DeltaMike

^Consider getting a utility that separates out GPU time eg fps4VR. Since "chill" doesn't affect that, locking your frames at 45 (which is possible in adrenaline, in a couple of ways) may prevent dropped frames but it introduces "judder," which is a stroboscopic effect that occurs when you pan the camera. Noticable looking out to the side when you're flying low, or looking at the horizon when you do a barrel roll. So these are your options 1. Lock your frames at 45 and deal with judder. 2. Use motion vector and live with those artifacts 3. Let your fps run whatever it wants and deal with synch issues 4. Run DCS on your secret Los Alamos supercomputer at 90hz and enjoy! All depends on what drives you the least crazy. Game is playable no matter which one you choose

If your number one priority is to be combat effective in DCS, and you have the GPU to drive it (3080 or better), the G2 serves well and is well worth putting up with its technical complexity. Especially if it's on sale, although I'd say it's an outstanding value at list price. If you find yourself saying, yeah but it's not pretty, and I don't want to buy a 3080, wtf is WMR4SVR and will it play Lone Echo? Then, no, that's a terrible price.

Good to know. Bad news always gets the most press huh.

You could watch winwing for sales. They've been discounting their F18 stick ever since their f16 gear came out. I don't know if it's as nice as virpil but it's head and shoulders above the warthog. I paid about as much for the whole stick as I would have for just the virpil base.

Controller conflict? Something similar happened when I left my Xbox controller plugged in

Which is usually a good idea. Up to date bios has been a really good idea lately, especially on the AMD side. Up to date GPU driver is usually a good idea. Windows updates haven't seem to have hurt anything lately but ya never know. Windows 11 has been problematic. When I installed steam vr a couple of months ago even steam said it wasn't compatible with Win11 Worth checking the Startup tab in Task Manager from time to time, and deactivate everything you can live without. Assuming nothing wrong with the system, I'd consider stress testing the GPU (eg Unigine Superposition) -- watch thermals, make sure it's benchmarking where it's supposed to -- and CPU (I use Prime 95). Note, it is possible to get a bad GPU. Sometimes you can prove it by checking VRAM (good article on that here.) Moving on to VR. VR has always been really picky about what USB ports you use for what. Mobo upgrades have helped a lot in that regard, lately. Still. The Reverb in particular is gonna have a USB port it likes, usually one on the back that's wired direct into the mobo. Once you find one it likes, use that same one every time. I still have to unplug it, and plug it back in fairly frequently. And then there's the charlie-foxtrot that is the WMR/WMR4SVR/SVR/DCS ecosystem. I'm not even sure what to tell you about that. There's a dark art to figuring out what to start when, in what order, and how. Skatezilla's "DCS Updater" helps a lot here. It's sort of a "DCS Dashboard" you can access from under the hood if you want to. Among other things, you can turn off Steam Home and let the Updater boot up the runtime and DCS, It's not perfect but it does a better job of that than I do. The "repair" utility is a lifesaver. Even when it works, it doesn't always work. I don't know about you guys but I have to start over at least once every session. Like, just reboot and start all over. Thud's guide is a little hard to follow but it's worth the persistence. Not possible to go wrong if you can follow it. Literally everything in there is important, and was no doubt learned the hard way.

That's the artist's craft. You want to make a technical subject hyper-realistic you gotta include the lingo. Explain only what needs explaining, otherwise it gets really tedious and potentially condescending. Trust the artist, he'll get you where you need to be. Faulkner once said of Hemingway that the man lacked courage; never used a word that might send the reader to the dictionary. Let that not be said of Baltic Dragon

Raven One is just outstanding. I found the book first, the whole time thinking, "somebody should import this into DCS." Great minds think alike eh The voice acting took this to a whole nuther level in terms of professionalism and polish; well done. Did you see RedKite's walkthrough for mission 9? Check out the trailer in his intro. There's a briefing for ya. Powerful stuff. Have you talked to Mover? Matt Jackson's "Crisis in the Desert" series is pretty darn cool -- scratches that Clancy itch especially if you remember "Red Storm Rising" -- and the first book, to a great extent takes place on our PG map. That would be a monster of a project I imagine but dang. You could charge more. People drop a lot of money on this hobby, it won't kill em to pay up for something that makes it worthwhile to strap on that headset.

