Higher weapon accuracy test

[Chris]

@xcomnaut - Xcom1994 really did not have one shot kills on the majority of aliens. It would frequently take 3-4 shots to kill them.

[StellarRat]

@xcomnaut - Xcom1994 really did not have one shot kills on the majority of aliens. It would frequently take 3-4 shots to kill them.

It depended on weapon if I remember right. Once you started using the heavy alien weapons you could one shot some critters. Lasers usually took at least two hits. On some aliens I remember it required 5 or 6 hits sometimes. Also, the alien missile launchers could one shot a whole herd of aliens or XCom troops. The real key was getting the high level armor so your troops could survive a shot or two.

[Rogueywon]

I've just fired up X-Com1994 to test! (I've got a few games on the go).

Early missions with ballistics, I'm averaging 2 hits to kill a Sectoid or Floater. A mid-game save with ballistics and lasers vs Mutons has me taking far more hits to kill them - one particular Muton doesn't die until the 6th hit from a ballistic rifle and lasers are averaging 3 hits.

Late-game mission with plasmas - heavy plasma is indeed getting 1-hit kills on Sectoids. I don't have a late-game save on a convenient Muton mission, so can't check that.

In TftD, Lobstermen could have whole clips of harpoon ammo pumped into them before they'd go down.

Edit: Dear god it takes a lot of shots from a ballistic rifle to kill a Cyberdisk. But a single rocket does the trick.

Edited July 31, 2013 by Rogueywon

[Ishantil]

Yes, Sectoids and Aquatoids were the most fragile of the species, followed by Floaters. Once you got Heavy Plasma Rifles, only the larger aliens such as terror units and Mutons/Lobstermen were problematic.

In any event, one hit kills with a good weapon, such as human designed lasers or plasma rifles would be okay on lower ranked stuff. The main problem I have is that it takes me several turns with 4 guys to hit an alien enough times to kill it. So the increased lethality that I was speaking of was accuracy based, not weapon strength based.

[xcomnaut]

Actually most aliens died in 1-2 hits if heavy plasma were used. Laser rifle took 1-3 shots, mostly 2-3 compared to heavy plasma. Only Mutons, Cyberdiscs Sectopods and some leader/officers of Ethereal could survive 3 hits of heavy plasma. Laser weapons were less powerful but very accurate so few more hits to kill wasnt a problem. Xenonauts made aliens much more powerful. Lets nerf them a little For example, nerf alien health so they die as easy as humans. Aliens still will be powerful enough not to make game too easy.

[Ishantil]

My experience with the aliens is entirely based on two species; Sebillions and Sectoids. So I cannot accurately say how well the aliens' health is balanced.

[StellarRat]

Sadly, I don't think we've gotten far enough in the "Balancing Phase" to even take into account alien health, armor, numbers, or much else. 90% of the effort so far seems to be spent on balancing the weapons against each other. Plus there is the economic balancing that isn't close yet too. It needs to be done, obviously, or the game would suck, but the rate of progress really makes me think we won't have a balanced game until Jan. or Feb. 2014. I hope I'm wrong. BTW, I'm not saying anyone is slacking just that progress seems slow.

To me it seems that a great deal of the balancing could be done purely mathematically with computer simulations, then given to us to test. The weapons, armor, and hit points outcomes could certainly be simulated quickly and assuming Goldhawk had a goal in mind, the XML could be adjusted until the simulation was giving the desired results.

Edited July 31, 2013 by StellarRat

[lightgemini]

Im up to lowering aliens health if it is balanced with armour, so higher alien tiers make lower weapons tier quite less effective/useful. So you could have a Caesan soldier with 50-60hp but 40-50 armour.If armour degradation does work lesser weapons would still kill him but need a lot more effort than a more powerful weapon. Also it opens the door to "super units" that have too much armour to be hurt and needs lots of hits to degrade armour or the use of a weapon specialised in AP.

Now that I remember it, when stats were flatened to begin balancing, the armour value was aded to the hp of the unit so it was lower health on previous builds.

[xcomnaut]

This accuracy test made assault rifles pretty much usable. I even started to use pistols. After playing a bit I noticed one interesting thing. Was it intentionally made or just overlooked, not sure. Single shot accuracy were increased significantly for humans and aliens, auto fire increased very little for humans but left unchanged for aliens. Because aliens love auto fire, they left unbuffed lol

[lightgemini]

Well I will put some reports after playing some important fights with high accu.

Got first terror mission against androns+drones, and a base defense against caesans. High accu made this fights awesome, I won both fights but I had to work my &&& to achieve it, not just use standar cheesy tactics, and losses were high as bad tactical decisions were punished with fiery plasma death. Same can be said for the aliens.

Here is a quick log of the evolution of the terror mission:

Battle of Karthoum.

