Realistic Ballistics

[NonWonderDog 0]

Now if someone would just fix the ridigilous bounce values of hand grenades.

I've halved the "deflect" value of the hand grenades in the ballistics mod, but I didn't really test it. They should bounce less often, at least, even if they continue to bounce just as crazy high. 40 mm grenades should bounce a bit less often, too. As to bayonets: they could probably be fudged, but I can't do the models or animation.

As to AI: their efficiency should be unaffected by the small arms ballistics. The weapon aiming code is pretty smart, and distanceZoomMin/Max are not fooled by airFriction changes. Since the AI uses the same kind of routines, it has no problem. All the assault rifles still have point-blank ranges of 350 m anyway. I haven't tried picking on the AI from 600 meters or more, but I never did that in the stock game either.

The rocket and missile ballistics may cause problems for the AI, particularly the AT4 and RPG-7v. I think it might be something to do with the "maxSpeed" parameter, but the AI consistently aims too low with these. I've limited the AI to shots within 250-300 meters with these, so they at least have a chance of hitting something.

I don't even know what happens if the AI tries to use a TOW missile. I anticipate bad things.

As stated in the comments, this will only affect weapons that use stock ammo. Most addon weapons should use stock ammo, as far as I'm concerned, but I'm sure there are some that don't. There is no formula I used to come up with the values, unfortunately, so I can't give much advice. I just used trial and error until the game trajectories and real trajectories overlapped.

If you want to use that range card in mils, just multiply everything by (5*pi/54). I'll update the second post with one in mils in a bit. The one in the SightAdjustment topic won't work right, as it's for the stock ballistics.

Bergmania: I set the RPG rounds to self-destruct after 4.5 seconds, at which point they've reached 915 m. They're launched at 115 m/s like they should be, and they have about one second of thrust time. They might have a bit too much impulse, but I couldn't find much hard info on the rockets that wasn't written by Americans in the 60's. The tandem-charge rockets won't go nearly as far.

I tried to model the trajectories on PG-7V rockets and PG-7VR rockets, because that's what they're called. The game actually has the 85mm PG-7V rockets do as much damage as 93mm PG-7VL rockets, however, and the 64/105mm tandem PG-7VR rockets actually look like PG-7VL rockets when they're on the laucher, even though they magically turn into PG-7VR rockets in flight. It's a mess.

By default the rockets were launched at a ridiculous 5 m/s and had an incredible amount of thrust in order to make up for it. They only went 700 m, too, since the drag was set so high. To make matters worse, the tandem-charge grenade (which weighs at least 75% more) had the exact same trajectory. They were just wrong.

The trajectory is still way too flat, because the grenades are modeled in ArmA as having active control surfaces! Nothing I can do about that, I'm afraid, short of scripting the entire flight.

[Törni 0]

Quote[/b] ]As to bayonets: they could probably be fudged, but I can't do the models or animation.

It is strange that close combat is missing from a tactical first person shooter. All it would have needed: hit with a fist/knife, stab with a bayonet, hit with a weapon. Pistol could equal hitting someone with a fist.

There are zillions of guns, but not the very basic way in which humans fight. Well maybe in Game 2...

[Tactician 0]

The rocket and missile ballistics may cause problems for the AI, particularly the AT4 and RPG-7v. I think it might be something to do with the "maxSpeed" parameter, but the AI consistently aims too low with these. I've limited the AI to shots within 250-300 meters with these, so they at least have a chance of hitting something.

IIRC, maxSpeed is the key to get AI to use rockets correctly. Ran into the same problem with WGL. I tweaked the maxSpeed by trial and error.

[wolfsblut_ 0]

... At 800m, it will take about 8 hits to kill someone. ...

Great work this mod. But this is a bit unrealistic i think...

[twisted 128]

this is a great mod. thank you.

however i think it could go further. by this i mean be a source of REALISTIC airfriction and speed values for a wide range of calibers.

i have edited the many weapon mods i use to reflect the realistic values you have here. but some i had to make up but if you have a list of values then it would be a great help (especially for other common calibers like that used in ak47, LUAPA, etc)

[CameronMcDonald 146]

Marry me, mate, this is excellent.  