Which is a really great implementation imo. Knocked it out of the park actually. I only really use vaicom pro for the tanker, the carrier and the kneepad, but it's worth the price of admission just for that. And for Sarge, after we had a little talk about the "wheel socks." Let's just call it "training." He got the message.

We can certainly agree, if there's a difference, there's not much of one... by the same token I'm not noticing much of a performance hit going from say 512 to 1024. I also haven't noticed nearly as much performance hit from the mirrors as I used to. Old farts like me have been making that recommendation for years, and it appears, it's not as relevant as it used to be. So perhaps we should say, far as cockpit display resolution is concerned, 90% of the time you're gonna be looking at MFD's, and it doesn't really matter what you set the display resolution, they are still gonna look crappy in VR. So dialing it down to 512 improves your render times, dial it on down Do we know what the F18, or Harrier MFD's are rendering at, in game?

And that's really kind of the bottom line for this iteration of DCS and this generation of HMD. Wanna drive 9.3m pixels, at native resolution plus MSAA, ya need a 3090 (or maybe a 3080ti?) ... Next step down to a 3080 or 6900xt you can either have native resolution, or MSAA, but not both. Next step down (3070/6700) you just can't drive that many pixels. (But can still have an enjoyable experience using upscaling technology) Glad to see HMD designers finally thinking about how to get the most out of your GPU. Finally. The G2 is the last and best of a breed, I think. There's a gold mine of data in your thread. I'd be interested to see an "infographic" going through each of the DCS graphics settings and showing how much each one affects GPU time, and how much CPU time proportionally. The absolute numbers will vary depending on the system, but I'll bet the proportions don't vary all that much. I may make a pass at that this weekend if you don't mind me processing your data... G2? Or Oculus? I was never able to get my RIft or Rift S off every-other-frame no matter what I did. There are places in Adrenaline that'll lock or limit your frames, typically marketed as power-saving features. Dunno of NVIDIA drivers do that.

Andrew -- think about your mapping strategy. Think about mapping gear, flaps, speed brakes, weapons, sensor control to the same place for every airframe. Beyond that, in VR we tend to make liberal use of the clickable cockpit. I use a mouse, some use a trackball, some use controllers -- several options. I only really map switches that are behind me cuz I'm getting too lazy to turn around looking for them. It is helpful to map gear and flaps, for a long time I mapped em to a hat switch on the throttle using a modifier and still think that was the good solution Link for your toolkit which has been discussed at length in other threads here ---- VpR81: see pic, I don't *think* its placebo but I can talk myself into just about anything

Have him read Raven One. Give him an idea of what kinda missions people fly, how strike packages are put together etc. Plus it'll blow his mind when he finds out he can get the raven one campaign and fly those missions in dcs

Point taken: Andrew will probably want mirrors off while tuning. The "1024 every frame" thing is kind of an apocryphal recommendation, albeit one based on an apocryphal reference: RES. OF COCKPIT DISPLAYS. Set the resolution of cockpit displays such as mirrors, targeting video indicators, multifunction displays, moving maps, etc. Note that higher resolution settings can negatively impact the smoothness of game play. ...which is still in the manual ten years later fwiw. Regardless. Theory is, give the VRPerfKit some information to work with, and let it do its magic. (ETA: I don't know what resolution the Hornet DDI's are coming across, I don't think it's even close 1024. Maybe 800x600? Only so much you can do with those things, especially if you're blowing em up to iMax proportions. Which is fine. Bodes well for upsampling strategies.)

Same. Orion isn't simply sufficient, it's exquisite. Fly with your fingers not your fist.

For the most part. Chief is kinda dense. "wheel trucks?" Wheel CHOCKS!" "Wheel socks?" I just want to climb out of the jet, take my helmet off and look the guy in the eye. Wtf is a wheel sock? Seriously. I want to know. But when you hear "rrrrrogerball" it makes it all worthwhile don't it

Andrew -- you might like it more if you set your textures to high and undersample. Google up fholgers vrperf toolkit, it actually works pretty darn well. Set steam global 100%, per app 100% and pixel density 1.0. Let vrperfkit do the work. Start with stock settings. RenderScale determines performance. Somewhere between 0.5 and 0.77 is your nirvana. Forest visibility should match your vis range, both should be medium more or less. You may or may not be able to keep the shadows. Set cockpit displays to 1024 and max out anisotropic filtering. Turn civ traffic all the way off and put gamma a smidge higher. Report back. Feel free to PM if you want.

