Ground Combat Balance Discussion Build V21 Experimental 4!

On Androns

I've said a couple of times that an awesome combination for Androns was low HP/high armour. Well, after having shot a lot of Androns I'm starting to rethink my position on that. Having carried out some experiments I think that Androns could stand to be put on the same HP "track" as other races.

The thing about low HP/high armour is that the number of shots needed to kill an Andron is indistingishable from the number of shots needed to kill a Sebillian of a similar class. When playing on Veteran you don't see that armour is absorbing most of the shot - all you see is that 2-3 rifle shots kills an Andron Guard, just as 2-3 rifle shots kills a Sebillian guard, so one of the Andron's primary racial advantages doesn't shine through. I contest that you could replace the armour values with the HP track of.. say.... Sebillians and you wouldn't see any difference. So, I think that if armour is going to be an advantage for Androns they need to be put on the "main" HP track of other races so they can benefit properly from armour. Given that we don't see Androns until Corvettes, in which time it's possible to tech to both Wolf and lasers with 30 scientists I don't believe the player would be at a significant disadvantage facing an HP+ Andron.

I don't really see the need to lower the accuracy of the weapons, given we're reducing the accuracy of the aliens themselves. A 15 point accuracy decrease is a 20-30% accuracy decrease for most of them.

@Max, I agree. Androns need to keep their armor and get a serious HP buff. At the moment they don't feel that terrifying. I'd like them to feel somewhat "unstoppable"... I'd like androns to always use most of their TU's to move towards the xenonauts and not just stand far away and shoot. It would scare the sh#t out of me. "I can't stop them, they just keep coming!"

On Hollywood-esque Cars

Could the spectres for cars and vans be set to explode on destroyed? I know it isn't the most realistic, but having a car just vanish when it's taken enough punishment seems quite limp.

The big gas tanks should explode too, but they just vanish with explosion sound when shot.

Jackal armour is already cheap and quick to manufacture, and doesn't require alien materials. I'm not keen to change any of those things. Again, alien weapons only do 80% damage on the Normal mode, so it depends what we're trying to balance the game for really. Perhaps it needs a slight stat boost, but I'm not willing to entertain much shift really - Jackal armour should give you only average chances of survival where you would otherwise die compared to unarmoured troops on Veteran. Skipping it should be a valid choice, if you're willing to take a few extra soldier deaths instead.

Skipping Jackal is a choice because Jackal does hardly anything against a plasma rifle (65 damage 20 ArMt)

Assuming a soldier with 55 HP, an unarmored soldier has a 34% chance to survive a plasma rifle hit, a Jackal-armored soldier has a 57% chance, and a Wolf-armored soldier a 100% chance. A Jackal-armored soldier is only 23% more likely to survive a plasma rifle hit than a naked soldier. Wolf is only a month off and offers not only a 100% chance to survive, but a 20% chance to take no damage at all.

A marginal 23% chance for my rookies to survive a single plasma rifle hit is not worth $20,000 and 9kg. For comparison, a naked soldier with 65 HP has a 49% chance to survive compared to the Jackal armored soldier's 57%.

The problem may not even be that Jackal's stats are too low. Alien weapon's armor mitigation is very high. Against anything bigger than a plasma rifle (ArMt 30, 30, 50, 99, 60) the jackal armor is mitigated to such a degree that the soldier might as well be naked. Even the light drone's gun has ArMt of 80.

Sebs could be a bit higher numbers, less hp, less accuracy. Androns could be in lesser numbers with more hp and AI to charge you.

As for the accuracy formula:

Range Penalty = ((Weapon Range * 2) - Target Range) / Weapon Range (100% if target range <= weapon range)

Unit accuracy value is multiplied by 1.2 if the unit firing is crouching (and it caps at 100 if crouched).

Recoil Penalty = Weapon Recoil - Unit Strength (0 if no recoil or if strength > recoil).

Unit Accuracy = Unit Accuracy - Recoil Penalty

Unmodified Accuracy = Unit Accuracy * Range Penalty * Weapon Accuracy for the shot

Cover Hit Chance = Flat Cover % (if object) or Unmodified Accuracy * NoTargetPenalty (if unit).

Cover Hit Chance = Cover Hit Chance / 0.8 if the unit firing is crouching. (Not displayed on screen for object cover, but displayed for unit blocking.)

The highest cover chance of any blocking unit/object is used for the shot.

Terrain Accuracy = 100% - Cover Hit Chance - (Tile Distance through smoke/gas * Accuracy Reduction for each smoke/gas) Minimum of 0%.

Chance to Hit Target = (Unmodified Accuracy * (0.8 if target is crouching) + Short Range Hit Bonus) * Terrain Accuracy.

