dated 06/21/2006 02:00:00 · driver 10.0.26100.8972 (06/21/2006) |
| Pagefile | 23,552 MB allocated · peak used 517 MB |
| GPU scheduling (HAGS) | not configured |
| Monitor | Resolution | Refresh | Desktop position | Connection | HDR | Size |
|---|---|---|---|---|---|---|
| (no EDID name reported - generic/DVI monitor)primary | 3840 × 2160 | 32 Hz | 0, 0 | connector | supported, off | — |
| Device | Class | Connected | Windows may power it off | VID:PID |
|---|---|---|---|---|
| TP-Link Bluetooth 5.4 USB Adapter | Bluetooth | yes | yes | 2357:0604 |
| Bluetooth Low Energy GATT compliant HID device | HIDClass | yes | yes | 0204:B386 |
| Bluetooth Low Energy GATT compliant HID device | HIDClass | yes | yes | 0232:0001 |
| Logitech Virtual Hid Device | HIDClass | yes | yes | — |
| Logitech Virtual Hid Device | HIDClass | yes | yes | — |
| vJoy Device | HIDClass | yes | yes | — |
| Wacom Virtual Hid Driver | HIDClass | yes | yes | — |
| USB Root Hub (USB 3.0) | USB | yes | no | — |
| USB Root Hub (USB 3.0) | USB | yes | no | — |
| USB Input Device | HIDClass | yes | yes | 046D:C534 |
| USB Input Device | HIDClass | yes | yes | 046D:C534 |
| USB Input Device | HIDClass | yes | yes | 046D:C534 |
| USB Input Device | HIDClass | yes | yes | 044F:0405 |
| USB Input Device | HIDClass | yes | yes | 046D:C534 |
| Generic SuperSpeed USB Hub | USB | yes | no | 0BDA:0411 |
| Generic SuperSpeed USB Hub | USB | yes | no | 0BDA:0411 |
| Install | Microsoft Store |
|---|---|
| Last launched | 2026-08-17 07:06 (settings last saved) |
| Quality preset | Low · VRMedium in VR |
| Anti-aliasing / upscaling | DLSS · DLSS AUTO · FSR PERFORMANCE · frame generation FSRFG |
| V-sync | off |
| Community folder | 7 packages (7 valid) · not loaded by sim, empty |
| Add-on tools installed | Navigraph Charts 7.1.35 7.1.35 · Navigraph FMS Data Manager 1.8.4.0804 1.8.4.0804 · Navigraph Simlink 1.1.22.0602 1.1.22.0602 |
| Window | Mode | Resolution | Render scale | Pixels | On monitor |
|---|---|---|---|---|---|
| Main window | windowed | 1140 × 821 | 100% | 0.9 MP | saved at -1520, 183 — no live monitor there |
| Extra window 0 | borderless fullscreen | 2560 × 1417 | 100% | 3.6 MP | saved at -2560, 23 — no live monitor there |
| Extra window 1 | windowed | 2560 × 1417 | 100% | 3.6 MP | (no EDID name reported - generic/DVI monitor) (3840×2160) |
8.2 MP per frame across 3 windows — the whole Windows desktop is 8.3 MP, so this is 99% of it.
Upscaling (DLSS / FSR) lowers what the GPU truly renders before scaling up to these sizes, so this total is the output figure.| Device | Profiles | Newest saved |
|---|---|---|
| Keyboard | 5 | 2026-06-26 |
| Logitech Flight Yoke | 7 | 2026-03-07 |
| T.A320 Pilot | 11 | 2026-08-13 |
| T-Rudder | 1 | 2025-08-22 |
| vJoy Device | 1 | 2025-08-06 |
| Install | Microsoft Store |
|---|---|
| Last launched | 2026-07-20 13:09 (settings last saved) |
| Quality preset | High · VRDefault in VR |
| Anti-aliasing / upscaling | TAA · FSR BALANCED |
| V-sync | on (1) |
| Community folder | 8 packages (7 valid) |
| Add-on tools installed | Navigraph Charts 7.1.35 7.1.35 · Navigraph FMS Data Manager 1.8.4.0804 1.8.4.0804 · Navigraph Simlink 1.1.22.0602 1.1.22.0602 |
| Window | Mode | Resolution | Render scale | Pixels | On monitor |
|---|---|---|---|---|---|
| Main window | windowed | 2560 × 1440 | 100% | 3.7 MP | (no EDID name reported - generic/DVI monitor) (3840×2160) |
| Extra window 0 | windowed | 2560 × 1369 | 100% | 3.5 MP | saved at -2560, 23 — no live monitor there |
