This is a simulationist rules-heavy design question for JAGS. I realize many/most people here aren't familiar with it (and our sever's down until the weekend). If you're familiar with GURPS or Hero think along those lines.

My problem is with the complex and unsatisfactory JAGS Auto-Fire rules (described in brief below).

THE BASICS
Automatic Weapons in JAGS have a number of "modes" of automatic fire:

1. Spraying bullets in an attempt to hit a hard to hit target ("Wasting rounds for a better chance to hit")

2. Trying to hit a target with multiple rounds for more damage.

3. Supression fire over an arc.

4. There's been toying with indirect machine gun fire for a "beaten zone" effect--but it's not so important.

RESOLVING AUTOFIRE ATTACKS FOR CASE 1
1. There's a chart: the more you "waste" the greater your plus to hit.
2. You only hit with one shot for simplicities sake when doing this.

RESOLVING AUTOFIRE ATTACKS FOR CASE 2
1. There's a roll to hit (how well was the weapon originally aligned with the target).
2. There's a "control roll" for how well you either keep it on target. This is a X-or less roll with negative modifiers for a control roll for the weapon and number of shots fired (i.e. controlling an Ingram is easier than controlling a from-the-hip M60). Long bursts are harder to control than short ones. There's a chart that tells what percentage of the rounds fell "on target"
3. The number of shots that hit is grouped: the first round does 1x damage +1/3 base damage per round (so a 4 round burst that hits does 2x base damage).

RESOLVING AUTOFIRE ATTACKS FOR CASE 3
1. Draw an imaginary line several yards across (your arc of fire) at some distance. The GM prevents "one inch lines of coverage" or other attempts ot break the system. This line is to be drawn either over the targets you wish to hit or over, say, brush you wish to "supress"
2. Divide number of rounds by square (hexed) yards.
3. determine chances of a bullet being in a yard with a target.
4. Roll to hit with a base 9- (slightly less than average).
5. If targets are further away or closer than the line the GM works out what the chance is of their being hit based on the arc covered.

PROBLEMS
1. These are the most complicated, byzantine part of the rules. I don't have a book where I'm posting from and *I'm* hardpressed to regenerate tables and such by memory.
2. Autofire weapons could be treated as simply bigger guns--but if you're attacking an especially hard to hit target (an Aliens alien, say) it's nice to have a better chance to hit due to spraying an area.
3. The math for case 2 works out so that the more rounds you fire the lower your controll roll--so much so that assuming average rolls it's about the same to fire 8, 16, or even 32 rounds at someone (your total number of shots go up--but your control roll drops. I'm going to fix this by giving autofire weapons a basic control factor (a negative) that is NOT based on how many rounds are fired but is assumed to be logical based on the weapon.

REQUEST
1. I'm familiar with Hero (whose autofire rules I pretty much like) and GURPS. Is there a game out there that has elegant sim-friendly rules for handling automatic weapons at anywhere near this level of detail?

2. As simulating every single bullet in a burst is unsavory ... but allowing a spread of fire is desirable tactically, can anyone think of a unifying mechanic or idea that could be used to handle "batches" of fire?

3. I'm toying as treating any automatic gun as 3 different weapons (or so). Essentially listing different damages and modifiers to-hit based on how its used. For example, in case 2, simply treat the base damage as higher than single-shot and assume that if the user hits s/he controls the weapon well enough to hit with some number of bullets--not get grainy enough for control rolls and figuring out exactly how many rounds hit (maybe base it on skill level).

If that were the case has anyone considered how they'd model a weapon fired as a spray of bullets without tracking individual bullets?

Thanks,
-Marco

I think Cyberpunk 2020 had some fairly reasonable autofire rules.

(Rummage rummage rummage)

Here we are, "Automatic Weapons", p. 96. It doesn't track individual bullets per se, it talks about ". . . for every 10 rounds fired at such-and-so range, modify Attack Total . . . " so that looks like your control modifier. It uses a correlation between degree of success and # of bullets on target.

For suppressive fire, the gunner actually creates a sort of "save vs. hail o' death" roll for the defender. This makes sense to me; the gunner is not trying to hit something specific so much as increase the difficulty for anyone foolish enough to hang out in his cone of death.

"Spray 'n' pray," even in semi-auto mode, is notoriously ineffective.