If Chance to Hit Target > MaxChance, Chance to Hit Target = MaxChance.

If Chance to Hit Target < MinChance, Chance to Hit Target = MinChance.

Then, random roll from 1-100. If roll > Chance to Hit Target, you miss.

If you miss and if your roll is > Unmodified Accuracy, your shot scatters and rolls a different trajectory.

If you miss, each blocking unit/object it passes, it rolls to see if it hits it based on the cover hit chance.

OK, I was ill on Friday so I've been catching up with this thread now. Thanks for the maths lesson, all. It's important we're all working with the same assumptions before we analyse the performance of the various weapons. Perhaps the more normal / aimed shots need increased accuracy, particularly as the aliens are receiving a 15 point accuracy penalty across the board.

Thank you! Really though 70 accuracy (63 with .9 modifier) is way too high for a guard. A 15 point nerf to accuracy sounds perfect IMO.

In addition to the global alien accuracy nerfs... I do agree with StellarRat though that snapshot TU/hit efficiency vs aimed shot TU/hit efficiency needs to be looked at for mid/long range. I personally think you should nerf snapshot accuracy instead of buffing aimed shot accuracy. The only time an alien should take a snap shot instead of an aimed/burst outside of 5 tiles is if he burned up his TUs moving.

I disagree however with his contention that burst fire should be a close range only weapon mode and only be good for suppression outside of that. Midrange is where burst fire was meant to be used IMO. Semi accurate fire that suppressed if it doesn't kill sounds perfect to me. Aimed shots should only be for long distance (think yellow tiles) shots, or if you are conserving ammo.

As for the accuracy formula: ...

Great!

I've started a spreadsheet with chance-to-hit calculations per range:

https://docs.google.com/spreadsheet/ccc?key=0AkNMPlaGZUcLdDlUYWlUeUhlQ2c2OFlZZW04WXZxN2c&usp=sharing

The only missing details:

How does "NoTargetPenalty" work?

How does "Short Range Hit Bonus" work?

Once that is in, we can just copy the sheet for each shot-type, add some graphs and compare

Short range hit bonus works by adding the bonus after modification (which is +12) for every tile the target is within 5. Ergo, within 5 tiles +12. Within 4 +24. And so on until 1 tile distance away, which would be +60.

Also, called it. (Sort of)

The NoTarget penalty only affects units that the player is not aiming at. If the bullet passes through a tile containing a unit that you are aiming at, it will hit it 100% of the time. If it passes through a tile containing a unit you are not aiming at, it will hit it 25% of the time.

Basically there used to be an exploit where you could aim at the tile just behind the target and hit much more consistently due to the bullet scatter not deviating very far on a "miss" shot, it's just intended to fix that.

Nice spreadsheet too. It needs the burst fire accuracy values, plus accuracy levels for up to four "zoom levels" for snap shots too if it's to be fully usable though. Turns out the accuracy calculation has got fairly long by now!

Short range hit bonus works by adding the bonus after modification (which is +12) for every tile the target is within 5. Ergo, within 5 tiles +12. Within 4 +24. And so on until 1 tile distance away, which would be +60.

Also, called it. (Sort of)

Thanks, Added

And of course, in a discussion about balance of hit chances you need numbers

Nice spreadsheet too. It needs the burst fire accuracy values, plus accuracy levels for up to four "zoom levels" for snap shots too if it's to be fully usable though. Turns out the accuracy calculation has got fairly long by now!

Aimed and snap are just straight copies of the table with different "Weapon Accuracy" values and burst a copy with the "two shots" columns removed. What do you mean with "up to four zoom levels"?

Sebs could be a bit higher numbers, less hp, less accuracy. Androns could be in lesser numbers with more hp and AI to charge you.

As for the accuracy formula:

Range Penalty = ((Weapon Range * 2) - Target Range) / Weapon Range (100% if target range <= weapon range)

Unit accuracy value is multiplied by 1.2 if the unit firing is crouching (and it caps at 100 if crouched).

Recoil Penalty = Weapon Recoil - Unit Strength (0 if no recoil or if strength > recoil).

Unit Accuracy = Unit Accuracy - Recoil Penalty

Unmodified Accuracy = Unit Accuracy * Range Penalty * Weapon Accuracy for the shot

Cover Hit Chance = Flat Cover % (if object) or Unmodified Accuracy * NoTargetPenalty (if unit).

Cover Hit Chance = Cover Hit Chance / 0.8 if the unit firing is crouching. (Not displayed on screen for object cover, but displayed for unit blocking.)

The highest cover chance of any blocking unit/object is used for the shot.

Terrain Accuracy = 100% - Cover Hit Chance - (Tile Distance through smoke/gas * Accuracy Reduction for each smoke/gas) Minimum of 0%.

