Leaderboard

What is it? This mod aims to address two specific issues with the current implementation of contact dots in DCS: Dot size is always one pixel, which means smaller dots at higher resolutions, and that the lower your resolution, the easier it is to find contacts. Dots are rendered too far. As long as an objects model is not culled, it will get a dot drawn on it. DCS' draw distances can go out to >40 miles, making it possible to see the dots of an aircraft before even your radar can pick it up. This is accomplished by adding these new rules to how dots are rendered: The size of the dot gets bigger with screen resolution, using the reference resolution of 1920x1080. The dot becomes fully opaque at distance of ~6 miles. The dot is completely transparent at a distance of ~18 miles. When between those two extremes, the dot will fade at an exponential rate. Download The mod is available on the DCS user files here: https://www.digitalcombatsimulator.com/en/files/3330454/ How to install Unpack the "Bazar" folder in your DCS World root directory. This will overwrite the "dots.fx" file. By default, this is located in "C:\Program Files\Eagle Dynamics\DCS World". As of the May 18 2.8.5.40170 openbeta patch, this NO LONGER PASSES IC. FAQ So what does this actually mean? 1. It means that dots are no longer visible from extreme ranges. You won't see planes 20-50 miles away. 2. Playing a high resolution (in this mod, defined as a screen height of >1080) you should have roughly the same "dot acuity" as somebody playing at 1920x1080. This corrects the common practice in DCS of people reducing their resolution to make the dots bigger. Are ground vehicles easier to see? Generally, no. At high resolutions, the dots become rectangles, since half the 2x2 dot is now under the ground. The dot also fades with distance the same as aircraft (the shader can't tell the difference), so you won't be able to spot vehicles from 30 miles like you could before. Why did you pick the ranges you did? Experimentally gathered data, from a paper that is infamous on this forum and won't be named, found that T-38s were spotted around 4-6 miles, depending on conditions. Knowing where the target already is can boost this detection distance by about 5 miles. Based on that information, I tweaked the opacity values and formula such that I was able to consistently find a dot around 4-6 miles, and could find and track distant dots (which are faded to ~50% transparency by this point) at around 10 miles, when I already knew where to look. Are dots completely invisible past 10 miles? Not completely invisible, but they are extremely faint. You'd have to already know they were there to find them, and they are very easy to lose. By about 15 miles they are basically impossible to see. Do I need to turn on labels to use this? No, the dot system in DCS is completely independent of labels. There is some confusion around the label system having a couple "dot" settings, but what those do is draw a label with a little . over the target. Labels are (at present) not obscured by clouds or the cockpit frame. The dots in DCS meanwhile are a completely separate function and do not interact with the label system at all. They are always on, cannot be turned off, and cannot be modified by players without messing with the shader itself as this mod does. To eliminate the chance of confusing labels with the dots, I recommend turning off labels completely when testing this mod. What makes 1080p special? In my experience, the dot size at 1080p resolutions is big enough to be useful, but not so big that it becomes distracting and strange looking. Therefore it was chosen as the reference resolution for which I wanted higher resolutions to have parity with. Does this mod do anything if I already play at 1080p? The only difference you'll notice is that contact dots fade away as they get further. Does this mod work on ultrawide? Yes. The only thing the mod takes into account for sizing the dots is the vertical resolution. Does this mod address the ability to see dots through clouds? No. I did make a brief attempt to see if I could fix that, but it's likely something that Eagle Dynamics will have to fix themselves the correct way. If I zoom into a dot that's far away, does that cause it to fade in? No, the dot opacity is based on a hard distance calculation. FOV has no effect on the opacity of dots. Does this mod address the exploit of raising the FOV to max in order to enlarge dots? It does not. The old impostor mod I made many years ago did this, but I wanted to keep this mod as simple as possible. I might add it in later if there is demand. As with the impostor mod, I'd likely just fade the dots over some field of view. What happens if I play at a resolution under 1080p? Dots will still increase in apparent size at resolutions lower than 1080. Initially I wanted to either try and make the dots "subpixel" by rendering them appropriately smaller at low resolutions, but this isn't feasible without engine changes. I also tried fading the dots proportionally when under the reference resolution, but this created problems with flickering models. The biggest advantage that low resolutions used to give was seeing dots from tens of miles away. Since the dots are now guaranteed to fade with distance, I figured it was best to just let the low resolutions keep their slightly larger dots. Why can I see missiles now? Unfortunately the only information I have to work with in the shader is the object's position. The shader is completely unaware of the model that is under it, so all the logic applies equally to all visible objects, regardless of their size. For objects larger than the average fighter, this works itself out and isn't an issue. However for smaller than average objects such as missiles, this can look a bit strange. Fixing this requires Eagle Dynamics to provide more information to the shader such as an object size. Does this mod pass IC? As of the May 18 2.8.5.40170 openbeta patch, this no longer passes IC. Update v1.1 See the below link for more information.

