Shot Miss Scattering

Ah ok, I thought my maths was just letting me down and I was missing something subtle

Another question is that if all weapons have a melee mode but it will have equal accuracy and likely much lower damage than simply shooting at point blank range will that melee mode ever be used?

Edited April 7, 2013 by Gauddlike

Ah ok, I thought my maths was just letting me down and I was missing something subtle

Another question is that if all weapons have a melee mode but it will have equal accuracy and likely much lower damage than simply shooting at point blank range will that melee mode ever be used?

Stun damage?

I missed around with some formulas today and I think this one seems to work pretty well.

Chance to hit = Base weapon chance - ((Range - (Weapon Effective Range +1))^3 / 4)

The base weapon chance is the chance in the weapon xml for the type of shot you're doing (snap, auto, aimed.)

You'll notice that the past the weapon effective range accuracy falls off rapidly, while accuracy increases as you get in closer than effective range. Using this formula you won't have any misses at point blank range (unless the hit chance is capped at 99% or something.)

The cap is 95%.

How does the soldiers accuracy factor in to that?

The cap is 95%.

How does the soldiers accuracy factor in to that?

I hadn't thought of that, yet...

OK, I included the soldier accuracy:

Chance to hit = Soldier Accuracy * (Base weapon chance - ((Range - (Weapon Effective Range +1)^3 / 4))

So, basically with this formula the weapons accuracy effect is about neutral around the effective range and it yields basically similar results to the current linear formula at max. effective range. Once you get somewhat further or closer than the effective range they become rapidly more or less accurate. It's not perfect, but it does mimic real life fairly well. It definately solves the point blank range problem. The only person that couldn't hit at that range is really crappy shooter with an inaccurate weapons, but why would even have someone like that in your squad, right?

Here are some example results for the shotgun (eff. range = 7) for a soldier with an accuracy of 50 (.50) base chance for snapshot = 90 (I think):

Range = 14: 18%

Range = 13: 29%

Range = 12: 37%

Range = 11: 41%

Range = 10: 44%

Range = 9: 45%

Range = 8: 45%

Range = 7: 45%

Range = 6: 47%

Range = 5: 49%

Range = 4: 53%

Range = 3: 60%

Range = 2: 72%

Range = 1: 89%

So, as you can see even a short range weapon in the hands of a barely competent soldier will hardly ever miss at point blank range.

Here are some example results for the assault rifle (eff. range = 18) for a soldier with an accuracy of 50 (.50) base chance for snapshot = 75:

Range = 26: 0% (or whatever the minimum is 1%?)

Range = 25: 11%

Range = 24: 22%

Range = 23: 30%

Range = 22: 34%

Range = 21: 36%

Range = 20: 38%

Range = 19: 38%

Range = 18: 38%

Range = 17: 38%

Range = 16: 40%

Range = 15: 45%

Range = 14: 53%

Range = 13: 64%

Range = 12: 80%

Range = 11: 95%

Range <= 10: 95%

Of course this formula can easily be modified to increase or decrease the range effects.

BTW, the formula in the Wiki is wrong: Shooter Accuracy * Accuracy Modifier * (Range to Target / Weapon Operational Range) = Unmodified Accuracy

It should read:

BTW, the formula in the Wiki is wrong. Shooter Accuracy * Accuracy Modifier * (Weapon Operational Range / Range to Target ) = Unmodified Accuracy

If the you use the wiki formula accuracy INCREASES with range and that obviously can't be right.

Edited April 8, 2013 by StellarRat

I've seen 104% from a sniper on a roof.

I don't recognise the use of ^3/4 at the end of your formula Stellar, what does it represent?

Decided to have a play in excel with the current formula as it appears to be quite flimsy.

Using your shotgun example of 50 soldier accuracy and 90 shot accuracy.

Assuming the 95% cap applies it would be (wiki formula output in brackets added to Stellars results):

Range = 14: (22%) 18%

Range = 13: (24%) 29%

Range = 12: (26%) 37%

Range = 11: (28%) 41%

Range = 10: (31%) 44%

Range = 9: (35%) 45%

Range = 8: (39%) 45%

Range = 7: (45%) 45%

Range = 6: (52%) 47%

Range = 5: (63%) 49%

Range = 4: (78%) 53%

Range = 3: (95%) 60%

Range = 2: (95%) 72%

Range = 1: (95%) 89%

Which is fairly reasonable.

The linear accuracy drop worried me though so thought I would try a few more examples.

Increasing soldier accuracy to 75 (not unreasonable) gives a 25% shot accuracy out at 19 tiles with a shotgun.

