Testing Impact of Monitor Res on Graphics Card in VR

[flameoutme]

I see some posts that say where your monitor resolution is set at has no impact on VR performance.

 

Out of curiosity, I loaded up fpsvr, launched the m-2000 in free flight caracus. This at my normal resolution 2560x1440.

 

Then I shut down DCS and steamVR, changed the screen resolution to 800x600 (in windows, not in DCS preferences), fired up steam VR, then DCS, then the same flight.

 

Then I repeated both.

 

I am running the ValveIndex at 80mhz. CPU is i-9700K and 32GB ram. 1080 card, 8gb.

 

Note that I normally run with most bells and whistles shut off. I do have high textures. My PD in game is 1.3.

 

My take away is that while the FPS was 40 no matter what the resolution I was running, there were a couple of notable differences.

 

First, reprojection rates were 1-2% higher in the 2544 x 1440 resolution. Note that my monitor refresh rate is 59.9.

 

Second, the VRAM use was about .3-.5 gig higher. I am giving a range because they varied by test.

 

I know that just two go throughs does not a scientific result make, still, I do think there is an impact. Even though fps is the same, reprojections are higher, and that means the accurancy of frames is less.

 

How much difference this makes in game play I can't say. Still, for me, I always kick Windows into low gear when I fly VR because I want the smoothest experience possible.

[Eaglewings]

Interesting find. I try this and see.

I have only changed dcs game resolution in dcs graphic option and not the windows resolution to see no noticeable impact.

[Taz1004]

Similarly, if you have dual monitor plus VR and if you do have on-board graphic card, plug your second monitor to the motherboard.

Edited September 14, 2020 by Taz1004

[Lange_666]

@Taz: For DCS only or for everything?

[Motomouse]

done that, been there, not worth your time

 

If you want to reduce a resolution and gain true benefits, it is not the monitor resolution you should reduce ;-)

Edited September 14, 2020 by Motomouse

[Harlikwin]

Long time ago when I using a 980m (mobile yeah baby) and CV1 rift, I did notice a minor improvement setting DCS as low as it could go, but it was like a few FPS. But back then every fps mattered.

[flameoutme]

done that, been there, not worth your time

 

If you want to reduce a resolution and gain true benefits, it is not the monitor resolution you should reduce ;-)

 

Well yes and no. It depends. For me, I find that in my ValveINdex @ 80Mhz I am right on the precipice of creamy smooth flight with acceptable gauge clarity vice micro stutters. Micro stutters just kill immersion for me. So, in that case, every 'ounce' helps. Besides, it only takes a few seconds to kick it into monitor low res.

 

But if I were flying with more bells and whistles turned on, such that micro stutters were already in play, then I agree there's no use in bothering with it. Or if I had more horse power to begin with and have no sign of micro stutters, then it would not matter either.

 

As for lowering VR resolution/pixel density, yeah that would work, but I am at the lower limit right now on gauge clarity, so it's a non-starter.

 

Can't wait for my 3080!

Edited September 15, 2020 by flameoutme

[speed-of-heat]

An observation, microstutters are, in my experience, usually things outside of dcs... Moving to a 3080 Is unlikely to change them...

[Bob_Bushman]

I'm my personal unscientific testing

 

I have noticed a lot less microstutters if I decreased my cpu frequencies from a stable 4.9 (i7 8700k) to around 4.7.

 

Every test etc ran fine on 4.9 so by any measure it was stable.

 

But dropping down a few hundred Hertz though did nothing to impact frame rate an reduced stutters.