THE PROBLEM: Games in SteamVR work very well with my Oculus, with no discernible lag, up until some seemingly random time (maybe every 10-20 minutes) where I get an unplayable intense amount of stutter / judder / jitter. The screen moves laggily and jerkily, such that the sides of the HMD screen have black rectangular edges and the game scene looks distorted. The frame timing skyrockets to over 16ms for Application (other) for CPU, and Application (scene) for GPU. This happens in Superhot VR, for example. Things like Dreamdeck and First Contact I've never experienced this lag or stutter / judder though. Don't know if it's a difference between SteamVR and Oculus Home, because I don't have any other games in Oculus Home to test.
MY SPECS: Windows 10 (UP TO DATE AS OF 8.20.18), freshly installed I7 7700k 16 gb ram Gtx 1070 8GB with the latest Nvidia 398 drivers H170 Pro Gaming Motherboard 1 tb HDD Toshiba DT01ACA100 (7200 RPM) Corsair water cooling systems for cpu Running Oculus rift, touch, and two (or three sensors) Problem persists despite headset and sensors in about every variety of USB configs I could think of: HMD to 3.1 or 3.0 usb, and two sensors in 3.0 OR 2.0, or three sensors with two in 3.0 and the third in a 2.0
I've prioritized high performance on the system I've disabled Windows' ability to regulate USB power for energy saving I've disabled Superfetch (MS background operation) Might uninstall MS Telemetry (because I see it sometimes takes up Disk usage)
I've taken a screenshot of my Frame Timing from the moments this problem occurs to shortly after it stops due to me Alt-Tabbing out of Superhot VR. There are also screenshots of the detailed frame timings https://imgur.com/a/4fneM8l
The persistent unplayable stutters ends usually because: 1. I alt-tab out of the game, I alt-tab back in and it is usually gone 2. I start task manager or JUST go to control-alt-delete screen 3. I minimize the display window of the VR game shown on my monitor (but even if this display window were minimized to begin with, the lag spikes still occur) 4. It just ends on it's own after a long while? I think?
Other possibly helpful diagnosis notes are: 1. The headset, as detected in Oculus Home, gets a orange warning message of poor tracking quality (I see that if I whip the HMD off my head when this problem occurs, but it quickly disappears, so I don't know how truthful this is). No sensors are getting touched or moved, and most of the time they are tracking great (so I doubt lighting or reflectivity is an issue) 2. Task Manager CPU, ram, GPU, and Disk, are usually all low in utilization. Usually around 40% for CPU and GPU, lower for RAM, and very low for disk except when loading which spikes Disk to 100%, but this drops back down to very low by the time the jitter starts (like 10 minutes later) 3. I did see that in one instance, the 3D performance of the GPU in task manager was at 100%. 4. Im not running any extraordinary graphics settings — just playing with whatever is default. I've tried "Allow interleaved reprojection" on and off, and "Enable Always-on Reprojection" on and off, with maybe some improvements, but the problem still occurs.
I would love to fix this because it makes for repeated sad nausea inducing breaks in the immersive experience. Everything runs flawlessly like I said, and then after maybe 10 or 20 minutes, with no warning (and it’s not even caused by especially graphics intensive sequences) but seemingly randomly it will start to judder / stutter. I feel like my comp specs are up to the task of running a game like Superhot VR no problem...i see people with worse machines doing so online?
Unsure if this concerns my GPU, my CPU, too much / uneven USB power draw, driver issues, something with my sensors / HMD, or something else entirely....
I had issues like that before but not in VR, try doing a fresh install of your graphics drivers and also download a program named Driver Booster, maybe you are missing some of the Microsoft Visual C++ Redistributables. That worked for me in the past, i hope it helps you 🙂
Most obvious cause for such behavior is pile of dust in GPU fan or broken GPU fan. If cooling is not adequate, GPU downclocks and this causes reduction in performance and in the end - fries the GPU.
However as this is so obvious and OP seems like a person who knows this stuff, this probably is not the solution for the problem. Just had to state the obvious 😉
Also nowdays I've been using https://us.msi.com/page/afterburner or similar software, which allows user to create own fan profile for GPU cooling. Recently I've encountered several graphics cards, which have really interesting cooling profile by default. Fans seem to start spinning when GPU reaches certain temperature (usually around 50C), but up to that point they are standing still. In my custom profiles fans always spin at least @ 30% speed, which keeps the GPU 10-20C cooler when no real load is applied. This will extend the lifespan a lot... For further reading; https://www.overclockers.com/overclockings-impact-on-cpu-life/
Btw could you enlighten me about "MS Telemetry". Could it be the culprit for spinning external USB drives up every 60 minutes with no apparent reason (while usb selective suspend setting=disabled)?
Get Help
Did this answer your question? If it didn’t, use our search to find other topics or create your own and other members of the community will help out.