Jump to content

Dragon1-1

Members
  • Posts

    5107
  • Joined

  • Last visited

  • Days Won

    2

Everything posted by Dragon1-1

  1. You can tune it for your type of stick to remove the deadzone, but in general, CAT III is heavily limited, particularly compared to how things work in older jets. Viper has small wings and a single engine, the FBW system allows it to be safely flown with a very heavy payload, but you pay the price for that. Flip the switch to CAT I and it suddenly becomes a very different plane. Also due to small wings, the Viper doesn't like being slow, particularly up high. Keep the speed up and AoA reasonable, and it flies just fine. If you get behind the curve, though, it turns into a pig. In particular, if you're trying to fly the CRUS page parameters, both max range or max loiter put you in a rather unfun flying regime. Good for ferrying, or getting back to base, but don't use this for anything tactical.
  2. Also, can the automatic turnaround be turned off when the mission is complete? After all, if you're going to shut down the plane and not take off again, it'd be immersion breaking for no good reason.
  3. I wonder if it's because it was the first basket tanker on the US side. The model was recently updated, but the old one was much older than the other two. I'm not quite sure on the timeline, but Hornet missions are, for most part, quite old, as well. I wouldn't be surprised if it was simply all they had to use for it.
  4. Also, I'm not sure which tanker this is, but try to learn on something that isn't the S-3. That one is the hardest tanker to take gas from, mostly because the envelope is tiny. Learn on a KC-130 or a -135MPRS, then progress to the S-3.
  5. It's very much not realistic, and it's not even that difficult to stay connected. IRL, a boom exerts some force on the jet that resists any drifting, and while a hose does not, the envelope is reasonably generous on most drogue tankers. It's challenging, but most pilots, once they get it, don't have problems with it.
  6. Mind you, unlike in DCS, a real enemy will not stay "marked" like that for long. Since most of them know what that plume of colored smoke is for, IRL they'd make every effort to be anywhere but at the position indicated by smoke. There are plenty of ways to lay smoke, but it needs to be combined with suppressive fire for best effect. Sometimes you'd mark your own position instead, or even a spot on the no man's land, and then talk the aircraft on using smoke as a reference.
  7. At that distance, TGP. Tomcat and later Phantom versions had a TV camera for this exact purpose, and on the Viper, TGP does that. VID is much easier if you can zoom in on your target.
  8. Worth noting that EGI is the newer system, despite the Hornet being technically an older jet. Vipers did get the EGI, too, it's just that the ANG ones (and ED has modeled an ANG jet) got it after 2007. EGI would have been able to spool up in half the time it takes INS to do so, for instance. So our Viper is still stuck with equipment not far removed from that on Gulf War era Block 30, while the Hornet has what would have been the most advanced navigation system available in 2002. The problem with INS is that it predates GPS. As such, when the GPS unit was added to the Viper, it had to be wired into the INS system somehow, and that was accomplished by coopting the "fix" procedure. While it would automatically take GPS fixes and generally try to keep its position accurate to within 200ft or so, it's not the same close coupling with GPS you see in the Hornet. The jet navigates using the INS, and it uses the GPS to keep it reasonably accurate, but you do not get the pinpoint accuracy you get with EGI. It's always a good idea to get your target on TGP so that you're sure that you're dropping on what you think you're dropping on, anyway. One mistake in data entry and you could drop on a hospital across the street instead of a terrorist hideout.
  9. It does not "confirm" a single damn thing. As a rule, don't ask AI about anything specialist, it spews BS way too often, and the more specialized the subject, the more often it's wrong. When it's correct, it's by accident. The real reason that oil dilution is applied before shutdown is completely unrelated to any of the above. You have to remember that you're injecting fuel into the oil sump. Fuel burns, and much more easily than oil does, at that. As such, it was recommended to idle the engine until the oil temperature got below 100F before applying the diluter. In theory, in a hot engine, the fuel will evaporate. In practice, the oil sump would occasionally catch fire if that rule was not followed. Injecting the fuel on startup, when the engine was warming up, not cooling down, would have a greater risk of fire.
  10. Yeah, the forum staff sometimes do that when they want to pretend it's no longer an issue. In some cases, there had been many locked threads on the same, "fixed" issues. It doesn't mean anything was actually fixed.
  11. Radios are implemented as, well, radios, so non-diegetic music setting won't affect them.
  12. No, you are not going to. In fact, rewriting ME would likely save you even more in long run, considering your modules typically include missions, too. Right now, ME is difficult to use, the scripting is brittle and putting together anything but simplest missions is an exercise in frustration. Both campaign creators and users would be well served by an updated editor, and you could probably drop a lot of technical debt as well, helping the coders. Time spent working on a replacement would be well spent indeed.
