Jump to content

DeltaMike

Members
  • Posts

    856
  • Joined

  • Last visited

Everything posted by DeltaMike

  1. Not playing DCS is out of the question. Me, I'd slap that 1060 in there and see what happens, I suspect you won't have any problems -- it's on a point system, I had to swap mobo, CPU, RAm and SSD all at the same time to trip the copy protection -- but if it does put you over the edge, I'd deregister first before I re-install the 2080, and then re-register once it's in. You get 9 or 10 shots at it and even then they give you more if you wait a while, I really think you're gonna be OK
  2. Not playing DCS is out of the question. Me, I'd slap that 1060 in there and see what happens, I suspect you won't have any problems -- it's on a point system, I had to swap mobo, CPU, RAm and SSD all at the same time to trip the copy protection -- but if it does put you over the edge, I'd deregister first before I re-install the 2080, and then re-register once it's in. You get 9 or 10 shots at it and even then they give you more if you wait a while, I really think you're gonna be OK
  3. With your background, the Hornet may be the easiest full-fidelity sim to get into. Everything is automatic. To give but one example, you already know how to make a coordinated turn. If you didn't, and had done your training in a Hornet, you would have no idea what that was. Don't get me wrong, it's fun to fly. You'll be amazed what this thing can do. Even from a sim viewpoint, if you understand avionics, you'll find it's way easier to just operate buttons in the cockpit than it is to remember key bindings. There's only one weird keybinding, which you use to hook into the catapault. You need key bindings for things like the menu, chat window, kneeboard, stuff like that. Otherwise it's all a matter of just flying the thing. And it's obvious to me that the system was designed to reduce pilot workload and to reduce training time. It's an amazingly well engineered system. The only downside to the DCS hornet is, it still doesn't have an automatic "respond or die" system. Makes it a little awkward in multiplayer. "Why did you shoot me down?" "Couldn't tell if it was a box or a diamond, plus it kept flipping back and forth" etc. "Buddy spike. Don't do it. Don't do it!!!! ARRRGGGGHHHH" You'll get that a lot. You can do manual IFF interrogation and it's well integrated into a beautiful situational awareness scheme but still. Hornet drivers are eager to get TWS; the rest of us are dreading it. If you're really focused on air superiority, it's not your best choice, for a couple of reasons. For flying around, learning carrier ops, blowing stuff up, and tangling with the occasional MIG, it's pretty awesome, and you can be thankful the F15's and F14's are on your side (usually) so please don't shoot us.
  4. Yeah you know when I look at a flaps switch that says "full," "half," and "auto," I'm assuming that's what it does, but that just ain't so on the Hornet. Near as I can tell you're selecting modes. "Half" = takeoff mode, for example. Flaps half on the ground, flaps retract at 250, pitch trim button sets takeoff trim. "Full" = landing mode; flaps extend at 250 and pitch trim button dials in AoA. I don't think the Hornet has a pitch trim control in any traditional sense. In free flight, it trims itself. I think it does that on takeoff and landing too, just not very well. Meaning, you have to use pressure on the stick to maintain level flight above a certain air speed. Let's say you drop your landing gear and set flaps to full at 250kts. You're going too fast for the auto-trim, the nose wants to rise and you can't fix that with the trim button, you need to use forward stick pressure to maintain level flight. As you pass through about 180, the trim button starts to work. As you go slower and slower, you now need back pressure on the stick to maintain level flight. Now your trim button has some authority, you can use it to relieve pressure on the stick. Problem is, now you're flying around at a high AoA. You have a ton of drag, so now you need a ton of throttle to overcome that drag if you want to keep flying. The effect of that is, you go from being idle throttle, trying to find some way to get the jet to slow down. All of a sudden you need at least half and maybe closer to 3/4 throttle, trying to find some way to keep the jet in the air. You have to do three things at once: 1. Go from stick forward to stick back 2. Mess around with the trim 3. Go from idle throttle to 3/4 throttle All three have to be