Sorry, but when in a Rift S I can instantly see a fighter at more than 30 miles because there is a big fat dot appearing, something is wrong. Irl spotting a jet at this distance is insanely hard, borderline impossible if you don't know where to look. Meanwhile people at 4K have difficulties spotting anything at more than a few miles. Knowing that ED is satisfied with that is unbelievable to me.

Eagle Dynamics "modular" dedicated server installer can now be downloaded for free from our website. Stable https://www.digitalcombatsimulator.com/en/downloads/world/server/ The main differences from previous all-in-one installs are: Supports installation and uninstallation of the terrain modules No longer requires special command line arguments. To start the server you just need to run: .\bin\DCS_server.exe NOTE: the update, repair, install and uninstall operations are handled by .\bin\DCS_updater.exe application as usual. Switching the existing installation To switch the existing install of the dedicated server to the new-style install you need to execute the following command line once: .\bin\DCS_updater.exe update "@dcs_server.release" This will switch the update channel of your dedicated server to the new one and convert the list of modules from the single DEDICATED_SERVER one to separate modules. NOTE: switching back to the all-in-one installation is not supported. Installation of a module To install a module currently you have to use the following command line in the server install folder: .\bin\DCS_updater.exe install MODULE1_id MODULE2_id ... Example: .\bin\DCS_updater.exe install SYRIA_terrain MARIANAISLANDS_terrain Uninstallation of a module To uninstall a module currently you should use the command line as well: .\bin\DCS_updater.exe uninstall MODULE1_id MODULE2_id ... Example: .\bin\DCS_updater.exe uninstall CAUCASUS_terrain List of available modules Module name Install/Uninstall id Supercarrier SUPERCARRIER WWII Units pack WWII-ARMOUR Caucasus CAUCASUS_terrain Nevada Test and Training Range NEVADA_terrain Normandy 1944/2 NORMANDY_terrain Persian Gulf PERSIANGULF_terrain The Channel THECHANNEL_terrain Syria SYRIA_terrain Mariana Islands MARIANAISLANDS_terrain South Atlantic FALKLANDS_terrain Sinai SINAIMAP_terrain Kola KOLA_terrain Afghanistan AFGHANISTAN_terrain Iraq IRAQ_terrain Germany Cold War GERMANYCW_terrain Marianas WWII MARIANAISLANDSWWII_terrain

I don't understand that... there are obvious flaws with current ED implementation, and a guy makes a mod which, even if not perfect, improves things a lot, and ED is like "we don't plan to change anything". Just... why? It's minimal development time, and would improve one of the most important aspect of air combat for WW2 and early cold war!

Iskander SSM 1.1.1 released! Changelog Version 1.1.1 Fixed 9M723 collision model against projectiles Fixed 9M729 collision model against projectiles Bastion-P LBASM 1.2.1 released! Changelog Version 1.2.1 Fixed 3M55M collision model against projectiles Fixed K340P model disappearing in certain angles Fixed Monolit-B model disappearing in certain angles Fixed Monolit-B model missing rotating plate between radar and radar mount Project 22350 Admiral Gorshkov Frigate 1.4.1 released! Changelog Version 1.4.1 Fixed 3M14T collision model against projectiles Fixed 3M54T collision model against projectiles Fixed 3M55M collision model against projectiles Project 22800 Karakurt Corvette 1.4.2 released! Changelog Version 1.4.2 Fixed 3M14T collision model against projectiles Fixed 3M54T collision model against projectiles Type 052D Destroyer 1.2.1 released! Changelog Version 1.2.1 Fixed YJ-18 collision model against projectiles Fixed CJ-10 collision model against projectiles Type 055 Destroyer 1.3.1 released! Changelog Version 1.3.1 Fixed YJ-21 collision model against projectiles Fixed YJ-18 collision model against projectiles Fixed CJ-10 collision model against projectiles Type 45 Destroyer 1.2.1 released! Changelog Version 1.2.1 Fixed NSM collision model against projectiles BTR-4 IFV 1.2.1 released! Changelog Version 1.2.1 Fixed RK-2 Barrier collision model against projectiles Arleigh Burke Flight III / Ticonderoga CMP 2.2.1 released! Changelog Version 2.2.1 Fixed SM-6 AShM collision model against projectiles M142 HIMARS 1.3.1 released! Changelog Version 1.3.1 Fixed ATACMS collision model against projectiles Fixed PrSM collision model against projectiles Fixed PrSM AShM collision model against projectiles NASAMS 3 SAM 1.0.0 released! Changelog Version 1.0.0 Release version And by the way, I updated the site. Removed everything unnecessary and put the release log right in the center.

