Thoughts on increased randomness

[Mytheos]

I am really unconvinced we'd gain anything from adding hit locations or damage modifiers - can anyone explain the advantage it would bring in a concise way?

I am also quite amused by how hard done by people feel about the damage spread - all you cynics! It's always "the game cheated me out of my soldier" rather than the other 50% of the time where it is "I thought my soldier was toast but the RNG saved his ass!". No love for RNG?

At the end of the day, I think enforcing a level of randomisation like this means you need to plan your actions in the game in a more organic and realistic way - General Montgomery did not know for certain that 3 of his tanks had a 75% chance to defeat 2 of Rommel's panzers (though he knew a dozen had an even better chance...); any commander in the real world must plan for screw ups, reversals, unexpected foes and - even - surprise victories. In Xenonauts you should be trying to build a team that is resilient to unexpected losses.

I'm not opting for hit damage location, that seems somewhat out of scope and can be replicated with simpler mechanics.

But in reality if a player gets lucky during a round of shooting, they say wow that was too easy lol, and forget about it.

However if they get screwed with a below avg result, they will remember it, simple gambler's fallacy, so you are right more people are going to notice the negative side of things.

I was just suggesting tightening the damage min/max the closer you are to the target, or splitting the + range and -ranges based on distance. (Max distance +25% -50%) (Close Range +50% -25%) Preferably using a gradient.

-This would give an additional mechanic to help balance Close range combat/weapons.

-This would add more tactical depth to the AI (Them running closer on their turn and firering would be benefited more, and would aid certain Alien types using aggressive AI)

-Added sense of danger in close combat/UFO breaches because you know the max possible damages are higher.

-It would "feel" more right in the same sense that having a higher accuracy boost at close range reduces odd results of missing 2 out of 3 times point blanking something with a shotgun. (What the crap??! 3 shots I miss 1 and then hit for 32 and 35?? with a 30-90 damage range?)

-Reduces the impact of RNG, or rather gives the player more control over it.

-On 80 damage weapons, a 40-120 damage range is rather large, and you end up with a 3 poor shots = 1 great shot situation, which can translate to when it helps, feeling like easy mode, and when it doesnt feeling cheated.

-Creates more consistent gameplay, 4 Aliens? Crap! 1 soldier takes 4 shots, bam all dead. Wow that was crazy! 5 mins later...Crap 4 Aliens again! And the result is a 4 turn battle with a bunch of misses and low damage shots....

But at the end of the day seeing a soldier point blank something 3 times and hitting it in the foot is the same thing as missing 2 out 3 times at point blank range, and people were complaining about that forever until accuracy increased and Close Range Accuracy Bonus was put in.

Summary:

It better simulates the whole point of variable damage and what it represents, it's more realistic, its more intuitive, it allows more player control in managing risk vs reward, it creates an additional and complementary tactical layer that supports existing accuracy bonuses for close range combat, not to mention serves as another tool to balance close range weapons.

Edited August 20, 2013 by Mytheos

[Aaron]

This is sophistry. In real life, you're not planning down to the individual number of rounds fired from a magazine by troops in the assault against an unknown number of combatants over absolutely unknown terrain, either, because real enemies don't have X HP, and different guns don't do Y damage.

This is, however, the case in Xenonauts.

I don't see how the level of detail possible in planning/simulation has any bearing on whether chance should play a role in the outcome? Are you suggesting that the commander in Xenonauts plans things down to an atomic level and all actions are thus fully deterministic?

[Voeter]

Fun discussion. Can't resist putting in my 2 cents

TL;DR: Add a range modifier to weapons to keep the overall system intact, but add a bit of realism and make the game "feel" better

I feel like we're discussing whether you prefer playing a game like chess over a game of poker. In chess 100% of what you do makes 100% of the difference in the outcome of your game/battle. In poker it's a lower percentage because part of it is caused by the sheer randomness of chance. IMHO the latter suits and simulates a "real" battle a -lot- more. Why? Because it accounts for a lot of "real" variables not directly accounted for in the calculations the game makes itself: a strong NNE wind blowing right as you fire your shot, angling it away a degree, a quick jerk-like reaction from the intended target, causing the bullet/laser/plasma bolt to end up 2 inches away from the intended target, etc. etc.