precise, with the hard part being the throttle. If you don't give it quite enough throttle, the aircraft sinks rapidly (some say too rapidly, I wouldn't know, never flew a real Hornet). If you give it too much throttle, and speed up, your nose climbs. If you come from a general aviation background, either in sims or r/l, you get this pounded into your head: don't fly on the wrong side of the power curve. Don't fly down there where you need almost full throttle to maintain level flight. Main reason for that is, you don't have any way of knowing what your AoA is, no idea how close you are to a stall. If you run out of throttle, there's a tendency to want to yank back on the stick, and after a certain point, if you do that you aren't flying any more. But, I guess that's how you do it in the Hornet. Good news is, you know what your AoA is, so you won't be flying that close to the edge of disaster. And the aircraft will trim itself to a given AoA (up to a point) so if you're Johnny-on-the-spot with the throttle, you can keep the thing under control. Just gotta have control over the throttle. It's a great system actually, it seems magic to me. The beauty of it is, once you get it set up right, of all the things you have to mess with, you don't have to fiddle with trim. I find it amazing the way you can dial in an AoA regardless of what you're doing with the throttle. The only price you pay is that brief moment of thrashing around as you're transitioning between 250 and 160kts. I don't think there's an easy way to fix that. You just have to train your left hand to be where it needs to be without having to re-invent the wheel every time you land. Try this. RCTL+ENTER. That'll give you a visual cue as to throttle position. See if that doesn't help
  5. Re vertical movement. Problem is if you pitch up you slow down, and vice versa. Once you're hooked up you slide forward a little, slack the hose, that gives you some wiggle room. You're flying in formation with the tanker. The position of the basket is merely a constraint on the geometry of the formation, and the length of the hose is a constraint on your precision. I can't fly with enough precision to nail the basket, I have to snag it on the way in. The basket is not my target, the target is an area of space about six feet further in If you're aiming for the basket you'll miss most of the time, when you do hit it, you'll disconnect almost immediately. If you're aiming for your space under the tankers wing you'll miss the basket most if the time. But if you do hit it, you'll stay connected
  6. Haha coxy that would have been me. Thought I was dead but I looked around and said hm, I'm still flying sort of, and my guns still work, so... take that! The tacview is hilarious, all these bullets come flying out of nowhere... it was the ghost of Delta Mike Hard to tell if that's a bug, maybe. Bigger the jet the harder it is to knock it out of the sky, F15 is a big jet. OTOH if you take one bullet in the Hornet it always seems to hit your O2 doesn't it? Maybe the damage model needs tweaking. In any event, DCS often thinks you're dead when you aren't quite... when that happens you drop off the f10 map, which makes it kinda hard to limp back to base sometimes eg when your nav systems are dead. Long as your radio is working the AI ATC can vector you in, but a human controller can't help you. Then there's the supernatural AI thing. Personally I think Coxy is supernatural too but maybe it's just me. (Great flying bro)
  7. First watch this Key points: - Learn how to get into waypont data screen - Learn how to enter a waypoint Then watch this Key points: - Designate waypoint as "A/A WPT" which = bullseye - Learn how to get your ownship position, and bearing/range to target using radar screen and TDI cursor Using bullseye is fairly easy conceptually. It would be like if you were my wingman, and you had a tendency to take off on your own. Which he does. "Where ya at?" "Um, 10 miles from the airport flying south." "Dangit I'm 10 miles from base flying east. OK, I'll turn right and look for you on radar." You call for a picture, Overlord says theres some crafty J11's at bulls 180 for 40. You glance at the radar, see you are bulls 270 for 20. (Note all these directions reference the radial FROM the waypoint, the direction you would be facing if you were flying away from the waypoint.) So you turn right and ask for dope. Navigating with precision to a particular point relative to bulls (or any navigation point) is a little trickier, I'd get the basics down first.
