HP's Reverb VR Pro Headset

[doraymon]

So far I kept PD always at 1 and last flight was with 150% and MSAA 2x.

With this setting the picture quality is barely palatable and I notice a loss of fluidity only when rolling fast or moving the head quickly.

 

Overall I have to say that I am a bit underwhelmed by the picture quality, I wish someone had told me "forget the monitor resolution" before I spent the money...

Obviously in VR everything else is much much better, landings are easier, following a target is easier, and so on...

Edited May 18, 2020 by doraymon
grammar check

[Taz1004]

I get biggest visual improvement from the in-game PD. Turn off all anti-aliasing. SteamVR PD to 100% and in-game PD to 1.4 gives me best image to performance. In-game PD hurts the performance more but looks better than Steam PD. Again, it's preference thing as I can live with jaggies but not blurriness. And at higher PD, jaggies get smaller since supersampling is a method of anti-aliasing.

 

And if you have dead pixels, I'd return it.

 

Stuttering also deals with what you're doing. If you have IR display on MFD on boresight and making a hard turn, dragging that boresight along the terrain, of course you would get stuttering.

 

I just discovered new Vsync settings in Nvidia control panel. Donno when those were implemented as I haven't checked that tab for a while. But Adaptive (half refresh rate) seems to help reducing stutter at low fps. And Virtual Reality pre-rendered frames set to 2.

[Hammer1-1]

You can hard limit your FPS in nvidia control panel by selecting the max frame rate dropdown list. That actually helps free up some headroom in your video card for other small things.

[imacken]

Bit of a moot point in a VR thread!

[VirusAM]

I get biggest visual improvement from the in-game PD. Turn off all anti-aliasing. SteamVR PD to 100% and in-game PD to 1.4 gives me best image to performance. In-game PD hurts the performance more but looks better than Steam PD. Again, it's preference thing as I can live with jaggies but not blurriness. And at higher PD, jaggies get smaller since supersampling is a method of anti-aliasing.

 

And if you have dead pixels, I'd return it.

 

Stuttering also deals with what you're doing. If you have IR display on MFD on boresight and making a hard turn, dragging that boresight along the terrain, of course you would get stuttering.

 

I just discovered new Vsync settings in Nvidia control panel. Donno when those were implemented as I haven't checked that tab for a while. But Adaptive (half refresh rate) seems to help reducing stutter at low fps. And Virtual Reality pre-rendered frames set to 2.

 

 

 

It seems better, just because it is calculated in a different way, so Pd 1.4 rendered resolution is > then ss 140% most people still fail to understand this, and it is a counterintuitive.

 

Pixel density, is calculated by multiplying the screen diagonal to the multiplier...while supersampling is just a multiplier for the resolution....two very different things.

So it is normal that with PD@1.4 you have better quality image then steam SS@140%

[Taz1004]

It seems better, just because it is calculated in a different way, so Pd 1.4 rendered resolution is > then ss 140% most people still fail to understand this, and it is a counterintuitive.

 

Pixel density, is calculated by multiplying the screen diagonal to the multiplier...while supersampling is just a multiplier for the resolution....two very different things.

So it is normal that with PD@1.4 you have better quality image then steam SS@140%

 

I am VR content developer so I know the difference. They may use exponential or linear multiplier, but what it does is actually same. Renders high and downsample.

 

In game PD of 1.4 is much better than Steam SS at 500%. Steam SS stops doing anything after certain point. Which makes sense.

I know some people suggested turning down in game PD to below 1 and jack up Steam SS to 300%. They do not have correct understanding of the two.

Edited May 19, 2020 by Taz1004

[doraymon]

I am VR content developer so I know the difference. They may use exponential or linear multiplier, but what it does is actually same. Renders high and downsample.

 

In game PD of 1.4 is much better than Steam SS at 500%. Steam SS stops doing anything after certain point. Which makes sense.

I know some people suggested turning down in game PD to below 1 and jack up Steam SS to 300%. They do not have correct understanding of the two.

 

 

Thats interesting.

May I ask you how you would proceed to reach the best combination of the two?

I’m not asking to give me a value as this will depend on the specific hardware but just a method to reach the optimum.

 

 

Sent from my iPad using Tapatalk

[Taz1004]

I know some people will disagree with me so I made a quick example out of 2D photo. Just took an image I found on the web, scaled up 200% and back down. 500% and then back down. You will see big sharpness jump from 100% to 200% but 200% to 500% is minimal.

