biggydeen wrote: ↑10 Aug 2021, 03:37
Well this looks like the exact same issue I have. The best way to test is by playing the smaller snow map called Fjell 652 (conquest). On this map, all gameplay is buttersmooth without any dips on my system (Steady 140 fps without dips). If you also expierence buttersmooth gameplay on this map without stutter then you have the same issue as me and I can explain what is happening and how to fix this.
I usually play Pacific maps and I haven't played Fjell in a long time but you're right, the map is known for running better than other maps and the gameplay is indeed the smoothest and frametimes most consistent, I immediately noticed tracking enemies was a lot more satisfying.
biggydeen wrote: ↑10 Aug 2021, 03:37
On other maps, FPS sometimes drops below 140. At that exact moment I notice some kind of stuttering. And the best way to explain this is that the monitor is not receiving a steady pace of images to process. This is not a problem of the monitor nor g-sync. G-sync will not fix any stuttering caused by bottlenecks in the system. I thought for years g-sync would offer buttersmooth gameplay @high fps but this is simply not true. That's why I never found a solution to my problem. I've had this problem with every system I had, with completely different hardware. The reason why I never found a fix is because it was never related with g-sync or any other setting. It's simply caused by the fact that if you do not feed your monitor with a steady pace of frames, you will notice some kind of stuttering if this pace is not steady anymore.
And although the guys from BB mentioned this before, it just hit me when I watched this vid:
There is definitely a difference in using Gsync but when we're talking about these microdips then yeah becomes very hard to chase that final step going from 9/10 to 10/10 smoothness.
I'm familiar with Battle(non)sense, I've watched many of his videos in pursuit of the best settings.
But in that video he simply talks about the "lowest FPS you can maintain" and that's where our issues stem from, it's not the FPS we can maintain, it's the absolute lowest microdips we don't notice with a basic frame counter when it constantly shows over 200 FPS.
biggydeen wrote: ↑10 Aug 2021, 03:37
Do not be fooled by the FPS you are getting when uncapped. Sometimes, fps will simply drop below 140 on bigger maps. For battlefield games this might be related to the CPU. I always have multiple cores @100% load no matter what map or FPS I get. I can only assume that cores @100% load all the time will cause hickup's somewhere. Just look up some vid's on YT with your specs, you will always see that the 1% low is below 140 eventhough avg fps is 200+
I've tested this extensivly over the past few days. And I always notice on the smaller maps in BF1 and BFV everything is buttersmooth because the frametime line is just a complete flat steady line. But when playing on the bigger maps, sometimes things get out of hand on the screen and is creating a cpu bottleneck causing fps to dip below the cap. And even just 1 frame below cap means inconsistancy in the frame pacing and that will be noticable as stutter.
You describe the exact same issue I have so i'm pretty sure it's the same problem. I've tried every setting there is and nothing will fix this. Try capping the FPS to 100 and see if that resolves the problem (it should).
It looks like this seems to be the case.
I just jumped across a few different maps lowering the cap down to 120 or 100 and this removes about 80%+ of the spikes in the frametime graph but at the same time this doesn't make the experience better because now I've removed about 30% (40 FPS) of the frames from my monitor and that causes noticeably less smooth gameplay overall.
It might be more consistent from the framtime perspective but it's also less enjoyable as a whole.
biggydeen wrote: ↑10 Aug 2021, 03:37
I think I finally understand now that there is no fix other then lowering settings to get a higher 1% low or lowering the FPS cap at or below the 1% low. Do note that battlefield is using 100% cpu all the time and not utilizing GPU that much. So in other games, when the GPU usage is at 100% you might have a different scenario. For example, when playing control at max settings my GPU is running 99% all the time. And although my frames are inconsistent I don't notice any stuttering. So be aware that the frostbite engine is more sensitive to this due to high CPU load or something.
Let me know if your experience the same thing.
The only thing different in our situations is some of your CPU cores hit 100% usage which is absolutely not the case with my system.
I just checked on multiple maps and I've never seen a single thread hit over 80% usage ever, most of the time they're in the 30-70% range and there's about 14 threads loaded during gameplay with rest in single digits.
But I'd say the fix is in better hardware, maybe not now but down the road. New Intel and AMD CPUs are coming in a few months.
If 1% and 0.1% lows are what's causing this issue then obviously a better CPU or RAM would boost those lows.
Overclocking RAM definitely makes a difference on Ryzen systems, I've noticed some small improvements with higher RAM + Fabric frequency and tighter timings which I'm still in the process of tuning since it's really time consuming for a RAM newbie like me.
I wish I had a 10900K/11900K system to try out, even though they might get slightly lower average FPS, I'm curious how they do with the lows.
It's very difficult to find this info online, the big reviewers can't do multiplayer test for obvious reason of pure randomness of action and people that post gameplay on YT all have very different specs/settings etc.
Aside from that I don't know what else would make a difference, if going 240Hz+ would bring any improvement.
Yes the dips would still be there but there would be more overall smoothness, just like the frame cap at 100 reduced spikes but decreased smoothness maybe the opposite happens at higher fps - bigger spikes but more buttery overall, a compromise.
I've read that higher Hz "break up" the tearing into smaller sections so they're easier on the eyes, but I'm not sure if this is true, maybe the monitor experts could chime in on this one.
halp
