Dear VR DCS users…
Teach me about the MSAA Mask in the VR options.
What about the size? Is it the AA enabled area I set, or is it the no AA area?
Does it affect performance?
Dear VR DCS users…
Curious here too. My hunch, and that’s all it is; depends on your headset? I’ve played with it on my O+ and see no difference. Now, I’ve not tried to actually reduce the size. Saying all that, I haven’t played with it since this feature came out. So, lets see…
I’m using the shader mod so that’s probably a factor, too.
UPDATE: Ok, yeah, it does do something (Shader mod = off) and it is helpful, visually. As far as the value goes, I assumed it was a fraction of the entire rendered frame but, on my gear, that’s not quite correct. While 1.0 will AA the entire image, 0.1 (lowest mine will go) still AA’s a reasonable amount in the center of view. Appears like more than 10%.
My test case was: F16, inside a hangar at Incirlik, where I can see another, open, hangar across from me. All those arcs (trusses inside/underneath the hangar roof) really show the effect. You can adjust this on the fly (menu open), making the results more apparent.
FPS wise I notice no difference (in the above, static, situation) from 0.1 to 1.0. However, seems you have to have MSAA set in Systems - that mask setting needs something to work with. That does of course put a big dent in my FPS. On the order of 25%-30% hit. I’ve learned to live without MSAA.
I haven’t used the shader mod for over a year… Mostly because of the hazzle of installing it every update and because VR performance in DCS got better.
Maybe I should have a look at it again…?
Has someone made an installer for the shader mod yet?
How long has this been a thing?
The MSAA Mask in DCS? Since this spring, I believe. But like @jross, I have learned to live without MSAA. Just thought I’d check if it makes any difference.
Why not use a mod manager like OvGME?
On my rig it does make a difference. Most noticeable on the Syria map (the most demanding one I have). I turn a lot off just to, hopefully, get some headroom:
MSAA = off
Ground Shadows = off
Cockpit Shadows = depends on my mood
Shader Mod = Installed (except for the test above - the VR MSAA mask has no effect with the Shader mod installed).
Doing acro over Aleppo in the Hornet I’ll stay at 45+ with the above.
Stock shaders it’ll drop into mid to low 30’s, sometimes even high 20’s. This is with very little else going on.
When the dust settles on the G2, and latest graphics card shenanigan’s, a new system is coming . I’d like to say dropping $3000, to replace my 3 yo system, will double my DCS performance. Just not that confident, yet. Still watching, waiting.
Then they’ll add dynamic Wx and fancy clouds - be back where I am now
Yeah, but that will only install the files, not remove the old shaders. You need to manually delete the shaders in each terrain folder.
Batch file? Gosh, haven’t written one of those in a long time. They still work I’m sure. Just a thought. In fact, darn it, I need to write one. Shouldn’t take more than a few minutes
On a whim I played with SteamVR’s SS setting. Mine defaults (auto) to 150%. Always has. Something about the reported ‘bug’ with SteamVR SS and the G2 - needs be set to 50% - made me adjust mine.
I actually got back some performance taking it down, way down. To 20%. Was doing this on the fly and didn’t notice anything ugly happen (fpsVR app showed it was in fact updating on the fly). At least, not uglier that before. Out in the countryside there’s no noticeable diff. Airfields and cities look as always. Shimmering messy…but not worse!
Thought I knew what SteamVR SS did but maybe not. Or drivers/software updates have improved things since the last time I twirled that knob.
That fpsVR app, while not free, is coming in handy. At least when, like me, your gear is on the edge; you can see the results of diff. settings on memory, GPU, and CPU work.
[EDIT] Ok, I was wrong: SteamVr SS being applied while DCS is running does nothing. Even though fpsVR showed it was changing, and I did see a small, but noticeable, FPS increase. However, on a re-start 20% looked, well, really bad. Search me.
Did you notice any difference there?
I’ve read about this ‘bug’ that sets the 100% resolution way too high. I experienced it with my Rift S too. Then I read that it isn’t a bug, but something to do with pixel fill in the middle and that Steam VR needs to render more pixels to be able to fill the full resolution of the headset… I don’t know.
To add to this discussion, I wonder is how SS and PD affect each other, and if one preferable over the other, with regards to image quality.
Does anyone, like me, think that there is much more haze effect in VR, which IMO, negates the increased resolution? Unsure if that is intended, or a consequence of how DCS implements VR. It feels like perpetual haze (good beer name?) looking out beyond a mile or so.
Great beer name!
I can’t compare to 2D since I’ve been flying DCS exclusively in VR these last few years.
I have been swapping between a higher PD and using MSAA. I was hoping MSAA Mask would let me get away with less flickering but at a smaller performance cost…
@Troll Your link enlightened me - did not suspect that’s what SteamVR SS was really doing. So, I’ve returned it to ‘auto’. Seems to be the best overall setting: visual quality vs performance.
I did find it strange that something was happening when I adjusted SteamVr SS as DCS was running, and thought I noticed a very slight performance increase, with no visible change (and fpsVR reported it changing so something was being communicated) - but upon a restart with those much lower settings it looked terrible - imagine DCS in VR at 320p!
@jross, bear in mind that I haven’t seen an official comment from HP on this and people say they are working on tuning this in Steam VR… So I guess things may change for the better.
I hope so. What that link said kinda makes sense. I always intuited that it was simply scaling the image (pixels) uniformly.