Armed Assault weapon ballistics

[frederf 0]

It's weird that it only calculates the deceleration of the bullet from its initial speed on not its current speed. No amount of air friction should make a bullet stop and turn around the other way obviously.

If a bullet only lost 10% of its speed this would make a good approximation with reality, but if the bullet loses 50% of its speed during flight then it's not.

[Cormega 0]

I wonder why they didn't just use the ballistics engine from VBS. It claims to be 99% accurate to real life, why not just use that? I don't understand "dumming" the game down, assuming this is what happened.

[max power 21]

Interesting fact: The force of drag on an m80 7.62 bullet is 70 times the force of gravity at mach 2.5.

The calculations based on bullet init speed are no doubt to save the work of making the processor check the bullet speed every instant, decreasing the overhead of the operation. But, so long as the initial speed is sufficient, the bullets will never turn around in the air.

[majoris 10]

Is there any way to just slow down the bullet in the config?

It's not realism, but it's probably a little bit better than the red lasers my M4 has been spittin' out lately.

[Iron+Cross 0]

Quote[/b] ]if it is done through engine instead of extra script, the engine load will not be too much

When you say engine do you mean, in the CFG file of the ammo/weapon?

or in the actual game engine? if so how would one put it in the engine? (if poss)

& if the ballistics Engine in VBS2 is still in ARMA how would we activate it or do we do this through our Config files for our addons? (sorry im always the last one to catch on to a joke)

Quote[/b] ]Another thing to consider is how well the AI corrects for it.  If the AI can't hit anything at distance because of the correction, I don't know how useful it will be to actually implement.

AHHHHH I think this happend with ICP's RPG's the AI couldent compensate & it was a comprimise between the AI & the Iron sights for player... this is maybe why 2d sights are better in some cases....??? or would it be possible to ajust for this?

if so im not to shure could some one take a crack at explaining this? (sorry im so retarded i know..    )

p.s how the hell do i add names to qoutes?

[frederf 0]

The calculations based on bullet init speed are no doubt to save the work of making the processor check the bullet speed every instant, decreasing the overhead of the operation. But, so long as the initial speed is sufficient, the bullets will never turn around in the air.

1. No doubt to save CPU time.

2. The bullet doesn't have to turn around in mid air to be wrong though. Constant deceleration vs speed dependent deceleration will never match at all ranges. If you assume the constant deceleration equal to the maximum deceleration then the bullet will be going far, far too slow at range. Conversely if you assume a moderate constant acceleration then the bullet is going too fast at first and still too slow at long ranges.

If we could get a linear, quadratic, or arbitrary function deceleration we could get better and better results. Of course I know you know this Plantiff.

If we're talking about rifle bullets than we can pick some reasonable range to "zero" the constant deceleration, say 500m or wherever 5.56 switches from its "lethal" to "Less lethal" modes, then at least that effect will be present. If we're talking about artillery then realistic trajectories are not really that important. You lob the shell into the air, see how far it goes, and make a firing table that only works in the wonky physics in game.

What would be funny is using a constant deceleration model that the artillery round might backtrack right back to the firer.

The AI issue is a significant one. I'm not entirely sure how the AI corrects for bullet drop right now. If they do it in an editable, robust way then there's no reason they can't be "reeducated" for a more complex ballistic model.

Here's an example of adding a name to a quote. If you press the quote button next to the post you want to quote it gets added auto.

<table border="0" align="center" width="95%" cellpadding="0" cellspacing="0"><tr><td>Code Sample </td></tr><tr><td id="CODE">

[fireship4 0]

Im not so sure it works the way I said it does anymore. I've done a few experiments and changing sideairfriction doesnt seem to do anything - the bullets seemed to have been turning around just because they were fired at low speed (5 m/s) so maybe there is an arbitrary air resistance factor. I'll keep trying.

EDIT It seems I have uncovered, a detective I am, something lame. It looks like the bullets follow an arbitrary parabola, and the flight of a bullet is not computed at all. Here is my evidence:

http://i11.photobucket.com/albums/a155/fireship4/tracer.jpg

Try it yourself - just set the initspeed of a magazine for any weapon to 5 or something, making sure the gun has tracers. The bullet will fly upwards out of the top of the gun and hit the ground a few meters in front of you.

Im guessing there is a diff parabola for the different simulations - ie shotshell, etc.

[max power 21]

This is probably because the rifle is still zeroed for 500m. Try zeroing it at 0 m.

[fireship4 0]

No, zeroing moves the sights as far as i know - I know you mean in game but i had thought that was what how they did it there too.

If you think it is this, maybe you are right, but I have never seen a command to change zeroing of guns.

So instead of discovering something good, Zoidberg has DESTROYED EVERYTHING AAAAAAAAAAAAAAAAAAAAAAA

[max power 21]

The distance for which the rifle is zeroed is called 'distanceZoomMin / distanceZoomMax' in Cfgweapons. These values should be the same if you don't want the rifle zero to change at different zoom settings.

[frederf 0]

If a very slow bullet seems to head straight up and straight down then it's rather clearly prescribing the vertical and horizontal speed profiles separately.

[Iron+Cross 0]

at this stage of advanced talk (remember were not all rimmers)

all i can say is.. fireship4 i take it you like futurama?

[fireship4 0]

yah. And red dwarf. Who doesn't.

[max power 21]

If a very slow bullet seems to head straight up and straight down then it's rather clearly prescribing the vertical and horizontal speed profiles separately.

That's possible too, but I think also it's possible that the algorythm that tries to find the correct trajectory breaks when the bullet can't possibly make it to the range at which it's supposed to arc through the pipper again. We'll have to wait for more information before we can verify that.

[fireship4 0]

Ok, seems setting the zeroing to zero fixes the up and down problem, sill unclear whether airfriction makes a difference though. Testing...

[max power 21]

I told you so

[fireship4 0]

Can't seem to get sideairfriction working. I've tried loads of stuff, including trying to use the shotmissle simulation instead. I have no idea how i managed to make the bullets come back towards me, im starting to think I imagined it.

[TheElite 0]

bullets dont aknowledge side airfriction in arma engine i thought ?

to get them back it is the deflection angle 180 degrees.

i had fun with it dodging my own tank rounds and law rounds some time ago .

[tj72 0]

Are the bullets in Arma using a particle engine or something object or expression based?

Could a new type of bullet particle be made that can access custom arrays perhaps?

[Iron+Cross 0]

any code examples or u keping that under wraps?

[fireship4 0]

TheElite, no I didnt set the deflection to 180, and they werent fired at a wall anyway, but into the air.

[kestrel7e7 0]

All you realism fanatics are playing the wrong game.

VBS2 is where it's at.

[Rustman 0]

All you realism fanatics are playing the wrong game.

VBS2 is where it's at.

True..but for some it's far better to buy the 40 to 50 dollar program and attempt to quasi-program it themselves than to buy the $1500 program with it already built in. I can't blame them for that at all. $1500 is a unbelievable price to pay for any software. That being said...I bought VBS2.

[fireship4 0]

I think Arma may turn out better with mods than VBS2...

[dslyecxi 23]

All you realism fanatics are playing the wrong game.

VBS2 is where it's at.

Good luck trying to find large groups of players in VBS2. The price of admission alone will prevent that from happening on the civilian side. I know that if I were to switch to VBS2, I'd lost 98% of my group in the process. It doesn't matter how much realism there is when you aren't able to play with the people you want to play with.