100 soldier accuracy gives a 25% shot accuracy out to 25 tiles.

Ok so that is a little more worrying.

Wonder what a precision rifle looks like...

Swapping the shotgun for a precision rifle in the last example then, 100 soldier accuracy, 150 weapon accuracy (current maximum zoom) means that the shot accuracy doesn't drop below 95% until 42 tile range and is still at 25% when aiming at a target 155 tiles away.

Interesting results really.

Wonder if this is the actual formula used after all?

Edited April 9, 2013 by Gauddlike

Elevation is supposed to provide a bonus to accuracy, that may not be limited by the cap.

I don't recognise the use of ^3/4 at the end of your formula Stellar, what does it represent?

Decided to have a play in excel with the current formula as it appears to be quite flimsy.

Using your shotgun example of 50 soldier accuracy and 90 shot accuracy.

Assuming the 95% cap applies it would be (wiki formula output in brackets added to Stellars results):

Range = 14: (22%) 18%

Range = 13: (24%) 29%

Range = 12: (26%) 37%

Range = 11: (28%) 41%

Range = 10: (31%) 44%

Range = 9: (35%) 45%

Range = 8: (39%) 45%

Range = 7: (45%) 45%

Range = 6: (52%) 47%

Range = 5: (63%) 49%

Range = 4: (78%) 53%

Range = 3: (95%) 60%

Range = 2: (95%) 72%

Range = 1: (95%) 89%

Which is fairly reasonable.

The linear accuracy drop worried me though so thought I would try a few more examples.

Increasing soldier accuracy to 75 (not unreasonable) gives a 25% shot accuracy out at 19 tiles with a shotgun.

100 soldier accuracy gives a 25% shot accuracy out to 25 tiles.

Ok so that is a little more worrying.

Wonder what a precision rifle looks like...

Swapping the shotgun for a precision rifle in the last example then, 100 soldier accuracy, 150 weapon accuracy (current maximum zoom) means that the shot accuracy doesn't drop below 95% until 42 tile range and is still at 25% when aiming at a target 155 tiles away.

Interesting results really.

Wonder if this is the actual formula used after all?

^3/4 = to the cube divided by four...but you must have already figured that out if you put it in Excel. Under the current system, at very long ranges most of the weapons are going to do negligible damage even if a hit is achieved. Also, your sniper rifle example appears to be using a super solider (100 accuracy). One has to bear in mind that kill shots over 1 mile away have been made by expert US snipers in Afghanistan. So, 155 tiles is NOTHING (1.6 X 155 = 250 meters or about 1/5 of a mile). Of course, at that range it's unlikely it will do any damage using the current damage formulas. Obviously, Chris has compressed the ranges greatly, but that's OK. Why does my proposed formula make sense? Well, I've studied weapons for a long time and that's just how they work. If you look at actual accuracy charts they are far closer to my formula than a linear formula. There is a lot of published data on this. If you take to a deeper level, I'm going to get a little physics/philo here, when you double the diameter of a circle the area increases by 4x. Think about that. So, now everything is actually spread over 4x the area. Do you really thinking someone/something can cover 4x times the area with only 1/2 the loss in capability? From a mechanical and energy perspective the numbers don't add up. Edited April 9, 2013 by StellarRat

Yeah, I really don't have a problem with someone with capped accuracy taking an aimed shot across the map. If you manage to keep that soldier alive for that long, you've earned it. Besides, you still need LOS.

As I pointed out Stellar, those numbers are using the current wiki formula.

It has very little falloff at range.

Your formula would be very different judging by the examples you posted.

I reckon yours is a better option for that reason.

I couldn't get yours to work in excel though, trouble getting the last part to work I think as the numbers didn't match up with your example.

The whole point in using the super soldier is to find out how the formula works at the extremes.

It isn't so much the massive range that I feel is an issue with the wiki formula, it is the fact that the accuracy is maintained over such a long distance.

I understand the damage will be decreased and no suppression will be applied but I think that adds to the problem rather than solving it.

Most players won't look deeply enough to realise that their weapon that states 95% accuracy at 40 tiles will be ineffective at that range.

I doubt those expert marksmen you mention hit their target for no damage and the enemy didn't even notice the impact.

The non-linear falloff makes more sense as it would be less misleading as well as more 'realistic'.

By the time the damage falls to the point where the shot is ineffective the accuracy would (probably) have made the shot difficult.

I doubt those expert marksmen you mention hit their target for no damage and the enemy didn't even notice the impact.

The non-linear falloff makes more sense as it would be less misleading as well as more 'realistic'.

