rocket spread info is missleading

[dario_gaston]

How much is one tile supposed to be in the game? because when rocket spread indicator says "2 tiles" the center of the explosion is usually at least 4 floor squares from the target square.

[Dlareh]

The indicator is very misleading. It is not telling you the max spread, as would be intuitive.

What it's basically saying is that on average the rocket will land that many tiles away from where you're aiming, which is much different information.

[MaxAstro]

The way it is displayed is fairly counterintuitive. The devs mean "+/- 2 tiles" to mean "a number of tiles away from the target on average equal to 2, plus or minus some amount". For many people, however, "+/- 2 tiles" reads as "plus two tiles or minus two tiles from the location you target". I know it took me a long time to stop thinking of it that way.

Even though it would be less accurate, there would probably be less complaints about the scatter readout if it simply read "2+ tiles", because that would calibrate people's instinctive expectations better.

[Tuhalu]

There are two numbers. The number on your cursor and the number on your hotkey bar.

The number on your cursor represents the average deviation from the targetted location. This is based on your accuracy and the distance to the target (the further away you aim, the higher the accuracy penalty).

The number on your hotkey bar is max scatter. This the most you can deviate from the targetted location. Without Fire in the Hole, this is 6 if you have 1 action left and 4 if you have 2 or more actions left. Fire in the Hole reduces max scatter by 2 (so 4 and 2).

Both numbers are important to know. The lower the average deviation, the more likely you are to hit close to your target. The higher your max scatter, the more you can miss by if you are unlucky.

Note: The game determines how far the rocket scatters by making an accuracy check for every tile of max scatter on the rocket. It then randomly picks a direction and scatters the rocket onto the closest tile in that direction and at that distance.

[Manifest]

How much is one tile supposed to be in the game? because when rocket spread indicator says "2 tiles" the center of the explosion is usually at least 4 floor squares from the target square.

In a simple explanation, it tells you the average. So even if there was a 50% chance it would hit dead-on, and 50% chance it would scatter four tiles, it would still say "2 tiles" on average.

[LordYanaek]

The way it is displayed is fairly counterintuitive. The devs mean "+/- 2 tiles" to mean "a number of tiles away from the target on average equal to 2, plus or minus some amount". For many people, however, "+/- 2 tiles" reads as "plus two tiles or minus two tiles from the location you target". I know it took me a long time to stop thinking of it that way.

Well, i don't know what "many people" think as i'm not in their mind but to me ± always meant "more or less" which is the same as "approximately".

It could probably be changed thought if it's not clear for some people as it's always best to be clear for as many players as possible. However ± have the advantage of being very short in every possible localization.
You can probably change it yourself if it bothers you BTW, just replace

Code: Select all

m_sAverageScatterText="± <XGParam:StrValue0/> Tiles"

with

Code: Select all

m_sAverageScatterText="Avg. <XGParam:StrValue0/> Tiles"

in LW_Overhaul.int and you should be good (thought i haven't been able to test it right now).

[Bullett00th]

it's criminally counter-intuitive atm.

on a side note I have to say that imo Pavonis have this unhealthy obsession with ridiculously inaccurate rockets.
It's their mod and their choice but it's sometimes so inaccurate that it completely breaks any suspension of disbelief. that or makes you think the rocketeers all have eyesight issues

[sadh]

The rocket scatter mechanic is, in my opinion, the one that puts you most at risk of squad wipe on the Alter of the RNG Gods(TM). Going loud with a rocket strike from concealment against a powerful enemy pod can put you on the fast track to Flawless... or sign your death certificate. With so few rockets available, they often need to be a critical part of your mission strategy, but the excessive scatter makes them among the least reliable tools at your disposal.

A recent mid-May campaign of mine ended thus: A-Team with ballistic weapons in concealment, Shinobi advances full move ahead to reveal 2 patrolling enemy pods that had converged to nearly the same spot (7 enemies total, 6 of which are within the rocket blast radius). I advanced several soliders toward the pod to maximize hit percentage and put them on overwatch, leaving a Gunner and a couple others available to handle crowd control or cleanup in the event the rocket strike went bad. My Technical hadn't moved but was a fair distance away, scatter listed as about 2.2 or 2.3 if I recall. He had to fire directly over the top of most of the squad. I went loud with the rocket strike and it did indeed go bad, *really* bad. It came up 4 or 5 tiles short, missing all enemies entirely, detonating instead near the van several squad member were using for cover (I know, I know...). The van exploded, killing one solider and leaving another bleeding out. 2 other soldiers immediately panicked. Only 1 overwatch soldier was still alive and available, and he missed when the pods scampered to form a rough semi-circle around me. So 2 dead or dying, 2 panicked, 1 out of moves and 1 lonely Specialist standing there with a smoke grenade and a look of abject terror on his face. He ran for the hills and called for the Skyranger. On the enemy turn, a drone incapacitated one solider and another was flanked and died gruesomely. One turn later, the squad was completely gone; 5 minutes later, a new campaign began.

Ok, I'm not blind to the poor strategy I employed on the mission. I got overly excited at the prospect of hitting 6 enemies with a single attack and acted rashly. Using the van for cover and firing a rocket over the squad to a spot so close was dumb and it got me exactly what I deserved. But the point is this: the success/failure of that entire campaign rested on a single RNG roll and that is TOO MUCH to put on something completely out of our control. I'd like to see the rocket scatter lessened slightly. Not so much that it makes it guaranteed, but enough that the rocket can be relied upon at least a little bit. As it is now, I'm hesitant to go loud with a rocket at all, preferring a grenade or sniper shot.

[DerAva]

Well, i don't know what "many people" think as i'm not in their mind but to me ± always meant "more or less" which is the same as "approximately".

