MSFS2020 CTDs with latest Nvidia Driver Generations (580 & 591) + PMDG 777

User Support Hub Crashes (CTDs)
,
1

Hello,

I’m consistently experiencing CTDs specifically with the PMDG 777 (200ER) on long-haul flights (3–5+ hours). This issue did not occur on older NVIDIA drivers (pre-560 generation), and only began after updating to the newer driver generations.

Current setup:

  • MSFS fully updated

  • PMDG 777 fully updated

  • Addons updated

  • DX12 enabled

  • NVIDIA driver: 591.95

  • Issue: CTD after several hours in-flight, often without warning

Important context:
This behavior is isolated to the PMDG 777. I do not see the same instability with other aircraft or in shorter sessions. The pattern strongly suggests either:

  • a long-session memory / VRAM handling issue, or

  • an interaction between newer NVIDIA driver memory management + PMDG’s WASM / systems depth

What’s confusing is that the older 560-era drivers were rock solid for me, but I’d prefer not to roll back permanently because the newer drivers bring meaningful DX12, DLSS, and performance improvements for other applications and sims.

I’m trying to understand:

  • Is this a known regression in newer NVIDIA drivers with MSFS + complex WASM aircraft?

  • Has anyone isolated a specific driver version that is both 777-stable and modern-feature complete?

  • Are there confirmed links between WASM memory limits, long-haul duration, and driver behavior?

At this point, the PMDG 777 is effectively unusable for long-haul on my system unless I downgrade drivers, which feels like a workaround rather than a fix.

Would appreciate any insight from others running long-haul 777 ops on recent drivers.

i9 13900k | RTX 4080 16GB VRAM | 64 GB RAM | 2TB M.2 SSD

Thanks

UPDATE: Updating BIOS fixed all of my CTD issues (yay). It was an i9 13900k microcontroller issue afterall.

1 Like
2

i9 13900k…You have the latest BIOS for your motherboard?

3

I use the 300 a lot and I don’t have any problems.

I’ll test it with the 200.

4

That’s an interesting question….i mean it was out of date so I just updated it. What makes you think this is seemingly a BIOS + CPU issue when a CTD is due to VRAM usage?

5

The 13900 and 14 series were bugged in microcode. A CTD can be related to more than just VRAM.

6

Driver NVIDIA 591.95?
The latest driver released is 591.74…=/

7

I had the same issue until I rolled back to November’s driver. I found a memory leak issue with the new driver. Performed great for several hours and then CTD. Monitored allocated memory and it continuously increased until over the maximum and eventually crashed to desktop. Rollback to November driver solved it.

8

I did extensive memory-leak tests with the new 737-900 beta (virgin sim and OS) and the latest driver this weekend (six 10-hour sessions) and could not detect any memory leaks. I do not know how you measure memory leaks, and I do not know how you can point out the graphics driver as the cause of your memory leak.

Before making such claims, I think it is good to outline the testing conditions.

Mathijs Kok
PMDG

9

Well, without changing anything other than rolling back my GPU driver and repeating the same flight, and not experiencing a memory leak…. I’m sticking with that. Didn’t claim it was a PMDG issue…