This is a very frustrating response to something that is clearly broken. There is no reason spotting dots should outperform radar, and there is no reason for spotting to be easier or harder depending on your screen resolution.

I'll check out the mod, we can put it forth to the team even as an option if everyone things it is worthwhile. Thanks! PS I added a poll on here, if you have tried it, please vote. Thanks.

I can spot where this thread is going... Because I fly in VR

Welp back to playing at sub 1080p for massive dots you can spot 30+ miles out rip The fact that you can spot planes before your radar at lower resolutions definently sounds like something working as intended

I don't think spotting is fine until we can get resolution parity. It cannot be intended that the optimal way to spot targets at a distance is to reduce your resolution below native. This mod has shown that we CAN get resolution parity, and I think it's worth acknowledging that maybe at least that aspect of this mod should be the default behavior. It seems like a clear improvement to spotting; even if the team says spotting isn't broken doesn't mean it can't be improved.

During testing of the C-RAM I found a rather annoying bug. Apparently a lot of my antiship and cruise missiles are imperviable to projectile hits from weapons like CIWS, machine guns etc. I've solved the issue and will soon be releasing fixed versions of all assets affected by this.

This is a very old image, but the dot system hasn't changed since I took this screenshot. I think that's bad.

Newy, wrong file. This is dots.fx. Can ED at least improve the dots before removing our ability to alter them? Maybe implement the mod as a permanent change so that playing at a higher resolution isn't a negative?

Damn, I almost thought it had been released this weekend. I refresh the DCS News page numerous times/day to see if this has been released yet, and DCS teased me a couple of days ago with a news item saying "Get Ready For Normandy 2!" Well, I am ready, and I'm not rich, but I could not care less how much it costs, because whatever they're charging is too little. I mean London? AND Paris? Holy #@$#$ing #$@$@$! Can you even imagine how time-consuming and tedious that has been. And I still love Normandy 1. Just the other day I found Mont Saint-Michel ... couldn't believe they had it modeled in detail. As for me, I'll re-buy Normandy 1, Re-buy Channel 1, Re-buy Persian Gulf, Re-buy South Atlantic, lose my existing missions, lose my virginity again (yeah ... "again" ... that's my position and I'm sticking with it), dance at your third-cousin's wedding, and eat my own pinky toe if that's what it takes to get this released. Name your price, I'm paying it ... even if I gotta shake my money-maker to do it. Please, take my money. Soon. The sooner the better. Thank you.

Here's my experience with this mod in VR with a Valve Index at full resolution with no anti-aliasing. Setup I have a mission set up with co-alt F-5s at 0.25nm, 0.5nm, 1nm, 2nm, 4nm, 8nm, 16nm, 24nm, and 32nm. They're offset by a small angle at each distance to give visual separation. Please see the attached mission. I start the mission and close the briefing by clicking the X at the top right, instead of clicking Fly, to keep the game paused while I look around. Observations Without the mod and without any zoom, the 0.25 and 0.5nm contacts are clearly visible models. The 1nm contact is a very flickery model/dot. Contacts at 2, 4, 8, and 16nm are faint dots. I cannot see the 24nm contact or any beyond it without zoom. With VR Zoom, the 1nm contact becomes a clearer model, and the 2nm contact becomes a flickery model/dot. The 24nm contact becomes a faint dot. With VR Spyglass Zoom, the 4nm contact becomes a flickery model/dot. The 32nm contact becomes a faint dot. With the mod without any zoom, the first 0.25 and 0.5nm contacts are clearly visible models. The 1nm contact is a very flickery model/dot. So far no change from without the mod out to 1nm. Contacts at 2 and 4nm are clearly visible dots, more clearly than without the mod. The contact at 8nm is a fainter dot. With VR Zoom, the 1nm contact becomes a clearer model, and the 2nm contact becomes a flickery model/dot. With VR Spyglass Zoom, the 4nm contact becomes a flickery model/dot. The 8nm contact remains a faint-ish dot at all zoom levels. Conclusion Overall, I like the mod. However, I would much prefer dynamic scaling to smooth the transition between dot (or scaled-up model) and model. DCS's dot to model transition point makes spotting very challenging as the model blinks in and out of existence, and this mod does nothing to address this existing DCS behavior. That said, this mod is a step in the right direction. spotting-test2.miz