Chance to Hit Target = (Unmodified Accuracy * (0.8 if target is crouching) + Short Range Hit Bonus) * Terrain Accuracy.

If Chance to Hit Target > MaxChance, Chance to Hit Target = MaxChance.

If Chance to Hit Target < MinChance, Chance to Hit Target = MinChance.

Then, random roll from 1-100. If roll > Chance to Hit Target, you miss.

If you miss and if your roll is > Unmodified Accuracy, your shot scatters and rolls a different trajectory.

If you miss, each blocking unit/object it passes, it rolls to see if it hits it based on the cover hit chance.

Most of it you can get from the Xeno wiki, and the rest you can piece together from this forum.

Where did you get all that?

Xenonaut debugger logs using the tool Chris posted on another thread:

Regarding the accuracy calculations, try playing the gc_editor.exe version of the game while running the debugger: www.xenonauts.com/devfiles/xeno_debugger.exe

That will post debug information on the numbers the accuracy formula is using for each shot.

Xenonaut debugger logs using the tool Chris posted on another thread:

The NoTarget penalty only affects units that the player is not aiming at. If the bullet passes through a tile containing a unit that you are aiming at, it will hit it 100% of the time. If it passes through a tile containing a unit you are not aiming at, it will hit it 25% of the time.

Basically there used to be an exploit where you could aim at the tile just behind the target and hit much more consistently due to the bullet scatter not deviating very far on a "miss" shot, it's just intended to fix that.

Is there a way to make bullets that pass target tile to not hit it 100% without causing that exploit? I just hate the visual effect where missed bullets always hit the tile under targets feet. It looks completely fake and gives an impression that everybody is aiming each others feet. At worst it gets really comical when two units shoot each other multiple times at close range and all the bullets stop at their feet and I'm like "shoot higher dammit!"

All the techincal stuff aside here are my casual gamer impressions of the ground combat AI at this point. I felt it was pretty balanced for the first 3/4 of my game (set to Normal). Then, upon reaching the abiblity to produce Sentinal Armor, Plasma weapons, and Fusion Explosives the aliens kick my ass in every battle. I went from losing 1 or 2 guys (usually new recruits) every few mission 8 man team to losing everyone or almost everyone.

I currently have a 6 man team with 4 vets (Commanders and Majors), a scimitar, and 2 Privates, and they can't clear out a downed carrier, battleship, terror site, or make it halfway through an alien base. If I do make it there's only 1 guy that usually gets lucky and makes it out - of course I usually start over because it's even harder to make it through a mission with an entire team of Privates.

My strategy is usually pretty effective. The scimitar scouts ahead for areas with cover, and my team moves ahead 3 at time with 3 always covering. The problem is that the aliens are now always crack shots and almost always score a hit - even on those behind cover.

I was in an alien base and had 4 guys in a room (4 had already been killed), all behind some type of cover. Two aliens came in the room, and one shot the closest guy in predator armor 3 times and killed him. The second alien shot a guy in Sentinel armor on the other side of the room behind a crate 2 times and killed him. The private with them panicked and started out the next turn with 0 TU (so he was shot dead), and the last guy left was vet in Sentinel armor that was already weakened. I was just over half way through the base at this point so I just gave up.

I haven't completed a terror mission, downed UFO, or base attack in days! I have 2 bases, one with 2 marauders, a foxtrot, and a condor, and another with a marauder, corsair, and 2 condors and there are so many freakin' UFO's I'd need 6 bases filled with marauders to even put a dent in them. I'm bleeding money and have to airstrike every UFO just to stay afloat.

Basically the game went from a medium difficulty to overwhelmingly hard within a matter of 1 game year!

All the techincal stuff aside here are my casual gamer impressions of the ground combat AI at this point. I felt it was pretty balanced for the first 3/4 of my game (set to Normal).

I currently have a 6 man team with 4 vets (Commanders and Majors), a scimitar, and 2 Privates, and they can't clear out a downed carrier, battleship, terror site, or make it halfway through an alien base.

Basically the game went from a medium difficulty to overwhelmingly hard within a matter of 1 game year!

The farthest I made it before was to the MAG weapons about 6 months ago. At that time I was devastating the aliens on the ground with a 3 man team of Majors in Sentinel armor. I could thwart terror missions, infiltrate bases, and clear any downed UFO with those guys. Now it's just downright painful. A few minutes ago I just had an armored Sebillian shoot a guy through a window. Then when I tried to return fire at the same angle through the window my shots were blocked by the wall, so apparently only the alien can get that line of sight and not me. Then he ran out of the building, threw a grenade that exploded and damaged 2 of my guys, shot another guy dead, then ran back in the building. During that exchange only one of my 3 guys facing him reactioned fired. Ugh. Now I'll start over and wait for 5 minutes while the map reloads.