Well, when someone tells me "I'll pick you up at 8:00, +/- 5 minutes", then I'd expect them to be there at some point between 7:55 and 8:05. I'd expect something like ~2 tiles, not +/-.
I mean, these kind of symbols do have a predetermined meaning.

[Alketi]

Yes, it's criminally misleading. It was a cost tradeoff of a seemingly minor UI artifact that turns out to be quite infuriating if you rely on its apparent meaning. I filed a bug report and attached a video to illustrate how pissed people get when they're mislead (a perfect, blind user-test).

The discussion and video was here: http://www.pavonisinteractive.com/phpBB ... 17&t=25693

However, once you realize that the ± unicode character was, no doubt, chosen to avoid having to translate "AVG", and once you understand its "intended" meaning, then you forgive and move on and hope that the discussion is remembered for the next UI decision.

[dario_gaston]

I am ok with the rocket scatter mechanics, I found them overpowered in vanilla and I liked that the nerf wasn't the old and dumb damage nerf. Maybe they are too nerfed, but still useful. Thing is: I also like to play on calculated risks; if I fire a rocket it's because I think that any of the possible outcomes is good enough.
Knowing now that the new number is not the actual scatter but an average, I can recalculate my moves.
And still, the max/min with the average scatter are still poor data to base desitions on. How is that data distributed? is the likelyhood of "missing" by one tile the same as the likelyhood of "missing" by 6 tiles?

Suggestion: the average is a very poor statistic to convey scatter, I'd suggest change that number for the upper limit of a 90% confidence interval.

Ideally: modify the highlighted zone so that it highlights in one colour the tiles that will surely get hit by the blast, and in another colour the tiles that might, or might not, get hit by the blast. That would provide users a very intuitive way of calculating the risks of shooting a rocket.

[RookieAutopsy]

Well this thread certainly explains a lot I too always assumed the figure given was the radius of how far it could miss by, rather than what it will almost certainly miss by, possibly double that. It now makes a lot of sense why rockets frequently land outside of that radius.

I do however like taking Fire in the Hole as it makes a big difference in the usability of the rocket without impacting a tank/flamer build by much.

[Jacke]

Suggestion: the average is a very poor statistic to convey scatter, I'd suggest change that number for the upper limit of a 90% confidence interval.

That's simple to estimate. 90% confidence is about the full max deviation in most cases: 6, 4, or 2 tiles. Even firing without moving with Fire in the Hole perk needs 95 aim to have 90% confidence of 0 tiles deviation and 70 aim to have 90% confidence of 1 tile or less deviation.

With current LW2 rockets, you have to plan they're going to max deviate to be sure.

[MaxAstro]

~glances over thread~

...Aaand that's why I said "many people".

[Sir_Dr_D]

it's criminally counter-intuitive atm.

on a side note I have to say that imo Pavonis have this unhealthy obsession with ridiculously inaccurate rockets.
It's their mod and their choice but it's sometimes so inaccurate that it completely breaks any suspension of disbelief. that or makes you think the rocketeers all have eyesight issues

This I agree with. I have never like their implementation of rockets. Rockets have the following against them:
1): there is that scatter roll
2): there is a damage roll
3): there is the fact damage decreases as you go away from the center

have both 1 and 2 make no sense. If you have scatter, rockets should always do max damage in the center. As really the damage rolling of grenades, is mostly to simulate scatter. Where the grenade lands in reality would determine its damage.

ANyway grenades having perfect accuracy but rockets having scatter just don't make sense. It should really be the other way around.

[Dlareh]

Weapons have both 1 and 2 (separate aim and damage rolls), so I don't see why rockets shouldn't.

I don't particularly like Pavonis' implementation of rockets at all, especially the ridiculously misleading +/- indicator, but they're limited by what they can do in a game mod so I understand the constraints somewhat.

I agree that XCOM 2's grenade's accuracy is really comical by comparison. The way you can precisely calibrate the bounces and land on exact pixels is hilarious. But well that's vanilla behavior and not Pavonis' fault, at least they tweaked the damage in ways that make some sense, and the new Sapper is actually decent for cover destruction.

If Pavonis were to mod grenades to be less accurate (which they probably considered at some point during development), a lot of players would be pissed and turned off by LW2 as a mod. So I'm betting they realized they couldn't get away with changing much and made do with only changing damage.

[Sir_Dr_D]

Grenades only need to be slightly inaccurate, so that if one of you solidier's is standing just at the edge, there is always a chance they could get hit. They really just need one tile of scatter chance. That would make them accurate enough, while not having them be so ridiculously pinpoint accurate.

And I really belive that a max one tile of rocket scatter, is all is needed as well. It just doesn't mesh with other abilities otherwise.

[Tuhalu]

Rockets have the following against them:
1): there is that scatter roll
2): there is a damage roll
3): there is the fact damage decreases as you go away from the center

have both 1 and 2 make no sense. If you have scatter, rockets should always do max damage in the center. As really the damage rolling of grenades, is mostly to simulate scatter. Where the grenade lands in reality would determine its damage.

ANyway grenades having perfect accuracy but rockets having scatter just don't make sense. It should really be the other way around.

You can mostly fix #1 by only making Rocketeer type Technicals out of high accuracy rookies, giving them Fire in the Hole at Lance Corporal, putting an Aim PCS on them and not moving before firing your rocket. 2 tiles of max scatter and more likely 0-1 is not bad at all.

You can fix #3 by taking Tandem Warheads at Staff Sergeant.

Once you have all that stuff going for you, rockets become extremely reliable and you have a very reliable rifle as well (due to all that accuracy).

#2 feels like a valid argument. I'd like to see the damage range on Rockets tightened up a bit. The current range is 2-7 (2-9 with Bigger Booms!), which is massive. 3-7 or 4-7 would definitely feel better.