7.2 MP per frame across 2 windows — the whole Windows desktop is 8.3 MP, so this is 87% of it.
Upscaling (DLSS / FSR) lowers what the GPU truly renders before scaling up to these sizes, so this total is the output figure.| Device | Profiles | Newest saved |
|---|---|---|
| Keyboard | 3 | 2025-10-20 |
| Logitech Attack 3 | 1 | 2023-01-13 |
| Logitech Flight Yoke | 2 | 2024-11-22 |
| Logitech Force 3D Pro | 2 | 2023-07-06 |
| T.A320 Pilot | 1 | 2025-10-20 |
| T.Flight Hotas X | 1 | 2023-09-30 |
| T-Rudder | 2 | 2024-11-22 |
| vJoy Device | 2 | 2024-03-31 |
| Install | C:\X-Plane 12 |
|---|---|
| Version | X-Plane 12.3.3-r1-eaab172a (build 123300 Intel 64-bit, Vulkan a4add999f62f4fecdc0c8b529856075d013c7693) |
| Last launched | 2026-08-27 09:52 (Log.txt timestamp) |
| V-sync | off |
| Master sound | OFF |
| Lateral field of view | 66.7° (X-Plane default is 60°) |
| VR | VR enabled: no (OpenXR: no) |
| GPU as the sim saw it | NVIDIA GeForce RTX 2060 (54e3000000) (driver 591.86.0, Vulkan 1.4.325) |
| How the last run ended | !! CRASHED - the log carries X-Plane's own crash footer. |
| Crash dumps uploaded to Laminar | 7 |
| Nav data | AIRAC cycle 2608 (custom) |
| Plugins | 7 enabled · 13 parked in disabled\ |
| Custom scenery | 27 entries in scenery_packs.ini |
| Window | Mode | Resolution | Render scale | Pixels | On monitor |
|---|---|---|---|---|---|
| View 0 | windowed | 1547 × 905 | — | 1.4 MP | (no EDID name reported - generic/DVI monitor) (3840×2160) |
1.4 MP per frame across 1 window — the whole Windows desktop is 8.3 MP, so this is 17% of it.
Upscaling (DLSS / FSR) lowers what the GPU truly renders before scaling up to these sizes, so this total is the output figure.| Axis | State |
|---|---|
| pitch | assigned on a connected device - OK |
| roll | assigned on a connected device - OK |
| yaw (rudder) | assigned on a connected device - OK |
| throttle | assigned on a connected device - OK |
| Install | C:\X-Plane 11 |
|---|---|
| Version | X-Plane 11.55r2 (build 115501 64-bit c22f0eb21d049f66fcba233c37e65c845605fc40) |
| Last launched | 2026-08-17 07:02 (Log.txt timestamp) |
| V-sync | off |
| Master sound | OFF |
| Lateral field of view | 106.4° (X-Plane default is 60°) |
| VR | VR enabled: no |
| GPU as the sim saw it | NVIDIA GeForce RTX 2060/PCIe/SSE2 (driver: 4.6.0 NVIDIA 591.86 (460/0)) |
| How the last run ended | clean exit ("X-Plane has shut down" footer present) |
| Nav data | AIRAC cycle 2608 (custom) |
| Plugins | 8 enabled · 7 parked in disabled\ |
| Custom scenery | 12 entries in scenery_packs.ini |
| Window | Mode | Resolution | Render scale | Pixels | On monitor |
|---|---|---|---|---|---|
| View 0 | windowed | 1280 × 720 | — | 0.9 MP | (no EDID name reported - generic/DVI monitor) (3840×2160) |
0.9 MP per frame across 1 window — the whole Windows desktop is 8.3 MP, so this is 11% of it.
Upscaling (DLSS / FSR) lowers what the GPU truly renders before scaling up to these sizes, so this total is the output figure.| Axis | State |
|---|---|
| pitch | assigned on a connected device - OK |
| roll | assigned on a connected device - OK |
| yaw (rudder) | assigned ONLY on a device that is NOT connected right now |
| throttle | assigned on a connected device - OK |
| Context | no simulator was running — these numbers are an idle baseline |
|---|---|
| CPU load | 8% |
| Memory in use | 23 GB of 32 GB (73%) |
| Commit charge | 36 GB of 55 GB limit (66%) |
| Graphics | 2% used · VRAM 1,985 of 6,144 MiB · 39°C · 15.15 W |
| Busiest programs | NGenuity2Helper (6.8% CPU, 80 MB) · VBCSCompiler (6.7% CPU, 399 MB) · dotnet (2.1% CPU, 139 MB) · powershell (0.5% CPU, 279 MB) |