EDIT - If you'd care to generate some values for .32 and .45 ACP, as well as .44in Magnum, I'd be massively pleased. But good job anyway mate!

[Biggus 0]

Marry me, mate, this is excellent.

EDIT - If you'd care to generate some values for .32 and .45 ACP, as well as .44in Magnum, I'd be massively pleased. But good job anyway mate!

What are you making in .32, if you don't mind my asking, Cameron?

[TrunkzJr 0]

Another Question. Does this mean that AI will be even MORE Accurate then what they already are in long distances?

[CameronMcDonald 146]

What are you making in .32, if you don't mind my asking, Cameron?

I'm not making anything, mate, but I just needed values to apply to Tiger's vz61 Skorpion, which fires .32 ACP.

[guttersnipe 1]

Great addon, much entertainment to be gained from the use of the properly adjusted range card & binos.

Just a quick editing quetion  - how do I use the range adjusted weapons in the editor : when I adjust a soldiers loadout using the regular addmagazine/addweapon commands the range adjustment ballistics no longer applies (ie only works with the default dolsier loadouts) - is there a way to call the ballistically corrected weapons? cheers

*edit*

traced the problem to a corrupted file during re-pbo of the ballistics pbo - my bad, sorry!

[max power 21]

I'm concerned with the bias you have going on here.

You're taking rifle values from the latest 5.56 rounds and pitting them against Russian bunk from the 70s. Then you take empirical American .308 data and plug it into a calculator and get a value that messes with your world view, so you artificially tweak it. I think that it would be better to use the best evidence you can find for making decisions, rather than the 'America's weapons are awesome, the calculator lies' school.

That said, I'm very pleased to see someone adopting such a rigorous attitude towards this kind of game content. Being able to use weapon sights properly gives an extra dimension to gameplay and rewards the serious user. Excellent work.

I have a ton of dispersion and other ballistic data that I have accumulated. I would like to compare it with yours, if you are at liberty... not so that I can pick apart your mod, I just want to see a source that's different than mine to see how well they agree. Where can I go to find your ballistic dispersion data?

[Q1184 0]

I personally have no problem with this '70s russian vs modern US ammo as the '70s cartridge (7N6) is still widely-spread in the army. The new cartridges are supplied in small amounts in few units, so they are still quite rare. As for the .308 vs 7.64x54 thingy it indeed looks strange, maybe the author should  consult more sources to find out whether it's the calculator error or something else.

Btw since the 7.62 ballistics now correspond to sniper ammo, does this mean the medium machineguns like PK, M240 and their tank variants also use match grade ammo? That's a bit weird

[max power 21]

I personally have no problem with this '70s russian vs modern US ammo as the '70s cartridge (7N6) is still widely-spread in the army. The new cartridges are supplied in small amounts in few units, so they are still quite rare. As for the .308 vs 7.64x54 thingy it indeed looks strange, maybe the author should consult more sources to find out whether it's the calculator error or something else.

Well said.

[jerryhopper 286]

hey this is nice to see!

i have seen an sight adjustment earlier! that could be handy here i guess?

http://www.flashpoint1985.com/cgi-bin....t=65706

[Sevan 0]

this sounds great!!!!!

I've always noticed the ballistics in the game were a little off. Nice realism choice to model soviet ammo as if it is from the 1970s. I say it adds to the realism, because it's not unusual that the SLA would be using 70s era Soviet surplus.

I like playing as the SLA, and even if it gives a slight disadvantage. I always perfer realism over balance.

[big 0]

hey this is nice to see!

i have seen an sight adjustment earlier! that could be handy here i guess?

In his second post he mentions it, including a cheat sheet for the diff weapons in MOA and mils.

On topic now, really interesting stuff there. I first noticed your addons cause Alpha Squad used them

But couldn't find a "proper" topic for it till now.

[Jinef 2]

This community needs more people like you NonWonderDog.