 

 

This is "basically" (I'm over simplifying) what SteamVR SS is doing. Taking the image the game renders, upscale with filter, then downscale with sharpen filter. This is obvious because if you crank the SteamSS all the way to 500% it'll say some 4500x5000 pixels. There's no way the game engine can render 3D at that resolution. SteamVR is only doing 2D Supersampling.

 

In game Pixel Density however, is rendering 3D at higher resolution and downsampling. This is why in-game PD takes higher frame rate hit. But is at higher quality.

 

2D Supersampling is fast but the limitation is as shown in that image above. It's effective up to certain point because it can't create something that wasn't there in the first place.

 

Best process is to get the highest in-game image quality vs frame rate as you can. Whether you choose to use MSAA or higher PD is up to you. Then back it down one notch and supplement with SteamVR SS. But once again, image quality is highly subjective. You have to find your own.

Edited May 19, 2020 by Taz1004

[imacken]

It seems better, just because it is calculated in a different way, so Pd 1.4 rendered resolution is > then ss 140% most people still fail to understand this, and it is a counterintuitive.

 

Pixel density, is calculated by multiplying the screen diagonal to the multiplier...while supersampling is just a multiplier for the resolution....two very different things.

So it is normal that with PD@1.4 you have better quality image then steam SS@140%

 

I don’t know how many times this has got to be clarified.

 

SteamVR number is based on total pixels and PD multiplier has to be applied to each axis. So, a PD of 1,4 gives you 1.4*1.4=1.96 times the number of pixels, equivalent of SteamVR 200% (well 196%).

 

1.4 does not relate to 140% in any way.

[Waxer]

and that it is plugged into the graphics card ... not the motherboard ...

 

Yeah:you are not using the HDMI or DP of your motherboard are you? You have you use a spare port on your graphics card. Otherwise the headset will be using the iGPU of the CPU.

[VirusAM]

HP's Reverb VR Pro Headset

 

I don’t know how many times this has got to be clarified.

 

SteamVR number is based on total pixels and PD multiplier has to be applied to each axis. So, a PD of 1,4 gives you 1.4*1.4=1.96 times the number of pixels, equivalent of SteamVR 200% (well 196%).

 

1.4 does not relate to 140% in any way.

 

 

 

Exactly what I wrote...

We wrote the same thing so why did you correct me?

I wrote that pd 1.4 is not the same of ss 140% but is much more.

I didn’t do your calculations but the concept is there.

Maybe you didn’t read me correctly

[imacken]

Exactly what I wrote...

We wrote the same thing so why did you correct me?

I wrote that pd 1.4 is not the same of ss 140% but is much more.

I didn’t do your calculations but the concept is there.

Maybe you didn’t read me correctly

 

I wasn't correcting you particularly. I was just expressing some frustration that people don't seem to get the difference in the numbers. It's answered over and over.

 

Also, I just wanted to set out exactly what the relationship between the PD number and the SteamVR one - again! You need to square the PD value (and multiply by 100) to reach the SteamVR %age.

[doraymon]

May I ask if there is a place where I can find the users settings with regards to:

1) WMR for Steam - Visuals (High or Very High?)

2) Screenshot of DCS graphics settings

3) Did you enable MotionReprojectionMode editing the default.vrsettings file?

 

If not, would anyone mind sharing?

 

As there are so many variables and one life only I was planning to study those settings along with the hardware of the user (usually in signature) to cut down the variables...

[speed-of-heat]

check my signature

[Hammer1-1]

May I ask if there is a place where I can find the users settings with regards to:

1) WMR for Steam - Visuals (High or Very High?)

2) Screenshot of DCS graphics settings

3) Did you enable MotionReprojectionMode editing the default.vrsettings file?

 

If not, would anyone mind sharing?

 

As there are so many variables and one life only I was planning to study those settings along with the hardware of the user (usually in signature) to cut down the variables...

Start here and tweak to what makes it work for you. Its not up-to-date, but itll get you in spitting distance to where you want to be. The shader mod helps a lot.

https://vr4dcs.com/2019/09/10/reverb-settings-for-dcs/

[imacken]

Start here and tweak to what makes it work for you. Its not up-to-date, but itll get you in spitting distance to where you want to be. The shader mod helps a lot.

https://vr4dcs.com/2019/09/10/reverb-settings-for-dcs/

There are so many things outdated and effectively wrong in that old page, I feel it is of limited use now. It can confuse new users a lot, as options have changed significantly in the 8 months since it was written.

