DeltaMike

Ruddy -- can I ask an aerodynamics question? Trying to wrap my head around the physics. Let's say you call the ball and find that your AoA is perfect and you're right on the glidepath and your speed is perfect. Would it be fair to say, you can't change one of those things without changing another? Would it be more fair to say you can't change one without changing all of them? Are you saying, then that you prefer to come in a little fast and then nail your speed/AoA during the flare? If so, that would be a shallower glide path, no?

^Personally I've found that undersampling makes it worse. OP, you might consider a sharpening filter. Fholger's universal VR shader (see this thread) helped me a lot. Fholger's VR Performance Toolkit is worth a look if you must undersample to keep your framerate up. You get your choice of three sharpening protocols. It works. Don't forget your in-cockpit controls for contrast and brightness. Essential for FLIR and moving map.

vkb t-rudder ftw Fine gear, reasonably priced, has everything you need for the hornet except brakes which are easy enough to map. Plus they don't take up a lot of room. I don't use em *much* up in the air but they are helpful for lining up on close approach or when tanking up. Problem with using the stick to make fine yaw adjustments is, you get three planes of movement every time. Sure you go sideways, but you also lose altitude and then you have to start messing with the throttle... easier to give it just a little bit of rudder and wait a second or two for things to line up. Also easier to regulate your turns to base and final. There are times when it's appropriate to be yanking on your stick, and times when it's not, capisce?

Something is upscaling the image -- maybe WMR I guess? At an effective PD of .8 or .9 WMR doesn't do a terrible job imo.

oops gotta change that, whole new system with 6900xt that vega punched above its class but that was then... OK, let's compare apples to apples: 1. SVR 100%, multiplier 100%, PD 0.9, vrperfkit off 2. SVR 132%, multiplier 100%, PD 1.0, RenderScale 0.77 Both function and look exactly the same, which is what we would expect -- we are driving the same number of pixels. GPU time 11.8 either way. Moderate amount of pixellation and shimmering, DDI's are reasonably clear. Key to the whole banana is to run #2 twice -- once with the advanced supersampler filter on, and once with it off. I did that, and I think it might help a little in terms of pixellation and shimmering. Maybe. It's clear to me that undersampling to any appreciable extent -- a PD or RenderScale of less than about 0.7 -- makes the picture look awful. Terribly blurry and pixellated. Everything looks more or less OK if you only undersample a little, and it's open to debate whether vrperfkit adds much of anything beyond the sharpening filter. Question is, to what extent does the advanced supersampling filter work. If it's working as we hope and suspect, then cranking up the steam resolution settings may help, even if the total number of pixels rendered is the same. We need to go back and really crank it up. Reproduce the Jabbers experiment, keeping total pixels (or "effective PD") constant and only vary the advanced supersample filter. Looking not so much at GPU time -- which we expect to be the same -- but at jaggies and shimmering. Worth looking into I think. I've noted that a moderate amount of oversampling (ie resolution at 132%) doesn't affect spotting distance. Undersampling and then applying MSAA destroys it.

Some more tests. 1. Steam thinks I should run at 132%, which is ~3600x3600 virtually rendered. Beautiful image, virtually no shimmering, as supersampling is used as anti-aliasing. 2. If I dial pixel density down to 0.8, end of the day I'm rendering at about native resolution. There is some shimmering but my times are good and the game is playable spotting-wise This is where I usually fly (my actual settings are 100/100/1.0. More than one way to skin this cat). 3. I installed vrperfkit and ran it at suggested settings all the way around (fixed foveated rendering is off, of course). Likewise it winds up being about native resolution, so there isn't really any net undersampling. The shimmering is there, so there's no real anti-aliasing either. All I'm really using here is the sharpening filter, which I will admit, looks kinda nice and didn't cost me anything. (I should probably select "CAS" rather than "FSR," I guess) I think the thing works as advertised -- FSR at Ultra is virtually indistinguishable from native resolution, and it runs a heck of a lot faster. Doesn't add much to the 6900xt unless you like the sharpening filters, which are starting to grow on me.

Could be, if you have SS, PD and MSAA all cranked up, which is the temptation with the rift.

Both (usually) refer to over sampling or "super sampling." Lets say your headset has two teeny monitors in it, one for each eye, each with 2000x2000 pixels. If you increase PD to 1.5, you are now rendering 3000x3000 or 9m pixels per eye. Which is more than 200% far as steam is concerned. Steam is measuring total pixels rendered, PD is looking at pixels rendered *per axis.* Back in the day we used supersampling to trick the eye into believing you were looking at a high resolution world through a screen door (long as you kept your head moving). That's less of a thing now that we have high resolution headsets. Now everybody is all stoked to undersample, using fancy protocols to scale things up to native resolution attractively. Current technology allows you to render at say 2.4 million pixels, scale it up to 4m pixels, and you'd be hard pressed to tell the difference. Brave new world we live in. At any rate. Keep your sanity. You might have as many as four numbers working with (or against) each other to determine how many pixels to render, all basically doing the same thing. Just change one at a time.

