@Interslice Not 100% sure I can make Sunday, but is it OK to sign up now or should I leave until last min? (sorry, still unfamiliar with SGP)
Either/or John. Whatever suits! There's usually a few in and a few out on race day.
AC RSS GT @ Road Atlanta - Sunday 19th June 2022
- Thread starter Interslice
- Start date
Assetto Corsa Racing Club event
Either/or John. Whatever suits! There's usually a few in and a few out on race day.
Haha yeah that's a big hit! Msaa supersamples all "solid objects". I don't have much knowledge about what these solid objects are and what it's called professionally, but if you disable temporal anti-aliasing (TAA) in csp, you'll see fences, white lines, grass, reflections and some other stuff having a lot of pixel crawl.
Everything that's still smooth is rendered at 8x the resolution!!!
So setting it to 4x will already free up your graphics card quite a bit!
Thanks for that info, makes sense!
I also always use an fps limit when using vsync. Somehow with a limiter things are smoother.
Btw if you want to improve your input lag by quite a lot, you should install riva tuner statistics server (I'd recommend the latest msi afterburner download from guru3D, which comes with rtss and will give you a lot of possibilities.
(if you don't change anything in afterburner, no overclocking/undervolting/fan curve etc will be applied!! But you'll get the hardware monitoring graphs).
And then limit the fps to 59.97.
Or more accurately: Google "vsynctester" and there should be a website with that name. It will test and read out the exact Hz of your monitor.
Limit the fps at -0.03 of that.
It keeps the frame buffer empty and although, in theory, you get a one frame long stutter once per second, I never noticed it..
The input lag is massively improved though!
AC benchmark is a good way for graphics card optimizing! Not so much for the CPU, but very good for graphics card!
Afaik it also disables vsync automatically so if you disable the limiter, you'll get maximum fps and can compare the results very easily.
Instead of going through them one by one, I'd run the benchmark once, then load my preset and check if the fps go up massively or not.
And then re-load your preset and go through it one by one.
Always good to make a first test to see what to expect.
Well that seems to have sorted that out
I've attached a notepad file of my quick testing results instead of cluttering up this thread any more than I have already. Biggest difference was indeed seen by turning down 8x MSAA. While not making a huge difference to the average and minimum framerates, running at 2x or 4x MSAA completely stopped the stutter at Road Atlanta. As I worked through the CSP settings some things had more effect than others but I arrived at settings that now work very well for my system and look great.....all in the notes.
Tried Riva Tuner but I seem to be far more sensitive to frame tearing than input lag so it's back to ingame vsync and 60 fps limiter.
THanks again Rasmus.
I've attached a notepad file of my quick testing results instead of cluttering up this thread any more than I have already. Biggest difference was indeed seen by turning down 8x MSAA. While not making a huge difference to the average and minimum framerates, running at 2x or 4x MSAA completely stopped the stutter at Road Atlanta. As I worked through the CSP settings some things had more effect than others but I arrived at settings that now work very well for my system and look great.....all in the notes.
Tried Riva Tuner but I seem to be far more sensitive to frame tearing than input lag so it's back to ingame vsync and 60 fps limiter.
THanks again Rasmus.
True of course, but this was the only track that was causing big problems for me with my previous settings so my newly acquired extra headroom helps here.
Also, I run quite a low FOV/ seat position on a 27" monitor so I can maybe now look at getting something bigger with a higher res screen which means I might even be able to see apexes
Perfect! Nice to hear!
Imo the distracting aliasing and pixel crawl happens for the stuff that isn't affected by msaa anyway (shadows, fences, lines, cables) but TAA from csp applies to the final image output, which works wonders.
It's quite heavy on the gpu due to the temporal part which calculates the trajectory of the aliased pixels and blurs them in advance.
Massive game changer for me personally! I forced 2x sgssaa (sparse grid super sampling) via Nvidia inspector before but that smoothed less than taa while being more blurry at the same gpu load...
Yeah I never bothered to bench every setting. Especially since they sometimes interact with each other.
Great that you've found a good mix now!
I looked into your notes but you didn't make it clear when the things were active or disabled so I'm a bit confused haha.
Only thing I saw was that you disabled reflections FX while also testing the normal reflections rendering frequency.
About this:
Since csp 0.1.72 iirc, the reflections FX changed.
Without them, my cockpit is very bright, rollcage poles are super bright etc.
With reflections FX active, my cockpit looks way more realistic!
So for me it's more about the shadows and lighting in the cockpit, not about "reflections".
And with this change, they also created a smoothing functions that makes "one face per frame" look super smooth without stuttering.
So if you test "one face per frame", always activate reflections FX and also make sure that you're on one of the latest csp versions.
For me, static is cool. Smooth and improving fps. But you only see the same reflection all the time.. Bridges aren't shown, buildings aren't shown.. I couldn't live with that downside.
You have to keep vsync enabled!!