Team of 8 with 2 LMGs, 1 shield+pistol and the rest rifles with medpacks and some nades.

Vs

Team of 17 Andron guards and lightdrones.

-Due to laziness and wanting to skip past UFO scouts, I entered battle with no rockets or snipers.Even the Squad leader with the C4 was in the hospital. No laser weps either, only ballistics.

-Turn1 2 androns+1 drone guard the chinook exit. With no rockets at hand team fails to clear the ramp and duck inside covered by the shield guy. Side guys rush out to cover.

-Aliens shoot the crap out of chinook interior. Half team supressed and shield guy killed.

-Team clears the ramp area and deploy around chinook. Androns march towards them shooting.

-In the shooting the cover around the chinook gets destroyed.Many wounded. Guys at front retreat inside the chinook.Androns keep marching forward and start to cross the street.

-Since it takes too many hits to kill the androns and theres no solid defense in place, I fear this will end in full team wipe out "Drang Valley" style. The team evacuates the chinook and run for their lives to the building at their left. 2 guys trail behind the main group.

-Androns are at the chinook. Team in cover in the left building covers the last 2 guys rush trough the street. The team then retreats to the left border behind the building and ducks in there.

-Androns marches towards left building. Some xenonauts cross the street towards the bus station/bank and get in cover.

-Androns engage and kill the guys at the bus station/bank. Firefight continues and Androns get killed together with more xenonauts.

-Team get to the bus station/bank. Some guys need to take ammo from the fallen.More androns appear. Following firefight ends with more guys killed. Kseniya Kirova has a "hero of the day" moment, killing 2 androns at close quarters with the last one getting killed with a snapshot of 21% that hits.

-The last guys left clear the last Andron that seems to be panicked and doesnt move from were it is.

Final Victory---- 3 guys remaining with wounds that will take time to heal. 17 Aliens killed.

[lightgemini]

This accuracy test made assault rifles pretty much usable. I even started to use pistols. After playing a bit I noticed one interesting thing. Was it intentionally made or just overlooked, not sure. Single shot accuracy were increased significantly for humans and aliens, auto fire increased very little for humans but left unchanged for aliens. Because aliens love auto fire, they left unbuffed lol

My update got updated because I left some autofire unbuffed. Redownload again and get the fixed file. Then start to eat some fiery plasma death lol.

[Anon_Spartan]

Are you talking about the accuracy problems in the experimental builds? I know that when I played Version 4, I could not hit any alien who had so much as a pebble in front of him, no matter how close I was or from which direction I fired. Even point blank shots deviated into the rock.

[xcomnaut]

My update got updated because I left some autofire unbuffed. Redownload again and get the fixed file. Then start to eat some fiery plasma death lol.

Trying updated file. So far no deaths

[xcomnaut]

Accuracy test was a great success on improving weapons overall. I let myself to dream on and this is what I thought of. This is my vision how to improve human weapons even further. I'll take assault rifle as example.

<Dream>

Short version:

Ballistic - Damage: vanilla(v19.5), accuracy: 50% of what Aaron suggested in accuracy test.

Lazer - Damage: vanilla(v19.5), accuracy: 100% of what Aaron suggested in accuracy test.

Plazma - Damage: +50% of vanilla(v19.5), accuracy: 50% of what Aaron suggested in accuracy test.

MAG - Damage: vanilla(v19.5), accuracy: 100% of what Aaron suggested in accuracy test, BONUS: +25% Rate of Fire

Long version:

Ballistic Weapons Vanilla(v19.5) | Ballistic Weapons Upgraded

Damage:

<type="kinetic" damage="30" mitigation="10"> | <type="kinetic" damage="30" mitigation="10">

Single Shot Accuracy:

<Set1 ap="20" accuracy="30" /> | <Set1 ap="20" accuracy="35" />

<Set2 ap="30" accuracy="60" /> | <Set2 ap="30" accuracy="75" />

<Set3 ap="40" accuracy="80" /> | <Set3 ap="40" accuracy="100" />

Burst Fire Accuracy:

<BurstFire ap="40" accuracy="20" /> | <BurstFire ap="40" accuracy="25" />

Damage: vanilla(v19.5), accuracy: 50% of what Aaron suggested in accuracy test.

-------------------------------

Lazer Weapons Vanilla(v19.5) | Lazer Weapons Upgraded

Damage:

<type="energy" damage="50" mitigation="18"> | <type="energy" damage="50" mitigation="18">

Single Shot Accuracy:

<Set1 ap="20" accuracy="30" /> | <Set1 ap="20" accuracy="40" />

<Set2 ap="30" accuracy="60" /> | <Set2 ap="30" accuracy="90" />

<Set3 ap="40" accuracy="80" /> | <Set3 ap="40" accuracy="120" />

Burst Fire Accuracy:

<BurstFire ap="40" accuracy="20" /> | <BurstFire ap="40" accuracy="30" />

Damage: vanilla(v19.5), accuracy: 100% of what Aaron suggested in accuracy test.