AMD works fine with oculus, I used to drive a rift s with a vega56. Easy headset to drive. Clean drivers, tune card, clear shader files. Start tuning in sp, pick a mission. Free flight in Caucasus minimizes CPU and ram bottlenecks so that's where I'd start. Textures high, vis range medium, shadows off, msaa off, civ traffic and grass off, trees medium, af 16x. PD 1.0. Start there and adjust supersampling in ott, you won't get much so don't push it. ASW on. Don't fiddle with pd and ss at the same time, they do the same thing. PD gets out of hand quickly, leave that at 1 and see if you can get ss up to 110 or 120 in ott Some of the above advice has to do with steamvr runtime, which is not relevant to oculus. I don't even know if you can use steamvr with oculus, but if you can, don't. Agree completely with svsmokey's recs. Also I agree, rift S doesn't look great with those settings but it's playable, which is more than we can say for msfs right. ETA: I looked it up, I guess you can use SVR. Having used both, I'd say the oculus has the better runtime. Fewer moving parts, much better user interface, fewer crashes and "window out of focus" issues, better motion reprojection (in some ways).... Generally harder to screw up. I'd still be using it if I could

Running this through Jabbers' spreadsheet, where "effective PD" is a measure of how many pixels total we are driving. (Actual in-game PD is kept at 1.0) F18, cauc free flight instant mission. Open FSR stock: SVR 100%, multiplier 100%, FSR downscale value 0.7, MSAA on. 100/100/1 refers to bone stock native resolution without the FSR mod, MSAA off. Final entry is: SVR 100%, multiplier 120%, FSR downscale value 0.7, MSAA off. MSAA just crushes performance. Clearly way better off using nikoel's strategy. I wouldn't say it's perfect in terms of getting rid of the shimmering, although it's about the same as running native resolution and I feel it looks better than 100/100/.7 even with MSAA on (although I think I'm getting a little too aggressive with the sharpening). Note, spotting distance (for an S3) is about 8 miles in native resolution, and is less than five miles at 100/120/.7. MSAA off for both trials. Still not completely sold on FSR although if ED sees fit to implement it -- with, for example, the HUD and DDI's coming through at native resolution -- I could see the utility

It appears FSR scales like pixel density, so we can plug it in to jabbers' spreadsheet to see if we are making progress or not, performance-wise. I think performance is gonna correlate with total pixels rendered, but it is very interesting how you get there. Essentially using supersampling as your anti-aliasing technique, no? In other words, take care of those jaggies *before* fsr upscales the image.... Very interesting approach and I'm eager to give it a try.

Excellent writeup. TED shows us that the 6900XT's performance is really close to 3080 -- within margin of error it seems -- and given the prices out there, I imagine we will be seeing a lot of 6900XT/G2 systems coming on line in the coming year. Threads like this are invaluable to new users. Couple of comments: 1. Overclocking AMD GPU's isn't as easy as it sounds -- those things have a mind of their own. But it's a pretty good mind. Best advice to a noob is, keep it cool. Every cooling strategy starts with fans. As Lynyrd Skynyrd says, "Turn it up." There's nothing good about a quiet rock n roll song, or AMD GPU. 2. Other than fan speed, the most important setting in Adrenaline is your minimum GPU clock, which must be set a couple hundred less than your maximum. Keep your GPU clock in a narrow range to avoid frametime spikes. 3. There is no substitute for pixels-per-degree. Back in the Oculus Rift days, we twisted ourselves in knots trying to turn a blurry mess into a playable game. Supersampling, MSAA. It's a wonder my old Vega 56 didn't catch on fire. We don't have to do that with the G2, and maybe we shouldn't. Agree, people should at least try running at native resolution without anti-aliasing and see if they like it. 4. AMD handles up-sampling pretty well. In other words, if you must undersample, the game still looks pretty darn good. I'm not sure I can perceive a difference between running the FSR mod at 0.8 vs setting SVR to 80%, just flying around. And SVR at least can be adjusted "on the fly." (Theoretically you could undersample while you're on the deck, and then crank it back up again when you fence in so you can spot those pesky MIGs) 5. It's not totally clear to me one accomplishes anything by running SVR at 120 and FSR mod at 0.8. Assuming those things add up, 1.2*0.8=.96. See #3 above. I'm not saying the FSR mod is superfluous -- haven't quite made up my mind yet. Let's just say it's *really* close. 6. Anisotropic filtering is a freebie and it really helps a lot. Crank it up. 7. In my opinion, MSAA has aesthetic benefits but it doesn't enhance game play. Especially for ACM. Your best spotting distance will be native resolution with MSAA off. Second best is FSR mod with MSAA on, which works OK but it's harder to spot a blob than a dot, except against a clear sky. Undersampling with SVR alone will hurt your spotting distance (because of "pin cushion" or barrel distortion). 8. "Stutter" can either be regular or irregular in rhythm. (eg disco is "regular", breakbeat is "irregular"). Irregular stutter can be reduced by tuning your card and adjusting DCS settings. I believe locking your frame rate at 30, 60, 45 or 90 will also reduce irregular stuttering. Regular stuttering -- what film makers call "judder" -- is pretty much inevitable. Fly at 500 feet, 500 knots, roll over and look straight down at the ground and you'll see what I mean. Motion reprojection will reduce that, at the expense of introducing artifact in text (eg instruments, DDI displays) To a certain extent we need to pick our poison. -- For ACM, I'd suggest getting rid of the eye candy, run at native resolution, and turn MSAA off and motion reprojection on. -- For flying at night, with your head down in the cockpit, I'd turn motion reprojection off and maybe lock frame rate at 45. -- For watching the sun go down over the desert, I'd max out shadows, undersample to about 80%, and turn MSAA on. So pretty! (Although I rather imagine MSFS might be prettier, we'll see) That's for jet jockeys of course. Warbirds and helo's have different needs I imagine.