That said, I really love a good amount of randomness in the damage you cause with your shots, simulating (in essence) the part of the body you hit. This randomness in -no way- diminishes the tactics needed to be successful, considering both you -and- the enemy gets the same (dis)advantage from this system. It only causes the need for you to adapt your strategy to be successful and (of course) that it is more difficult to get through every mission with all your xenonauts alive. In fact, if there is -one- thing I really love about the x-com(like) games it's the fact that even though you might have a really good chance to be on top in any given situation it could all go to hell and you lose Rambo "Predator" Terminator you lovingly crafted into the badass he is today. I honestly feel like there's a lot more emotion and thus fun in going all-in with a pair of aces with a chance of losing against someone than knowing that the next move with your queen will net you the checkmate 1000/1000 times. It might give you the satisfaction of the win, but -not- the thrill of knowing you could actually lose that queen and have to rethink your strategies. If you do -not- like the first option, there's always the option to save and load.

Now I feel most, if not all of you agree that at least a certain randomness in the game is wanted and the main gripe I picked up over the last 10 pages is the fact that some situations don't "feel right". (actually I mostly have this with the grenades in xenonauts, but I guess it happens

I think there's a fairly simple solution that works well to compensate any mentioned gripe, and that is to add either to hit modifiers per weapon for short, medium and long engagement ranges or both to hit and damage modifiers. The latter option might seem the better choice for some (a shotgun or rifle does more damage at point blank range than it does at 15 yards, right?) but actually it will only lower your chance to hit if you don't want your average damage done to become insane (ie. a shotgun almost always hitting AND instagibbing an alien). Here's a small table explaining both situations, where the average damage done per hit is still the same as is current per the wiki (ie. 25 dmg for the ballistic rifle and 35 for both shotty and sniper rifle):

The important parts are the hit chance and the damage done per shot. I hope most of the table is self-explanatory but here is a breakdown for the shotgun why this "feels" right and/or is realistic: At small range the shotgun is outfitted best to hit and do it's awesome damage potential. This is explained by the highest dmg/shot on that range for any weapon AND the fact that it does the most damage/shot (= absolute damage) along with the sniper rifle. A sniper rifle might do the same damage per hit, but of course it's much harder to hit at this range with the weapon and optics designed for medium to long range engagements.

These numbers are of course arbitrary and you can fiddle around with the to hit % and absolute damage a bit if wanted/needed. I'd say keep the current to hit chances for the medium range and adjust for short and long ranges per weapon so the damage/shot changes appropriately.

Good idea or not?

@ Aaron & Chris: is this something you guys would be willing to consider implementing or should we drop thinking along these lines?

[edit] a couple of typo's and such.[/edit]

Edited August 20, 2013 by Voeter

[Ishantil]

What I meant was a random hit location table to simulate the variance in combat. What makes a wound survivable and not, that sort of thing. It makes the combat a bit more interesting in that you can have different kinds of effects based on where you got it (for both the player and the Aliens). But it brings another layer to the game play, and I'm not sure if it's right for this game. I'd like to see it in there, but that's personal preference.

I've always liked the Battletech hit location table. In our case, we would not differentiate between right, center, and left torso locations. Additionally, 2D6 skews naturally towards the results of 6, 7, and 8. A headshot and an automatic critical (being roll sof 12 and 2 respectively) are both 1 in 36. http://www.sarna.net/wiki/CBT_Tables#BattleMech_Hit_Location_Table

Additionally, you could apply modifiers to this table based on circumstance. If I were going to make the table for this game, I would probably rearrange it so all the leg shots are low numbers and all the arm shots are the higher numbers, that way you could use a modifier to skew towards and and low shots (someone behind half cover isn't as likely to be hit in the leg, so grant a bonus of +1 to the result before checking the table).

