Weapon Inertia & Sway Feedback (dev branch)

[13islucky 10]

Also, some of you that are struggling may have to lower your sensitivity settings, at least until you get used to it. It's incredible to see people do 180s in an instant and complaining that the guy that turned slower managed to kill them anyway. It is inertia, after all.

[machineabuse 11]

I wrote a ticket on the tracker with my thoughts on the Weapon Inertia system in dev build. Frankly the issue isn't that it makes or doesn't make the game harder, but the fact that it is an abstraction that contradicts 2 markmanship fundamentals; namely cheekweld and proper sight alignment. As far as I can tell deflection isn't currently modeled properly. For the most part, most shooting scenarios one will encounter in ArmA are not all that demanding.

What would be a real test of this system would be to see how it stacks up in a skeet shooting scenario without optics, an activity that in reality demands the shooter to track and lead fast moving targets smoothly while keeping the barrel and sights properly lined up.

Trap guns are both long and heavy and yet are some of the most agile pointing guns you will find. If this implementation of weapon inertia can serve that purpose then I'd say that its flaws can be overlooked.

[Bouben 3]

I wrote a ticket on the tracker with my thoughts on the Weapon Inertia system in dev build. Frankly the issue isn't that it makes or doesn't make the game harder, but the fact that it is an abstraction that contradicts 2 markmanship fundamentals; namely cheekweld and proper sight alignment. As far as I can tell deflection isn't currently modeled properly. For the most part, most shooting scenarios one will encounter in ArmA are not all that demanding.

What would be a real test of this system would be to see how it stacks up in a skeet shooting scenario without optics, an activity that in reality demands the shooter to track and lead fast moving targets smoothly while keeping the barrel and sights properly lined up.

Trap guns are both long and heavy and yet are some of the most agile pointing guns you will find. If this implementation of weapon inertia can serve that purpose then I'd say that its flaws can be overlooked.

What are you talking about? Are you aware of the fact that your sights are always perfectly lined up if you put them up in the game? The misalignment happens only when you hip fire.

[machineabuse 11]

What are you talking about? Are you aware of the fact that your sights are always perfectly lined up if you put them up in the game? The misalignment happens only when you hip fire.

Misalignment is when your eye, the rear sight and the front sight are not lined up. In effect, every time you turn with the "weapon inertia" system currently in Dev build.

Yes, I am aware that the sights are aligned when you raise them, no the sights are not always perfectly aligned when they are "put up" if you are turning and no, evidently misalignment is not exclusive to hip firing anymore.

I hope that categorically answers your questions.

[chortles 263]

Here you go:

Is something supposed to be wrong with the ARCO?

As I said before: I wanted a comparison/contrast between your and djotacon's videos with the same optic (and both modes thereof) on the same weapon. ;) That and it's also useful to see how a top-mounted "backup" collimator compares against a dedicated, standalone one. Thank you for your time!

@ Machineabuse: But they were going for an abstraction in the first place and said as much in the OPREP... "we decided to avoid blindly adhering to the specific mechanics of physical simulation" ("a more blinkered pursuit of 'realism', which mouse/keyboard peripherals can never truly provide") because "while we don't think that our mechanics represent the perfect physical simulation of real life weapon handling, we are confident that they enable authentic in-game situations, where skill, knowledge, and experience are vital." You read it and yet you went ahead.

Edited August 30, 2014 by Chortles

[Bouben 3]

Misalignment is when your eye, the rear sight and the front sight are not lined up. In effect, every time you turn with the "weapon inertia" system currently in Dev build.

Yes, I am aware that the sights are aligned when you raise them, no the sights are not always perfectly aligned when they are "put up" if you are turning and no, evidently misalignment is not exclusive to hip firing anymore.

I hope that categorically answers your questions.

You are right. I double-checked just now and have to apologize for misinformation.

[machineabuse 11]

You read it and yet you went ahead.

Indeed. Because I personally do not feel it represents an authentic weapon handling experience as is the goal of the system. Others are free to think otherwise and I wouldn't hold it against them or you for that matter :)

We all want the same thing in the end; for the experience to be a good one. I believe the current implementation has limitations that may hamper the experience when it comes to tracking fast moving targets; such as shooting the tires off a moving vehicle as it speeds past you in the street (without blindly blasting in the general direction of the tires.). Which is why I suggest that a good test of such weapon agility is probably skeet.