Let's try to turn this around: from a sim perspective for a given burst, you have a given level of control that is actually exerted, and the defined target area, and the resulting actual bullet density over that area. Can we extract a single mechanic from these common factors?

Simplest case: one round. Difficulty is based upon getting that one round into an area of such-and-so size; presumably you have this covered. Density rule is "invisible" here, since you've got it defined at one.

Three-round burst: this maybe should be covered separately since it is a separate mode of fire on the real thing. But it's a special case of full auto regardless.

Difficulty for autofire should be based on target area (bigger is easier), how many rounds you want to squeeze off (I think it's easier for the player to say how much, even if it's a touch unrealistic) (more is harder), and the other usual modifiers. That's only two, not bad.

Then determine the actual target area according to how much they made it by (I dunno if JAGS does degrees of success, but if you're using typical dice resolution, it will be there to be read). If you're using hit location and the target is one person, not too hard to figure out the density and treat it appropriately (I think). Or if it's suppressive, the success turns around and becomes a factor in the difficulty of popping up, running across, or whatever, and *not* getting splattered--miss the roll by X, get hit by Y bullets, roll damage, thankyewverymuch.

I'll try to get a look at JAGS and be more specific soon. Been a while since I knocked the rust off the ol' sim ciruits. (Hey! Mike!)

--JB

I can't remember what game it came from, but one of my favorite suppression fire mechanics worked something like this:

Pick a hex, pick a number of bullets fired into that hex, determine the status of any targets in that hex.

If the targets in the hex are behind cover capable of stopping the rounds fired, then they are not effected...period, no roll, no anything.

However, the number of bullets put into the hex (or inch of frontage if tabletop) established a difficulty number. 1 per bullet, 2 per bullet, 3 for the first, 2 for the second, 1 for each additional...what ever number selection gets a range within the normal range of difficulty for rolls in your game. We liked the diminishing returns model best because it allowed single "covering fire" shots like those seen on cop shows to be effective because the initial bullets had a proportionally bigger number. Also you could build in penelties for automatic guns with notoriously bad recoil, or when fired by weak or untrained individuals simply by dropping the diminishing returns by 2 points per additional bullet instead of 1.

At any rate this number meant nothing unless the target in the hex attempted to do something. Depending on the action system this can be based on the number of actions used or just GM judgment of the type of action. The more actions used, or the more exposed the action was the more added to the score. 1 roll was made.

So basically the concept was: I lay down suppression fire, as long as the bad guys just sit there and hide nothing happens...easy. If they decide to pop out and do something (run, shoot back, etc) they make a roll against whatever score I accumulated with my fire. Success (for me) means they got tagged.

What would happen then is we'd use numbered chits (some with 0s) and spread them over the line of the arc we were fireing into hidden. The opponent wouldn't know which hex had rounds and which didn't until he made his move...or made some "battlefield veteran / tactics" type roll and then he'd get to uncover a chit.

But the bottom line of our thoughts on the subject was if you stayed down you were safe, if you got up you survived based pretty much on luck.

Maybe this isn't interesting, but I did some really fast rules for automatic fire I thought was ok. They can be used for any system that has "hit of x+ on a Dn"

It works like this:

You want to fire x bullets. You roll (x*2-1) Dn. Pick the x lowest numbers to see if they hit or not.

For a simple example. Let's say you hit on 5+ on a D6. You want to spray a guy with 10 bullets. You then roll 19 dice (ooops) and keep the 10 lowest numbers.

Calculating this effect for the general case we see that for hit on z+ the number of dice lower than z will be (x*2-1)*(z-1)/6. So number of hits on average would be x-(x*2-1)*(z-1)/6.

For our 5+ example that yields a negative number. For 2+ on the other hand. We have 10-19/6=~7 bullets would hit.

Depending on the rest of the game this could be acceptable (if 2+ is close range like a few meters, 7 bullets would be a good estimate)

If you spray at long range though, you won't have much chance of hitting at all.

It's (the system) not so complex, so it doesn't give much more detail than that.

I don't know if it's useful.

Marco wrote:
> If that were the case has anyone considered how they'd model a weapon fired as a spray of bullets without tracking individual bullets?

There were several threads on and about this topic on the RPG.net Game Design forum. You might like to check that out.

I like that idea alot PF. I've used the roll multiple dice keep the worst mechanic many times, but never thought of applying it to autofire.

Why z*2-1 instead of just z*2?

