bandit648

I've been troubleshooting an issue I have with intermittent stutter for months now. Generally I see a quick split second drop down to 30-40 fps and then back to 60. This can happen on an empty map with no units present. At first I thought it was related to static clouds but I've later discovered that the clouds just make it worse and are not the cause. While running my test track I setup perfmon to monitor my disk and memory activity and I found that the exact same time I get the frametime spike I get disk and memory read spikes as well. Apparently DCS is choking while trying to read terrain information as I move through the virtual world. I was able to prove this with a simple test. I setup my test track that is just floating at 400 knots over the ocean towards the terrain off in the distance. I get frametime spike and stutter at the exact same time, every time. If I run the same mission but without moving and just holding at 0 knots I get no stutter. The game is stuttering when trying to read terrain information ahead from disk and memory. Cycling F11 view makes it better but doesn't completely eliminate the issue. I've tested my memory for 5 hours of memtest86 v8 and no errors. This may just be a DCS engine issue but I'd sure like to know if it's something on my end that can be fixed. Does anyone have any ideas? This is my test track floating towards the terrain @ 400 knots This is the same track standing still over the ocean I've attached the track I've used for testing. If someone who knows how to use MSI Afterburner could run this track and record the results I'd really appreciate it. For me, the track stutters with frametime spikes every time at about 20 seconds in. CA_benchmark_clouds.trk

SOLD on eBay for $670.

I'm selling my EVGA GTX 1080 Ti FTW3. The card has nearly 2 years warranty left on it through EVGA and includes all original parts/packaging. I will do $600+shipping on it with direct sale through paypal. Shipping in USA only. $670 Buy-it-now through ebay. Here is my listing on ebay: http://ebay.us/5acALt?cmpnId=5338273189

GTX 1080 Ti for sale, PM sent.

The other component we're forgetting here is the issue only persists INSIDE the cockpit.. so it has something to do with cockpit glass AND dynamic weather reflections :doh: :doh:

Use the link on the first post of this thread, it works.

Maybe the dynamic clouds use a different shader fx file. I've been through all of them trying to figure out a way to lower the dynamic cloud level and haven't figured out which one controls dynamic clouds yet. I've had pretty good results messing with the cirrus clouds through the \Bazar\shaders\clouds\cirrus.fx file though.

6. Dynamic clouds get a higher density in stormy conditions vs static clouds in level 8. 7. Static clouds can be set to different base levels, while dynamic clouds cannot.

Something I just noticed about this issue.... I watched the video of water flickering on my laptop and it's not noticeable. My laptop display is 4k@60hz. I wonder if the way DCS is rendering reflections is triggering this negative affect on certain types of monitors. My monitor is 32-inch LG-32GK850G 1440p 144hz G-Sync. I've tried different resolutions, refresh rates, gsync off, etc. but the issue persists. What type of monitors are you guys running? **NEVERMIND- I just watched the video in this thread on my laptop and it flickers as expected. My video doesn't possibly because of the way I recorded it in fraps.

Have you changed anything in regards to system config or settings?

It's frustrating because I can guarantee you this is a bug and not user created or system configuration related but there are only a few of us complaining about it. So, it's unlikely to get any attention from the dev's. The priority appears to be on the Hornet right now and not anything else. If you look that the bugs and problems sub-forum of the Hornet the dev-team is all over that, flagging posts with "[REPORTED]" and other tags indicating that they are looking at the issues. This forum is VERY quiet in comparison... :shocking:

Yes, so the issue is caused by reflections produced with dynamic weather clouds. I've seen a few threads here and there detailing this issue since around June of last year, but no response from ED. It's difficult to get attention on issues like this as ED seems to consider them low priority. You could try opening a ticket but it's unlikely get any attention. The best thing you can do is try to identify the exact conditions that produces the flickering. I think it has to do with certain lighting conditions so the only solution is to try and avoid those conditions.

I agree with Goa... no change in smoothness in latest update for me. I remain capped at 60 fps :thumbup::thumbup:

What I meant by that is the premium on the RTX cards is mainly for ray tracing, which I just don't feel is worth the cost in my use case. I pretty much only play DCS right now and don't play hardly any other games. That's a nice overlock you have there on your 2080 TI :thumbup::thumbup:

On my rig I'm running MSAA at 2x and view distance at ultra. I can maintain 60 fps in pretty much any situation. Upping MSAA to 4x and/or view distance at extreme just kills it though. Sure I can run 60+ fps just flying around at high altitude. As soon I get low over heavy forest or city fps start tanking. You guys make me jealous with your 2080 TI's. I'd love to upgrade to latest and greatest but just can't bring myself to pay those prices for ray tracing that I don't care about at all. Right now I'm going to try to squeeze some more perf out of my 8700K and 1080TI. Planning to delid my 8700K next week and go for 5.0+ overclock.

I started a thread about this recently also. Are you using dynamic weather in that video? In my experience I can only reproduce the flickering with dynamic weather clouds in certain situations. Static weather clouds don't do it.