-------------------------------

Plazma Weapons Vanilla(v19.5) | Plazma Weapons Upgraded

Damage:

<type="energy" damage="70" mitigation="25"> | <type="energy" damage="105" mitigation="25">

Single Shot Accuracy:

<Set1 ap="20" accuracy="30" /> | <Set1 ap="20" accuracy="35" />

<Set2 ap="30" accuracy="60" /> | <Set2 ap="30" accuracy="75" />

<Set3 ap="40" accuracy="80" /> | <Set3 ap="40" accuracy="100" />

Burst Fire Accuracy:

<BurstFire ap="40" accuracy="20" /> | <BurstFire ap="40" accuracy="25" />

Damage: +50% of vanilla(v19.5), accuracy: 50% of what Aaron suggested in accuracy test.

-------------------------------

MAG Weapons Vanilla(v19.5) | MAG Weapons Upgraded

Damage:

<type="kinetic" damage="90" mitigation="32"> | <type="kinetic" damage="90" mitigation="32">

Single Shot Accuracy:

<Set1 ap="20" accuracy="30" /> | <Set1 ap="15" accuracy="40" />

<Set2 ap="30" accuracy="60" /> | <Set2 ap="22" accuracy="90" />

<Set3 ap="40" accuracy="80" /> | <Set3 ap="30" accuracy="120" />

Burst Fire Accuracy:

<BurstFire ap="40" accuracy="20" /> | <BurstFire ap="30" accuracy="30" />

Damage: vanilla(v19.5), accuracy: 100% of what Aaron suggested in accuracy test.

BONUS: +25% Rate of Fire

---------

This not only differentiate various tiers of weapons but will let to use them together. In some situations, lower tier weapons will be better than higher tier.

</DREAM>

[Aaron]

How are people finding precision rifles with this change? The quick accuracy formula I used for this tended to give them fairly enormous potential accuracy, I'm wondering if that's too much?

Sadly, I don't think we've gotten far enough in the "Balancing Phase" to even take into account alien health, armor, numbers, or much else. 90% of the effort so far seems to be spent on balancing the weapons against each other. Plus there is the economic balancing that isn't close yet too. It needs to be done, obviously, or the game would suck, but the rate of progress really makes me think we won't have a balanced game until Jan. or Feb. 2014. I hope I'm wrong. BTW, I'm not saying anyone is slacking just that progress seems slow.

To me it seems that a great deal of the balancing could be done purely mathematically with computer simulations, then given to us to test. The weapons, armor, and hit points outcomes could certainly be simulated quickly and assuming Goldhawk had a goal in mind, the XML could be adjusted until the simulation was giving the desired results.

We are a lot further along than you are giving us credit for - weapons obviously get most of the discussion because they are the primary way players interact with the ground combat missions, which make up the largest chunk of the game. However in actuality every single variable in the game that controls anything of consequence in the ground combat/geoscape has been reset according to an overall design plan for the game, and now we are at the stage of broadly nudging those initial values to improve specific aspects of the gameplay - this accuracy buff for example - but after that it's really just tweaking. Geoscape financing is probably the final big thing to fix, but it's also a system with relatively simple inputs so that won't be a difficult task.

Also, in order to have a useful "simulation" we could use for balance we would literally have to re-implement and reasonable facsimile of the game (else it would be worthless) in some language. That is both a huge design and technical challenge, akin to making a whole new game. I have worked at a games company in the past who had help from Microsoft doing that sort of simulation based testing, and it was only possible because Microsoft already had a heavily staffed department dedicated to that sort of thing across their entire range of software - not something you chuck together on the quick (and expect it to be of any use).

[Rogueywon]

Yeah, I think the impact that Geoscape economy balancing will have is generally under-estimated. If you don't believe this, try tweaking the .xmls to do a couple of playthroughs with different economic settings. In many ways, the impacts of geoscape balance on the tactical game are bigger than many of the tweaks that have been made to weapon/unit balance directly.

Deploying against a Corvette with a squad in full wolf armour, armed with laser weapons is very, very different to going in with a squad in basic/Jjackal armour and ballistics.

[xcomnaut]

How are people finding precision rifles with this change? The quick accuracy formula I used for this tended to give them fairly enormous potential accuracy, I'm wondering if that's too much?

