Latest Build Update rFactor 2 - 1.1110 Now Available!

See your first screen shot,it’s says auto for gpu,click it and select your 1060,I have lower performance if I leave it set to auto
I did the test with the 1060 selected, when I opened the rfconfig to take the screenshot it reversed himself to auto.
I will do another test tomorrow, but I remember that there was no difference for me.
Thank you for the tip.
 
I did the test with the 1060 selected, when I opened the rfconfig to take the screenshot it reversed himself to auto.
I will do another test tomorrow, but I remember that there was no difference for me.
Thank you for the tip.
At your session settings and specific at weather settings you have select "storm" weather for the 1st 2 weather slots
Storm has the most heavy fps wise rain effect
you can choose the light rain and gain a lot fps (that is my 1st tip to you)
ALSO do what patchedupdemon told you about your graphic card
with your computer you can do good fps
not sure if you upgrade your cpu to newer standards you can go better.
personally i experience perfect fps with my combination which is ryzen 5 1600 cpu and rx 580 4 giga gpu
8 giga ddr4 @2666 ram
 
For VR users especially Oculus Rift users, it is important that you not have Oculus home running in the background. All you need to do is fire up steam and then steam VR to run rfactor2 in VR. Having Oculus home running in the background seems to significantly decrease smoothness and frames per second. At least for me.
I have Oculus running as administrator and don't start it ,only Steam but if i check Task Manager there are a few Oculus things running when I start Steam VR ! But not Oculus Home even when I had the OH Beta 2.0 on .
 
one last thing to people with low fps....
do you guys running the game with Supersampling or multisampling ?
supersampling will kill your fps badly
also i notice supersampling isnt give better graphics at all
for RF2 i suggest multisampling or Adaptive multisampling.
 
My system is low enough on the "totem pole" to not have seen any noticeable difference either way. But, something I found interesting that definitely drove the point home about being CPU-intense: 25 AI with Endurance Series mod at Silverstone and I could still squeeze out 80-100FPS (depending on area of circuit) with graphics turned down quite a bit. However, it was about as far as I could reliably go without the physics going out of real-time. i5-750 w/GTX560 and 4Ghz DDR3 RAM. My PC isn't dead yet...but to run larger fields and/or be able to be visually better than rF1-based sims, I'll need a new setup for sure. Hah!
 
What does this mean? Ive heard it before but never understood it

It means physics at any given time are given certain time buckets to get their calculations finished within. If that fails, the entire thing collapses and everything goes into slow-mo. The reason it does this is so that physics are consistently same on every CPU, i.e. there is no laptime gain in buying a faster CPU. The only thing to be aware of is to not go over this realtime threshold limit. But I have a feeling 99% of issues here are related to graphics, not physics. Physics as far as I know haven't changed much over the past few years, so basically you can still run rF2 in realtime on an old AMD FX system unless you add hundreds of AI's.
 
@stonec Physics have definitely edged up a little over the years, main step was the new CPM and tyres with higher number of nodes on them that started a couple of years ago. Not a huge difference, but depending on content (and that includes some stock) I can now encounter out-of-realtime from the occasional CPU spike. But bearing in mind I'm currently on a quad core overclocked to 2.8GHz and DDR2 ram at 800MHz, I think it's fair to say my hardware is reaching the end of its useful life. For rF2.

Dx11 is the nail in the coffin for systems like mine, it drives the processing up just enough to cause much more frequent slow downs.

But I doubt anyone with even 20% better CPU performance would have noticed a change over that time, so it's pretty insignificant.
 
My slo mo-ments were due to my processor getting hot. New cooler and it's been fine since, and the new build/patch/whatever seems to have stabilised it a little more. Ran 35 cars at Silverstone on Tuesday no problem.

But I still find it odd that people with 'better' processors than me are experiencing a backwards step so to speak.

But then, I find AMD Nomenclature easier to understand than Intel ones.
 
@stonec Physics have definitely edged up a little over the years, main step was the new CPM and tyres with higher number of nodes on them that started a couple of years ago. Not a huge difference, but depending on content (and that includes some stock) I can now encounter out-of-realtime from the occasional CPU spike. But bearing in mind I'm currently on a quad core overclocked to 2.8GHz and DDR2 ram at 800MHz, I think it's fair to say my hardware is reaching the end of its useful life. For rF2.

Dx11 is the nail in the coffin for systems like mine, it drives the processing up just enough to cause much more frequent slow downs.

But I doubt anyone with even 20% better CPU performance would have noticed a change over that time, so it's pretty insignificant.

The increased node number only affects the QSA portion of the tyre, right? That way you can define the tyre construction more accurately. How the tyre interacts with the road in the game is done with the realtime sections nodes which are 7 nodes across times something around the circumference.
 
My system is low enough on the "totem pole" to not have seen any noticeable difference either way. But, something I found interesting that definitely drove the point home about being CPU-intense: 25 AI with Endurance Series mod at Silverstone and I could still squeeze out 80-100FPS (depending on area of circuit) with graphics turned down quite a bit. However, it was about as far as I could reliably go without the physics going out of real-time. i5-750 w/GTX560 and 4Ghz DDR3 RAM. My PC isn't dead yet...but to run larger fields and/or be able to be visually better than rF1-based sims, I'll need a new setup for sure. Hah!
I can run 40 cars with an i3 2100, the game used to run out of real time as soon as the race started but in DX9 I could run even more cars with 0 issues, I wanted to show Marcell about it and switched back to build 1098 (I think was this one) since it had 0 DX11 in it, made videos of running 40 cars then went back to latest build... problem solved lol no more issues running 40 cars haha
Maybe try that in your PC too :D but 4gb ram is bad, rF2 easly use 5gb here
 
@mesfigas Want to check my set up, but being a noob at the moment. Where is this setting - had a quick look in my Nvidia 3d settings and SteamVR and nothing obvious? [Please be gentle o_O ]
i have Ati card and these are the most common settings
must be somewhere where you can see the antialiasing filters or anisotroping filters
maybe someone with nvidia can tell you where in the Nvidia contrla panel you can find it
Remember when you find it check the "multisampling" not "supersampling"
 
For VR users especially Oculus Rift users, it is important that you not have Oculus home running in the background. All you need to do is fire up steam and then steam VR to run rfactor2 in VR. Having Oculus home running in the background seems to significantly decrease smoothness and frames per second. At least for me.
I can't confirm that. There is no difference in Performance for me. The only thing is that ASW stops working correctly if needet when Oculus home is not running
 
If somebody has a moment could you try out what is described in this post? https://forum.studio-397.com/index.php?threads/inconsistent-fps.59894/

I don't think it is necessary to make settings match. Just go in your game into LRP No chicanes, probably any car is fine, I tried flat6 and megane. Just click race, observe fps idling in garage, add 1 ai, click race again and see if fps dropped. For me megane at lrp goes from 163fps to 81 after 1 ai added. Thanks! Post any answers in other thread please.

UPDATE: I got this all sorted now, no need for anyone to recreate test now. Thanks a lot mesfigas & Goanna :)
 
Last edited:
The increased node number only affects the QSA portion of the tyre, right? That way you can define the tyre construction more accurately. How the tyre interacts with the road in the game is done with the realtime sections nodes which are 7 nodes across times something around the circumference.
Maybe language barrier here, but you wont tell us/believe that a tyre in rF2 have 7 noes, do you?
 
Back
Top