I used to use Voice Attack back when I had a T16K, it was pretty slick. Still use it for comms but prefer to map switches to the throttle base now that I have one

1. The 60Hz flicker is horrible in WMR home; it's like a fluorescent light is going bad. I'd suggest trying it in-game, where it's not nearly as bad, before a person decides. (I still don't like it, but will admit, it beats the snot out of flying in pancake mode) 2. I wonder if forcing DX11 makes it work better with the fholger mod? Since I quit using FSR4VR (in favor of ReShade) I don't see any difference whether it's on or off 3. From what I can gather: Steam gangs up with WMR and sends a recommendation to DCS as to what size image to render. DCS can either comply, or not; you adjust that with PD. You can also edit the recommendation with Steam Global and Steam Per-App settings. Point being, all that stuff goes into the hopper. It doesn't matter how you get there. If you play around with Jabber's spreadsheet you'll see there are several ways of hitting your target, and which knobs you twiddle make no difference at all (says Jabbers -- who is a programmer -- and based on my experiments I'm inclined to agree). It is important to remember, if you send an undersampled image to the compositor, it'll be scaled up, with the artifact that goes along with that. If you send an oversampled image to the compositor, it'll be scaled down and in that case you get the advantage of using supersampling as an anti-aliasing technique. (Which is what I do). Note, "100%" in your Steam settings should be about 140% of your native resolution, depending on the headset. (Confusing, I know.) So if you want to run precisely at native resolution in the G2, your combination of Steam Global, Steam per-app, and PD should add up to 3100x3100 pixels. How you get there is up to you, and I'd bet my own money that it doesn't matter how you get there. The exception to that rule is if you use the fholger mod (FSR for VR, or VR Perf Toolkit). That takes whatever image size you've specified using Steam Global, Steam Per-App, and PD. Downscales that for performance, and then using some kind of dark magic, upscales it back to what you specified. That virtual image is sent to the compositor for processing. I think it follows, the full impact of the fholger mod should only be apparent if you send the exact right size image to the compositor (for the G2, 3100x3100) otherwise WMR has to scale the image again and I imagine that might dilute the effect of the fholger mod. Which is kind of not good if you're undersampling, but, TED's approach is very interesting... So I guess the strategy is to send a supersampled image to the compositor -- not all of which was rendered by DCS. Some of those pixels were synthesized by the fholger mod, right? Interesting, I'll have to try that.

The setting on your HMD depends on how big your head is in real life. The setting in DCS is how big your virtual head is in-game. If you set it small, you're like a little person barely able to see over the sill. If you set it big, you're like Shaq trying to squeeze into a teeny tiny cockpit. You can set it anywhere in between those extremes looks right for you, or just play with it (and I'll admit, it's about thirty minutes' worth of silly fun) or if you have OCD you can turn on the pilot body and adjust IPD so your body "looks right." Or do like most people do, ignore it

Taurus, or Orion? For the Orion, idle and idle cutoff are coded as button presses. Kind of hard to screw up the idle cutoff, you have to be against the stops for that to trigger. Going "over the gate" puts you into idle. Both work flawlessly. By the same token, you don't have to map either one. In which case you would go back to using the keyboard or the mouse. You might consider doing that if you're rough with the controls, although in general I wouldn't say Winwing is designed for that. It's like the difference between a Harley and a BMW. One you drive with your fist; the other, with your fingers. (That's coming from a guy who drives a Harley. Which does what it's supposed to, it just requires a firm hand, or foot. Kind of like a Warthog joystick, you gotta show it who's boss, right?)

^Interesting When I upgraded from a Rift S to a G2, I found that in-cockpit textures were exponentially better. DDI legibility was only marginally better. I was a little surprised -- the G2 is no slouch when it comes to pixels-per-degree (and neither is the Quest 2). But perhaps shouldn't have been. The DDI's max out at 1024, and perhaps don't tolerate being blown up to iMax proportions very well. To be clear I'd LOVE to have a Pimax 8K, just wondering what it would take to drive it. My 6900xt is working itself to death driving 8.8m pixels. I imagine the dang thing might catch on fire if I try to make it drive 16.6m pixels, holy cow. I like the fixed foveated rendering, I'm sure that helps but I got a nickel says most people are undersampling (as most people are with the G2). Based on my experiments, your best DDI legibility is gonna happen at native resolution. Not that I'm tooting the G2's horn. I wouldn't recommend it to anybody, it's a pain in the ass.* Quest 2 is a fine piece of kit, and there's not a dang thing wrong with it imo. Also, if I spent another four grand on this game, and my DDI's were still blurry, I'd be pissed. Pimax 8K: $2K. RTX 3090: $2K. ReShade: priceless. ------ *And I wouldn't trade it for anything. It's kind of like my cat in that respect.

