Qcumber

Nice. Very clever. I think I will steal your idea! How do you manage to integrate VR with the cockpit? Do you use pass-through? Or is it more a case of lining up virtual hands with the switches etc? What is your next project?

Truly amazing. I am jealous. When I retire I am going to build something similar, although likely the BF-109 or Spitfire. I have 3 questions. 1) How long did it take to build? 2) How difficult is it to get into? My simpit is very basic but it is a challenge to my hips and knees. 3) Does than fan sound more like the real Corsair engine than the sim?

Yes please!

The only way I can get decent results is with high supersampling. I have started using MSAA x2 with QVFR. Set VD to Godlike with 150% in the slider. Then set the peripheral resolution to 0.33. I get good aircraft sporting and only a slight flickering. Overall it's a better experience for Warbirds.

No. I'll give it a go. Cheers PS I am very grateful for moders but I still would like an official upgrade.

I want the Dream Air too but the extra 1k seems a bit steep. I'm not in a rush so I'll see how things look in Jan 26.

ED can we please* have a cockpit refresh of the K4? It's my favourite warbird but I think it needs some love and attention. Also would it be possible to have a choice of cannon. I understand that some units were fitted with MK 151/20 Motorkanone. A much better choice for dogfighting. Finally wing mounted cannon pods please*. *Bitte, bitte, mit einer Kirsche obendrauf

The Dream Air SE looks good too. Half the price of the Dream Air.

OXRTK still works fine. The turbo mode issue was related to something else.

Sorry for false information. Try NVPI.

You set use preset K with the NVIDIA app.

The only way to optimize your VR experience with DLSS is to make sure you use the most recent version and preset K. Then supersample to about 4800-5000 pixels. If you combine this with QVFR then you should get reasonable results. I find that there is only a slight smearing of aircraft against some backgrounds, low over trees for example. My biggest bug is spotting dots which look smudged compared to no AA or MSAA.

This is a longshot but is your GPU overclocked and are you using DLSS? For the last few updates (going back since DLSS 4 was released) there has been some instability with DLSS and manual overclocks.

You cannot get rid of the ghosting with DLSS in VR. Even with the latest DLSS version and preset K. The advice I gave above regarding the super sampling will minimize this but you will still notice a fuzziness in spotting dots and a slight bleeding effect from moving objects against some backgrounds, particularly low over trees and dense areas. In most situations though it looks very good. But this comes at a massive performance cost unless you use fixed Foveated Rendering with QVFR. Then you can push the centre region high and keep the periphery low. Not everyone can cope with that though. It would be worth giving it a go if you have not already tried. You have basically the same PC specs as me so you should get some decent results. I have been trying out using MSAA again. It's not as good for an overall smooth image but is much sharper when it comes to spotting and moving objects. As with DLSS you need to super sample in order to get the best results. The ideal world would be to be able to use QVFR with DLSS applied to the periphery and MSAA applied to the centre. @sleighzy I am not sure that this is even possible!? Could someone potentially work such "magic"?

In VR you need to push the resolution high to remove ghosting. I don't know what headset you use, or PC specs, but try increasing the resolution to about 4800x4800 pixels. I use Quad Views Foveated Rendering to increase the Foveated resolution to this level and it removes most of the ghosting. Spotting dots are still not great though.

Why do you always do these patches when I am about to go away. I mean, come on, can you not at least check my diary first!

Thanks. I've had a look at reviews and the Charlie is very heavy. About 1kg heavier than the TPR. It is also quite big (45cm x 45cm) and has carpet grips so good for keeping in place without a frame. As it's from Amazon I might try the Charlie and if it's not for me get the TPR instead.

Does anyone have any experience with Honeycomb Charlie rudder pedals? I can get them from Amazon for £250. I can also get Thrustmaster TPR for £430. Unfortunately I don't have easy access to Virpil or Winwing so these are not options for me. I know what I want to get but is the extra £180 worth it?

I've tried it and unfortunately it's not for me. I am sending it back. I am not saying it's bad headset and it will be a good upgrade for many users. However, I think the Quest Pro is a better option IF you can get one for a decent price. I paid £300 for my mine. I was lucky to get such a good deal for a mint headset that had hardly been used. There are still some bargains.