You get the idea.

[EchoFourDelta]

@Aaron

No, of course not. But in-game, you're able to plan down (and have to do so) to a much greater degree than in the example you gave, due to how few shots and how few actions we're taking in-game compared to a real-life scenario. Of course chance does (and should) play a role, but the current model is so wild in both directions that it just seems... off.

[cody522]

How about this then, the damage variance increases slightly if you go from snap shot, to normal, to aimed ? If you're aiming, the soldier more than likely will hit a vital organ than he would if he shot from the hip. I do not like the idea of this system entirely focusing on luck alone.

Anybody?

[Ol' Stinky]

Anybody?

What damage variance would burst have? I like the idea though.

[StellarRat]

You guys do realize that if shot placement were added and certain parts like the head increased damage that the damage variance will be MUCH LARGER than it is now? People will get their heads blown off from time to time and then you'll all be crying about "How could my best guys with mega-armor get one shotted from across the map! WHAA, WHAA!" If you don't like the variance now think about how bad it will with this mechanic added.

[Voeter]

What damage variance would burst have? I like the idea though.

I don't' date=' because the average damage you do will already go up simply because of higher chance to score a hit. I know my last post was really long, but at least check out the two damage tables. Here you see that if you want the average damage per shot to remain the same, the only way to increase damage done on a hit is to [i']lower accuracy[/i]. You -could- of course also lower average damage of the weapon to compensate for your idea, but that would simply mean that aimed shots become more and more cost-effective in -any- situation, eventually making snap shots and normal fire virtually useless at -any- range.

Currently using an aimed shot is only cost effective (in damage per TU used) at (near) maximum range of the weapon used. If you increase the (average) damage on aimed shots, keep it the same on normal shots and lower it on snap shots it very quickly. If you up the damage on aimed shots by only 10% (and lower it by 10% for snapshots) it's almost as damaging to use aimed shots as it is using normal or snapshots. Skew the damage by 20% and it'll always be best to use aimed shots. It's mathematically possible to change the system to incorporate your idea, but that would mean changing the basics of the entire system right now.

[EDIT]By the way, it might be important to note that I did not simple increase variance in my calculations as you mentioned, but I skewed it in favor of aimed shots (and against snap shots, with normal shots remaining the same as it is now). Simply increasing variance would also mean you have a chance to score a lot LOWER damage hits on aimed shots and that did not seem the intent of your idea.[/EDIT]

Edited August 20, 2013 by Voeter

[cody522]

What damage variance would burst have? I like the idea though.

The maximum amount of damage would stay the same since a snap shot could still hit you in the head. The minimum damage bar would raise though as the shot becomes Normal, or Aimed. Burst shot will probably be treated as a snap shot.

So something like this for a weapon that has a base damage of 60 with a +\- 50% damage variance.

Snap\Burst: 30-90

Normal: 38-90

Aimed: 46-90

[Ishantil]

You guys do realize that if shot placement were added and certain parts like the head increased damage that the damage variance will be MUCH LARGER than it is now? People will get their heads blown off from time to time and then you'll all be crying about "How could my best guys with mega-armor get one shotted from across the map! WHAA, WHAA!" If you don't like the variance now think about how bad it will with this mechanic added.

Well, technically if it's applied both sides, it's fair, right? And it actually doesn't have to be that way, depending on how you implement the damage location system.

You could also use aimed shots in the game (for snipers) or something. I dunno, it's all about how you want to do it. On the other hand, you can keep it simple like it is.

Notably, since the aliens almost always have better equipment, the game will almost always skew towards the aliens. Which is why the game feels more lethal with the increased variance. Notably, I'm not saying that's necessarily bad, just pointing it out.