By the time the damage falls to the point where the shot is ineffective the accuracy would (probably) have made the shot difficult.

Yes, those Taliban died. A .50 cal bullet is fatal for miles. And I agree completely with your reasoning about trying to match accuracy to weapon damage. The real question is whether Chris does and if he's willing to change the formula. This will probably be the last chance we get for such a change (pre-beta testing.) I'd also like to know what the actual formula being used for damage and accuracy is now because I'm getting the the feeling (like you) that more is going on than we see in the Wiki which is obviously wrong.

I think I missed a ")" in the formula. It should read:

Chance to hit = Soldier Accuracy * (Base weapon chance - ((Range - (Weapon Effective Range +1))^3 / 4))

Here is the Excel version: =ROUNDUP(A2*(B2-(POWER((C2-(D2+1)),3)/4)),0)

Where cell A2 is the soldier accuracy in decimal (.5 = 50), cell B2 is Base Weapon Chance (shotgun snapshot = 90), cell C2 is the range in tiles, cell D2 is the Weapon Effective Range in tiles. You can paste the formula into cell E2 which is the final chance to hit. There are some very slight rounding differences with my posting from a calculator, but it's close enough.

The only time this formula will have a problem is if any of the input values are zeros or negative which should be impossible.

Edited April 9, 2013 by StellarRat

I just made a modification to your accuracy formula that I think works well.

I noticed that in the original where you have it /4 that longer ranged weapons work better with a higher number there while shorter ranged ones need a smaller number.

I altered the formula so that it divides by the weapon range instead which gives a much lower output in general and a slightly shallower falloff, maybe 8 tiles between 95% and 1% instead of the 4 yours gave (with precision rifles).

The short range accuracy is practically unchanged unless the weapon has precisely 6 range.

Excel formula:

The graph compares the three accuracy formulae using a precision rifle with 80 weapon accuracy (between step one and two as currently balanced) and a soldier with 60 accuracy.

I also added the current damage formula for the precision rifle, not taking into account target armour.

Accuracy is capped at 1% to 95% as per the game and listed on the left side.

Range along the bottom axis goes up to 75 tiles, however the precision rifle doesn't hit 5% accuracy until the target is ~230 tiles away and will still be doing 4 damage per hit at that range, providing the target has less than 15 kinetic resist using current accuracy and damage formulae.

Not much I know but I thought it was interesting

Edited April 18, 2013 by Gauddlike

Is that graph for the precision rifle? That's a pretty steep drop-off at only 16 tiles. I thought the whole point of it was being able to accurately engage past sight range.

That is because it is a low accuracy shot and low accuracy soldier.

It drops down to 45%-50% at 26 tiles and maintains it out to 36, that is with only 80% accuracy on the weapon and 60 accuracy on the soldier.

The precision rifle goes up to 150 on the weapon and soldier accuracy goes to 100.

With both maxed out that 50% accuracy mark is out at around 40 tiles with the new formulae and 75 ish tiles with the original one.

That is because it is a low accuracy shot and low accuracy soldier.

It drops down to 45%-50% at 26 tiles and maintains it out to 36, that is with only 80% accuracy on the weapon and 60 accuracy on the soldier.

The precision rifle goes up to 150 on the weapon and soldier accuracy goes to 100.

With both maxed out that 50% accuracy mark is out at around 40 tiles with the new formulae and 75 ish tiles with the original one.

I actually thought about exactly what you did with formula, but I wanted to keep it simpler. Your change is good for the long range weapons. Just out of curiosity, do you have a damage/range graphic for plasma cannons? I'd be interested to see what that looks like since they are the "ultimate" alien weapon.

Plasma Cannon used by Andron Support:

For accuracy the x axis is percent, for damage it is a flat number.

I stopped the graph at 100 tile range.

Edited April 20, 2013 by Gauddlike

Plasma Cannon used by Andron Support:

[ATTACH=CONFIG]2074[/ATTACH]

For accuracy the x axis is percent, for damage it is a flat number.

I stopped the graph at 100 tile range.

Thanks for the graph. Basically, there is decent chance they can hit you from anyway on the map. It just seems wrong to me.

Assuming a basic armoured, low rank soldier they would probably be able to one shot you at 60 or so tiles as well.

Tougher troops would need to be a bit closer though.

Assuming a basic armoured, low rank soldier they would probably be able to one shot you at 60 or so tiles as well.

Tougher troops would need to be a bit closer though.

These ranges pretty much make flanking a close to useless tactic. Everything is basically a frontal assault just from different angles.