The Shaders mod helps some people a lot, but not everyone. Makes very little difference to me as an example.

[sze5003]

May I ask if there is a place where I can find the users settings with regards to:

1) WMR for Steam - Visuals (High or Very High?)

2) Screenshot of DCS graphics settings

3) Did you enable MotionReprojectionMode editing the default.vrsettings file?

 

If not, would anyone mind sharing?

 

As there are so many variables and one life only I was planning to study those settings along with the hardware of the user (usually in signature) to cut down the variables...

I have a 1080ti and a reverb. My dcs settings include PD at 1.0, everything except shadows set to high, draw distance all the way up, trees and clutters as you prefer, I use that slider to increase a little.

 

Play around with the shadows, this seemed to have a performance impact for me where I could see it by moving my head around vs how smooth it was in VR.

 

No msa but it's been a while since I've got back into dcs. No super sampling either in the dcs video settings.

 

Steam VR I use the wmr for steam VR beta app and I have the steam resolution sampling at 100%.

 

I didn't count fps but I mainly do single player stuff and I have not had any stutters or graphical issues.

 

You will have to play around with the settings until you get the head movement as smooth and as quick as you prefer. You will notice if you increase the settings or the dcs PD that when you look around your movement will seem slower. So then go back and decrease sone of the graphical settings until it's "lighter" to move around and smoother.

[Taz1004]

I wasn't correcting you particularly. I was just expressing some frustration that people don't seem to get the difference in the numbers. It's answered over and over.

 

Also, I just wanted to set out exactly what the relationship between the PD number and the SteamVR one - again! You need to square the PD value (and multiply by 100) to reach the SteamVR %age.

 

Then were you correcting me? SteamVR SS and in game PD is different method altogether. Not just difference in multiplier. Run PD of 1.4 and Steam SS of 100%. Then run PD of 1 and Steam SS of 196% (as you claim). Do you get same quality and frame rate? I certainly don't.

 

And I can run Steam SS at 500% which it says is 4500x9000 pixels. And I can run 500% at 30fps. Is my old GTX1080 able to render 3D at that resolution and 30fps?

[Taz1004]

May I ask if there is a place where I can find the users settings with regards to:

1) WMR for Steam - Visuals (High or Very High?)

2) Screenshot of DCS graphics settings

3) Did you enable MotionReprojectionMode editing the default.vrsettings file?

 

If not, would anyone mind sharing?

 

As there are so many variables and one life only I was planning to study those settings along with the hardware of the user (usually in signature) to cut down the variables...

 

For number 2, this is the most complete comparison in one place I found. Maybe there's something better.

 

https://forums.eagle.ru/showthread.php?t=202092

[doraymon]

Thanks everyone, very useful.

Just one thing, could you check question 1) WMR for Steam - Visuals (High or Very High?)

My system seems to be very sensitive to this, if I select Very High i get immediately a heavy drop in frame rate.

While it seems to handle better increases in pixel density. I am running PD @1.2 and I am constantly below 20ms except if I fly very low over trees.

[Hammer1-1]

There are so many things outdated and effectively wrong in that old page, I feel it is of limited use now. It can confuse new users a lot, as options have changed significantly in the 8 months since it was written.

The Shaders mod helps some people a lot, but not everyone. Makes very little difference to me as an example.

 

Still a great place to start. I used it a month ago and went from there.

[doraymon]

check my signature

 

Thanks, the only thing I wasn't able to read is the defaultvr.settings file, any chance to share the text with me?

Sorry to bother you mate

[speed-of-heat]

open it with notepad or notepad++

[imacken]

Then were you correcting me? SteamVR SS and in game PD is different method altogether. Not just difference in multiplier. Run PD of 1.4 and Steam SS of 100%. Then run PD of 1 and Steam SS of 196% (as you claim). Do you get same quality and frame rate? I certainly don't.

 

And I can run Steam SS at 500% which it says is 4500x9000 pixels. And I can run 500% at 30fps. Is my old GTX1080 able to render 3D at that resolution and 30fps?

 

I was talking about the maths involved in rendered pixels. not trying to make a comparison between the two methods. However, it is generally felt that SteamVR SS is a more efficient method of super sampling than PD. YMMV.

[imacken]

Thanks, the only thing I wasn't able to read is the defaultvr.settings file, any chance to share the text with me?

Sorry to bother you mate

 

Check my post here https://forums.eagle.ru/showpost.php?p=4251157&postcount=5176, it may be useful to you.