[Josip Broz Tito]

I think the system's fine as is. Shotguns need a tweak, but that's about it IMO. The 50% +- modifier is fine IMO for these reasons:

1. If you think about it, this turn-based system actually represents a short amount of time which doesn't lend to accurate shots at all times.

2. You're fighting cloned aliens designed to sustain wounds that would kill a human and keep fighting: notice how they don't bleed out. Their pain tolerance has likely been artificially boosted too.

3. The damage system in the game represents physical trauma AND pain tolerance. There's only so much dmg a body can take until it's unusable, so our soldiers will never have alien amounts of health (realistically at least). They will, however, become more resistant to the mental affects of pain and autonomic damage responses as shown by gradual health point increases.

So that means that the numbers we generate with the +-50% modifier already represent 'critical' and 'flesh' wounds. This is a proven system from the originals and has already led to more tense gameplay. It's forcing us to play more carefully- and that's a good thing.

As for losing soldiers seemingly randomly= there's always something that you could've done better. Whether that's making sure another soldier's there to cover the first, throwing a smoke at your feet, or just simply taking your time and never letting soldiers end a turn out of cover or un-croutched. There is some RNG, but the fun of the old games was learning to play better to minimize the risk.

At this point, I really feel that the aliens need the dev's attention to ensure that they will be varied and challenging.

[Chris]

Thus far we're quite happy with the effects of the 50% variance we've added. I think increasing it any more puts us in the realms of where luck may become a bigger factor than the equipment you have, but I think most people agree that we've seen a big improvement from the 20% it was previously.

It could be made more complex and and realistic but I don't really think it's necessary, and represents further investment of implementation and balancing time I think would be better spent elsewhere.

[Ishantil]

Fair enough, Chris. I'm pleased that there was a lot of debate about the subject. The general consensus seems to be that the 50% variance adds elements to the game. I would definitely say that it makes the game feel more like X-COM 1994.

[dmz]

Actually I count 6 people in favor of the current spread and 5 against (in one way or another) on this very thread. The pro group is larger, but by just 1 person. Anyway - if it will stay the way it is now, will we be able to mod it in order to have a weighted spread? Or, at the very least, mod the percentages?

EDIT: not to say this means it can't be stay as it is (despite the fact I'd love to see a weighted spread), just that it isn't most people.

[Mytheos]

Honestly it is too early to have this conversation in any matter other than sharing 1st experiences and talking theoretics.

It needs a little time being tested before anyone would make a change.

But as it stands, the pellets on the shotgun should be great to help balance them, and this replicates a random damage system effected by range, two thumbs up to the Devs for putting in the time for this.

However for all other weapons, there is little reason to ever want to get closer. Sure you get a slightly better accuracy bonus, but you have to spend TUs to move, and take takes away from shot #, which more bullets = more cover destruction and Suppression.

So even at a statistical disadvantage you still are better off just putting more bullets on the target in many cases.

Having a system that rewards getting close with both positive effects on random damage and accuracy, creates an interesting tension, and helps to buff close combat.

Tactics are a major reason why shotguns are hard to balance.

You can either take all your shots 1st and then send in the shotgunner, which often just kills the target, so you feel like you never send them in much and it would probably be better if they just had a rifle or Machine gun.

Or you send them in 1st and if they dont kill the Alien, you tend to have problems taking shots to finish it off without risking shooting them.

So before you knew the damage to expect, and you got it. You just had to win the hit rolls. Now you have to win the damage rolls and the hit rolls, and over 10 missions sure it'd average, but you put yourself at a much greater risk from many different elements...and dead soldiers = bad.

Its risk vs reward, and the risk is great and the reward isnt generally worth it. Variable damage nerfs close combat more than it helps, you just need to have more expected results if you want to run the risk, and now they are more varied.

Edited August 21, 2013 by Mytheos

[Josip Broz Tito]

I think that just means that shotguns need some attention. Now that I have wolf armor on some of my troops, the shotguns show promise.