  13. I'll add that modern 30mm rounds have the same problem. An A-10's strafing run should look like you're hosing the target area with high speed hand grenades (the amount of explosive is similar in both cases). Likewise with Russian aircraft guns, they fire big shells, too.
  14. No. Viper and Hornet have vastly different styles, but in a match between them, it'll always be down to pilot skill. The same will probably be true in a guns-only and AIM-9M only situations about those two and the Mirage 2000C (though I have less experience with that one, so I'm less sure here), but its lack of a HOBS option will do it in if the AIM-9X is in play. While I haven't heard many Mirage 2000C stories, I did note that Hornet drivers tend to have plenty of stories of kicking Vipers to the curb, and Viper drivers likewise boast of beating up Hornets in an about equal measure. MP dogfights on YouTube between the types seem to go either way, too. It's not a scientific analysis, but neither Viper or Hornet seems to skew the odds in favor of its pilot. For Mirage 2000C, again, I haven't seen enough fights involving it to say or not say the same. For what it's worth, while I like the Mirage, stories from its pilots seem harder to come by, and it's much less popular in DCS. That said, I do know that it should be approached as a peer level opponent by both Viper and Hornet, at least as long as HOBS is not in play.
  15. Honestly, putting the entire machine together is not much harder than installing the GPU four or five times in a row. In fact, the GPU and CPU are the only things that require any sort of skill (the GPU mostly if you're installing some chonker that barely fits inside the case). If you can put together a piece of IKEA furniture, you can likely manage to assemble a PC. The CPU requires knowing how to spread thermal paste without making a mess, but that's about the extent of the difficulty. In general, the sockets are labeled and keyed, so it's hard to plug something in wrong. The most fiddly part is typically connecting the front panel. PSU is some mounting screws and plugging in cables, RAM clicks right in, M.2 SSDs are a socket and a single screw. Fans just plug into the mobo and that's that, unless there is some fancy RGB involved. Worth noting, it used to be a lot more involved, especially on the software side. My dad was amazed when we put my old M.2 SSD into my younger brother's new rig and it started right up. No fussing with AHCI settings, installing drivers from another machine, or anything like that. I'm happy to say that Windows 10 pretty much eliminated those things that used to be the bane of PC builders of not so long ago. The hard part is figuring out which components you need, where to get them as to avoid paying through the nose, and what to do if something doesn't play nice. Once you know what you need, and all the parts have arrived, the procedure for making a computer out of them is remarkably simple.
  16. This seems to be a TV thing, because I never needed to do that, or heard about people doing that, on a computer monitor. I have one which runs at 144Hz and never had tearing running exactly at that value. FreeSync should fix any tearing issues, anyway. It probably helps that a gaming monitor is designed for gaming, while a TV is designed, well, for watching TV, which usually doesn't run at 120Hz. So I suspect a monitor is going to be better in that regard.
  17. Well, there's also the Bell 47 (the MASH helo), OH-6, the Twin Huey (any variant), the Phrog, H-34, CH-53, the SH-3... No shortage of major helos for ED or others to make. I do hope we'll get the Seahawk at some point, if only so that we might see the last of that FC1 era model that's currently looking out for carrier flyers.
  18. Not yet, but then, nobody has tried yet. Also note, the things you are describing are all different aspects of the game. Talking AI? BeyondATC has you covered (in the civilian sim, for ATC only, and for no small amount of money), and LLMs in general have already made it possible. Realistic destructibility is hard, but it's been done, too (doesn't help that, as RF: Guerilla devs found out, if you make physics realistic enough, you'll need to use real engineering to keep your buildings from collapsing). Putting that into a big world is just a matter of properly streaming everything, so that it's not all loaded at the same time. Flight models are probably the hardest part, and that will always require powerful hardware. However, let's wait for GFM, maybe it'll be good enough as to be indistinguishable in hands of the AI, while not bogging things down to much. Honestly, the only thing I can't see a near term solution for is wake turbulence from large bomber formations, and turbulence in general, since that is fluid dynamics, and those require some nasty maths that are very hard to optimize. Either way, the pieces all exist, they "just" need to be put together. Also, there's a flight sim being developed for professional users that works on UE5. All the tech that Bodycam uses will be available to them, as well. Since they are doing things from scratch, they'll be free from technical debt that bogs down DCS in many areas, and being professional software, it might be optimized for workstation CPUs, and even if not, they could still specify top dollar hardware. They probably won't have talking AI, because in a government setting it's easier to put humans in every role, but what you are describing might very well exist in a few years. The big problem with such a monumental project is less about the tech, and more about the sheer scale of dev resources you'd need to put something like that out. Plus, you need to actually want photorealistic, which many devs don't. Bodycam deliberately went for something that could (and probably will, given that the press occasionally put up ArmA2 screenshots as real pictures) be mistaken for, well, bodycam footage. We'll see where we'll be in, say, five years, but unless someone can get LLMs to do 3D modeling, texturing and rigging for games (highly unlikely given how complex this is), huge worlds rendered in photorealistic detail will likely remain the domain of AAA studios, even when the tech to render them as such becomes commonplace.