Having fired full automatic weapons quite a bit (mostly M2 .50 Cal HMG), I think I'm qualified to say a few things about them.

First, you don't aim individual bullets. Ever. Not possible. You aim the first shot, maybe, and then, at best, you use feedback from the fire to "walk" the stream of bullets onto the target. Then you try to hold the fire there. There are problems with this however, as you aren't supposed to fire long bursts. Mostly because it eats up ammo unnecesarily. The vast majority of rounds fired from automatic weapons miss their target, and you can run out of ammo fast. So readjusting your fire regularly is important. You just have very ittle control of the fire after the first shot. Even with the most stable of weapons.

The idea of automatic weapons is not to hit the target more than once. If that happens, it's a happy coincidence. But in combat one hit is considered a success. Usually after that first hit, you have effectively removed that combatant. Any system that doesn't take into account the idea that a character who is hit by a bullet is now probably a non-combatant is simply not realistic. Said character goes from being a combatant to, at best being a survivor. With a fifty cal, he's probably just hamburger. I've fired the damn things through (not into, through) vehicles, and they leave huge holes. I don't care where you're hit, even in a hand, you'll be reduced to just holding that hand and trying not to bleed out.

I feel that I can speak with real authority on this one, as I once saw a firing accident where, due to a badly set headspace and timing on the gun, a round ejected through the ottom of the gun instead of out the business end. The firer was standing in five ton truck with the gun mounted on a ring, and the round went through his thigh. I can easily say I've never seen that much blood in my life. Had there not been combat lifesavers on the scene, and a medivac chopper ojn the way instantly, he would have died. As it was, it wasn't pretty, and I heard he lost that leg (I didn't know him, he was in another unit).

Now that's a huge automatic weapon. But FN/FALs fire the equivalent of a 30-06 round, you know, the kind designed to kill deer with one shot. And the M-16 fires a .22 round, more or less, but at very high velocity. That velocity causes all sorts of tissue damage. I've never seen one hit anybody, but apparently it's not nice either. Many SMGs fire a .45 which makes a big hole, too.

My point? Nobody worries about needing to hit anyone more than once. It's just not necessary. Fire of automatic weapons is merely a matter of bringing the fire onto the target. The number of rounds that actually hit is somewhat random. At range it's likely to be no more than one. But, again, that's enough, unless you're in a movie.

In other words, combine one and two. Use somthing like Hero system: For each extra round give a +1 to hit, and for every four points over what you normally need to hit, you overkilled the target with an extra bullet. Thus, if you empty your 20 rnd magazine into a target, you will likely hit with about five rounds, all other things being equal. Adjust these numbers to fit JAGS. Does this make full auto weapons very dangerous at point blank? Yes, and so should they be. The guy who fires first at this range wins. Period. Remember penalties for non-aimed fire, cover and range when using these rules.

If there are several targets, roll randomly for which get hit by which bullets. Again, you aren't aiming each round, just the fire in general. Well focused fire will result in more hits, just not with any placement. You could probably add +1 or +2 to hit per target in the field of fire.

Supression fire is the most common use of automatic fire. In this case, just say that anyone entering the arc is hit on the 9-. Don't bother with the other calculations. Or just have the gunner roll to hit anything that comes into the supresion zone, even if it's not his "turn". Use the same method as above.

The easiest option is to make all auto fire work like supression fire. Establish a field, and then roll to hit each target that is in it or gets in it.

There really aren't many choices to make when you use full auto fire. At least not in a realistic game.

Mike

Marco,

The other thing that I've thought of about automatic firing has nothing to do with your chances to hit, but your chances of your rifle continuing to work. I've seen barrels glow red from automatic fire - it heats your rifle up like a son of a bitch.

Since the JAGS site is down, I can't grab the rules to figure out what the most common dice roll is in the system, but here's an example of what I'd do: for each combat round after the first that you continue to use auto fire, roll a d10. You have to roll over a number that gets progressively higher each round or your rifle overheats and malfunctions. For example:

Round 1: No roll.
Round 2: Roll over 1 on a d10.
Round 3: Roll over 2 on a d10.
Round 4: Roll over 3 on a d10.
etc.

You could have a table that correlates to how bad you fail this roll. Take the difference between the number you had to roll over and what you rolled.

0: Rifle jams.
1: Rifle is non-functional.
2: Round ejects through your leg (thanks, Mike!)
3: Rifle explodes.
etc.