I don't believe anyone who says they "never dip below 60 fps". There's certain situations in DCS that just cause the GPU utilization to get maxed and fps to drop. I don't recommend anyone with a 1080 TI or below to be running at Extreme view distance. With Wag's settings, all you have to do is fly low over a heavily forested area on the Caucasus map or over a city in PG and watch your fps drop into 50's or below. And that's on an empty map with no other AI units present. Other things like lots of static AI units or heavy clouds also hit hard. On my fairly decent rig I have to run view distance High, civ traffic off and drop MSAA to 2x to make sure I don't get drops below 60 fps in most situations. I wouldn't care that much but TrackIR becomes a stuttery mess below 60 fps. What we need is a standard set of tracks to benchmark with so we can get some real world results and see what kind of fps people are really getting. Maybe I should start a thread for that :thumbup:

Yep, you are absolutely right. I just did a test with DCS to see how this would go down on my system with 32 GB of ram. In general, DCS does not use the pagefile at all. Even when running a test mission I setup with over 400 units it didn't touch my pagefile. Memory usage peaked around 23 GB on the PG map. One thing to note is mem usage on PG is much higher than Caucasus map. Now, if I disable memory management in DCS by setting preloadRadius = 0 in the options.lua we can see what happens. I setup perfmon with pagefile, committed bytes and available mbytes counters. Then I loaded up a mission with around 100 units on the PG map and watched what happened. Shortly after starting the mission available memory started getting consumed until I reached 4 GB available. At that point the pagefile kicked in. So with 32 GB it appears the pagefile starts getting used when memory usage reaches 28 GB. At its peak it used 20% of my pagefile which was set to a system configured 4 GB. So, even under these EXTREME conditions the system only used 800 MB of pagefile. When it comes to pagefile advice I think people should just leave it as system managed/default. Windows used to match the pagefile to the size of physical ram but it doesn't do that anymore. On my system it's 4 GB. I've seen some people talking about dedicating 30 GB of SSD space for the pagefile when that is completely unnecessary for DCS or for any other reason really.

I've heard some say that the purpose of preload radius is to shorten load times for multiplayer. For me there is little difference in load time with DCS installed on an NVME drive. With min preload the PG map took 22 seconds and at max it took 29 seconds. I still think preload radius is doing something with memory management though. That's how it's described in the manual. If I force it off in options.lua then it effectively disables memory management and DCS uses all available until it runs out and then uses the pagefile.

The first question I would ask is how are you tracking your pagefile usage that makes you so sure it's being used. Many people on here confuse the commit charge with pagefile usage. Secondly, I can reproduce your exact scenario with DCS. Pagefile on, and DCS doesn't use it. Pagefile off and DCS doesn't use it. (w/32GB RAM) If I force off preload radius in options.lua, and run a heavy mission... THEN DCS uses all 32 GB I have and starts using the pagefile.

Actually it is correct. Windows doesn't just use the pagefile for no reason when physical memory is available. The primary function of the pagefile is for system crash dumps. If you have sufficient physical memory in your system it's totally fine to turn it off. If you don't believe it I suggest you monitor your pagefile usage CORRECTLY and then see for yourself. For anyone who would like to learn about pagefile usage on Modern versions of Windows I suggest you read this... https://support.microsoft.com/en-us/help/2860880/how-to-determine-the-appropriate-page-file-size-for-64-bit-versions-of

So it appears the "pagefile usage" counter is no longer available in current versions of MSI Afterburner. I think they changed the name of it to "commit charge" since that's what it was actually reporting. The pagefile usage counter seems to have caused a lot of confusion with people thinking their pagefile was being heavily used with it was actually just the commit charge. There's plenty of threads around discussing this. Here's a couple... https://linustechtips.com/main/topic/1009966-pagefile-usage-during-gaming/

I'm running 4.6.0 Beta 12 and I don't have any pagefile monitor. What version are you on and where do you enable the pagefile counter?

YES... this! We get little to no updates on work being done to the game engine so as a customer it becomes frustrating. Communication regarding updates to the Hornet have been great, but I'd like to see that same level of communication be applied to work being done to the game engine itself.

I use MSI Afterburner also... and it DOES NOT show you your pagefile usage. I assume you're talking about the "Commit charge" counter. That just shows the same thing as "Commited" memory in Windows Task Manager. It has nothing to do with your pagefile. Those counters go up to 20-30+ GB on my system also, with the pagefile DISABLED. This is why there is so much debate/confusion over pagefiles in DCS. People aren't monitoring the usage of it correctly. If you really want to see how much your pagefile is using you have to use windows Performance Monitor. 1.) Launch Performance Monitor>Select "Performance Monitor" in the left pane 2.) Click the + sign at the top, then find "Paging File" on the list of counters and select it. 3.) Add both Usage counters for your pagefile and click Add, then click OK. 4.) Change the graph type to "report" by selecting the graph icon at the top Now launch DCS and run some missions and tell me how much your page file is being used.