DeltaMike

Don't use the stick to keep the velocity vector on the threshold. Use the throttle to do that. More throttle moves the VV further down the runway. Less throttle moves it closer. While you're at it, you might try using the rudder to regulate alignment. Leave the stick alone. Try flying the pattern. If you have yourself all set up and trimmed out in level flight on the downwind, the thing will pretty much land itself.

To isolate the problem I'd suggest mapping the brakes to a button and see how they work. I haven't tried taxiing with toe brakes (and can't now because my VKB pedals don't have em) but "weak brakes" is a common complaint related to pilot error. Come in low and fast and it's easy to run off the runway. I would suggest first learning how to do a Navy landing (who needs brakes, right?) and then try to figure out how to grease her in if you want. Which is harder than it sounds, for various reasons. One way to experience a proper Hornet landing is to use ATC. So, about 8 miles out, slow down to <250kts, full flaps and drop the gear. Wrestle her into submission. Trim her out so the AoA indicator is bracketed (make sure your autopilot is off!) Then press ATC. Give it a minute to adjust. Then, just point your velocity vector at your planned touchdown spot, which should be about three degrees below the horizon (use your HUD ladder to estimate). Take note of your approach speed. It should be closer to 140kts than 180kts, which I would bet money is your problem. That should give you the visual and auditory indicators of what it's like to do an "on-speed AoA" landing. Next, let's try it without using ATC. Eight miles out, drop gear and flaps and trim her out so the AoA indicator is bracketed. Now, leave the stick alone. The flight computer will regulate your speed for you. You're gonna get yourself on the glide path using your throttle. If you're low, goose the throttle a little. If you're high, ease it off a little. You see, when you're flying at a high AoA, your controls are sort of reversed. You regulate speed with the stick -- nose down to go faster, nose up to slow down (in the Hornet, the flight computer handles that for you, to help you maintain a constant speed). You regulate altitude with your throttle. You can actually land any plane like that. They can install an AoA indicator in your Cessna if you want. Only problem is, it doesn't exactly make for a smooth landing. Which is fine. Hornet likes it rough.

You can rename one of them and it'll work If you're using a WMR headset like the Reverb, most people are migrating over to OpenXR. Same functionality, less overhead, fewer moving parts.

Pretty awesome if I do say so myself. I did a head-to-head a couple of weeks ago when I first tried the OpenXR mod, I ran it in stable first (so I wouldn't screw up my OB install, which is where I am most of the time) and then plugged it in to OB. They were exactly the same for me

Locking Chill at 45fps is not smooth. It's OK in the cockpit as long as you keep your head still; in other words you can rotate your head, but can't pan or strafe the camera. You should get a stroboscopic effect along the 3-9 line. That's why I don't do that. Don't recall if I get dropped frames every 8-10 seconds, I'll look next time I'm up. So, every 8-10 seconds your GPU is just standing there with nothing to do? So the CPU missed a deadline in there. Wonder if it's just a VR optimization issue. Sounds like your system is running beautifully. Take it from somebody who fiddles with this stuff for entertainment. Sometimes you just gotta learn to ignore stuff.

Lots of exciting news in the GPU world these days. The 4000 series from NVIDIA is on the way and Intel is throwing its hat into the ring. Question is, will it be enough? What will it take? Let's start with the premise that the Varjo XR-3 is where we want to go with this. The central resolution of your iPhone display, and the field of view of your Pimax. Each accomplished with its own display; so in other words, using a combination between optics and eye tracking, there's a tiny "retina" display that follows your eyeballs around. Your peripheral vision is treated to a 2880x2720 display, a bit higher resolution than your G2's central "sweet spot." Let's look at the total number of pixels displayed (see chart below). Keeping in mind that AMD's flagship GPU can just barely keep up with the G2, what kind of performance improvement must we demand of the next gen GPU's? If all you really want is a Pimax 8K, you need a 77% improvement over the AMD GPU. Probably not a bridge too far for NVIDIA; I imagine the 3090 is close and perhaps the 4000 series GPU's will get you there. For the Varjo? We need the new generation of GPU's to be almost two and a half times as fast and powerful as today's flagship GPU's. Odds of that happening? Those of us old enough to remember SLI, or who have built multi-GPU rigs for mining, think we know the answer. We don't ask much of our GPU's. They are just big dumb idiots to us (extremely fast idiots, fortunately -- especially to those of us who keep winding up with AMD GPU's). Why look for a Sumo wrestler when two linebackers will do, right? Interesting article about where NVIDIA (and AMD as well, I believe) are going with this. Note how the "broadcasting" technology might appeal to applications like DCS that are wrangling a lot of objects at once. I keep meaning to download a copy of XPlane to see what, if anything Vulcan has done for it. Not much, I hear. The "SLI of the future" sounds a lot more interesting to me. I'll believe the "it's just a few lines of code" thing when I see the results, but still. Exciting stuff imo.

