Taz1004

Below video, I dropped two CBU-105's with no crater model. Also showing resource usage graph. Skip to 2 minute mark to see where FPS drop starts to happen. You can see the GPU usage spikes when looking at the site of CBU impact. Even when there's nothing and smoke all dissipated. CPU also spikes somewhat but not as much as GPU. And towards the end you can see there are no craters. But still GPU spikes when looking towards and away from impact site. Below is same video WITH crater. Results are virtually identical. And as I noted above, this doesn't happen with MLRS which doesn't create crater. Obvious place to look at in my opinion is the code that creates the crater. I wouldn't be surprised if a lot of performance complaints are caused by this issue. Craters do not dissappear as far as I know. And it will accumulate longer you play a mission.

Click on the cog wheel and select 4K.

No. I suppose you can make it green. That's about it.

Strange I didn't get mention notification for this thread. Didn't mean to ignore. But I believe all smoke durations are locked away since 2.7 And I left VehicleFire.lua in my mod in case there's still something that uses it but new smoke are controlled by smoke.fx in Bazar and staticSmoke.lua in Config.

A10C-II was pretty much APKWS and HMCS too. And BS3 will be worth it just for bug fixes. Right now BS2 is unplayable other than maybe multiplayer.

I have related question. Most airfields have ILS only one direction I think. What happens if the wind's blowing the other way? Conditions requiring ILS are usually very heavy wind too.

I upgraded to 64GB today. RAM so cheap nowdays. But no change in Syria performance.

As I noted, you need to right click and open in new window from Chrome. And click on address bar and Enter. Or use Edge and ignore security warning. Warning is from no SSL on my web server.

I think the issue here is that FPS drop is there even without all the polygons.

It always was cumulative. Why I always recommend using the vertical resolution number in Per App. Percent value is different per headset and has no meaning other than to you.

It's not the crater graphic but it is related to crater creation. MLRS in the game doesn't create crater (which should be another bug BTW) and they don't cause stutter. And it's also related to number of craters. When I test with just 5 artillery, stutter was very minimal. But increased it to 15, it started to crawl. Thinking it's related to warhead but weapons are off limits since 2.7

It's not the crater graphic. I deleted the crater textures and still happens. I then removed the crater edm file altogether and there are no more craters but still FPS drop when looking at the site of impact. You can test out and let us know. Attached below will remove all craters from the game. CraterTest.zip Has to be some sort of leak in the code. For now, I just deactivate all artillery and avoid using cluster bombs.

I can see how it could be confusing. So I reformatted the first post.

Thanks Grimes. I hope you can pass on this note along with the bug report. In my track, if you change AI weapon from Vikhr to Kh-25ML, AI have no problem attacking with any AI skill level. Please test yourself. This is problem that existed with Ka-50 Vikhr for years. It just seems to have gotten worse since latest OB. Before 2.7 when we were able to mod weapons, if I put Hellfire on Ka-50 AI wingman, he had no problem attacking in any conditions. At least the Ka-50 issue is related to Vikhr.

Something's not right. It has no control over SteamVR setting.

You downloading Legacy version? Legacy is for old A10C. Not A10C-II. I didn't really understand what you mean by outside of the hog. But folder inside Liveries folder are independent.

Yes it works. You have to put the Liveries folder from the mod into \Saved Games\DCS\ So it should read \Saved Games\DCS\Liveries\A-10C-2020-CPT\Taz1004 Custom Aged\

Same? Between 100, 130 and 150? Sure it's working? FSR or NIS is not meant to be supersampling. You can try higher renderscale than 1 but really it's designed to be lower. It's algorithm is designed to clean up artifacts when images are upsampled from lower resolution images. There's nothing to clean up if it's downsampled. And SteamVR doesn't do any algorithm for upsampling. So in a way, you're basically not using FSR if you use higher than 1 render scale. As for the math, take your Per App SteamVR vertical resolution and multiply by Render Scale. That's what FSR is rendering at. Vertical resolution divided by Render Scale should be max. With some performance hit from FSR of course. Somewhere between the max and original resolution should be good balance of performance and quality.

Assume you mean video editing. Not sure if it's simple but DaVinci Resolve is most feature complete free video editor.

I'm pretty sure it's changing. Source

I did some further testing with all 4 AI skill level and Trained was only one that attacked the targets. Tested multiple times and Trained attacked them every time while others failed. Seems like their skill level is related to stand off distance. Ace stays furthest away and rookie gets very close. Ace and Veteran stays far away but seems like they can't see the target even within 6nm. Rookie just gets too close and runs away. Trained is only one that gets close enough so it can see the target but not too close to run away. Will test them more in other missions.

Yes, the latest version from today. I forgot to attach track for the second video with crash. I'll attach it here. Ka-50_AI2_Crash.trk

Well, can't really prove it unless I am streaming all the time. Believe or not, in this second video, the AI wingman did ran into me during gameplay. But shows going under in the track playback.

Right, this is different than ReShade. It's not sharpen filter. It does come with its own sharpen but it's very rough sharpen. I turn its sharpen way down and use ReShade sharpen instead. If your SteamVR vertical resolution is 3000 and set render scale to 0.67, your frames will be rendered at 2010 vertical resolution and then upsample to 3000. But even with upsampling algorithm, it will look worse if you don't compensate. You need to compensate by increasing SteamVR resolution. How much you increase will depend on whether you want quality or performance. In this example, if you increase SteamVR resolution to 4477 vertical resolution, you're basically back to where you were. FSR will render at 3000. There will be performance hit from upsampling algorithm however so I wouldn't recommend going that high. And you probably don't need to.

As I showed on AI Bug thread with video, there are inaccuracies even in 2.5 minute track.