Oculus Rift with DCS World Discussion

[cichlidfan]

I actually changed the way I do this now,...

 

The point is, since DCS autodetects a VR headset, it should be automatic.

[Rocky49]

Reading this, I almost feel like we're using two completely different VR products. I'm wearing contacts that correct my vision back to nearly 20/20; I ran the Oculus Debug Tool, which said pixel density was at 0.0; I changed that to 1.5 and 2.0 and it didn't seem to make any difference whatsoever to either Home or DCS. I ran the Performance Overlay tool and it seems I'm averaging 22.5 FPS in DCS, pretty poor, but I wouldn't think image quality would depend upon framerate.

 

And nothing I'm seeing in DCS would even remotely qualify as "crisp display panels", so I really don't understand what's going on. It feels more like sitting close to a a large 720p TV. Good for you for apparently getting better results than I, though!

 

Well there is a lot to love with the cv1 if you have the horsepower to use it. I suspect your 2500 chip is just not cutting it. I f you look at my specs I have pretty new gear, only 16 months old. VR is tasking for state of the art equipment so if your gear is a few generations old, you can have issues. I can run DCS at 90 FPS with some tweaking but still can read most of the gauges with just a glance, If I need to really have a close look, I just use zoom function. BTW I usually wear glasses to use my monitor, but not necessary with my rift

[SkateZilla]

The point is, since DCS autodetects a VR headset, it should be automatic.

 

It Detects it if you tell it too.

 

Either way, DCS Still uses the same options.lua whether you enable or disable VR,

 

The only way to use 2 separate profiles that you can edit and save without having to re-edit every time you want to switch is to use 2 separate options.lua files.

 

and teh ability to simply add a argument to the shortcut to enable a separate saved games folder is easy, and enables you to have a separate folder for main settings, input settings, screen shots, logs, tracks etc etc.

[cichlidfan]

It Detects it if you tell it too.

 

Actually, it detects it by default, unless you tell it not to.

[dogbite]

Well there is a lot to love with the cv1 if you have the horsepower to use it. I suspect your 2500 chip is just not cutting it. I f you look at my specs I have pretty new gear, only 16 months old. VR is tasking for state of the art equipment so if your gear is a few generations old, you can have issues. I can run DCS at 90 FPS with some tweaking but still can read most of the gauges with just a glance, If I need to really have a close look, I just use zoom function. BTW I usually wear glasses to use my monitor, but not necessary with my rift

 

I admit GregP's 2500 is less than ideal. my 2550k runs DCS ok but I can't run as high settings as you I am sure, however his not seeing any difference from the pixel density adjustments is not related to cpu power. Even if he had framerate issues he should still be able to get sharper instruments.

[GregP]

I admit GregP's 2500 is less than ideal. my 2550k runs DCS ok but I can't run as high settings as you I am sure, however his not seeing any difference from the pixel density adjustments is not related to cpu power. Even if he had framerate issues he should still be able to get sharper instruments.

 

Although, I have my 2500K running at 4.8 GHz, which should without question be faster than say, a stock i5-4590 running at 3.3 GHz. I'm running the latest 368.69 on a single 980Ti on Windows 10 (dogbite, aren't most people running Win10 these days?).

 

Anyway, I continue to run more tests. I've at least confirmed that in Oculus Home, I can get the Pixel Density changes via the Debug Tool to take hold, as verified by the Layer overlay that shows it. But I struggle to actually see any improvement in image quality in Home.

 

Similarly, in DCS, using the in-game Pixel Density setting, I can go all the way up to 2.5 and not see any noticeable image improvement at all. My VRAM usage in MSI Afterburner goes up a bit, which makes me think the PD is taking hold, but the actual image is no sharper.

 

As an example, sitting on the runway in the A-10C (where I get about 60 FPS), I cannot read the word 'temp' on the engine temp gauges, nor can I read the coordinates of the TDC on the left MFCD that appear above the bottom row button labels. And the UFC labels under the buttons, for example STEER, are not even clear, despite them being very close to my virtual head position. The side console button and switch labels are blurry mess, and really don't improve no matter how far I lean into them.

 

I've also got all kinds of god rays on any text that contrasts sharply with whatever is behind it; the CDU on the right MFCD, for example, has terrible god rays as I move my viewpoint slightly left and right of its center.