I ran another test using your settings. The QP resolution is lower at these settings but still looks better than the PSVR2. The cockpit is sharper than I managed but is still a bit fuzzy compared to the QP. It is similar to what I saw with the Pico 4. I tried a few runs over Syria, and the oversaturation is less noticeable compared to a green map like Normandy. BTW thanks for these settings. I like them. I have tried lots over the years but did not have much success with the edges. 0.33x0.33 with 0.33 edges works a treat. QVFR config smoothen_focus_view_edges=0.33 sharpen_focus_view=0.33 turbo_mode=0 horizontal_focus_section=0.33 vertical_focus_section=0.33 peripheral_multiplier=0.33 focus_multiplier=1 debug_eye_gaze=0 debug_focus_view=0 Here is an excerpt of the log for the PSVR2 set at 150 in Steam VR 2025-09-14 16:20:06 +0100: Found option 'smoothen_focus_view_edges=0.33' 2025-09-14 16:20:06 +0100: Found option 'sharpen_focus_view=0.33' 2025-09-14 16:20:06 +0100: Found option 'turbo_mode=0' 2025-09-14 16:20:06 +0100: Found option 'horizontal_focus_section=0.33' 2025-09-14 16:20:06 +0100: Found option 'vertical_focus_section=0.33' 2025-09-14 16:20:06 +0100: Found option 'peripheral_multiplier=0.33' 2025-09-14 16:20:06 +0100: Found option 'focus_multiplier=1' 2025-09-14 16:20:06 +0100: Found option 'debug_eye_gaze=0' 2025-09-14 16:20:06 +0100: Found option 'debug_focus_view=0' 2025-09-14 16:20:06 +0100: Eye tracking is supported 2025-09-14 16:20:06 +0100: Recommended peripheral resolution: 1938x1978 (0.330x density) 2025-09-14 16:20:06 +0100: Recommended focus resolution: 1938x1978 (1.000x density) 2025-09-14 16:20:06 +0100: Stereo pixel count was: 70370048 (5872x5992) 2025-09-14 16:20:06 +0100: Quad views pixel count is: 15333456 2025-09-14 16:20:06 +0100: Savings: -78.2% 2025-09-14 16:20:16 +0100: Session is using quad views 2025-09-14 16:20:16 +0100: Edge smoothing: 0.33 2025-09-14 16:20:16 +0100: Sharpening: 0.33 2025-09-14 16:20:16 +0100: Turbo: Disabled Here is an excerpt of the log for the QP set to 150 in VD. 2025-09-14 16:44:56 +0100: Found option 'smoothen_focus_view_edges=0.33' 2025-09-14 16:44:56 +0100: Found option 'sharpen_focus_view=0.7' 2025-09-14 16:44:56 +0100: Found option 'turbo_mode=0' 2025-09-14 16:44:56 +0100: Found option 'horizontal_focus_section=0.33' 2025-09-14 16:44:56 +0100: Found option 'vertical_focus_section=0.33' 2025-09-14 16:44:56 +0100: Found option 'peripheral_multiplier=0.33' 2025-09-14 16:44:56 +0100: Found option 'focus_multiplier=1' 2025-09-14 16:44:56 +0100: Found option 'debug_eye_gaze=0' 2025-09-14 16:44:56 +0100: Found option 'debug_focus_view=0' 2025-09-14 16:44:56 +0100: Eye tracking is supported 2025-09-14 16:44:56 +0100: Recommended peripheral resolution: 1520x1592 (0.330x density) 2025-09-14 16:44:56 +0100: Recommended focus resolution: 1520x1592 (1.000x density) 2025-09-14 16:44:56 +0100: Stereo pixel count was: 44457984 (4608x4824) 2025-09-14 16:44:56 +0100: Quad views pixel count is: 9679360 2025-09-14 16:44:56 +0100: Savings: -78.2% 2025-09-14 16:45:04 +0100: Session is using quad views 2025-09-14 16:45:04 +0100: Edge smoothing: 0.33 2025-09-14 16:45:04 +0100: Sharpening: 0.33 2025-09-14 16:45:04 +0100: Turbo: Disabled

I have a lot of experience with VR headsets.over the last 6 years. Rift S, Pico 4, Quest Pro. My PC is a new build. Only 3 months old. I keep mods to a minimum. I based my calculations on the base panel resolution of 2000x2000 and wanted to make sure that I was comparing like for like. With the extra x1.5 from QVFR the equivalent resolution of the foveated region is 4800. That should be more than enough. I did try going higher to about 6000 but there was no further improvement. In this case both the PSVR2 and the QP had the same (within a few percent) resolution. Ideally I would have compared the QP with Steam VR but I couldn't get QVFR to run properly. I did try at 120Hz. The reason I chose 90Hz for this test is because it is the highest the QP will go. Maybe I have a bad example of the PSVR2. But to me the colours looked very bright and over-saturated and the chromatic aberrations were bad in the outer edges. Have you used a Quest Pro? I'll run another few runs and save the logs. Then I will do the same using your settings.

I haven't had chance to do the testing I wanted to. The main reason to use XRFrameTools was to measure game performance. I miss-typed in my previous message. I meant frametimes. I played with the two headsets comparing them at the same resolution and same settings in QVFR. As like-for-like as possible. PSVR2: Steam resolution set to 160% for about 3200x3200 pixels. Refresh rate 90Hz. QP: Virtual Desktop set to Godlike (3027x3216) pixels. Refresh rate 90Hz. I tried to use Steam link for this but had issues running QVFR. QVFR for both PSVR2 and QP: foveated region 0.25x0.25 at x1.5, periphery at x0.5. DCS settings were the same for both. I can post them but they don't really matter as I am not including any performance numbers. The PSVR2, even in the sweet spot, is not as sharp as the QP. Not by a long way. Cockpit resolution with the QP is near perfect. With the PSVR2 it always looks slightly blurred. I can read most things in the cockpit without leaning in too much but it never gets as sharp as the QP. Landscapes and distant objects look soft, even mushy in some cases. I have tried AA-off, MSAA, DLAA and DLSS (quality and performance; latest version; preset J and K) and varying degrees of sharpening but it just makes the landscape look very artificial. I just cannot get an image that compares with the QP. The colours on the PSVR2 are very oversaturated. To the extent that there is some colour bleed between objects. There is also a lot of chromatic aberration which can make some objects in the peripheral vision look quite garish. I know I could use Open XR toolkit or Reshade to reduce the saturation but this would not help with the fringing. It is also another XR layer and I prefer to keep these to a minimum. There are other more minor issues that I am not keen on such as the overall build quality, the face gasket (a magnetic one would have been much better), the lack of integrated lens protectors for glasses, the design of the controllers and the Steam VR interface. I realise that the display port connection should give a better experience but for me the Quest Pro has much better image quality. I wish they had released the QP with a Display Port. I think I will send it back and save for a Pimax Dream Air SE.

I am going to take another look today. I'll try to summarise my set up and findings here with more structure. If I get chance I will measure the game latencies using Fred Emmott's XRFrameTools comparing the QP with the PSVR2. Then there will be a more objective comparison if anyone else is considering switching from the QP to the PSVR2.

Good point