Also, just got the Valkyrie drop ship. A waste of money! It provides no cover when sending in troops. I dropped into a downed battleship zone in the desert and had no cover to place my guys behind, so I just shrugged and clicked end turn. Well, my scimitar got destroyed in shot, then 3 out of 7 of my troops got killed before my turn started. Sigh, restart. 5 minutes later after the level loads I re-place my guys. Scimitar gets destroyed, and my guys kill the 2 aliens with reaction fire. Yay!

I make it to the battleship with only 1 human casualty, so I think I'm doing good so far. I decide to teleport into the side rooms from those platforms outside. 2 guys on 1 side, and 4 on the other. I'll secure those rooms and meet in the middle. After I teleport in and end my turn an alien runs into the doorway and kills 1 of my guys then runs out and another one runs in and kills another guy. In the other room an alien runs in and throws a grenade and damages my other 2 guys. Then a second alien steps in and shoots them both to death with 1 shot each.

I end my turn with the loss of 4 guys! When my next turn starts my commander has panicked and is useless while the noob in the Predator suit keeps it cool. I walk him a couple of steps and shoot an alien standing outside the door and kill it, finally! However, TU's are gone, he panics on the next turn, and the aliens kill him. 7 guys dead in 2 turns = no fun.

A few minutes ago I just had an armored Sebillian shoot a guy through a window. Then when I tried to return fire at the same angle through the window my shots were blocked by the wall, so apparently only the alien can get that line of sight and not me.

Yeah, there's something wrong in the LOS system when dealing with walls, corners and windows at the moment. Often I can see an alien behind a corner but can't shoot it or, just like you described, I can't shoot alien but it can shoot me without a problem. Very annoying and I really hope GH can iron these incosistencies out.

Basically, you need to prove to me that it's advantageous for the AI to burst fire with a plasma rifle rather than take an aimed shot. The numbers are:

<Set1 ap="25" accuracy="85"

<Set2 ap="30" accuracy="100"

<Set3 ap="35" accuracy="120"

<BurstFire ap="40" accuracy="40"

I'll look at the post again tomorrow, but as far as I'm aware three shots at 40 accuracy using burst fire gives you 120 cumulative accuracy for 40 AP spent, whereas you can get one shot at 120 cumulative accuracy for 35 AP with an aimed shot. Therefore the aimed shot is better. You need to prove my maths wrong before I understand the problem, if there is a problem.

EDIT - having read your mathematical proof above (using admin magic), what you're saying seems to agree with what I'm saying. An aimed shot is a more efficient use of AP than burst fire is, and two snap shots have a cumulative 80% chance of hitting with at least one shot - making them the same as burst fire.

However, burst fire is 10 AP cheaper than two 2 snap shots, so why would the aliens ever use two snap shots? And an aimed shot is 5 AP cheaper than burst fire, but is 4% more effective according to your figures.

Therefore, what I'm asking you is why you think it would be beneficial for the AI to not be using aimed shots. We can discuss what's actually happening in the game once I understand the assumptions you are making. Everything I've seen so far suggests that the AI should be taking aimed shots in preference to any other action.

Right, now that my spreadsheet is more or less finished(I don't know the min and max tohit chance), lets come back to this.

The following assumes a shooter with accuracy 0.6.

In the third tab I've graphed the expected damage as fraction of the weapon damage, versus the range, for 1 aimed shot, 2 snap shots and a 3 round burst.

The result: at point-black range (1, 2 or 3) the 3rd burst is best. At rang 4 to 36 the 2 snap shots have better expected damage. After that the burst takes over again (due to the minimum hit chance)

At no point is the aimed shot best, and most of the time the aimed shot has exactly the same expected damage as the burst shot.

Conclusion: For damage output 2x snap is best, and burst and aimed are the same.

In the second tab I've graphed the chance to hit any shot for 1 aimed shot, 2 snap shots and a 3 round burst. It's mostly similar, with 2x snap shot being the best most of the time. Burst is actually worst most of the time.

Conclusion: For finishing blows on nearly dead enemies 2x snap is the best and burst the worst. Though considering the small difference between 2 snapshots and 1 aimed shot, and that 1 aimed shot is 15AP cheaper than 2 snap shots, it's probably better to use 1 aimed shot to finish something off.

When you increase the shooter accuracy to 0.8 interesting things happen with the Chance to Hit. A single aimed shot has the highest chance to hit for normal range, but at long range two snapshots become better.

Yeah, the later parts of the game are less well balanced simply because less people reach them and therefore there's less comment on them. We will be doing full playthroughs for our final balancing though so we'll get there eventually - especially as I'm currently testing the final mission.