The devs have said they exist for closed-in spaces. Not many closed in spaces exist in the game atm. In the previous build I used carbines on a few soliders that I would send into UFO's to clear them and they worked well for that purpose. That alone has made it hard to test them.

Alien bases could incorporate more maze-like areas, aliens could use buildings more often, and UFO's could receive slight interior randomization to encourage/force you to get in there.

Regardless, without randomization we would lose the X-Com feel (which is the whole point of Xenonauts), tension, and a chunk of realism.

Sure you shot the sebillian with a rifle...that doesn't mean it should always do the same damage, especially if it didn't get hit in a vital area. If a person gets shot in real life the same damage isn't always 'dealt' to them. Fleshwounds exist, now they're actually represented in game.

As for soldiers dying: X-Com and TFTD have always punished players for exposing their soldiers. Start using snap shots to make sure you can get your guys back out of alien LOS and always engage multiple soldiers against one alien at different angles. That's what we had to do in the old days. The game's harder until you get armor, just like the OG and TFTD.

Now as for a possible theoretical compromise: The 50%+- could be checked against a soldier/aliens accuracy stat to determine an overall tendency for the modifier. A person w/ 75 accuracy might receive positive modifiers to dmg more often than a sebillian w/ 50 accuracy. This keeps 'fleshwounds' and a sense of realism while allowing for more vitally aimed shots for veterans. Seems like an idea I'd be comfortable with. What do you think of the above idea, Mytheos?

[StellarRat]

The devs have said they exist for closed-in spaces. Not many closed in spaces exist in the game atm.

I've used the shotgun to great effect inside UFOs. When every shot is less than 7 tiles they come into their own. With the addition of smoke that blocks all fire and LOS they are much more valuable.

That being said, the AI seems to show not much interest in occupying buildings in field combat and I don't understand why. They generally give the best protection and allow for keyhole shots out the windows and it's where the civilians (aka "targets") like to hide. If we had to clear more buildings, the game would not only be tougher, more tense and more interesting, but the shotguns, grenades, C4 would be much more useful too. I think GJ needs to take a look at his AI routines and see why the aliens don't like houses much. In the OG they loved buildings and that was usually the most fun/dangerous place to hunt them. Goldhawk has a lot of effort into designing beautiful building interiors, yet I hardly ever have to enter a building to win a scenario.

Edited August 21, 2013 by StellarRat

[Josip Broz Tito]

Agreed with the building comment, I also look forward to more Z-Axis combat.

Notsomuch in the OG, but TFTD had a lot of combat with enemies higher and lower than you. This game needs that, especially in alien bases.

[GizmoGomez]

With Floaters popping up all the time, and with Flying Armour, there was plenty of Z Axis combat for me.

I second the request for more building-friendly aliens. I want to breach walls. (Of course, the C4 is oftentimes much too powerful to breach human buildings; at least without blowing half the building apart. Being able to set a tiny charge that blows one or two wall sections would be awesome; there's a mod out that I've been meaning to try out that does just this.)

[kabill]

I've used the shotgun to great effect inside UFOs. When every shot is less than 7 tiles they come into their own. With the addition of smoke that blocks all fire and LOS they are much more valuable.

That being said, the AI seems to show not much interest in occupying buildings in field combat and I don't understand why. They generally give the best protection and allow for keyhole shots out the windows and it's where the civilians (aka "targets") like to hide. If we had to clear more buildings, the game would not only be tougher, but the shotguns, grenades, C4 would be much more useful too. I think GJ needs to take a look at his AI routines and see why the aliens don't like houses much. In the OG they loved buildings and that was usually the most fun/dangerous place to hunt them. Goldhawk has a lot of effort into designing beautiful building interiors, yet I hardly ever have to enter a building to win a scenario.

+1. Lack of in-building combat is one of my major disappointments. I always assumed it was because it's difficult to place spawns in (random) buildings as a result of possible prop/spawn overlap, but actually this shouldn't be any more of a problem then placing spawns on any other submap.