I don't know if this helps, but it should add some danger and realism to auto fire.

Clinton raises a good point. If I was going for a simulationist mechanic, I'd also make that roll somehwat dependant on conditions. So it's more easy to destory your barrel in hot conditions (ie desert).

If I recall correctly, the SA80 has exactly this problem, the plastic shielding on the barrel melts if fired on auto in hot climates, rendering the gun pretty much useless.


-Matt

Couple of points:

One mechanic I did see that did treat bullets in a burst individually, without bogging down was the Twilight 2000 / Dark Conspiracy rule set (from GDW). Pick up as many D6 as your fireing bullets. Lose some for deleterious conditions - longer range, large recoil etc. Roll the remainder. Every 6 is a hit on the target. Pick up the ones that missed. Roll them. Every 6 represets a bullet that hit something else 'interesting' in the target area, evenly distributed.

Supressive fire, and spraying an area worked similarly (Roll the D6, every hit is a bullet that hits a possible target, evenly distributed).

Caveats: No place for skill to apply. Later versions of Dark Conspiracy (2nd Ed) had skills that allowed you to fire more bullets before taking recoil penalties. It did (does) however, have detailed rules on recoil, which became the limiting factor on autofire. I suspect that that's not ideal for a generic system (Consider recoilless weaponry).

However, other than that it's really elegant.

Clinton R Nixon wrote:
The other thing that I've thought of about automatic firing has nothing to do with your chances to hit, but your chances of your rifle continuing to work. I've seen barrels glow red from automatic fire - it heats your rifle up like a son of a bitch.

On the multiple-hit thing - sounds like it is not, in general, realistic. Certainly the 2-3 times I fired an automatic weapon, "aim" meant "make sure the business end is pointed at the target" and that's about all I could manage. But I remember the Mekton (yup, there goes realism :-) rules for autofire and volleys of missles were fairly elegant - something like each point/every two points (or something) you roll over what you needed to hit at all, you get an extra damage roll. I'll review the rules sometime over the weekend and see if it could actually be used here.

Or maybe some other Mech/Tank Combat/Air Combat game would have some interseting rules. Certainly, single vs. multiple autofire hits on a vehicle frame are probably more significant than multiple hits on a human body.

And if JAGS is meant to support (e.g.) firing an auto weapon at a Dragon, the issue does seem relevant.

Gordon

Thanks for the input so far (and more is certainly welcome if anyone has any good ideas).

1. I'm pretty experienced with autofire myself (silenced M-11, M16A1 and A2 (3shot), M-60).

2. JAGS won't go into the low-level of abstraction of jamming more in the desert--the GM could impose those limitations and would be highly advised to warn players with experinenced characters.

I do think an autofire mishap option might be neat. I'll explore it.

3. Overkill *is* an issue. One round of an assault rifle in JAGS is deadly to a normal person. It isn't deadly to a super or (as noted) a dragon. So multiple shots on target is part of the game (as well as part of cinematic action which JAGS assumes is a "common" standard). Forex: 9mm weapons are about 10-40% less likey to perform a 1-shot stop to a normal man than in real life.

Thus far I think that "defining a zone of fire" and assigning single bullet hits is a good idea. Because arcs and curves are cumbersome I might go into wide-spray (weapon waves wildly over a wide area), controlled kill-zone (medium area), and tight burst (all or most of the bullets fired at a close target).

I think that I'll treat tight bursts as simply more damage (i.e. not worry about exactly how many rounds hit--just give single shot damage and "tight burst" damage based on skill level of the attacker (so a few more columns in the weapons table, perhaps).

Since the amount you hit by has a heavy determination of damage already I don't want to compound that with "the more you hit by, the more roudns hit"--that was the reason for the control roll in the first place.

Thanks, it's been great!
-Marco

In my reply to Marco's post, I've rearranged Marco's writing so I can write better answers (I hope! :) ).

There's two ways of simulating automatic weapons in a game. The Hollywood or cinematic way, as shown in many movies, and the realistic way. Of them all, I think the realistic way is the simplest.

For an automatic weapon like an SMG, assault rifle and so on:


• Single shot fire hits a single target and wounds;
• Burst fire hits a single target and kills;
• Single shot at a group, is usually ineffectual (wounding at most one person);
• Burst fire at a group, wounds the front people in the group and suppresses the rest;
• Automatic fire misses the target and empties the magazine. The target/s are suppressed for a while.