 

At this point I suspect it's some combination of 1) too high expectations, 2) my own eyesight, 3) possibly something just wrong with my Rift in particular. So many people across different forums seem to think that playing with PD radically improves image quality, and I'm just not seeing it at all. Really frustrating. At least I have also found a small few who aren't seeing any difference.

[dogbite]

May be your eyes. may be your Rift. I have never read of anyone not seeing the diff before. I just updated DCS 1.5 now that the patch was released today and not only is it's pixel density @ 1.5 better than the debug, it is also less of a hit on performance than the tool and appears to be even sharper. It may be only one mans opinion but this is excellent and definitely noticeable.

[whitav8]

@GregP

 

You say:

"As an example, sitting on the runway in the A-10C (where I get about 60 FPS), I cannot read the word 'temp' on the engine temp gauges, nor can I read the coordinates of the TDC on the left MFCD that appear above the bottom row button labels. And the UFC labels under the buttons, for example STEER, are not even clear, despite them being very close to my virtual head position. The side console button and switch labels are blurry mess, and really don't improve no matter how far I lean into them.

 

Great example - I can't read them either. Many of the textures and fonts are not legible unless you really lean in - those are some of them. They need to be redesigned for VR. Bigger textures and fonts like for the main instruments (ADI/Altimeter/VertVel,etc..)should seem fairly clear but the Screen Door Effect as well as resolution and texture filtering/antialiasing all play a real part in being able to read elements easily.

Edited July 9, 2016 by whitav8

[Teej]

Reading this, I almost feel like we're using two completely different VR products.

 

FWIW...you list 2 980Ti and 22.5 fps.

 

I'm running a single 980Ti and with no pixel density adjustment I spend most of the time locked in at 90 fps. turning it up close to 2, I get drops down to 60fps or so.

 

Neither DCS nor the Rift are very good about SLI support AFAIK...not sure if that's throwing you off.

[Johnny Dioxin]

I've seen posts on reddit by people who tried to use contacts due to the situation with wearing glasses in the Rift and having a lot of clarity issues. Maybe this is your problem.

[hansangb]

Or a lot about how people will defend a questionable investment, after they have made it. Though, I must admit, I am very happy with the significant investment that I made in Facebook after they bought Oculus.

 

cichlidfan, to each his own. Some with projectors will say monitor people are stupid for not adopting bigger screens. 4K crowd will say projector people are stupid for sacrificing clarity over size. Pit builders will say physical realism trumps all and people with just HOTAS are stupid. I think VR immersion trumps stupid eye candy every time.

 

And everyone would be right. It's not about defending questionable investment. After all, someone who's struggling to make ends meet will think people who plunk down $2K for a game computer are making questionable investment. So again, to each his own.

[Kaiza]

Although, I have my 2500K running at 4.8 GHz, which should without question be faster than say, a stock i5-4590 running at 3.3 GHz. I'm running the latest 368.69 on a single 980Ti on Windows 10 (dogbite, aren't most people running Win10 these days?).

 

Anyway, I continue to run more tests. I've at least confirmed that in Oculus Home, I can get the Pixel Density changes via the Debug Tool to take hold, as verified by the Layer overlay that shows it. But I struggle to actually see any improvement in image quality in Home.

 

Similarly, in DCS, using the in-game Pixel Density setting, I can go all the way up to 2.5 and not see any noticeable image improvement at all. My VRAM usage in MSI Afterburner goes up a bit, which makes me think the PD is taking hold, but the actual image is no sharper.

 

As an example, sitting on the runway in the A-10C (where I get about 60 FPS), I cannot read the word 'temp' on the engine temp gauges, nor can I read the coordinates of the TDC on the left MFCD that appear above the bottom row button labels. And the UFC labels under the buttons, for example STEER, are not even clear, despite them being very close to my virtual head position. The side console button and switch labels are blurry mess, and really don't improve no matter how far I lean into them.

 

I've also got all kinds of god rays on any text that contrasts sharply with whatever is behind it; the CDU on the right MFCD, for example, has terrible god rays as I move my viewpoint slightly left and right of its center.

 

At this point I suspect it's some combination of 1) too high expectations, 2) my own eyesight, 3) possibly something just wrong with my Rift in particular. So many people across different forums seem to think that playing with PD radically improves image quality, and I'm just not seeing it at all. Really frustrating. At least I have also found a small few who aren't seeing any difference.