If you like to design your own missions, Liberation is a God-send. It automates a great deal of drudgery and does an amazing job of it. Having done that, I see what a challenge it is to get the AI to do what you want it to do. You kind of have to take what it's willing to give you and under what circumstances. Take CAS for example Cobras do a fantastic job. Take names, kick butt. Harriers seem intent on getting themselves shot down. Your wingmen have an amazing ability to find ground units. Take a section up, cut em loose. You can work with the JTAC or provide high cover. Just reel em in before they all get killed. Otherwise it's a long lonely ride back to the carrier. First couple of times playing Liberation it was like the Keystone Kops out there. It was hilarious (and a sense of humor goes a long way here). Now that I'm kind of figuring out the rules of the game, I've grown to really appreciate the dynamic campaign. I spent hours yesterday not flying, just playing the game itself as a turn based strategy game. They did a really good job. They should charge more. Part of it is, I put hours into developing a Straight of Hormuz campaign. It was designed to show that those who don't recall the history of the Dardanelles campaign are doomed to repeat it. As it turns out, that's the exact scenario of the PG "Scenic Route" scenario in Liberation, to the point where I strongly suspect that was the designer's intent precisely. The result (in my hands at least) is what I expected so far at least. And... dang. I put hours into that project. Now all I have to do is press a couple of buttons. I don't take DCS as seriously as I used to. If I did, I'd still be doing MP.

I've been delighted with my Orion F18 throttle, that is some fine engineering. I really like the finger lifts and the new switch layout; well done. I would definitely consider upgrading the base depending on the price.

There's a certain rigmarole involved in updates. I'm not saying this is a best practice or anything, I've only been doing DCS about four years or so. That said, here's what I do: Read release notes and get all excited. Especially for the most obscure aircraft in my collection. Which is most of them. Double check that I have the current .ini files in my mod manager library, and then un-check all my mods Download and install update through Skatezilla Clean the install, delete whatever random trash has accumulated. Check for updates for my mods and take care of that Re-install mods Delete my shader files Pull up Adrenaline and make sure my GPU is still tuned. (Half the time it isn't, ymmv) Get cup of coffee Drink coffee while DCS is starting up. Expect a bunch of stuttering and fiddle-farting around the first couple of times you fly while DCS is shaking off the willies. Log on the forum and praise the developers for their hard work. The latter is optional of course. As is the cup of coffee, just realize, this goes a lot smoother if your team-mates aren't sitting there idling on the ramp while you're doing all this stuff Set your minimum GPU clock 100-200 below your maximum. I found this helpful

PSU: Use this calculator. Don't go cheap on your PSU. Way I look at it, you have two options. Get a "cheap" PSU and buy about twice the amount of wattage you need, or buy a really good PSU and cut things a little closer. How well it's certified (is it bronze or platinum certified?) gives you some idea of quality but be sure to read reviews on your finalists. CPU: You want maximum single-core performance for DCS. Buying more cores doesn't help you any. AMD ships their CPU's optimized for single core performance and they work fine right out of the box with DCS. Seems like a lot of people are overclocking their Intel CPU's to get the most out of DCS. If you're gonna go Intel, read up on that and regardless, make sure you understand what it'll take to keep the CPU cool. Hint: not the stock cooler. Even for AMD. (I'm using a Noctua tower cooler, not particularly expensive and it's fine) GPU Let's start with the proposition that a 3080 will run a Reverb G2 at native resolution -- about 9.4m pixels -- but not with high graphics settings. You get your choice of shadows, clouds, or MSAA basically. But only one of those. A Quest 2 (7m pixels) will be easier to drive. A Pimax 8K (16.6m pixels) will be harder, and perhaps impossible to drive. Nothing is gonna run that thing at native resolution. I mean you can run it at lower resolution if you want. A Pimax at Quest 2 resolution? Not a horrible choice I guess -- the peripheral vision would be cool -- but is the cost expenditure worth it? DCS is a game of visual detail. You need to be able to see your instruments and displays, and see the bad guy. That's why the G2 is so popular. For all its faults, it's the least expensive way to get the best central detail. Theoretically the Pimax 8K will give you the same central resolution as the G2 and on paper it would be a great choice for DCS. If you can figure out how to run it without lighting your computer on fire. For the money, I'd do Varjo all day long. But then I don't mind fiddling with stuff, ymmv. (It's still an early-adapter product.) You wanna bend DCS to your will? Consider Quest2 or G2 with GPU to match (say 3070(ti) and 3080(ti) respectively) and take the win, for now (it'll be obsolete in two years anyway). With the money you save, invest in some fine controllers that you'll still be using ten years from now. Winwing or Virpil HOTAS, MFG pedals, a button box or two. Some nice monstertech chair mounts (every fighter jet needs a swivel seat and cupholders). A decent gaming keyboard and mouse with macro buttons. You can then fight them on the beaches and landing-grounds, fight them in the fields and streets, and never surrender.