Wondering where in the scheme of things OpenFSR does its undersampling, and how that affects barrel distortion and central resolution. Near as I can tell, Steam SS settings account for barrel distortion. So let's say we are using a G2. We want the game to render at 2100x2100 approx. Steam suggests actually rendering at 3000x3000 approx due to data loss that occurs as a result of barrel distortion. In other words, once the barrel disortion is completed, we are back to native resolution, but with pixel density much higher at the periphery, than in central vision. You gotta upsample to preserve pixel density in the center. Using Steam SS settings, I do seem to lose some resolution in the center if I undersample (as measured by spotting distance) regardless of whether I do this in global, or per-app settings. Using the FSR mod, my spotting distance seems to be about the same as running 100% SS -- it's a blurry mess, but it's there -- which suggests to me that undersampling occurs before barrel distortion correction. So in other words, steam knows we are using a G2 and calls for 2100x2100 from the game. It upscales this to 130%. Barrel distortion downscales this right back down to 2100x2100 and then Steam feeds that to your headset. Would it be fair to say, then, that FSR hijacks this at the outset, only calls for 80% of native, then steam upscales that by 130%, downscales it right back down to 80%, and then FSR handles upscaling it up to native resolution before feeding it to the headset? If so, I imagine that's why you would use FSR over steam SS, no? That said, the game doesn't look terrible at Steam SS 80%. Something is upscaling the image, I imagine that something is the GPU, and it's not doing a terrible job of it.

Kinda surprised your Quest 2 works. If it does, I'd stick with it. G2 is better on paper but it seems like most people with anything less than a 3090 are undersampling anyway. Also the user interface for the G2 is a pain in the gonads. Also the motion reprojection works better in the Oculus (ahem "Meta") products IMO (I went from Rift S to G2)

https://vkbcontrollers.com/?product=gnx-sem-side-extension-module

For prop planes, it's kind of nice to be able to split the throttle, use one side for the gas and the other for the prop. I've tuned my axis for the hornet (83-84) and on the two planes I've tried (yak52 and T51) popping out of "afterburner" puts my prop right at 80%. Typical cruise setting for either plane isn't anywhere near the AB detent, it only comes into play on takeoff. I wouldn't go out of my way to buy an Orion for prop planes, but it works... WW does sell a collective that goes on that base, and it looks like it's calibrated from detent to detent, in other words they take the detents out of the picture. There's at least one youtube review out there

PSA: don't forget to set your minimum GPU clock, it should be 100 less than your max clock. I've been trouble shooting spikes all morning, average GPU time in the cauc map is ~14ms but it was spiking up to 30ms a couple times a second. Setting a minimum clock speed sorted all that out. My experience with tuning this card for DCS is, so far is, most important thing is the fan profile; second most important, minimum clock; and it's not clear that anything else I did made any difference.

Agree it generally looks about the same. I don't lose spotting distance with the fsr mod tho. It does introduce its own artifacts.