I have always been looking for solutions to ArmA's unrealistic engagements through the use of AIdispersioncoeff and dispersion; I was hoping to make humans and AI to just throw more bullets downrange. The revelation that damage is affected by speed is really quite promising. More realistic firefights with a lot more bullets and suppression are my goal, as well as the need for higher echelon support units for the rifle platoon. In open combat, rifles and machine guns are quite innefective at killing people, they can just keep people in one place long enough for a tank or artillery to get some heavy firepower onto it. Whereas I imagine this would make urban combat the meatgrinder it is in relation to open warfare.

Very good news, now we just need CE3, Arty and the Zeus server back.

I will be very interested in this for implementation into a gameplay/realism modification we are doing. If you have no objections of course.

[chops 111]

Yes! Yes! Yes! This is what I'm talking about! More stuff like this! Fantastic work!

I don't like the term realism as I would qualify as an A grade desk jockey in any real military. Immersion is what I'm after!

More technical stuff like this repels the "Why can't I jump?" crowd too

[NonWonderDog 0]

Back from vacation, and there are a couple things to clear up.

I have no magic formula for this. I wish I did. I have no idea what kind of crazy units airFriction is actually in (it's negative, so it's presumably some multiplier of velocity). As such, I cannot "generate" values for any specific caliber. Nor do the values I've come up with follow any obvious internal logic. 9x19 needs significantly more airFriction than the slower 9x18 to match the calculated trajectories, for instance, even though 9x19 should have a better ballistic coefficient. I think this is because ArmA is crap at transonic drag, but I don't really know the reason.

All I did was match game trajectories to calculated "real" trajectories by trial and error. I could do the same for any bullet with enough data, I suppose, but the more data you can supply the less I have to slog through.

Specific answers:

I designated the 7.62 "notracer" ammo as match-grade ammo. This ammo is only used by sniper rifles, and sniper rifles only use this ammo. The M240 and PKM have entirely different, quite poor, ballistics.

Information about new Russian 5.45 armor-piercing ammo does not exist on the English-searchable internet. I don't know the designation of these rounds either, so I can't search for it in Russian. If you happen to know, say, the projectile weight and G1 ballistic coefficient of these rounds, I would happily model them. Otherwise there's no chance. The "crap from the 70's" is still in very widespread use in the Russian army, however, and I doubt very much that our little podunk island would have stuff the Russian army can't afford. (That doesn't mean it's not crap ammo, of course. It's still crap ammo. The Russians are very dissatisfied with it.)

Eight rounds of 5.56 to kill someone at 800 meters is probably unrealistic, but just think to yourself that the ceramic plates in the body armor are stopping the first six rounds. What's really happening is that ArmA is rather shortsightedly scaling damage based on kinetic energy (or velocity squared), and 5.56 has realistically lost 70% of its velocity and 90% of its kinetic energy after 800 meters (the stock bullets would only lose 40% of their velocity or so after 800 meters, which is wrong). Damage should be similar to default settings within normal engagement ranges, and it's nearly impossible to hit someone at 800 meters with an M16, so it's not too much of a problem. 7.62 might be a problem at longer ranges, since I didn't spend as much time with it. The .50 caliber rifles should now be deadly at two miles or something, and capable of disabling light vehicles in five shots or so; don't worry about them.

After looking further into the M24, I believe it should have a maximum range of 1250 m with the scope set to +70 MoA elevation, and that's what I'm trying to calibrate the ballistics to. This requires a ballistic coefficient very slightly higher (about 2%) than that provided by Sierra for the 125 grain MatchKing. I might have increased the ballistic coefficient (only by 3%, not an unbalancing amount) in the current version for the wrong reasons, but I'm pretty sure I was on the right track.