Again, if this system can be made to do that as a reasonable facsimile of what can be achieved in reality (i.e. "authenticity" right?) I don't really have any real further concerns about it.

You are right. I double-checked just now and have to apologize for misinformation.

It's all good bro :) It does say you stay in "a non-english speaking country" below your name so I just gave you a straight answer.

[chortles 263]

Indeed. Because I personally do not feel it represents an authentic weapon handling experience as is the goal of the system. Others are free to think otherwise and I wouldn't hold it against them or you for that matter :)

Thinking otherwise then. ;) Just keep in mind the devs' explicit differentiation between "authentic" and "realistic" and the former being the goal, not the latter (due to their claim of the latter being impossible with mouse/keyboard-only).

Which is why I suggest that a good test of such weapon agility is probably skeet.

Thankfully there are at least two missions to this effect for Arma 2/OA; they'd all need converting for testing Arma 3, but what the heck. :D

[roshnak 41]

Indeed. Because I personally do not feel it represents an authentic weapon handling experience as is the goal of the system. Others are free to think otherwise and I wouldn't hold it against them or you for that matter :)

We all want the same thing in the end; for the experience to be a good one. I believe the current implementation has limitations that may hamper the experience when it comes to tracking fast moving targets; such as shooting the tires off a moving vehicle as it speeds past you in the street (without blindly blasting in the general direction of the tires.). Which is why I suggest that a good test of such weapon agility is probably skeet.

Again, if this system can be made to do that as a reasonable facsimile of what can be achieved in reality (i.e. "authenticity" right?) I don't really have any real further concerns about it.

Out of curiousity, do you have a suggestion for an inertia system that would not potentially limit the player's ability to track fast moving targets? I only ask because the other two that have been proposed have the same potential.

[machineabuse 11]

Out of curiousity, do you have a suggestion for an inertia system that would not potentially limit the player's ability to track fast moving targets? I only ask because the other two that have been proposed have the same potential.

Well, I personally feel that lagging the weapon slightly behind the cursor is enough to do it quite well as it keeps the sights always aligned. The major advantage is that it properly simulates what occurs when you bring the weapon to an abrupt stop; you will overshoot and then track back onto the target. With heavier weapons that would be pretty noticeable. That's I feel is the better system BUT...

... I'm guessing that BI went with the current solution as it's camera based; so the weapon itself has no inertia but your POV does. (I know this from looking directly down at the stock using TrackIR) It's a much cheaper (in coder hours) and expedient way to get a "kind-of" weapon inertia simulation going without physically introducing motion to the weapon. The current math looks like it just tracks the velocity and direction of the mouse inputs. *Still love you BI ;)*

Being that is the case I would say that a compromise solution may work best; have it so that there is misalignment during the acceleration-deceleration(velocity/time) phase of the mouse movement and not merely velocity. So abruptly trying to break the inertia of the weapon at rest causes misalignment as does trying to bring the weapon to a sudden stop. So if you are smoothly tracking a moving target the sights should stay aligned even if you are tracking it pretty darned fast.

Edited August 30, 2014 by Machineabuse

[blackpixxel 53]

You are right - weapon inertia is just a shift of the camera, there aren't any new animations.

What I really like is that Holo- and Reddot-Sights are now a real advantage compared to ironsights, because they always show you where your bullet will hit, even when the camera is offset to the muzzle axis. So you should be able to track targets while turning with them much better than with ironsights.

[bad benson 1733]

Out of curiousity, do you have a suggestion for an inertia system that would not potentially limit the player's ability to track fast moving targets? I only ask because the other two that have been proposed have the same potential.

reading this made me think that it would be great if holding your breath would influence inertia. like when you try to track a moving target you would increase tension in your body and control breathing and the result would be more aiming accuracy. i've never fired any military grade stuff but i know that when i did skeet shooting and paint ball i instinctivly did that.

sorry if this has been suggested already i quite honestly didn't have the time to read all the new posts.

side note: i still hope there will be some visual inertia applied to the hip fire mode like a fraction of what was shown in that bug video where it was really overdone. it almost seems to me that it's already there and can be seen if you take a close look at the closer part of the iron sights but i think, if that is the case, it needs some kind of multiplier for the hip fire mode to make it more visible.

it's not so much a gameplay thing but i feel liek this would contribute immensely to making you feel the weight of the weapon a lot more and improve the whole "feel" side of things.