FSR should work best, I think, if your Steam and DCS settings combine for a resolution of about 140%, depending on the headset. For a G2, we are talking about a 3100x3100 image. Anything else dilutes the effect of FSR (because WMR is going to be scaling the image after FSR is done with it, resulting in Lord-only-knows-what) My recommendation: 1. Start with Steam global resolution 100%, per-app 100%, and pixel density 1.0. Go for a fly. 2. If your system handles this with ease, you get to try antialiasing. You can use supersampling or MSAA, your choice. 3. On the other hand, if this kicks your hardware's butt, it's worth trying FSR. Keep your Steam and DCS settings exactly the same, let FSR do the scaling. I'll bet you like it; I find it basically impossible to tell the difference between native resolution and FSR running at a RenderScale of 0.8 or better, on the G2 at least. There's gonna be some jaggies though. The extent to which the FSR mod is better than merely dialing down your settings, and letting WMR handle the scaling, is debatable. I don't think WMR does a half-bad job meself. Try one against the other and see which one you like best. If it's all the same, I agree with the sentiments in this thread: keep it simple. 4. If you have to dial your effective pixel density much lower than about 0.7 -- regardless of whether you're doing this with FSR or with any combination of Steam or DCS settings, you might need a new GPU. (Or a new HMD... I'd rather drive a Quest 2 well, than a G2 poorly). Put it this way, an effective PD of 0.5 looks pretty gross any way you cut it Sharpening is another matter altogether. It will not eliminate jaggies, will not improve performance, and it introduces its own artifacts. Aesthetic effects only. Speaking of aesthetics, I'm not at all sold on the idea of combining undersampling with MSAA. It's fairly pretty I guess, good for sight-seeing, but isn't good for gameplay and IMO would be easier to just buy a copy of MSFS or XPlane. The combination of supersampling and MSAA is a different matter. Research suggests that's the optimal combo, but not having a NASA supercomputer, I wouldn't know. Net of everything, I'm led to the conclusion that there is no such thing as a free lunch.

Ya know I think I'd rather fly in a real p51 than a real f18. Seems like that would be about the most fun you can have with your clothes on

Let's talk about aerodynamics in general, before talking about the Hornet specifically. 1. A given aircraft won't always stall at the same speed, but it'll always stall at the same AoA. That's the theory behind the on-speed/AoA approach, fly it down at just above stall speed, pull back on the stick when you're six inches off the ground, done. 2. When you're flying around at high AoA like that, throttle no longer controls speed, it controls vertical velocity. You use pitch to control speed. Thus AoA = speed, under those conditions. I think it follows, then, that if you want to control both forward and vertical velocity, you are locked into both a given AoA AND a given glide path. Yes? No? So I think what Ruddy might be asking can be bifurcated, with the general aerodynamic question being, can you land "A" plane (such as an A4 for example) with a lower AoA ... well, really, under any circumstances. Obvious answer is "yes," you land it just like you land an F15, come in fast and flat and sorta gradually approach your target AoA as you flare, and set it down nice and easy Air Force style. With the related question being, can you do that on a carrier, and the answer is probably no... there's no guarantee of even catching the 1 wire cuz there's kind of a tendency to bounce over it. Maybe if you catch it at the last possible second, and suffer the scorn of the LSO? Still, the meatball is useless there, so it'll be all eyeballs... Then the other question is, can you do that in an F18? Which is a slightly different question because the flight computer kinda "locks in" the AoA for ya. Which I suspect, but can't quite prove to myself, locks you into a glidepath as well. Given how the system is set up, I think there's only one way to land it really. Which is good, I guess. All you gotta do is control the throttle and you can't miss. The A4 is almost like that, but not quite. You gotta work both the throttle and the stick, in other words you have to be mindful of both your forward and vertical speed. One way of looking at it is, the Hornet regulates your speed for you, all you have to do is stay on the glideslope If so, perhaps the answer to Ruddy's question is, "no." But, I'm not sure cuz my brain is melting