For a LMG and bigger, that's in place:


• Single and burst fire misses the target (if aimed carefully, they can hit);
• "Walking" the fire onto a target, hits the target and usually kills the target at the same time;
• Automatic fire suppresses a target area for a while;
• Continuous fire suppresses an area, while the firing continues, and empties the belt. Then the barrel needs to be replaced! That's why there's a gunner's mate in MG teams, he carries the spare LMG barrel and also spots for more targets, while preparing the ammunition belt.



A LMG and bigger that's not in place, or fired from the hip, is ineffectual. Just cross off the ammunition.

This is based on information from real world experts, who have fired real weapons on the range and/or at real people. They include a US Delta Force soldier, a US Army Engineer, a British soldier and several NZ reservists.

On to the cases...

Marco -- Case 1 wrote:
1. Spraying bullets in an attempt to hit a hard to hit target ("Wasting rounds for a better chance to hit")

RESOLVING AUTOFIRE ATTACKS FOR CASE 1
1. There's a chart: the more you "waste" the greater your plus to hit.
2. You only hit with one shot for simplicities sake when doing this.

Marco -- Case 2 wrote:
2. Trying to hit a target with multiple rounds for more damage.

RESOLVING AUTOFIRE ATTACKS FOR CASE 2
1. There's a roll to hit (how well was the weapon originally aligned with the target).
2. There's a "control roll" for how well you either keep it on target. This is a X-or less roll with negative modifiers for a control roll for the weapon and number of shots fired (i.e. controlling an Ingram is easier than controlling a from-the-hip M60). Long bursts are harder to control than short ones. There's a chart that tells what percentage of the rounds fell "on target"
3. The number of shots that hit is grouped: the first round does 1x damage +1/3 base damage per round (so a 4 round burst that hits does 2x base damage).

PROBLEMS
The math for case 2 works out so that the more rounds you fire the lower your controll roll--so much so that assuming average rolls it's about the same to fire 8, 16, or even 32 rounds at someone (your total number of shots go up--but your control roll drops. I'm going to fix this by giving autofire weapons a basic control factor (a negative) that is NOT based on how many rounds are fired but is assumed to be logical based on the weapon.

Marco -- Case 3 wrote:
3. Supression fire over an arc.

RESOLVING AUTOFIRE ATTACKS FOR CASE 3
1. Draw an imaginary line several yards across (your arc of fire) at some distance. The GM prevents "one inch lines of coverage" or other attempts ot break the system. This line is to be drawn either over the targets you wish to hit or over, say, brush you wish to "supress"
2. Divide number of rounds by square (hexed) yards.
3. determine chances of a bullet being in a yard with a target.
4. Roll to hit with a base 9- (slightly less than average).
5. If targets are further away or closer than the line the GM works out what the chance is of their being hit based on the arc covered.

Marco -- Case 4 wrote:
4. There's been toying with indirect machine gun fire for a "beaten zone" effect--but it's not so important.

Marco -- Request 1 wrote:
1. Is there a game out there that has elegant sim-friendly rules for handling automatic weapons at anywhere near this level of detail?

Marco -- Request 2 wrote:
2. As simulating every single bullet in a burst is unsavory ... but allowing a spread of fire is desirable tactically, can anyone think of a unifying mechanic or idea that could be used to handle "batches" of fire?

Marco -- Request 3 wrote:
3. I'm toying as treating any automatic gun as 3 different weapons (or so). Essentially listing different damages and modifiers to-hit based on how its used. For example, in case 2, simply treat the base damage as higher than single-shot and assume that if the user hits s/he controls the weapon well enough to hit with some number of bullets--not get grainy enough for control rolls and figuring out exactly how many rounds hit (maybe base it on skill level).

If that were the case has anyone considered how they'd model a weapon fired as a spray of bullets without tracking individual bullets?

Marco wrote: 3. Overkill *is* an issue.

I have to go back and look over the game again Marco before I comment. t I just wanted to say I appreciate the work you put into JAGS!

Its also great to have you here on the forge. Besides someone has to take the simulationist camp around here ;)

Valamir wrote: I like that idea alot PF. I've used the roll multiple dice keep the worst mechanic many times, but never thought of applying it to autofire. Why z*2-1 instead of just z*2?