Accuracy vise sniper rifles are very good now. Maybe a little too good. My suggestion would be

<Set1 ap="40" accuracy="80" /> to <Set1 ap="40" accuracy="80" />

<Set2 ap="50" accuracy="150" /> to <Set1 ap="50" accuracy="120" />

<Set3 ap="60" accuracy="180" /> to <Set1 ap="60" accuracy="160" />

[Ol' Stinky]

How are people finding precision rifles with this change? The quick accuracy formula I used for this tended to give them fairly enormous potential accuracy, I'm wondering if that's too much?

The sniper's aimed shot seems a bit pointless with how good standard shot is. A rookie gets 72%-90% to hit. I think a soldier reaches the to-hit cap by the time he hits 63 ACC.

I'd use something closer to the original values once you put hypervelocity back in.

Deploying against a Corvette with a squad in full wolf armour, armed with laser weapons is very, very different to going in with a squad in basic/Jjackal armour and ballistics.

I've been fighting Corvettes in jackal/basic and man, it's a bloodbath with the increased accuracy mod. Although that might be down to me being bad at games. It's cool.

Edited August 1, 2013 by Ol' Stinky

[kabill]

I've been playing with the vanilla accuracy for 19.5 (i.e. with the short-range accuracy buff but not the general buffs) and I've come to the conclusion that I have a preference for the full accuracy buff rather than just short-range modifiers. I thought that the short-range bonus might make enough of a difference on its own, but this doesn't seem to have turned out to be the case (for me, anyway).

I'd agree with Stinky's point about the sniper rifles, though. The final accuracy tier seems irrelevant. It would be useful if they didn't get hypervelocity, since it would give you a better shot against targets in cover still. But if you're ignoring cover, it doesn't matter how high your accuracy is above 95% (ah - unless there's still a plan to add a penalty to hypervelocity).

[xcomnaut]

20+ missions done using accuracy buff. My team: 2 scouts with shield/pistol, 2 assault rifle users, 1 sniper and 1 machinegun. I have greater success compared to v18 1 sniper and 5 machineguns. Sniper rifle is deadly when there is no cover in between but when cover is used there is no way to get 95%, mostly 60% while assault rifle gets 30-40%. Still I suggest accuracy steps 80/120/160 instead of 80/150/180.

[Ol' Stinky]

I've been playing with the vanilla accuracy for 19.5 (i.e. with the short-range accuracy buff but not the general buffs) and I've come to the conclusion that I have a preference for the full accuracy buff rather than just short-range modifiers. I thought that the short-range bonus might make enough of a difference on its own, but this doesn't seem to have turned out to be the case (for me, anyway).

I'd agree with Stinky's point about the sniper rifles, though. The final accuracy tier seems irrelevant. It would be useful if they didn't get hypervelocity, since it would give you a better shot against targets in cover still. But if you're ignoring cover, it doesn't matter how high your accuracy is above 95% (ah - unless there's still a plan to add a penalty to hypervelocity).

Even when you do get higher than 95%, it's not all that helpful; the penalty from cover seems to be applied after everything else, so when dealing with 50% cover, the most you'll see from a non-hypervelocity gun is ~48%. At least, that's what the accuracy gauge in-game seems to suggest. I think the only thing it helps with are targets out of range, but since out of range targets don't suffer suppression and recieve less damage, that's not as amazing as it might sound at first.

I agree about the short range buff not being enough on its own, although it's nice to be certain of hitting a wall point blank. (;

Edited August 1, 2013 by Ol' Stinky

[Mytheos]

What file is the short ranged accuracy buff in?

[Max_Caine]

You're spot on, Stinkster. Cover and intervening obstacles are counted after base accuracy is calculated. This is because cover/intervening obstacles are handled differently to the base accuracy calculation. It used to be that an intervening object has a "stopping chance", which is incorporated into the to-hit roll, and cover is also incorporated into the final to-hit roll as well. As an example, if you have a squaddie whose base accuracy is 95%, and you shoot across an obstacle with a stopping chance of 50%, the "chance to hit" is still 95%, but if the RNG comes up with a value between 46-95, the shot would hit the intervening object. If the target was behind cover with a "cover save" of 30% as well, if the RNG was between 11-46, it would hit the cover, and between 46-95 it would hit the intervening obstacle. A shot only misses if it fails to hit the target, any cover the target is behind and any intervening obstacles.

However, things have changed since this formulae was first implemented. It's not quite as true as it once was (this original formula was why a squaddie would turn and hit the squaddie next to him), but it's true enough to be a rule-of-thumb.

-------

Mytheos, it's in config.xml.

[kabill]

Ah, fair enough. I figured it would cap at 95% *after* all other modifiers. In which case, the final tier on the sniper weapons is actually useless for almost all soliders!

[Mytheos]

@Max_Caine

ShortRangeHitBonusPerTile = 15

So that means Short range = x, at at that distance, you get +15 accuracy, and another +15 per tile closer?

And what is X?