Yep, that's by design.

Maybe, but not anytime soon as it's super low priority. Reading some of the "reviews" of this update is amusing. The placebo effect is real and I'll leave it at that.

Agreed, amazes me how ED can think dots and spotting are not broken while they completely are, especially on higher resolutions. And what fascinates me, is that these are long lasting problems which would require a pretty easy temporary fix while waiting for a definitive solution (like the bomb splash damage) or even better, have already been patched up by the community. I'd rather have slightly too easy spotting, than the situation we face now where you can literally lose track of an object in front of you just because it has passed through your canopy frame for a second (problem is particullary bad in the Mig-21 and F-14 for example) or because you glanced at your instruments. And I don't buy the "spotting is hard IRL" thing, since living near an airport makes it pretty clear to me how "hard" spotting but more notabily tracking airplanes is.

New vid:

It's a dream come true to see ED taking interest in this topic, as the current spotting situation is the single biggest issue affecting those with high resolution displays! I play multiplayer at 1080p on a 4k display because of the spotting issue, and whenever I jump into a single player campaign and crank the resolution back up I'm astounded by how beautiful this game actually is! My preference would be to keep the resolution scaling aspects of this mod but discard the more aggressive culling. 18 miles is too short a distance for aircraft to vanish completely - pilots like Chuck Yeager claimed to be able to see enemy aircraft that were 50 miles away (although this was probably in a best case scenario with the sun reflecting off them). In any case, 18 miles is much too short and would have a dramatic effect on gameplay in servers like Enigma Cold War. In summary, fix the resolution scaling issue which causes a massive disadvantage at high resolution and leave the rest as is!

In my mind the ideal solution from a user experience standpoint would be to have a series of sliders in the special options to adjust the TDC sensitivity on all formats: HUD ATTK AZEL SA HSI TPOD GRID Because although the new speed does make it easier to do things like bump ranging on the ATTK format, its going to be absolute hell to try and enter MGRS with it, at least unless you have a force transducer for your slew control. It seems that instead of trying to cater for all controllers and ending up with a one size fits none solution, ED should just let the player decide what they want, especially when it comes to issues like this that are wholly dependent on hardware being used.

I'm speechless.

Tests have shown that changes in the signal strength of the incoming radar do not have any effect on the rwr system. This happened in the last patch before this function worked. Test conditions: Two aircraft irradiate each other with airborne radars and change the distance between them. Test results: The RWR system refuses to perceive the signal strength completely. RWRbroken.trk

Just want to throw out there my 2 cents. Using a Pimax 8kx (so 3840x2160 roughly per eye) I cannot spot anything until it is roughly 3 miles away on average. That is pretty much unusable. A friend of mine uses the Quest 2 and can always see me at 10+ miles. This makes even forming up a challenge for me and he has to talk me on. Going for the best picture possible should not result in something tantamount to punishment. (Edit: I know it isn't actually punishment and not meaning to be hateful in anyway. Just a very frustrating user experience) EDIT 2: After playing with the mod for a few hours here are my thoughts, Outside of 4~6 miles, there is little difference. Inside of that there is a huge difference, I am able to spot things off at the edges of my peripheral vision that I previously wouldn't have been able to see looking straight on. I will do more testing tonight and see if my opinions hold up.

This has nothing to do with the labels. It affects the dot. I don't think dots in isolation are a particularly good solution to spotting, but they're all I have to work with. The old impostors were much better, since you could read a silhouette and aspect from them, but we all know how that went. They were flawed, but the problems were totally fixable. It cannot. It's only a shader.

Das freut uns natürlich. Wenn du uns jetzt noch sagst, was die Lösung war, können wir auch noch was draus lernen.