  8. That didn't work for me either, I'm in VR so it depends on where my head is, if I move my chair a little bit it throws my references off. One one level, a lot of it is peripheral vision, I think you get more of a feeling of relative velocity in your peripheral vision. Also I find I tend to go where I'm looking. I would agree that looking at the pod seems to get everything pointed in the right direction, although I wouldn't say I'm staring at the pod, that's just where I'm looking. On another level it's reacting before the stimulus, it's a matter of anticipating what you have to do. One way of thinking about it is, everything you have to do, you have to undo. If you move the stick to the right you're gonna have to move it to the left. If you advance the throttle you have to pull it back too. So you never just do one thing. Hm I'll move to the right a little. Now I'll move forward a little. No. Stick goes rightleft. Throttle goes backforward. Forwardback. Backforward. Little teeny movements in pairs. Also breathe. When I got frustrated with refueling I shot the tanker down. When I got bored with shooting I learned how to nudge it offcourse by whacking the tail with my nose. When I figured out how to do that, I was able to refuel. Go figure.
  9. Question in my mind is whether or not we can even do that. Most headsets come with 90hz monitors, for various reasons. To achieve that, both the CPU and GPU have to do their work in about 11ms. Very difficult in DCS. At the other end of the scale, you need to keep your latency at about 20ms or less to avoid the puke factor. So what you're really shooting for is 45fps, hoping your GPU can take on its share of things like vis range and shadows, do whatever anti-aliasing you want, and turn on all those pixels in less than 22.2ms. It's harder than it sounds.
  10. So far they don't seem difficult to kill if you can get within 15nm. That's a big if. I think it's worth taking a little time to practice missile evasion, first things first. Doesn't take a lot of time. Practice your split S, which is very handy; and your barrel roll, which comes in handy in a pinch. Practice notching, although personally I find it difficult to notch six missiles. Learn to pat yourself on the back for getting back to base alive in MP. The key I think is to be just a little less persistent than the Tomcat drivers.
  11. Get an SU33, and instead of doing what you're supposed to -- flying straight into our missiles from 40mi away -- sneak around in the mountains, disappear off radar, dodge shots you're not supposed to be able to avoid, and then pop up out of nowhere with those infernal ET missiles.
  12. hands down the t16k. Stick is a little more stiff than I like it but it loosens up after a while. Very accurate. Throttle needs dry lube and you want to back out the tension screw as far as you can, that mitigates sticktion considerably. The throttle joystick is nice. Easy to set up in DCS. The pedals are good enough, they don't have very high resolution but they appear durable enough. Look at what they are charging for rudder pedals out there; even if you get tired of the stick and throttle, if you keep the pedals you aren't losing much even if you upgrade the stick and throttle later.
  13. Good question. In the Linux world you can wire up as many GPUs as you want for compute applications, but for graphics sli and crossfire was the only game in town and I guess that's not a thing now. Thing is, it's not totally clear that a 2080ti is enough for the next gen headsets. AMD has proposed dual GPU setups for VR -- one for each eye -- and plans to work this in to vulkan. We will see how it works. VR is all about the render time and we don't have much time to work with, you fix one bottleneck and another comes up. But yeah I'm with you, I'd much rather hack together a GPU cluster than to start shopping for research/professional GPUs, at whatever cost, hoping it'll work
  14. Thanks for this. Takes work to generate hard numbers but they are very useful. One question in my mind is, if you have enough RAM to run high preload, would that allow you to squeeze a little more vis range in there. So instead of going for max fps, if you said OK 60fps is enough, how can I get the most out of that? If preload radius would make a difference My prediction is, it might smooth things out a little as you transition into a busy area, like going from desert to city. I'd be surprised if it gave you enough headroom to add eye candy. Also, go team red
  15. The ryzen5 2600 is no slouch, mine didn't overclock well but I've seen it hit 3.9 during stress testing right out of the box. For less than $300 for that and a halfway decent b450 mobo, you don't even have to wait if you don't want to
  16. I can maintain 45fps with shadows off, vis range medium. Currently running PD 1.2 and msaa2. Gotta undervolt which is easy with latest crimson driver.
