RPG-7 too accurate

[gazzthompson]

what MOA should a RPG have ? unless you already said and i missed it , alot of MOA talk :S

[nedlands1]

what MOA should a RPG have ? unless you already said and i missed it , alot of MOA talk :S

Actually I think it's fairly good value at the moment.

In short:

The "current" (aka PR v0.856) RPG probability of a hit (PH) curve compares fairly well with the RPG's real life curve (@ 0 MPH) as can be seen in Figure 2 (see below).

"MOA talk":

My original estimate was based on the worst accuracy needed to get a 100% hit rate on a 3.1m, circular target at 125m. The thing is that on the graph the probability of getting a hit roughly halves when you double the range (at least when you compare PH(~125m) to PH(~250m). In BF2, if you double the range you double the diameter of the cone of fire and consequently quadruple the area where rounds can land in. If you assume that the rounds are distributed in a uniform fashion then you quarter the chance of a hit (based on a target which fits within the cone to begin with).

The shape of the curve that you can produce in BF2 is different to the graph and the number of variables you can alter is limited. This means that you can't get the curves to perfectly match. If there was a formula for the real life curve then you could fit the curve that exists in-game to it using the least squares method but since I don't have that I've done it geometrically.

So the probability of a hit on a circular target with a uniform distribution is a piecewise function which is dependent on a number of variables. It's pretty much target area divided by the area of the base of the cone of fire and all of that multiplied by 100 to convert it into percent. When the target area is larger then the area of the base of the cone of fire, there'll be a 100% probability of a hit (since PH > 100% makes no sense in this context). See Figure 1 for the piecewise formula.

Figure 1[1]

Now if you assume that the tank's body can be approximated by a circle and you use the 0.856 deviation values you get a curve which indicates that the accuracy of the RPG is too high at close range and almost perfect at long range when compared to the 0 MPH crosswind curve (see Figure 2).

Figure 2[2]

Now if you use my original estimate of 85.24 MOA you get a curve which indicates that the accuracy of RPG is almost correct at 125m and 500m, but way off in between in the case of the 0 MPH crosswind curve (see Figure 3). However it is very close to the 7 MPH crosswind curve at all points except for very close range.

Figure 3[3]

Errors

• The shape of the tank was taken to be circular for ease of calculation. The actual area is somewhat larger and not circular. Thus the actual probability of a hit will be increased in general. The above piecewise distribution can't be used with such an area unless the whole target is within the cone of fire (top picture of Figure 4) or the cone is fully within the target (bottom picture of Figure 4). In the case that you have part of the target within the cone and some gaps (middle picture of Figure 4), the formula is wrong since it's the actual area within the cone of fire that is important, not the total area of the target.

_____Figure 4

• The distribution of the rounds was taken to be uniform which may not be the case. The distribution may well be uniform in the case of the angle of the chord passing through centre which the rounds land on (eg rounds are just as likely to land along the 3 o'clock position as along any other position) but the distribution along the said chord could be normal (ie rounds tend to bunch towards the middle).
• The RPG's drop wasn't taken into consideration which results in the calculated curve having inflated probabilities, especially at long range (the extent is dependent on the skill of the player of course).

Notes

The formula was neatly set out in Openoffice Math, the graphs were made in MATLAB and scaled and superimposed in Photoshop CS3, for those that may be interested.

[1] Where PH is the probability of a hit (%), Rtarget is the radius of the target (m), Dtarget is the distance to the target (m) and theta is the angle between the centreline of the cone of fire and the edge of it.

[2] Rtarget is 1.55m (=3.1m/2) and theta is 0.5 degrees (= 60 [MOA] x 1/60 [degrees/MOA] x 1/2).

[3] Rtarget is 1.55m (=3.1m/2) and theta is ~0.71 degrees (= 85.24 [MOA] x 1/60 [degrees/MOA] x 1/2)