  19. https://store.steampowered.com/app/2406770/Bodycam/ On the software side, UE5 is getting there. VR tech is not quite there yet, certainly not running displays with resolution matching that of a human eye, but this is within reach. In fact, apparently headsets are closing in on that. Pimax is working on one with a 6K display for each eye. It'll probably be out of most peoples' price range, but this could very well be the point where you start seeing diminishing returns. I believe we'll see that endzone within our lifetime. Note that Bodycam is apparently playable on a midrange gaming rig, so this level of photorealistic rendering is thanks to advanced technologies, not brute force. If you can build a rig that can push that kind of graphics in 2x6K, I think you could make something that's very near to being indistinguishable from reality. Now, for DCS, its rendering technology is somewhat behind UE5, to say the least, and it's using DX11 which is very inefficient in VR, so it'll take some more time. I hope it'll happen eventually, though.
  20. Honestly, your post did look like you were being an ass and questioning the basic competence of people involved in the accident. Even if it was just bad timing, you could have worded it better. They didn't use those little kneeboard maps like what we have in DCS? I know they are a thing in single seaters, but I guess that doesn't necessarily extend to aircraft big enough to have a navigator.
  21. As a matter of fact, any gaming-oriented CPU will blow typical office applications out of water. So even if you do use the PC as a "daily driver", as long as gaming is the only thing you need a good CPU for (as is the case for most people), it makes sense to maximize performance in gaming. If you are doing something that requires top level CPU performance and do not have a dedicated workstation for the purpose, then it obviously makes sense to focus primarily on that, especially because it's usually what makes you money, but that's a whole other set of requirements. IMO, the X3D tech is here to stay. The 9800X3D is rumored to boost to perfectly adequate 5.2GHz, and even with Vulkan, DCS will ultimately be running on the 8 P-cores on Intels as well. The benefit of a few tenths of GHz the Intels have on it (Core Ultra 9 boosts up to 5.6GHz, and that's one tier up from 9800X3D) and the E-cores is highly unlikely to outweigh the large L3 cache. It is a niche (it always was, with middling performance in most areas that weren't graphics-heavy gaming), but it's a niche very relevant to DCS users, particularly those with 4K screens and/or VR. In particular, I can see it being heavily relevant to VR going forward, because there's so little room for compromise there.
  22. Power lines are supposed to be on every detailed map. Theory and reality coincide most of the time, but sometimes they do not, particularly when infrastructure has recently been added or removed. There are ways to mitigate it, but they don't always catch everything. Power lines are particularly insidious because while the pylons are plenty visible (and in fact, great landmarks to navigate by, if not using GPS), the wires themselves are not. This is a good reason not to get too low over roads, too, because power lines often follow them.
  23. I'll stop you right there. I, and probably most people, have no desire to fuss about with manually setting core affinity for every single thing that's currently running. With a P-core/E-core architecture, you have to do this to ensure some part of DCS doesn't get shunted off to an E-core. Similarly, that's why for a primarily gaming machine I'd prefer 9800X3D and its predecessors over higher tier equivalents - they split the CPU between v-cache/no v-cache, requiring you to tell the CPU which process goes where. Ultimately, while either of those can provide a slightly better performance, for me at least the convenience of not having to deal with Process Lasso is worth sacrificing a little bit of clock. Also, you left out the bit where the v-cache itself provides a major performance benefit in DCS, probably more than enough to beat the E-cores. Even the humble 5800X3D punches well above its weight in applications involving heavy 3D and large textures, such as gaming. I'm not convinced that offloading background stuff from the main cores offers gains of quite that magnitude.
  24. For what it's worth, current generation Intels have durability problems, and they do not have anything like the X3D technology. A few tenths of GHz on the clock will help less than the copious amount of L3 cache on the X3D CPUs, particularly in DCS, where there's a lot of graphics to do. Also, while Intels generally have more cores, some of those are much inferior E-cores, which, if anything DCS-related ends up on them, will bog down the sim. AMD CPUs have the same total number of cores as equivalent Intels have good ones.
  25. Note that the 8K resolution won't do much for you if your hardware can't render it. Since DCS struggles even with 4K on top tier hardware, I doubt that we'll have a use for an 8K headset anytime soon. After the Vulkan version comes out and is optimized, it'll be a whole different ballgame, of course.
×
×
  • Create New...