 

What do you have textures set to? I think this makes a big difference for gauge readability.

[MBot]

VR leads to interesting new insights on cockpit ergonomics. I have a side-stick arrangement with my joystick on leg level, very similar to the F-16 layout. What I noticed is that this position severely restricts the ability to turn the upper body to the left and check six over the left shoulder. A central mount apparently allows for much more freedom to maneuver in your cockpit, including switching hand when looking aft-right. I seriously need a ground mounted, long-throw, central stick arrangement :)

[GregP]

Appreciate all the comments, guys. Taking them one by one:

 

Great example - I can't read them either. Many of the textures and fonts are not legible unless you really lean in - those are some of them. They need to be redesigned for VR. Bigger textures and fonts like for the main instruments (ADI/Altimeter/VertVel,etc..)should seem fairly clear but the Screen Door Effect as well as resolution and texture filtering/antialiasing all play a real part in being able to read elements easily.

 

OK, that's validating to hear then, that what I'm seeing is apparently not radically different from everyone else. I think I truly have been spoiled by 4K -- I've been running it on a 58" monitor for 2+ years now, sitting up close, so everything in the cockpit has been easily readable, almost real-life size.

 

FWIW...you list 2 980Ti and 22.5 fps.

 

I'm running a single 980Ti and with no pixel density adjustment I spend most of the time locked in at 90 fps. turning it up close to 2, I get drops down to 60fps or so.

 

Neither DCS nor the Rift are very good about SLI support AFAIK...not sure if that's throwing you off.

 

I should've been clearer: as soon as I hooked up the Rift, I disabled SLI in NVCP, so that shouldn't be causing an issue here. My earlier example of 22.5 FPS was also probably a poor one to use -- that was in the M2000C, with my pre-Rift graphical settings (e.g. Shadows ultra, etc.). In the later A-10C tests, I dropped Shadows to medium, etc. and just sitting on a runway in the Caucasus, I was seeing mostly in the 60 FPS range. In 2D, on 1.5.4OB, at 4K with most graphics options maxed, I can generally maintain 40-90 FPS.

 

I've seen posts on reddit by people who tried to use contacts due to the situation with wearing glasses in the Rift and having a lot of clarity issues. Maybe this is your problem.

 

This could definitely be it, because I have astigmatism in both eyes, which makes wearing contacts annoying -- they have to be in exactly the correct orientation in order to work. I imagine the perception of god rays gets worse with astigmatism too. And overall, even outside of the Rift, I see less well with contacts than I do with glasses.

 

What do you have textures set to? I think this makes a big difference for gauge readability.

 

High, so I don't think that's causing this.

 

I'm wondering at this point if DCS really is using the Pixel Density that I choose in the VR menu, because I really see no perceptible improvement in image quality when I change it. During the tests I did last night, I think I may have still had the Oculus Debug Tool open -- I wonder if its PD setting might have overridden that of DCS?

Edited July 9, 2016 by GregP

[Flamin_Squirrel]

During the tests I did last night, I think I may have still had the Oculus Debug Tool open -- I wonder if its PD setting might have overridden that of DCS?

 

Easy way to check. Open the tool (don't change the PD setting) and select the layer overlay. It should say 1.00. Run DCS, and it should change to 2.00 (or whatever you've set in DCS) once DCS loads.

[GregP]

Easy way to check. Open the tool (don't change the PD setting) and select the layer overlay. It should say 1.00. Run DCS, and it should change to 2.00 (or whatever you've set in DCS) once DCS loads.

 

Right, I was wondering if that would work, i.e. if the Debug Tool would notice when an outside app changed PD.

 

But, a few seconds after my previous post, I recalled that the previous day, I'd been changing PD in DCS before I even knew about the Debug Tool, and I was seeing no difference. So it's unlikely that the Debug Tool was overriding DCS's PD. Still though, worth verifying if DCS is applying its own PD correctly.

[Flamin_Squirrel]

Yes, I was suggesting the PD tool use as above to confirm that the PD option in DCS is actually working.

 

There does seem to be a difference in opinion, for whatever reason, on the effectiveness on PD; it's not just you.

[hansangb]

Right, I was wondering if that would work, i.e. if the Debug Tool would notice when an outside app changed PD.

