BI Developer REQUESTED for Ballistics Question

[Spartan0536 189]

Hello BI Team, I really really need your assistance here. I have been following your formula to the "T" regarding the calculation of Airfriction values in ArmA III, however it seems like I am missing some critical information. I am using JBM Ballistics for my external ballistics calculator, I am 100% accurate on my ballistics research on the round in question, there is 0 fault in my information.

What I need to know is the following....

1. What elevation do you use in your calculations?

2. What barometric pressure do you use in your calculations?

3. What temperature do you use in your calculations?

4. What relative humidity in % do you use in your calculations?

5. What is the atmospheric density set to in your ballistics calculations/calculator? (ie. 0.07509 lb/ft3 [cubed])

No matter how I calculate this I end up getting more drop than I should according to JBM regardless of distance, and either slower speed or faster speed than I should at any given distance outside of zero. This "anomaly" has to be attributed to my ballistics calculator not having the correct atmospheric density set correctly, this is the only thing I can think of outside of an engine error in terms of calculating aerodynamic drag and gravitational pull which should be a standard 9.8067 m/s2 [squared]. Thanks for any and all help, once I can get the data needed I can re-write all my work and ensure proper accurate ballistics for real projectiles in ArmA III, this will not only impact me but the BI modding community as a whole.

Edited May 24, 2015 by Spartan0536
Edited #5 to accurately reflect the question, there is no atmospheric density in ArmA III.

[x3kj 1247]

try 9.81 m/s² or 9.805 m/s² gravity acceleration, because thats the 2 default values used in most simulations.

Also, this is a game and not a dedicated bullet simulator - that means that the time steps executed for bullet calculation are limited, therefore inaccuracies might be caused by this alone. It also means that the physics used are reduced to describe the flight with reasonable accuracy. Aerodynamics are very complex, you won't see this fully modelled in a game, whereas i would expect it to be done in a ballistics simulation tool.

Edited May 24, 2015 by Fennek

[Spartan0536 189]

try 9.81 m/s² or 9.805 m/s² gravity acceleration, because thats the 2 default values used in most simulations.

Also, this is a game and not a dedicated bullet simulator - that means that the time steps executed for bullet calculation are limited, therefore inaccuracies might be caused by this alone. It also means that the physics used are reduced to describe the flight with reasonable accuracy. Aerodynamics are very complex, you won't see this fully modelled in a game, whereas i would expect it to be done in a ballistics simulation tool.

Here is the issue, BIS describes how to factor Airfriction on their Wiki page, it states in a section that you need to match the drag coefficient from a REAL ballistics calculator to the airfriction value in the game, this must be how they get their values, if so then I need to know what they have the atmospheric density set to in their calculations. No external ballistics calculator worth anything would not incorporate atmospheric density, it would defeat the purpose of the calculator in its entirety. Thus my questions still stand, its nothing complex, and if this is the way BIS gets their values then they must be using an external ballistics calculator, that implies that they have an atmospheric model in mind when they produce their numbers. Now if they do not and they just randomize their values I would be very disappointed but also not entirely surprised.

[x3kj 1247]

i'd think they just eyeball it with some testresults they found somewhere... because that's the most reasonable thing to suspect from a game developer imo.

In addition to that - yes, you can enter ballistics calculator values. That doesnt mean however that if you enter the exact value from a ballistic calculator, that you will get the same end result in the game.

Maybe they go at greater lengths in VBS...

[Spartan0536 189]

VBS is not a BI product, they are a completely separate entity.

Edited May 24, 2015 by Spartan0536

[x3kj 1247]

i know that...

Rule 1 of simulation (that everyone that teaches simulation will tell you): Every simulation is flawed. They only differ in how much they are flawed.

Rule 2: The guideline for the simulation should be to be as accurate as necessary and not to be as accurate as possible.

[Spartan0536 189]

Well I have been doing some extensive testing, it seems like BI is not using 9.8067 m/s2 as their gravitational pull, this is very disturbing, however I believe I am getting closer to figuring out their atmospheric density, or so I hope....

No dice here, and in reference to atmospheric density I am referring to an external ballistics calculator, not in game.

