I had this problem in the past and it has cropped back up. I have tried all variations of V-Sync, lowed resolution, lowered all settings to low, removed visible cars, removed AI cars and still it is an issue. When I pop open the historiogram it shows that my physics randomly max out, sometimes multiple times per lap. I have placed a video below to show what I mean. I have already capped frames to 60, and made pre-rendered frames=1. i have a hexacore AMD processor and an nvidia 750ti, with 8gb of ram, I also tried the fixes regarding the steering wheel refresh rate to no avail. I wish Skip Updates was still an option! Any ideas???
Try changing "Use thread" setting in controller.json. https://forum.studio-397.com/index....-1080-now-available.52208/page-16#post-853310
If you want less CPU usage, aside from what above poster mentioned, try to use DX11 instead. DX9 build uses more CPU resources for rendering.
It seems that switching to DX11 fixed my issues. But I lost my trackmap (not an insignificant loss) but I'd rather have a smooth game than trackmap
Hello, A friend had the same problem. I found the solution. Its antivirus was the cause. I made him cut and he has no problem. Its antivirus is avast. I had the same problem with Bitdefender. Hope to be able to help you.
Joekart, did he just disable it or uninstall altogether? I have Norton 360, never thougt that might be an issue.
Alas, the issue persists. Mostly during races after a lap or two. Has completely ruined my experience with rF2. May need to focus on AMS for a while until this is sorted. My Logitech DFGT works in every other sim and I have no issues running other high fidelity games on high settings. I get the CPU usage is a beast with the tire model but I cant figure out why I keep getting these random physics spikes. Sorry rF2 your getting pretty but I still cant use you.
Does it still persist with DX11 and the fix mentioned in post #2? Also you mention you have a hexacore AMD, but the number of cores doesn't matter much for rF2. What matters is single core performance, in which AMD was light years off Intel until their Ryzen series released this year. So an older AMD is almost guaranteed to be problematic in rF2 if you add a lot of AI and use custom heavy mods, and assuming you are running it at stock clock speed. IMO non-Ryzen AMD's need to be overclocked heavily to work well with most modern games, their single core performance is on par or worse than Intel CPU's from 2009.
Yes even with DX11 and usethread set to true or false it still occurs. Its strange this never used to happen to me. Unfortunately I have a non-ryzen processor. I have a FX-6300 overclocked to 3.8 gzh. Could probably go as far as 4.0 but with stock cooling not really a wonderful prospect.
Update: Saw someone mention adding procmask=5 to the command line in the launcher and it seems to have helped. Have tested different cars on different tracks both S397 content and mod content, and it has helped for now.
I had similar problems with my old FX-6300. After some weeks of trials I decided to switch to Intel. i5-6600k apparently solved all these issues, but what is a reasonable CPU usage in order to prevent ANY impact on the real time syncronization? I currently have between 50 and 60% purple bar in most track-mod combos, but is it enough? I'm asking because I have the dubt that an high (but not too high, ie you can't notice the stuttering) CPU usage could have an impact on lap times, especially online.
According to this post, procmask=5 will force your CPU to use cores (0, 2), that is, only two cores. I have a hard time understanding how it could help, but knock the wood as long as it works.
I know lots of people who play rF2 on FX-4xxx, FX-6xxx and FX-8xxx CPUs without the issue in the video so I'm not sure it's as simple as leaping to blame the CPU.
Did you make a specific profile for rfactor2 in the logitech gaming software. Better delete the profile and let rfactor2 only use the global settings. The logitech gaming software applies a damping algorithm if you make a specific profile for a game, but that damping algorithm only works properly if the game uses directinput, if the game uses xinput the algorithm uses a wrong axis offset which makes the damping feel subtractive instead of additive which makes the car feel upside down. (dampers feel compressed then when the car is in the air, and decompressed when the car is pushed into the tarmac) Should`ve watched the video first i guess, went to much with the title+description Anyway what i mentioned is true but not really applicable to your problem.
I have had this problems too with my FX6300. I´ve health it by put new heat paste under the cpu, i think it could be problems by overheating.