This is just SCANDAL that for all those years our beloved dev team done...not enough in this most important matter for combat sim!!! In real life spotting is dependent on light, weather (humidity and clouds) and distance. In DCS on...dots :(. For WW2 spotting is EVERYTING. Come on ED... Please.

Hello everyone, it's been a while since a new version of the program was released, mainly because I didn't find time to further code it. Lately however, I have found time to actually build a new version, with a fully remade interface that implemented all your requests so far. The changes include: New user interface with features like renaming waypoints, changing elevation and reordering (check the video below for more details) Support for exporting/importing files with waypoints Added keybinds for VR users Fixed FPS issues and other bugs Added support for the F16 Sufa mod and the Ka50 3 Please keep in mind this is a first release, please PM me any bugs or issues, and we'll get it sorted. Here is a video showing all the new features: To upgrade to this version, head over to the release found here Download the zip file from there. Then, extract it and replace your existing `TheWay.lua` file in Saved Games with the new one. Otherwise, the new version won't work properly. Then, run the installer, and that's all. If this is your first time using TheWay, follow the full install instructions here: instructions As always, thanks to all the people who've helped me so far with developing and testing. This is the way.

Mate, this is a dream coming true! Tally is the most frustrating aspect of DCS for me and you are on the rightest track... But I'd like to work with you on this as I think there's room for improvement! Some folks with experience tell me that a medium sized fighter can be seen at 15 NM on a good day (with effort...). I have tested the mod and it is a great improvement on a HP G2, but I think you can be bolder and bump up the visibility even more. I would, for example, keep 100% opacity up to 10 NM for further testing. Also, I have had strange instances where a jet in front practically disappeared at 2-3 NM (see example below). Anyway, care to work on this together? I am on the DCS Discord, please get in touch. Again, very well done!

"correct is as" -- how is this possible if the official manual completely denies this? Could you post some documentation on which you are making this module?

Outstanding work! Hopefully this becomes an option integrated into the base sim (or at least some iteration of it). Thanks Why485!

For me, it feels much more wobbly in the hover and slow speeds. I can still hover fairly easily but getting there is like trying to stay still while being submerged in water. Before the patch cyclic movement would produce instant desired aircraft reaction. Now, I need to counter my cyclic deflection to stop the aircraft movement as going neutral would not suffice (just like in space!). I'm not saying it's incorrect! Just different and it'll take time to get used to. One other thing I found much more problematic is strafing sideways. The Apache develops awkward roll movements that quickly get out of control even at speeds of 30kts even though it was fairly easy to go sideways at 50kts before. Again, this could be true to the real thing... Oh and now I need to push the cyclic slightly forward from the centre to stay in hover after a vertical takeoff (that centre of mass shift?).

Interesting and well done proof of concept. This mod shows that there is a lot that can be done to shift spotting in a more realistic direction, imo, and highlights what people have been saying about DCS spotting for years (regarding issues related to how dots are rendered and how it leads to unrealistic behavior). Overall this reflects findings that lead to the use of smart scaling in other professional/military and consumer sims, so it shouldn't really be surprising.

CCRP/DTOS/delayed CCIP can release at any angle (45, 60, 90, 120). The problem is that if you are near the maximum toss range and anything goes wrong which reduces toss range then you will get no solution because at no point is the projected impact close enough to the designated target to allow a release. You see that the solution cue gets really, really, really close to the FPM during the maneuver. You were just outside of toss range the entire time. Notice the "44" for predicted release angle which appears both before and after (but not while passing) 44 degrees. The max toss cue is based on a lot of assumptions that you may or may not comply with and let's be honest it might not be 100% accurate. It should also be noted that 45 degrees is not the maximum toss angle if the release point is above the target in elevation. I know we learned 45 is maximum in grade school but it's only correct for the same elevation beginning to end. If you try to toss something from 10,000' to 0' a 45 degree angle will have less range than a smaller angle. The toss range of 35 degrees is very very close to the maximum (like 90%) and has much better tolerance to ensure release. Plus you spend less time up above the deck getting shot at in seconds. In a perfect world one should be able to meet the assumptions exactly and release at 44.999999 degrees but that prediction is perfect and/or that pilot will achieve agreement with assumptions exactly are not high confidence.

Yes, that's one of the many reasons that I dislike Discord a lot ... pity that traditional Forums are dissapearing from the Internet.

This is a great change. You release so much content, this will help keeping up.