The 59.97 limit allows for vsync to stay active without noticeable stutters but at the same time keeps the frame buffer more or less empty, which improves input lag significantly.
If you disable vsync, the tearing indeed becomes an unbearable mess
Yep!
The only reason would be input lag.
But the good thing about simracing with a ffb wheel is that the wheel has a very high update rate and is independent of the visuals.
So your controls aren't affected by the input lag. You can catch the slide before your eyes notice it.
If you're playing a shooter, your mouse will feel like navigating through pudding with normal vsync.
Or if you're like me, playing some shooters here and there with vsync on for years and then changing to gsync, your mouse will feel like you've found a wormhole in the time-space-continuum
Wouldn't be better to choose another time? Exactly then the F1 Race in Canada starts.
It's the same time as every Sunday in 5 years
But since I'm following F1 this year, I agree that a change just for this race might be a good idea!
I can move it forward ten minutes. You can practice/qualy while the safety car is out
Extra 10 mins added to qualy. Race should run from xx:55 to yy:55 in reality.
The Almighty Snark
Premium
I've been looking forward to this one all week but unfortunately my muscles are waaayy too stiff and I am walking around like an old man today I checked and it would have been a fun Ferrari vs Lambo race today!
I checked and it would have been a fun Ferrari vs Lambo race today!
I've been looking forward to this one all week but unfortunately my muscles are waaayy too stiff and I am walking around like an old man today
I'm not gonna make it either
Not feeling well so I'll probably go to bed after the F1 race...
I hope you'll have a massive blast around this roller-coaster!
Thanks Lads. Get well soon Rasmus.
Damn, sorry @JamKart22. I didnt mean to take you out on lap one. My pedals slipped and I completely lost concentration. Could not fix the pedal positions without stopping and caused me to go off track so many times I was almost being lapped within 10 minutes. Sorry to spoil your race Colin. Have a good evening everyone.
I concur Brian, such a great track and car combo even with a small grid it was fun pushing around it for 45 laps. The gap I had was not one to be complacent with and I started to sweat towards the end of stint one as you knocked about 3 seconds of it in 4 laps. I must have had a better stop as I had regained a bigger gap when I came out despite losing time missing the pit entry and having to do another lap to get back round. I was expecting to see you a couple of seconds behind me at that point so I appear to have got lucky. Thanks for organizing
Last edited:
Shame more folks didn't get to try this. Road Atlanta in these is really really nice. You could really feel the quality of both the car and track working together.
Enjoyed the race alot. Myself and Tim had similar pace over a 30 minutes stint but he was much more consistent and seemed to make no mistakes. I made alot, costing 1 - 2 seconds each that I never got back. A few of them were code brown so i was happy enough to not loose any places and stay in P2 early on. Went for the fastest lap, late on, on the last lap or two. Finally that properly binned it through the esses
First 2 laps chasing @Fiberoptix. Closest I got
Thank to all for racing. Congrats to Tim for the win and @Goldtop for the P3
Looking forward to next week already
Enjoyed the race alot. Myself and Tim had similar pace over a 30 minutes stint but he was much more consistent and seemed to make no mistakes. I made alot, costing 1 - 2 seconds each that I never got back. A few of them were code brown so i was happy enough to not loose any places and stay in P2 early on. Went for the fastest lap, late on, on the last lap or two. Finally that properly binned it through the esses
First 2 laps chasing @Fiberoptix. Closest I got
Thank to all for racing. Congrats to Tim for the win and @Goldtop for the P3
Looking forward to next week already
Alfred Wayne
Premium
Yeah driving these here was I think my favorite track so far. Right on the edge of control.
As far as my race, spent the vast majority behind Tudor, managing to stay within a second or two. Decided to try and pit a few laps after him to have fresher tires at the end, but the strategy was for naught and I binned it with about 15 min left, and from there I couldn't find a rhythm again. Good pace Tudor
Bigger grid would have been nice, but there are worse cars and tracks to spend an hour in on a Sunday. Good pick Brian, and good racing everyone
As far as my race, spent the vast majority behind Tudor, managing to stay within a second or two. Decided to try and pit a few laps after him to have fresher tires at the end, but the strategy was for naught and I binned it with about 15 min left, and from there I couldn't find a rhythm again. Good pace Tudor
Bigger grid would have been nice, but there are worse cars and tracks to spend an hour in on a Sunday. Good pick Brian, and good racing everyone
Partick Eaton
Premium
I managed a two stop strategy, well I guess I need to check my fuel calculations! I had good pace, but not very consistant.
Great combo, would do it again.
I had a hike with my son and his dog planned for after the race. They showed up early and the dog nudged my elbow causing a spin. Should have kept him in the back seat.
Congratulations to the winners and thanks for hosting.
Great combo, would do it again.
I had a hike with my son and his dog planned for after the race. They showed up early and the dog nudged my elbow causing a spin. Should have kept him in the back seat.
Congratulations to the winners and thanks for hosting.