[blackpixxel 53]

it almost seems to me that it's already there and can be seen if you take a close look at the closer part of the iron sights but i think, if that is the case, it needs some kind of multiplier for the hip fire mode to make it more visible.

A problem is that you don't have any drawbacks with the current inertia implementation when usnig hipfire mode. You will still hit where the crosshairs are pointing at, there is no missalignment because the inertia is just a simple parralax camera movement. I am still not sure whether a BF3 like weapon-behaviour would be a better inertia system. in BF3 / 4 the weapon always rotates a bit slower than yourself in hipfire-mode, but in this game the bullets spawn in your eyes, so it does not influence you, it is just a nice visual effect.

Maybe they could implement something like this for Arma (Of course the bullets will spawn in the muzzle there). And maybe only in hipfire mode, since ironsight + slow weapon would feel very unresponsive. The technology to do this is already in the game. Just add an offset of freeaim/floating zone when turning tdepending on the speed and the weapons weight, and the weapon will rotate a bit slower than you.

[machineabuse 11]

Now that I've had a bit of time to think about it; the inertia system pitches the camera in the opposite direction to the movement of the mouse while simultaneously banking the camera in the direction of motion to achieve the misalignment.

I would venture that reducing the bank to be less than what is half of what it is currently while increasing the rotation deflection in the same ratio would make iron sights feel a long way towards being right.

I feel like the sensation that ought to be achieved is your POV catching-up-to-the-gun as opposed to the gun limp-pivoting in the view :)

[John Kozak 14]

Tried the new system - it's definitely an improvement.

There is still a small problem with non-3d scope sway, however. When the sway moves the aim point to the right and you move the scope a degree to the left, the moment you stop movement the gun instantly continues to move right, as if there was no inertia. If I move the gun sharply to the left, I would expect it to drift in approximately the same direction after.

Essentially, it seems, there are two different sets of velocity/acceleration, with just the resulting positions being summed.

P.S. Maybe I'm repeating a point already stated - skimmed the thread perfunctorily, possibly missed it. If that's the case, sorry, just my feedback.

Edited August 31, 2014 by DarkWanderer

[chortles 263]

P.S. Maybe I'm repeating a point already stated - skimmed the thread perfunctorily, possibly missed it. If that's the case, sorry, just my feedback.

From what I recall, "they" knew of it but went ahead with the dev branch release anyway -- didn't want to wait on 2D scopes before getting feedback on the rest of the implementation, I guess? :)

[blackpixxel 53]

The 2D-drift issue is not really something special for the 2D scopes. It is also there for 3D scopes and ironsights. When you turn and stop, the weapon will move a bit further in your rotation direction. This doesn't affect your own view, it stays there and just the gun moves a bit. With 2D scopes your view is fixed onto the weapon. And this is the reason why your camera is also moving a bit further.

[John Kozak 14]

The 2D-drift issue is not really something special for the 2D scopes. It is also there for 3D scopes and ironsights. When you turn and stop, the weapon will move a bit further in your rotation direction. This doesn't affect your own view, it stays there and just the gun moves a bit. With 2D scopes your view is fixed onto the weapon. And this is the reason why your camera is also moving a bit further.

Well, the problem is, with 2d scopes like SOS, DMS, etc. it actually drifts into opposite direction, unlike "3d" RCO/ARCO/ACO.

[machineabuse 11]

With even the 3d scopes the "inertia" system no longer applies because the narrower FOV will result in less misalignment relative to the amount of optic/scope/sight on screen. Whatever secondary system they are using for 3D scopes and 2D optics hasn't yet been updated with any behavior congruent with how unmagnified sights work.