It's a almost perfect mod that DCS players needed desparately at the moment. I am an not a fighter pilot but as an airline pilot, I can tell it's easier to spot an aircraft in reality than in my 2560x1440 screen. I do think the mod makes missiles too visible, the dot is almost as big as an aircraft when viewed form certain distance. Overall it's a good start to improve the DCS spotting if ED allows and include it in future update.

I humbly ask that you really press this issue home with the team and make sure that they understand this is going to be another PR stick in the eye if you simply lock a change like this behind IC without doing something to address the underlying issue its trying to fix. Taking some time to examine this mod in detail and potentially including it as a temporary improvement over the current single pixel system would go a long way towards making the community feel heard when they bring up issues that are low cost to fix and implement but they go unacknowledged by the team at ED.

Disclaimer : Not implying I know better and with respect, but I think that statement is not correct (or I misunderstood what you are saying). First of all, rapid/firm collective movement is never really needed during take off. Quite the contrary I'd say, slow and steady collective movement works best, especially when you're still learning the Apache behavior. The rotor rpm high comes from to fast/much collective, not from too little. @bradmick explains it very well in this thread here: @VR Flight Guy in PJ Pants; My first thoughts, when looking at your video, were: You might be a tad late on the cyclic correction (left) and then overcompensation when things start to go sideways (literally ) You might want to pull your collective more gently (looks like you might have some stiction going on there, as your collective inputs appear to not be very gradually). Best way I found to learn, is to gently pull collective until she gets light on her wheels and then, with wheels still touching the ground, make very small cyclic & rudder corrections to keep her steady/straight. (then again, it's very difficult to draw conclusions from a video. Also, as mentioned above, double check your controller axis) fwiw; I couldn't see your rudder input in the video, but I tend to give some slight left rudder input while increasing collective. At some point, you'll know how much just by feel and it'll all go automatically.

Mods will still work, but they should not be passing pure integrity check. I have mentioned spotting to the team already. thank you

We're working on it. The current problem is a consequence of fixing (or trying to) the previous oscillation issue.

I've never once touched this switch. What does it do?

DCS: Mi-24P by Eagle Dynamics Added: Special option for AP authority over pedals (if disabled AP won't be able to take control over anti torque pedals and move them without pilot input, although 9% authority over yaw control is still there. Disabling it making this system not realistic (as real system AP has full control over whole pedals axis) but due to simulator and hardware joysticks reasons we are putting it in) Would it be possible to ask to put a description next to the new option saying that it is activated for realism? Would be like this. PEDALS AUTO MOVE (On for Realism) This would prevent many people who seek realism from getting confused due to lack of knowledge. Thanks @BIGNEWY!

The Apache to me has strange hover characteristics compared to the other DCS helos. I would describe it as too much momentum (energy) post axis input, not enough damping. Like its large belly mass was not considered for stability. So it rocks back and forth, and sideways like a boat with a very tall sail, drifts and 'capsizes' (rolls) occasionally. Trying to transition from enroute speed into a hover (not those santised IGE hovers) with rudder input to face the desired enemy location, feels at times like this pic.

Thank you @currenthill- we've waited long enough for a C-RAM!

AviaStorm tornado dev update:

Been flying sims for 25 plus years, and DCS for the past 2 months.(only fly warbirds) I find DCS the most realistic WWII sim by far, I also fly real planes and one of those a warbird trainer, last week I offered here in the forums to take any DCS pilot who lives in WA state for simulated dogfight (have done the same with IL2 and taken a few people) this time no DCS pilot showed up, so took a friend. After flying DCS Warbirds for 2 months constantly, I have a good feel for most online dogfighting and have seen the impossible jerking maneuvers, which kills immersion. During my dogfight simulation last week, I tried to compare the real event to DCS, FMs and DM are pretty good in DCS. In VR checking six is pretty realistic, a bit easier than being strapped to a 4 point harness IRL, getting the head out of the cockpit is not (VR limits would be nice) Force feedback its something that will make things closer to real. And the most important is pilot fatigue, even in a simulated dogfight you start to get tired, and the speeds are between 200 and 350KPH, DCS warbirds are a lot faster and heavier, so more demanding, some sort of pilot fatigue red out blackout should be modeled to avoid those impossible jerky maneuvers, because really ruins immersion, the pilot is superhuman, blackout happens and its good, but jerking the plane positive Gs then negatives time after time repeatedly , its not realistic, we never see these in real WWII footage.