 

But, a few seconds after my previous post, I recalled that the previous day, I'd been changing PD in DCS before I even knew about the Debug Tool, and I was seeing no difference. So it's unlikely that the Debug Tool was overriding DCS's PD. Still though, worth verifying if DCS is applying its own PD correctly.

 

Maybe look into these guys? I'm going to order a set.

 

https://vr-lens-lab.com/product/lens-inserts/

[GregP]

Yes, I was suggesting the PD tool use as above to confirm that the PD option in DCS is actually working.

 

There does seem to be a difference in opinion, for whatever reason, on the effectiveness on PD; it's not just you.

 

Right, that was a great idea, and I just tested it -- yes, the DT does confirm that DCS is changing PD. And this time around, yes, I could see a steady improvement as I moved from 1.0 to 1.5 to 2.0 to 2.5, especially in the menus.

 

BUT, unfortunately for me, even at 2.5, I just wouldn't consider this a viable alternative to 4K. The immersion is absolutely stunning, I completely agree with that. But the tradeoff is just too great. VR on the Rift right now reminds of watching a video of some domed military flight sim from the 90's, or Falcon 4.0's original "3D" pit -- sure the immersion is great, but wow, the lack of detail just kills it for me.

 

Maybe look into these guys? I'm going to order a set.

 

https://vr-lens-lab.com/product/lens-inserts/

 

Yeah, I'd considered that, but even without the god rays etc., based on above I'm pretty it wouldn't make enough of a difference, at least to me. Especially as I actually did the above tests with my glasses on....

 

Oh well. Seems the bottom line for me is that VR simply isn't far enough along yet. Maybe in a generation or two, which hopefully will only be a few years.

[Flamin_Squirrel]

Understandable, it depends on your priorities. The aircraft you like to fly makes a difference too I think. Warbirds and some helicopters in particular that benefit the most from the increased depth perception also tend to have simple cockpits so work well.

 

If you're into planes like the A-10, I can see why you'd consider it not worth using. It's already easy to fly, but has a complicated cockpit with lots of small text. Losing the higher resolution to pickup ground targets is also a problem; not ideal.

 

Once the F/A-18C comes out, I doubt I'll fly much else; I'm concerned it'll be tricky with VR however. Learning the aircraft so I know where everything is without having to stare at the text will be key.

 

Roll on CV2...

Edited July 9, 2016 by Flamin_Squirrel

[cichlidfan]

After all, someone who's struggling to make ends meet will think people who plunk down $2K for a game computer are making questionable investment. So again, to each his own.

 

Which has absolutely nothing to do with what I said.

[average_pilot]

VR leads to interesting new insights on cockpit ergonomics. I have a side-stick arrangement with my joystick on leg level, very similar to the F-16 layout. What I noticed is that this position severely restricts the ability to turn the upper body to the left and check six over the left shoulder. A central mount apparently allows for much more freedom to maneuver in your cockpit, including switching hand when looking aft-right. I seriously need a ground mounted, long-throw, central stick arrangement :)

I was thinking just that a couple of days ago. I also have a side-stick arrangement and realized the difference between looking over your right or your left shoulder if you keep the hand on the stick. A problem from real life that passes unnoticed when playing on a monitor.

[Purg]

Yes, I was suggesting the PD tool use as above to confirm that the PD option in DCS is actually working.

 

Another way is to knock it down to 0.5 in game. You'll notice very quickly if the slider is doing anything.

[hansangb]

Which has absolutely nothing to do with what I said.

 

Sure it does. Average_Pilot mentioned that he was shocked more people didn't complain about clarity etc. Which he credited the power of VR.

 

You then said it's more about people defending questionable investment. Implication is that we are all "drinking the coolaid" and are like sheep. Is that not your point? My point is that to some, *all* of this is questionable investment.

 

I for one am having a ball and am willing to pay the price. Both in terms of $$ and lack of 4K clarity.

[SkateZilla]

I didn't forget about the VR/NonVR Settings QuickGuide Guys, I'm Actually trying to catch up on some other things, will post up a guide soon.

 

Also Debating if I want to Add VR / NonVR Launch Buttons w/ only the arguments to enable/disable to the DCS GUI Utility.

Edited July 10, 2016 by SkateZilla