You're completely right about C4, as well - when I first started playing 19.7, I had a mission where I was 'lucky' enough to get two aliens in a building with only one entrance. No way I could attack through it, so I used C4 to open up one of the walls. Probably the most useful deployment of C4 I've had.

[Josip Broz Tito]

I understand the argument against a breaching charge (just shoot the wall)...but it is on my want list right along w/ a plasma spike melee weapon. That's right: I want a lethal melee weapon. Trying to get some Terror from the Deep going in Xenonauts : )

The balcony rooms in alien bases in the OG were some of the best combat I had in that game.

[Mytheos]

I think that just means that shotguns need some attention. Now that I have wolf armor on some of my troops, the shotguns show promise.

The devs have said they exist for closed-in spaces. Not many closed in spaces exist in the game atm. In the previous build I used carbines on a few soliders that I would send into UFO's to clear them and they worked well for that purpose. That alone has made it hard to test them.

Alien bases could incorporate more maze-like areas, aliens could use buildings more often, and UFO's could receive slight interior randomization to encourage/force you to get in there.

Regardless, without randomization we would lose the X-Com feel (which is the whole point of Xenonauts), tension, and a chunk of realism.

Sure you shot the sebillian with a rifle...that doesn't mean it should always do the same damage, especially if it didn't get hit in a vital area. If a person gets shot in real life the same damage isn't always 'dealt' to them. Fleshwounds exist, now they're actually represented in game.

As for soldiers dying: X-Com and TFTD have always punished players for exposing their soldiers. Start using snap shots to make sure you can get your guys back out of alien LOS and always engage multiple soldiers against one alien at different angles. That's what we had to do in the old days. The game's harder until you get armor, just like the OG and TFTD.

Now as for a possible theoretical compromise: The 50%+- could be checked against a soldier/aliens accuracy stat to determine an overall tendency for the modifier. A person w/ 75 accuracy might receive positive modifiers to dmg more often than a sebillian w/ 50 accuracy. This keeps 'fleshwounds' and a sense of realism while allowing for more vitally aimed shots for veterans. Seems like an idea I'd be comfortable with. What do you think of the above idea, Mytheos?

I am for variable damage, and if it stays as is, better than nothing, even tho that positive comes with some negative if not balanced.

Making it stat based doesnt really cover the "point blank you shouldnt be hitting the guy in the foot 3 times as often as you put 3 in his head situation"

If and of itself its an interesting idea, but it doesnt solve it being more intuitive to have more damaging shots happen more often at closer range vs long range.

[StellarRat]

+1. Lack of in-building combat is one of my major disappointments. I always assumed it was because it's difficult to place spawns in (random) buildings as a result of possible prop/spawn overlap, but actually this shouldn't be any more of a problem then placing spawns on any other submap...

The spawn locations might be an issue, but really the aliens should be "drawn" to take up defensive positions in buildings even if they start outside. They ought to be seeing the buildings as highly desirable places to be. At GC start there is plenty of time for them to move from wherever they are into better positions (without being spotted 90% of time.) Only the really aggressive aliens (Androns, Flying Discs) would advance in open after the Xenonauts land or are spotted unless they had a significant advantage in numbers or firepower. Also, as I pointed out, the aliens get "points" for killing civilians and the civilians generally hide in buildings. Edited August 21, 2013 by StellarRat

[Josip Broz Tito]

I can see your argument for closer=more damage. You're starting to turn me a bit It would be a nice offset to the less-risk long-range combat that seems to work better this build (for me).

Super-ultra compromise: random variable in ode to x-com/for tension influenced by a range offset to damage and possibly influenced by the accuracy stat as well.

Another part of the reason I advocate the variable is that it leads to higher volumes of fire directed at the aliens. IRL, lots of bullets are exchanged in infantry battles. It feels right to have to empty clips over the course of a mission. In 19.6, I often got away with 1 magazine being used per soldier. Kinda felt wrong to me. 19.7, I'm using a magazine and part of another in many cases: for me, that feels better.