[bad benson 1733]

A problem is that you don't have any drawbacks with the current inertia implementation when usnig hipfire mode. You will still hit where the crosshairs are pointing at, there is no missalignment because the inertia is just a simple parralax camera movement. I am still not sure whether a BF3 like weapon-behaviour would be a better inertia system. in BF3 / 4 the weapon always rotates a bit slower than yourself in hipfire-mode, but in this game the bullets spawn in your eyes, so it does not influence you, it is just a nice visual effect.

Maybe they could implement something like this for Arma (Of course the bullets will spawn in the muzzle there). And maybe only in hipfire mode, since ironsight + slow weapon would feel very unresponsive. The technology to do this is already in the game. Just add an offset of freeaim/floating zone when turning tdepending on the speed and the weapons weight, and the weapon will rotate a bit slower than you.

pretty much agree on all of this. to be honest i think the whole "actual inertia, aka slower turning, would cause a detachment from your avatar"-argument from the sitrep thingy is just an excuse for doing it like they did as a visual trick, since as you described, other games that have more elaborate animation based implementations don't make you turn slower (afaik) either. it's not about the whole body, it's about your arms and the weapon dragging behind.

i too wish they would use a deadzone based approach. like changing the way the blend mask for the deadzone works right now and give the head/camera bone less weight than the weapon and arms. but still. i'm also happy with a the current approach as long as something is done about hipfire. i'm not even talking about "penalties" being the crucial part. it's just, like others have said, that now the fact that only ironsights/scopes have the effect creates a much bigger contrast to the gun movement in hip fire mode and thus makes it look more robotic than before.

[devilslayersbane 28]

Back on topic, fatigue in arma before the bootcamp update was a joke (unless you had a mod). Now it's something to think about, and worry about. Even when I play with it turned off, I still worry about it now. It's made me a smarter, better player and I'm fucking proud of that. This is something that I can use in real-world scenarios (I'm going into the military after college) and so it's nice to know that even if it's not entirely accurate, I have something to base my thoughts off of when it comes to fatigue and gear. Having a game without bugs is impossible, and the Real-virtuality engine has been known for it's bugs for a long time. I, personally, am glad that arma 3's release was relatively bug-free compared to arma 2's release. I also feel like the changes since release day have been very positive, even if they're not fleshed out right the first time. Fatigue, imo, still needs to be toned down slightly (like, just a sliver), but to say that content is more important when having more than one person inside that content breaks it, is ignorant of how much time it takes to make this content.

Edited September 1, 2014 by MadDogX
removed off topic quote + response

[machineabuse 11]

I feel that weapon inertia should scale with fatigue. The inertia values for weapons up to 4kg should be close to zero when you are fresh and be really bad when you are say at 70% fatigue. That at least accounts for the limp behavior. Any weapon greater than 4kg should have weapon inertia ramp up on an exponential curve.

I also feel that inertia should be worse while prone to represent that positioning the weapon becomes more laborious when you have both elbows on the ground.

Retrospectively, weapon inertia isn't the greatest name for this mechanic :)

[bigpickle 0]

I also feel that inertia should be worse while prone to represent that positioning the weapon becomes more laborious when you have both elbows on the ground.

Prone is the most stable fire position, its the most comfortable too, was taking you seriously till you said that.

[machineabuse 11]

Bigpickle; yes the most stable but the least agile. Weapon inertia isn't about stability, it's about limiting the ability to rapidly make gross movements of the muzzle and get on target quickly.

If me thinking that turning 90 onto a target is somewhat more difficult while prone is comedic then I apologize :) but I'm inclined to assert that sway is ArmA 3s expression of positional stability.

Edited September 2, 2014 by Machineabuse
typos. stupid touchscreen keyboards.

[maddogx 13]

Bigpickle; yes the most stable but the least agile. Weapon inertia isn't about stability, it's about limiting the ability to rapidly make gross movements of the muzzle and get on target quickly.

If me thinking that turning 90 onto a target is somewhat more difficult while prone is comedic then I apologize :) but I'm inclined to assert that sway is ArmA 3s expression of positional stability.

Rather than inertia, the prone stance is the one instance where I personally think a limited turn speed (even with the resulting "negative mouse acceleration") would be appropriate, because being able to do a split second 180° turn while prone is plain ridiculous. Just my opinion.