Aluminum Donkey

Holy smokes man, you like doing everything the hard way, don't you? $5 USB flash drive. Back up your files. Remove drive and place in desk drawer. Done! AD

I use a desktop stick (TM Cougar, similar to Warthog) and no curves at all. Feels great to me--the F/A-18 is fine with linear curves. The trick is to get over the tendency to "ham-fist" the plane around (excessively fast & coarse control inputs). Not that I do any aerial refuelling, I'm not that good :) I usually end up shot down before needing more fuel, and if I'm not, I have no missiles left anyway :) AD

If I really could only have 1, it would be the Su-25T, because the price is right, and the TF-51 doesn't have any gunz :) I'm sure there more than a few DCS buffs who spent all their bread on the computer itself, and only fly the '25T :) AD

Sure, it's called the Getting-Close-Enough-Before-Firing missile mod :) AD

Weta43 seems right to me. The nosewheel has trail--the wheels themselves are behind the steering axis, so when you crank them over and they skid, they produce more drag on the side opposite your steering input, and the plane turns the other way. Of course, doing this in real life would destroy the tires, which is why they don't do this in real life. But, the real F/A-18C may very well behave like this. Oh man, lots of discussion over this little detail... Let the ED dudes work on stuff that matters :) AD

Shibby, I have a similar system to yours, but I'm using a GTX 1070 vid card with 8GB of video memory. If you absolutely can't/won't fork over the cash for another SSD, then put Windows on your mechanical hard disk, and wipe the SSD and put DCS World on that instead. It's the only way it'll work decently. You also might want to put your Windows pagefile on the SSD as well. Not absolutely necessary, but it'll be faster. DCS needs so much transfer speed it isn't funny, and running it from an SSD is the only way to go. Windows will work fine on the hard disk. You're also using a 4GB video card which is very marginal, so a much heftier one is in the future unless you run on all low textures. AD

The Mi-8, and the UH-1H Huey, are by far the hardest to land, being conventional helicopters (I don't have the Gazelle, but it's probably a handful too). Everything else is a total milk run compared to the helicopters with tail rotors. Even carrier landings :) AD

Is DCS installed on your hard drive? If so, move it to your SSD if you have enough space. DCS constantly loads huge graphics textures while flying around, and the mechanical hard disk isn't really fast enough to keep up. Really bad stutters (long pauses) are a result of running out of RAM, and DCS will start swapping huge graphics textures to your hard disk pagefile. You should post your system specs in your signature (CPU, memory, graphics card) so we can easily tell if there's an inherent system limitation. Finally, DCS is always a work-in-progress. Even the 'release' version. It's just the way it is, and I'm not a programmer, but the sim seems to have major shortcomings in how it manages assets and memory. It's been that way for years and will probably never change, so you pretty much have no choice but to tolerate lots of small stutters during combat. So, post your specs and maybe someone here can help you out. AD

I usually just line up first with the gear and flaps down, play with the throttle to get the speed I want (roughly 130 knots), and finally, mess with the pitch trim until I hold a roughly constant pitch. Then mess with the throttle to keep the right speed. Then land that puppy. That's all there is to it :) AD

You don't need to. SSAA works/looks better than MSAA, so it uses more graphics card power than MSAA, and gives lower framerates. Running both at the same time is totally unnecessary.. AD

It isn't a problem, and as such, you can't 'fix' it. More expensive computer. ;) AD

There's an easier way to check engine thrust. A 1G acceleration is 9.8 m/s^2, or about 35 km/h per second (22 MPH per sec.) Load the aircraft lightly with no external stores, set autopilot to straight and level, fly at a low airspeed to minimize the effect of drag, and hit the gas. If the plane gains about 35 km/h per second, then it's thrust is about equal to its weight. Seems about right to me, any opinions? AD

It's an entertaining module to fly. Nothing like landing a narrow-span delta fighter with the wing loading of a steam locomotive to remind me of what a dreadful virtual pilot I am :) AD

They really, really have to sort this out. Switches and buttons that work like the real thing are nifty and all that, but it's a *flight* sim, and the flight model feels very incomplete to me whenever I start putting some gusto into the stick. I'm also not sure what kind of airplane it's supposed to be simulating :) Since the flight characteristics change wildly from time to time with various updates, it seems like they can't decide what kind of plane they're simulating! The problem with this is, when I'm flying the module, I'm always thinking, "Ok... this is kinda cool, but so what? Is this really a half-decent MiG-21 simulation, or a half-arsed one, or a half-brained one, or a half-baked one, or a... you know. Stability changes. Roll rate changes (a lot!!). Damping/oscillation tendency changed a lot. One iteration, it flies like a FBW aircraft. Next one, it's wobbling around almost like a WW2 fighter, minus the P-factor! Can't M3 agree for once on what a MiG-21 actually flies like, finish the flight model, and release it once and for all, instead of messing around for ages? AD

