You have to take the screenshot of temperature while playing, it's important to know if you're getting above 80°C And the PCI x4 is definitely a problem..
Thanks. So as I suspected, you have the GPU installed in the wrong slot, which gives you PCI-E x4 instead of x16. This will automatically reduce rF2 FPS by over 50%.
This is a overlay of my card to yours. My card is a GTX 1070 but it uses the same PCIe x 16 3.0. Note that my card is idle at this time so it reads PCIe x 16 3.0 @ PCIe x 16 1.1. When it's test loaded by the ? mark beside the interface readout it works at it's full performance of PCIe x 16 3.0 @ x16 3.0. I noticed your is reading PCIe x 16 3.0 @ x4 30. This should be I'd suggest @16 1.1 idle. So if I'm reading this correctly your connection to the CPU is only using 1third it's potential. Like what is written above you may have used a lower graphic slot on your motherboard giving you access to only PCIe x 4 3.0 not the PCIe x 16 3.0 your card is capable of. Top Slot A physical PCIe x16 slot can accommodate a x1, x4, x8, or x16 card, and can run a x16 card at x16, x8, x4, or x1. ... A single PCIe 1.0 (or 1.1) lane can carry up to 2.5 Gigatransfers per second (GT/s) in each direction simultaneously. For PCIe 2.0, that increases to 5GT/s, and a single PCIe 3.0 lane can carry 8GT/s. The physical slot on x4, x8, and x16 are all the same. If it says 2 PCIe x16 slots it means that there are 2 slots on the board that are capable of 16 lanes. If it says 2 PCIe X16 (and then in brackets says x16, x4 or x8, x8) it means when you have 1 slot being used it uses all 16 lanes for that 1 slot. Basically, The top slot on most motherboards are PCIe x 16 3.0. New ones anyway. Uses all 16 lanes memory transfer. The second slot down can also be a PCIe x 16 3.0 but more often it's a PCIe x 4 3.0 using only 4lanes maximum memory transfer. (Slower than a 16 lane slot).
OMG ...Problem solved. It was the PCIE slot. And i had the old gpu installed on the same slot for 5 months and i didn't know it was running on 4x instead of 16x. Thank you all for your help.
Yep, rFactor2 and R3E rendering engine are very dependable of PCIe speed. Not the case of other sims, like AC. Make sure using NVIDIA Profile Inspector, set "Power management mode" to "Prefer maximum performance" and "Multi-display/mixed-GPU acceleration" to "Single display performance mode" if you are not using a SLI configuration. I need to do this on my system to ensure x16.
Ok, will do that. I thought that is a rfactor2 problem because I did the Firestrike benchmark and the results I've got were what most people with the same gpu had.
Concerning the problems associated with performance and optimization in a general way, I hope that the improvement of the use of PCIe X16 2.0 is a priority.
There should be a check list guide were all these aspects of system configuration should be taken care, like: Memory size and paging settings CPU core parking and throttling PCIe speed Power and performance driver configuration OS active process, like antivirus, indexing services options In my case, it took me a while realise my issue was PCIe speed related. Unfortunately there is a lot of misinformation on the net that states PCIe speed is not a big factor at all. It results it is. Problem is not all software is being affected by it.
It doesn’t as part of the game . It’s a user issue as are many issues . That’s why many don’t have these problems
It's not really misinformation, as in most games x8 vs x16 is still making no difference at all. This is why, unfortunately, when this issue is brought up (here, in the rFactor 2 forum, where it is pertinent) people often dismiss it out of hand. It's a bit like dx11 vs dx9 performance ("this other game had a x% improvement when switching to dx11, so should rF2!"), and dx11 vs dx9 image quality before it. Something about what rF2 does means that PCIe bandwidth affects it probably before it really should. I wouldn't really label it a 'bug', it's more a characteristic. The question for me is that although it can be largely bypassed by ensuring the system as a whole can do 3.0 x16 (unless you're running SLI and each card is forced to 3.0 x8), if this underlying nature of the engine remains will it again have an impact as higher resolutions become more common? (i.e. will 3.0 x16 not be enough either?) But obviously it's not an easy fix, because it's been 3 years since the PCIe dependency was identified in this forum and nothing appears to have changed.
IMO, it doesn't matters what pci-e speed you have. 4x 8x or 16x but if you reach the bus limit. You can monitoring this with msi afterburner. If you have pci2 2x (extreme) but you are not maxing out the bus line (100% in bus usage in afterburner) the performance ingame should be the same.
No, this is wrong. I have benchmarked and come to conclusion that already at 30% bus usage the bus starts becoming a bottleneck. 100% bus usage is an idealistic number that could never be achieved with a real world game app, a bit similar as you cannot actually transfer 56 Mbit/s of data over a 56 Mbit/s WiFi connection due to overheads etc.
