My 3090 Settings for my G2 (2.7.11)

[speed-of-heat]

Off , absolutely mullers the frame rate .25ms off the GPU time test 14.

Edited September 13, 2021 by speed-of-heat

[skypickle]

On 9/9/2021 at 1:28 PM, speed-of-heat said:

I will always recommend leaving the PD at 1.0, and just change the CRM ... your pixels is very high for running at 68% set your general resolution to 100% which should be 3168 x 3096

 

On 9/10/2021 at 5:41 PM, speed-of-heat said:

I leave world scale set to OFF as per my guide.

 

your resolution in Steam is set to what is close to 150% for me  

 

as you can see... 

 

Since we both use a G2, and VR sets resolution based on the headset, why do have different pixel counts at 100%?

[speed-of-heat]

1 hour ago, skypickle said:

 

 

Since we both use a G2, and VR sets resolution based on the headset, why do have different pixel counts at 100%?

No idea. Something else is off, but, but if you do the math then you will find mine is “right” (e.g. adjusted for barrel distortion correctly) but regardless that will account for some performance difference.

[Calinho]

Can you test Nvidia low latency on NVCP?

[speed-of-heat]

I will give it a go, but, as the effect of Low Latency mode is to reduce the pre-rendered frames to 1, and as the default pre rendered frames is already one , im not sure it will have much impact ... 

[speed-of-heat]

3 hours ago, Calinho said:

Can you test Nvidia low latency on NVCP?

		Frametime in ms		Delta From Baseline
Test	Baseline	GPU	CPU	GPU	CPU	Changes
25	Test 21	10.70	7.39	0.18	-0.04	Shadows Reduced Frame Rate Impact MOD
75	Test 25	11.36	8.27	-0.66	-0.88	Baseline 2.7.5.10869 (as per Test 25)
82	Test 75	9.52	8.55	1.85	-0.28	Average for Baseline 2.7.6, FSR1.1, Windows 11, Hotfix Shaders
83	Test 82	9.61	8.65	-0.09	-0.09	NVCP Low Latency Mode ON Test 1
84	Test 82	9.69	8.70	-0.18	-0.15	NVCP Low Latency Mode ON Test 2
85	Test 82	9.60	8.30	0.01	0.34	NVCP Low Latency Mode ON Test 3
86	Test 82	9.63	8.55	-0.12	0.00	Average for NVCP Low  Latency Mode On

 

Also note whilst I have moved to Windows 11, all we are really interested in is the delta from the baseline. In this case functionally nothing, I'm not particularly GPU bound, it might make things very fractionally worse on my GPU, but TBH  you aren't going to notice the impact on my hardware.  

[speed-of-heat]

OP Updated 

 

Methodology now has a link to the current performance Track...

 

I also corrected a numbering typo which referenced Track 26 as opposed to Track 25, being a baseline, it was just the numbering that changed (the actual data referenced is unchanged

 

I will likely restructure the OP as it has gotten quite difficult to ingest and update.  It may take a while 

 

 

[speed-of-heat]

p.s. FWIW Frame rate limiting via NVCP has zero impact in VR 

[Calinho]

6 hours ago, speed-of-heat said:

		Frametime in ms		Delta From Baseline
Test	Baseline	GPU	CPU	GPU	CPU	Changes
25	Test 21	10.70	7.39	0.18	-0.04	Shadows Reduced Frame Rate Impact MOD
75	Test 25	11.36	8.27	-0.66	-0.88	Baseline 2.7.5.10869 (as per Test 25)
82	Test 75	9.52	8.55	1.85	-0.28	Average for Baseline 2.7.6, FSR1.1, Windows 11, Hotfix Shaders
83	Test 82	9.61	8.65	-0.09	-0.09	NVCP Low Latency Mode ON Test 1
84	Test 82	9.69	8.70	-0.18	-0.15	NVCP Low Latency Mode ON Test 2
85	Test 82	9.60	8.30	0.01	0.34	NVCP Low Latency Mode ON Test 3
86	Test 82	9.63	8.55	-0.12	0.00	Average for NVCP Low  Latency Mode On

 

Also note whilst I have moved to Windows 11, all we are really interested in is the delta from the baseline. In this case functionally nothing, I'm not particularly GPU bound, it might make things very fractionally worse on my GPU, but TBH  you aren't going to notice the impact on my hardware.  