Edited May 24, 2015 by Spartan0536

[bakerman 247]

RV doesn't use elevation, air pressure, temperature or anything of that nature. Only drag and gravity are used for shell ballistics, these forces are applied on every tick/frame.

i know that...

Rule 1 of simulation (that everyone that teaches simulation will tell you): Every simulation is flawed. They only differ in how much they are flawed.

Rule 2: The guideline for the simulation should be to be as accurate as necessary and not to be as accurate as possible.

^ This. A thousand times this.

[Spartan0536 189]

RV doesn't use elevation, air pressure, temperature or anything of that nature. Only drag and gravity are used for shell ballistics, these forces are applied on every tick/frame.

^ This. A thousand times this.

Bakerman, we have been over this, I completely understand that ArmA in its entirety has no bearing on atmospheric density or its properties, however what I am referring to is in the external ballistics calculator that BI uses to gather their numbers just like I do with either Hornady or JBM and now as I am looking into others which may be more accurate. I am just trying to follow their work as closely as possible so that I can transpose my data in the same manner they do theirs just with real data thus making it give an authentic experience. I am at the 99% threshold of what ArmA's base engine can do and I intend to squeeze that last 1% out even it its effects might only alter something 1/2 a %, I can't leave it alone until its done right. That being said it also seems like we discussed to POSSIBLY be an error with the community ballistics calculator, I can't wait to see how much work you have done. If it was not for you Bakerman and Olds I would not have gotten my work off the ground with any legitimate claims what-so-ever, you guys are the foundation to which I have built my work on, thanks a million times over.

Addendum: I am editing my original post, it seems in my lack of sleep I screwed up on #5 asking for the Atmospheric Density in ArmA III, which DOES NOT EXIST, I mean't to say in their ballistics calculations/calculator as that makes a difference.

Edited May 24, 2015 by Spartan0536

[blu3sman 11]

No way you can get within 99% accuracy. Drag functions in ballistic calculators take into account Mach effects, which in Arma world means that airFriction should change with projectile velocity.

Since airFriction is a constant, no amount of fiddling with it will solve anything.

[Spartan0536 189]

No way you can get within 99% accuracy. Drag functions in ballistic calculators take into account Mach effects, which in Arma world means that airFriction should change with projectile velocity.

Since airFriction is a constant, no amount of fiddling with it will solve anything.

I was not stating 99% accuracy in terms of a ballistics profile but in terms of doing everything I can to replicate how BI gets their values, using their methods to ensure an authentic experience, knowing the atmospheric density that they use in their external ballistics calculator will help me achieve that. The 99% threshold is in reference to how far I have gone to make sure I am following BI protocol in regards to how ArmA III "simulates" ballistics. However once finished I claim and will still claim that my ballistics are within a 5% threshold of the real deal, in some cases like pistol/smg bullets I can get it dialed in to about 3% maybe slightly less as the ranges are considerably shorter.

[yourimag1nat1on 10]

Sorry to hijack your thread, but could a Dev give a clear explanation of how Armor and Passthrough while they are at it, people have lots of different explanations, i would like an official one.

[pettka 694]

The air friction computation is mostly based on ballistic tables in case we have enough data (which is actually in most of cases) or some light-weight simulation in case if we lack some data (which means we use ideal conditions for the simulation). Worst case scenario is some heavier approximation based on closest results we have (it's pretty hard to get any solid data for underwater ammunition of SDAR for example). It may possibly disappoint You that there is a lack of heavy-weight simulation with various elevations, pressures and temperatures, but that is possibly beyond the threshold You have mentioned - more detailed data simply would not change enough in our simulation to justify resources spent :icon_twisted:

Sorry to hijack your thread, but could a Dev give a clear explanation of how Armor and Passthrough while they are at it, people have lots of different explanations, i would like an official one.