  17. Maybe coming in too shallow or too fast? Gotta drop em from up high, dive in steep, chop your throttle in the dive.
  18. I dunno, I don't think elevation *angle* is all that helpful. That angle will change constantly unless you're diving, he's climbing, and you're pointed at each other. In which case you don't need to change your radar elevation. In all other circumstances, you will. You need to know altitude, not angle. Move the TDC gate around the screen, look at the top and bottom numbers, that'll give you a good idea of the altitudes you're scanning at various distances. Angels 20 strikes me as kind of an awkward height. You aren't exactly sneaking around, but you aren't exactly taking the high ground either. If you think about it, it's a lot easier to set your radar up if you're on the deck, or at 40,000
  19. If you're scanning out to say 80nm you might get the feeling you are covering everything from ground level to angels 40. But in reality, if the target is closer than that, it's easy for it to be above or below the radar beam. Search range doesn't affect the geometry of the radar beam, just affects what shows up on the display. But, if you keep the search range close to where you expect the target will be, it'll keep ya honest
  20. It would make people puke due to a mismatch between visual and propioceptive (neck position) input.
  21. Last time I checked, which was about a week ago, everything seems to work if you make it a habit to set both manual fuse and electrical fuse, with the one exception being laser guided bombs, where you need EFUZ but need to leave manual fuse off. Note you have to be in a steep dive, helps to chop the throttle. You can only drop on the +, not the _|_. Wait for it. Put the target where the circle is, chop the throttle, roll in.
  22. And are you able to turn on shadows?
  23. If it's modern a/a combat you're interested in, I'd consider adding an air superiority fighter to the mix. You want something fast, with loiter time, BVR capability, good at spamming long-range active missiles. F15 is cheap, available now. Fast. Simple model that lacks nav or comm features, although there are work-arounds. F14 won't be cheap, not yet avail (soon, they say). Not all that fast but the Phoenix missile seems fast enough. And it's Navy, if you're interested in carrier based strike packages. Personally I think the Hornet's advantage is in close. It can fire AIM120C, and will have TWS eventually, but it's slow, doesn't accelerate very fast, hard to put enough stank on that missile to make it truly effective, and if you're running around in AB you run out of gas pretty quickly. In real life, you get tired of spending the money to purchase and maintain F14's, you get the idea in your head that the F18 can do it all, and keep telling yourself that. In our world, F14's only cost 60 bucks and never need maintenance. What's not to like?
  24. Finally nailed it. What helps: 1. Breathing What doesn't help: 1. "Now it places the lotion in the basket. It places the lotion in the basket! PUT THE X!*$%&$ LOTION IN THE BASKET!" Wax on, wax off. Stick right, stick left. Throttle forward, throttle back. Breathe in, breathe out. Never one thing, always two.
  25. If I'm understanding correctly, nvidia is supporting multi-gpu computing (hence the nvlink) but only for professional cards, people who are doing deep learning as a hobby are starting to get pissed at nvidia due to lack of support. But. That's for computing, not gaming. Question is, what's DCS asking of the card. It doesn't appear to be memory-intensive, for example boosting my memclock makes a huge difference with a compute task (mining) but has no effect on DCS; my main job now is to keep the dang thing cool and if I accomplish that, it does what I want it to do. I get the feeling the NVIDIA cards have stuff DCS can't use, like ray tracing and I don't *think* we use the tensor cores. If my impression is right, and it all boils down to compute units, the VII should give the 2080 a run for its money. And if DCS stays true to form, and remains hardware agnostic, then it should be a wash. THe kicker is, radeon hasn't given up on the crossfire concept, it's kind of their strategy for VR, to run a two-card system, one card for each eye. That does require developer support, but apparently radeon is trying to work that functionality into Vulkan, so... maybe someday.
×
×
  • Create New...