Greetings, The 21 is a nice module and all, but seems to have problems with its flight model. From the "digital stall" effect, to the lack of roll inertia, to the general lack of consistency across various DCS updates--it seems to fly like an entirely different aircraft every time DCS gets updated. It makes me wonder if the real thing flies anything like the DCS module, despite Leatherneck/M3 LLC's marketing to the contrary, including their bragging about having a real MiG-21 pilot on the dev team :) So, is the MiG-21 still essentially a pre-release version, despite having been available for quite a while now? Is it still being refined, or is it essentially abandonware? In other words, is M3 LLC still in business at all and working on finishing their so-close-yet-so-far module, or is what we see what we get, too bad so sad and thanks for your cash? AD

That certainly sheds some light! So, the data bar at the bottom in external views shows the real, airflow-over-wings AoA, and the cockpit gauge shows an arbitrary, wildly inaccurate and somewhat useless reading that at least still varies directly with actual air-over-wings AoA :) Thanks for the posted link... makes sense to me now. Still, considering how much time and effort (and bread!!) they pour into developing a fighter jet, you'd think the cockpit AoA gauge would be calibrated/corrected to show a usefully accurate value. It's not like it would have been difficult or anything. But, if this is the way the real MiG-21 AoA indicator behaves, then it's certainly the way the simulated gauge in DCS should behave too. Imagine if the speedometer in your car showed 100 km/h when you're actually going 50 :) AD

I'm sure this has been brought up, I just don't remember much about it... The AoA gauge in the cockpit reads twice the value of the info bar. So, external view says 10 degrees, cockpit gauge says 20 deg. Any idea why it was done this way? AD

You got it! Trim nose down for continuous level flight, but when you're engaging, trim up again and just hold the stick forward if you're going fast. That way you'll be able to reach critical AoA/max G. I'm pretty sure that's how real 109 pilots did it. AD

You're right. I've noticed this myself many a time. Hardly a real dealbreaker, but it's there. AD

Sub2K, I forgot to ask you something! When you overclocked your CPU, did you OC only the cores, or did you also OC the cache? Reason I ask, is that there are separate multipliers for the cores and the cache memory on a CPU. If you aren't sure, go into your BIOS settings and make sure you're also running the cache as fast as you can get it to go while being stable. If your cores are running at 4.4 GHz (which is excellent) but the cache multiplier is only, say, 28 or 30, it won't improve matters much. It might take more voltage than you're normally used to, but your CPU can probably handle 1.35V to maybe 1.4V without getting too hot. Let us know what happens :) AD

You have to select a weapon first--just selecting AA and Master Arm on won't do much. Configure your weapon select switch first, try selecting the gun and pull the trigger. See if it fires. If not, there's something else to sort out first (such as making sure your joystick trigger is configured properly--I only config the second stage, not much use for the first.) AD

Oh man, much as I love the new content, the good folks at ED are gonna have to get off the bowl and iron all this out soon. It's aggravating a *lot* of people, especially ones that spent more on their computers than I ever spent on a car :) AD

Congrats on your nice new card :) Dang man, hope you can get it all ironed out. Have you removed all your old graphics adapter drivers with something like DDU (Display Driver Uninstaller) and installed the latest ones for your new card? In your Nvidia Control Panel settings, in DCS World, is your power profile set to Prefer Maximum Performance? Is your Vsync turned off in Nvidia Control Panel? Sometimes that messes with things. Wish I could afford a 1080 Ti ;) Still, I'm getting far better framerates than that (50+ FPS) with my wimpy lil' GTX 1070, so it's not a DCS limitation. AD

Yeah, I just wiped out my NVMe drive and I'm reinstalling DCS... doesn't seem to make much of any difference. Imma stick with the Release version for now instead of OB, but haven't noticed much improvement yet. I did it just to try it. Now I've got 100 GB of stuff left to DL! :huh: AD

A better way of putting it would be that although ultra-realistic modelling is certainly possible, it wouldn't be allowed for use in a public-release entertainment sim. That's what I meant. Instead, we get a pretty good representation of how things generally behave. It's easiest to do with "generic" items, such a solid lead slugs or cannon shells that have no guidance system, which is why I maintain that for best realism as far as weapons performance goes, you're better off sticking with guns :) Radar, guided missiles of all varieties, and even the effectiveness of chaff & flares on various missile types is the kind of info that's not only hard to get, but is a political hot potato in some respects. For example, the latest Sidewinder X and AMRAAM D missiles are incredibly expensive, and if they're either extremely good or extremely bad at rejecting countermeasures, no-one will want the general public to know about it. We just assume the latest ultra-expensive Western equipment is un-beatably good, and the rest of the world is scrambling madly to catch up :) But you're right, their general flight performance such as acceleration, top speed at a given altitude, turning ability etc. can be reasonably accurately guessed based on fairly straightforward physics. AD