At this point i guess that NVCP changes don't work in VR, thanks for test. GPU frametimes on test 75 are very good, the delta from baseline 25 is from DCS 2.7.6 or windows 11 itself?

[speed-of-heat]

Mostly the big delta from 25 in terms of performance is fsr 1.1

I think the only change in NVCP that makes any real difference is prefer max performance.

Edited September 26, 2021 by speed-of-heat

[Calinho]

Just now, speed-of-heat said:

Mostly the big delta from 25 in terms of performance is fsr 1.1

I think the only change in NVCP that makes any real difference is prefer max performance.

 

Another interesting option in NVCP is MFAA, I think it doesn't work

[speed-of-heat]

Yep agreed, Lucas Did a bunch of work on that

[Calinho]

Maybe when DCS go to vulkan, ED can contact Nvidia to enable support to Resizable Bar. Another feature that DCS can benefit a lot is Direct Storage, but i don't know if vulkan support this technology

[speed-of-heat]

Vr DLSS would be better in terms of overall performance improvements… but that’s a discussion for a different thread.

 

I’m actually doing some rebar retests at the moment…

[speed-of-heat]

OP Updated Rebar retested with Release version of BIOS

[FupDuck]

Hey, after reading your first post for the third or fourth time over the last few months as I keep upgrading, I just want to double check: are you running the steam version? Or standalone and playing with the steam API? Thanks! 

[speed-of-heat]

Stand alone, and using steam just for vr

[Arcticfox]

Thank you for this, even though im using a 6900XT and a Oculus Quest 2 it still shows what can improve the framerates and what to avoid.

 

Kind regards,

 

[J-1775]

On 1/4/2021 at 2:21 PM, speed-of-heat said:

adding for the other two processes replacing dcs.exe with vrserver.exe and vrcompositer.exe

just to mind you, probably a mistype, 'vrcompositer.exe' should be 'vrcompositor.exe' to make a difference, shouldn't it?

[speed-of-heat]

Yep good spot, sorry completely missed that, if that’s the file name that’s what you need to change! Updated, you will note I spelt it correctly in the paragraph above. 

Edited October 9, 2021 by speed-of-heat

[dureiken]

Hi @speed-of-heat

 

why did you stop to use reshade ? does it impove colors ? or just sharpness ?

 

You achieve to have a perfectly smoot image with MSAAx2 or X4 or it's still "shining" on buildings and lines ?

 

thanks a lot !

[speed-of-heat]

@dureikenbecause of the increased amount of shimmer and the decrease in performance.. the decrease in performance was/is marginal but it is there, the sharpness is the very thing that increases the shimmer,  and its "HUGE"... so i stopped using it.

[dureiken]

Just now, speed-of-heat said:

@dureikenbecause of the increased amount of shimmer and the decrease in performance.. the decrease in performance was/is marginal but it is there, the sharpness is the very thing that increases the shimmer,  and its "HUGE"... so i stopped using it.

shimmer is what I call "shining" ? on lines and buildings, specially when sun is down for example ?

 

MSAAx4 is the only way to decrease that ?

 

thanks

[speed-of-heat]

i don't know what you call it; but probably... and MSAA was the only way i found to decrease it ... to something less "horrible" ... but you really are fighting the very thing you want... the very nature of the sharpness in the resharde makes everything shimmer...  MSAA "blurs" things... its sort of a pointless battle 

[dureiken]

6 minutes ago, speed-of-heat said:

i don't know what you call it; but probably... and MSAA was the only way i found to decrease it ... to something less "horrible" ... but you really are fighting the very thing you want... the very nature of the sharpness in the resharde makes everything shimmer...  MSAA "blurs" things... its sort of a pointless battle 

ok

 

I did exactly what is in your guide. Don't use reshade, only use FSR at 0.8 and sharpness to 0. Don't really understand what sharpness does if it doesn't make like MSAA ?