A bit oversimplified version is that if an object gets hit, the hit point gets damage according to ((hit of weapon) * (actual bullet speed)) / ((typical bullet speed) * (armor of the object class) * (armor of the hit point class)) and the total damage of the object is this value multiplied by (armor structural of the object) * (pass through of the hit point). There are several more factors taken into account, but this should be roughly enough to understand the properties. The bold parts are actual config properties :icon_twisted:

[bakerman 247]

A bit oversimplified version is that if an object gets hit, the hit point gets damage according to ((hit of weapon) * (actual bullet speed)) / ((typical bullet speed) * (armor of the object class) * (armor of the hit point class)) and the total damage of the object is this value multiplied by (armor structural of the object) * (pass through of the hit point). There are several more factors taken into account, but this should be roughly enough to understand the properties. The bold parts are actual config properties :icon_twisted:

Thanks for posting!

Edit: I missed the brackets :eek:

Edited May 25, 2015 by Bakerman

[Spartan0536 189]

Well then after my deliberations as of last night (of this post) I will alter my work to use ICAO/ISA atmospheric densities as a standard. Now all I have to do is wait for Bakermans bad ass ballistics calculator so that we can iron out any possible inconsistencies that may currently exist between JBM and ArmA (within reason of course).

[Jackal326 1181]

Thanks for posting, but is that correct? Looking at the following example of a simple bullet hitting a person the damage values seem exaggerated:

hitPoint damage = 8 * 800 / 800 * 2 * 2 = 32

total damage = 32 * 7 * 1 = 224

Unless I'm reading this wrong, your math is off:

8 * 800 / 800 * 2 * 2 = X

6400 / 3200 = 2

So then it would end up 2 * 7 * 1 = 14, not 224

That is unless I've missed some ( )'s out denoting the orders of multiplication...

[pettka 694]

Unless I'm reading this wrong, your math is off:

8 * 800 / 800 * 2 * 2 = X

6400 / 3200 = 2

So then it would end up 2 * 7 * 1 = 14, not 224

That is unless I've missed some ( )'s out denoting the orders of multiplication...

You are right, despite of me being an idiot. Obviously, the total damage should be divided by armor structural, not multiplied :icon_evil:

And there are two more things to take into concern:

• these rough approximations of the equation are in ideal state of hitting directly the hit point. Actual impact is on the fire geometry and distance from the point causes some interpolation to zero at radius of the hit point
  
• there is additional armor of vests / helmets for their respective hit points. As far as I remember, it multiplies the hit point armor
  

Still, this is a rough approximation, there is a bit more behind that (energy lost by the bullet on impact is converted into spheric volume which covers some spheric volume around hit point, the volume covered part is actual damage), but it should be pretty enough to know what's going on :icon_twisted:

[Spartan0536 189]

You are right, despite of me being an idiot. Obviously, the total damage should be divided by armor structural, not multiplied :icon_evil:

And there are two more things to take into concern:

• these rough approximations of the equation are in ideal state of hitting directly the hit point. Actual impact is on the fire geometry and distance from the point causes some interpolation to zero at radius of the hit point
  
• there is additional armor of vests / helmets for their respective hit points. As far as I remember, it multiplies the hit point armor
  

Still, this is a rough approximation, there is a bit more behind that (energy lost by the bullet on impact is converted into spheric volume which covers some spheric volume around hit point, the volume covered part is actual damage), but it should be pretty enough to know what's going on :icon_twisted:

pettka, I was able to work some "magic" as it were to figure out what a standard atmospheric profile is for ArmA III....

BI it seems has the speed of sound set to 330 m/s in the game according to the latest base sound files, now anyone with any decent basic science education can tell you that the speed of sound changes with atmospheric conditions which is based on temperature. So in lieu of this I ran some calculations...

I use what ISA proclaims as the standard atmospheric density at sea level which equals......

0.07648 lb/ft3 @ Sea Level

this equates to

-2.1 degrees centigrade

0% relative humidity

0.0m in altitude (hence SEA LEVEL)

Barometric Pressure of 28.147 Hg

ArmA III's speed of sound is configured for 330.0 m/s, using this atmospheric density with these parameters will give you those exact results, this is ArmA's ambient environment. THIS IS NOT STATING THAT ARMA HAS AN ATMOSPHERE IN GAME, THIS IS ONLY FOR USE IN EXTERNAL BALLISTICS CALCULATORS!