Still doesn't work all that great imo. Ergonomics and leverage are all wrong. With the WH, most of the work is getting it off center, and there is some downward pressure on the stick while you're using it. With an F18 style grip you're yanking it around by the neck, and it really fights back at the extremes, which is awkward and gives you the wrong kind of leverage. Not a slam against WW, the F18 grip by design belongs between one's legs basically (although I'm using it as a side stick, using monstertech chair mounts, and it's fine) The F16 style grip has a number of advantages, including a more ergonomic wrist angle and a hand rest, and might be worth a look if you absolutely must use the thing on the desktop. Try it and then see if you need a c-clamp or something.

Whoa WR that's a cool setup Back when I had room I kind of went back and forth on whether to play in VR or on monitor, I do think some kind of head tracking is important because I may be a noob when it comes to radar but I can dispense flares with the best of em. "OH there he is, he's right on our tail." Now that I am set up in a 8sqft space, it's pretty much VR the whole way lol. Recently got a G2 so at least I can read whats on the monitors... So the challenge is to operate the radar essentially blind, with a minimum amount of fumbling around. So there's the right hand stuff, most if not all of which I can map to joystick. Left hand functions, which is where the sensor control panel comes in. Here, complexity is actually a handicap, it would be quicker to just grab the mouse than feel around. And everything else for which I use the mouse. Was only half joking about the xbox controller, it actually works pretty well... iirc I was doing radar controls with the left hand, so the stick was az/el, click to center. Four-way was bars/az, Right hand was hcu and radar modes. CM on the bumpers. Don't remember how I mapped the two stage trigger. Not perfect but surprisingly not bad. But, my warthog has been gathering dust since my winwing stick came in, would be nice to put it to some use I guess. That said, your weapons control panel is the shiznit, that may well be worth developing the muscle memory. Just depends, the extent to which you do that stuff as part of pre-flight vs trying to deal with it when the fur is flying....

I'm pretty sure navy jets use the range. I know they did in the past; F4 and F14 pilots flew against migs based out of tonopah, often stopping at nellis to fuel up. I kinda get it. Drooping bombs on one patch of desert is about the same as any other. ED can't compete with xplane or msfs in the sight-seeing department and I don't think we want them devoting resources to that anyway.

That. is. impressive.

I dunno. All they have to do is get rid of that abominable joystick base and they'd have a decently competitive, if expensive high end lineup. But that base.... omg I'd rather use an xbox controller.

Haha check out this vid. In which a MIG-21 defeats an F-14 using a slick move. That'll be your callsign, once you get that move down: "SLICK!" (I'm currently reading Peck's history of America's Secret MIG Squadron which is highly enjoyable. Among other things, there's a great summary of the surprisingly rapid evolution from fluid-four to loose-deuce, which occurred during his watch. One of these days I'm gonna start up constant-peg server...) I imagine that move is overkill vs a human piloted F5; first one to the top loses and the F5 gets there pretty quickly (in my hands anyway). As for the AI... you know the problem with the AI, at first is its superhuman abilities. Once you learn how to exploit that -- and it is exploitable -- you wind up having to forget everything you've learned and start all over again once you start flying against human opponents. Same general principle applies vs FC3 modules; I personally feel it's best to seek out human opponents who fly full fidelity modules. Back when I "started over," I REALLY started over. Looked at doghouse plots till I was cross-eyed. Learned aerobatics in the Yak-52, TF-51, and C-101. Mostly in the latter which is a fine trainer. Being an Eastern Bloc sorta guy you might want to look at the L-39 instead, which I have not flown but I hear it's great. Watched this series. Read this book. Watched the moustache man video. Whole nine yards, and still have no idea what I'm doing. But at least I know I have no idea what I'm doing. Used to fly against a dude who went vertical at the merge every time. Drove me absolutely bonkers.

Getting back into DCS after a hiatus and seriously thinking about doing a deep dive into the F-14 for various reasons. Enjoy playing RIO although I need some work in that department. OK, a lot of work. Question. Any of you RIO's using a button box? Or interested in one (found a dude on Etsy who makes custom boxes for not much $) Primarily interested in an exact replica of the sensor control panel which can be used by feel in VR, for "left hand" functions (figuring my warthog has plenty of switches for radar and hcu modes, and also not against using a mouse when I'm not glued to the radar screen). Kinda leaning toward "KISS" principle, so, thinking I'll just have sensor controls on it. Well OK, maybe a launch button. Maybe a modifier button. OK maybe a TCS slave switch. As is often, perhaps always the case, mission creep is setting in. Seriously if you guys don't help me out, before it's all over with I'll have the guy building an entire left hand console and then I won't be able to find ANYTHING from under the hood and I'll just go back to using an xbox controller (which isn't terrible actually). What do you have, or would want to have, or recommend to have on such a box? Also gauging interest, maybe the dude will do a group discount eh. Looking at this guy, who lives near me. In fact I wonder if I know the guy... I might. Price is right, I'll say that.