Don't knock it. When my gas heater went out, I heated by RV with a 4xR9-290X mining rig one November in Albuquerque and made money doing it.

^and they say DCS won't run on more than one core I have a Beelink SER3 on the way. Will post numbers once I have it.

https://mbucchia.github.io/OpenXR-Toolkit/

So you're at 88%. Once you get to that level it might more efficient to simply run at full resolution (the scaling software does carry some overhead)

You shouldn't have to downscale. I'd try full resolution without MSAA and see how things look and run. Maybe go the other way, try a little supersampling instead of MSAA It's not really MSAA, it's probably just supersampling anyway Scaling introduces jaggies and shimmering. Kind of a death spiral. DCS doesn't really have a decent anti-aliasing solution Keep it simple

Here is a very good and understandable explanation of barrel distortion for those of us who are visual learners. Net bottom line is, you lose data as the compositor prepares the image for display. Sadly you just have to render more pixels than you can display. If you don't, it's not distortion you have to worry about. You wind up with a smaller sweetspot, a fuzzier sweetspot or both. I don't have anything against small fuzzy sweet spots in general, but I don't like em in VR at all. I do think WMR4SVR introduces some artifacts, shimmering among them. Whether OXR improves performance, and to what extent, seems to depend on your system. We are all scratching our heads trying to figure it out. My advice is, don't get too hung up on FPS. Not a great metric for VR to begin with, and some of the action might be happening beyond where your FPS counters can reach. @Rosebud47 what GPU do you have? Here's one organized way to troubleshoot your stutters. You'll need two utilities: OpenXR Developer Toolkit, which you get from the Microsoft Store (sounds like you already have that) OpenXR Toolkit Companion App, which you download from github I'd start with the Companion App, bring up the FPS counter, select the detailed output, and take her for a spin on an empty map (Caucasus free flight works for most of us). Nobody really cares what your frame rates are (welcome to VR) but we are interested in your GPU times. What are you averaging? If they are averaging much more than 20ms, look at your settings. Are you running ultra clouds, high shadows and MSAAx4? You're dreaming my friend. Read some VR optimization guides and come back down to earth. If you still can't nail 20ms after that (likely if you have a 3070/6700XT or less) you probably need to undersample. Two ways to do this. First, make sure your DCS Pixel Density is 1.0, don't mess with that. Then you can either: Adjust the slider in the OpenXR Developer Toolkit (from the MIcrosoft Store). This works just like the SVR slider. Remember fholgers VR Performance KIt? The one that lets you use FSR or NIS? That functionality is built into the OpenXR Tookit (from github). You just have to navigate through those infernal menus to find it. This scales differently from the SVR slider (and acts a lot like the pixel density slider). I'd try 77% which is "ultra" quality and even though the performance impact is like running Steam at 60% (or PD of 0.77), it looks a heck of a lot better. (NOTE, it seems to work for me but I don't go there very often; others have reported this was still a work in progress. If you don't like it, try again with the next round of updates) If your GPU times are averaging less than 20ms, you're golden. You have several options available to you to smooth things out. What you're supposed to do is, turn on motion reprojection ("always on") in the OpenXR Developer Toolkit (from the Microsoft Store). It definitely works but it causes a lot of tearing. You may hate it, or you may love it. Try it and see. The other option is to dial down either your settings or resolution shooting for a GPU time of somewhere around 16ms. Personally I think that looks OK without motion reprojection. VR is no hobby for a perfectionist. ALL of these strategies introduce their own artifacts. Welcome to VR! You just gotta figure out which one drives you the least bonkers. My advice is, don't forget to get out there and play some. Stuff that seems like a big hairy deal when you're tuning (or sight-seeing, which is why they made MSFS) may not matter much when the fur starts flying (which is why they made DCS).

I've read reports that Win11 causes problems with this mod. But moreover it doesn't seem like many people with 3090 love it all that much. Perhaps a testament to your VRAM bandwidth, which is pretty awesome when you think about it.

You don't have to uninstall SVR. Try Skatezilla's "updater." It's more than an updater, it's a DCS dashboard/launchpad. Among other things it makes it really easy to update and clean your install which might help. Good news is, if you follow the instructions in this thread it tends to work. Bad news is, if you don't.... Word.

Based on the images at the beginning of the thread that doesn't look like a performance issue. Are those images from the mirror? Yikes.