I took a few bits of information that I believed to be true, such as labeled "test" data (especially from Aberdeen) and photos of bullet-drop compensation reticles (an absolute bitch to work the math out on, so I might have made mistakes), and worked from there. I had to adjust a few BCs to fit data I do have, but I didn't really think twice about that. Most ballistic coefficients of commercial rounds are heavily exaggerated anyway, and they never specify the ballistics formula they used. On top of that, the standard G1 ballistics formula is incorrect at long ranges for boat-tailed rifle bullets anyway. With my ballistics it's certainly more accurate than it was, and I don't think you should expect too much more than that.

I have now changed the SVD ballistics after some more research. I had originally taken the values from the test data at 7.62x54r.net, but I just read his test methods...and I'm not going to take any more data from that site. Most other sites quote a BC of about 0.4 for the 7N14 load, which would make it very much inferior to M118LR at long range, but slightly better than standard M80 Ball.

Actually, now that I've found photos of the PSO-1 reticle, I've reverse engineered the ballistic coefficient from that. That gives me a ballistic coefficient of 0.387, which is actually a lot worse than I expected. I've already implemented this for the next version. The next version of the SVD won't shoot nearly as flat, but it should fit I now know of the PSO-1 cams and range chevrons to 95% accuracy.

I also finally found data on the Mk 262 Mod 1 ammo the Mk 12 SPR is supposed to shoot. I'll fill the 20 round magazines with that in the next version, and I'll try to make the 20 round magazines a bit more rare. That should make the SPR rather more useful as a sniper rifle, bringing effective range up to 500 meters or so. Again, Sierra's claimed ballistic coefficient of .372 doesn't fit the data -- a BC of .359 or so is much closer to the data I've found at gives me a believable maximum range of 1250 m at +90 MoA elevation.

If anyone wants to use this mod as part of anything else, feel free. I'd like a name-drop and maybe a link to this thread, but I'm not going to claim some silly intellectual property right on a bunch of tweaked numbers in somebody else's game. But do be aware that a new version will probably come out some time or another, in some form or another, that may tweak any of the values herein and hopefully include ballistics for large-caliber weapons.

Random observation of the day: men take half damage from all sources when prone. This is probably to compensate for the poor explosion modeling, but it has silly effects on rifle combat.

[funnyguy1 0]

Thank you!

BIS should pay people who improove their product, really.

[Jinef 2]

Your amazing ....

I love your commented Cpps, comedy gold

Thank you very much for allowing us to implement it, your amazing!

[NonWonderDog 0]

Big update in the first post!

The SVD ballistics are now quite a bit more accurate, but that's not the main feature.

[Canalien 0]

Brilliant!

This might be a stupid question but...

what are you using to get those range readouts?

ie... the 'Current Distance: 1300 metres" message

[NonWonderDog 0]

The range readout is from Kronzky's Portable Target Range, which I used to calibrate everything. I can't overstate how useful that script was when I was making this mod.

[Maximus-Sniper 0]

Hi mate.

Damn good work (sending over lot of beers)

I'm gonna test this new update

One thing that i thinking about:

Quote[/b] ]

#Damage depends on bullet speed in ArmA. I've tweaked the 5.56mm and 5.45mm rounds with this in mind. They should be one-hit kill at ranges below 50m, about the same damage as stock at 200m, and less damage past that. At 800m, it will take about 8 hits to kill someone. Damage for other bullets is probably still a bit off. Subsonic rounds in particular need some work

If i hit a man in his head at 1000 meter, will he die or do i have to put 8 bullets in his head?

Not sure about the effective range for M24, but i'm sure i bullet will do the job if u hit him in the head.

Please correct me if i thinking wrong here..

And the M107, pretty sure you need just one bullet to kill a man at 1000 meter, even if u hit him in the chest. Not sure how tick the body armor for the soldier is, but i know the M33 ball penetration:

Armor Penetration.

500 meters: 0.32 in (8 mm)

1,200 meters: 0.16 in (4 mm)

I'm gonna test your new update mate

EDIT:

Ok.. i did some fast test in the Target Range, and the soldier die with one bullet in his head at 1000 meters. And die with 2 bullets even if u hit him in the chest, arms and legs at 1000 meter.

And the M107 did his job all the way, hehe

Cheers