If you're using the stand-alone version try Skatezilla's app. I fire up WMR, make sure the headset is working, then peek out to launch DCS. If you're using the Steam version of DCS, I'm not even sure what to tell ya. I've heard of people launching from a desktop shortcut. Wonder if installing OXR globally (using the installer .exe) might work.

So if we go back to the original question, to summarize the relevant issues, and to include comments from other threads Ability to check six. Easier to do that with track-ir. Although I'll shout out to Monstertech for their chair mounts. Every modern fighter jet should have a swivel seat for this precise reason. There's some technology available for the VR folks (VRNeckSafer) but it's still a work in progress and makes some people puke. Don't get me wrong, I love my VR but this is a real disadvantage. Scuba mask isn't good for checking six, in the air or the water. Resolution, recognizing that sometimes less is more. A competitive player using track-ir might find an advantage in using a 1080 monitor, as opposed to 4K to make spotting easier. I imagine same principle might apply to VR. Easier to "padlock" using VR? Maybe? SA? Both track-ir and VR are good SA tools, is VR any better? Definitely for stuff like formation flying. Question is, how good is track-IR at building a world model? So you split, bandit decides to chase you. You break left and dive and tell your wingman to go get him. Question: after you level off, can you point to your wingman? Where is bullseye? Can you point to that? Is it any easier to do in VR than track-IR? Fiddliness and reliability? Maybe? Murphy's law is no joke How does cost factor in? Even if you think VR is superior (which is not at all clear to me) is it worth the expense? For competition? So many of us are really good at one or the other, helpful to talk to people who are familiar with all three (track-ir, VR, and IRL) for example. Not every comment is relevant to competition, although Tactical makes a good point about pets. Dogs are bad enough; cats walk on the keyboard and can figure out how to turn your computer off. And then there's "the claw." So, my conclusion is, it depends. Do you have a cat, or not?

Yeah motion vector shouldn't have anything to do with head rotation. Camera rotation is handled differently from camera panning. Motion vector, or in Oculus terminology "asynchronous space warp" reduces judder along the 3-9 line, or when doing a barrel roll. Asynchronous time warp uses sensors in the HMD to determine rate of rotation. To my understanding that's built in and always on. How do you go from looking at the left DDI to the right? If you rotate your head, you're using asynchronous time warp and it should work pretty well; if not, use the toolkit to adjust. If you pan or strafe our head from left to right, for example if you have to lean in to read the displays, one would hope that asynchronous space warp would take care of the judder. Although that's a tall order. The closer the object is to the camera, the more of a stroboscopic effect you're going to get, and even 90fps may not be enough

It just bugs me that a used 1080ti is going for about the same price as a new 3060ti or 6700XT. It was a great GPU, amazing really when you look at the VRAM and TMU count compared with its peers. Was able to keep up with the 2000 series during the Rift S era. But that was then. I've spent a lot more time looking at G2 than Q2 but that said, I haven't seen any recent reports on tuning the 1080ti for the G2. 2000 series GPU's are running into problems. On the other hand, @edmuss has shown that the 3070 works just fine with the G2 if you use the right tools. Net of everything the 3070 strikes me as a great value if you can get your hands on one without paying scalper prices. Quest2 should be easier to drive than the G2. I wouldn't hesitate to do Quest2 + 6700XT (which was my initial plan until mission creep set in) and I'd bet my own money a 3060ti could drive it. 3070 should be able to drive Quest 2 with one hand tied behind its back. Oh and wait'll you see what your flat screen looks like with a 3000 series GPU. Those things are very impressive imo. I feel you are on the right track. Get the Quest 2 and have some fun. Get wifey and the kids involved. There are all sorts of games, creativity apps, educational apps, immersive movies and such you can play right off the headset without a desktop computer. Meanwhile enter whatever lottery you can, get on whatever wait list you must, and lie in wait for a reasonably priced 3070 (or the equivalent) to come your way. That's what I would do. RE Q2 vs G2. I see the Q2 as the best of the consumer headsets; the G2, as the least expensive of the professional headsets. I'll hand it to Oculus, their ecosystem is fantastic. I dunno about the whole Meta thing, not sure I'd let my kids anywhere near the Metaverse. But otherwise they would have a blast with it and possibly even get some exercise heaven forbid. G2 is very cranky and dorkish and it still feels and acts like a prototype. "Dad! The Reverb is broke again! DAD!!!" Which is what I say to edmuss about once a week. (A lot of us do that. Thank goodness he's very patient.)

Sort of. Easier to just use FOV I think

Forest visibility should match with vis range, you should be able to dial that down to 50% or so. and should probably do the same with detail factors. Have you tried cranking up preload radius? Beyond that, welcome to DCS. What are your specs? Memory counts, 32gb makes a big difference in MP. If you have Intel CPU, might need to overclock (if not upgrade)