The Importance of Inaccuracy

Discussion in 'Planetary Annihilation General Discussion' started by ledarsi, October 19, 2013.

  1. ledarsi

    ledarsi Post Master General

    Messages:
    1,381
    Likes Received:
    935
    The general impression I have of weapons in PA is that they are highly lethal, which is good. However I am also noticing that they are, in general, very accurate. This could be an issue down the road, and introducing a certain amount of inaccuracy into most weapons could substantially improve gameplay.

    I am also assuming that Ants will have their direct-fire aim fixed to allow them to lead shots.


    Predictable Misses

    TA's physics simulation introduces another dimension for unit design; inaccuracy. Without projectile simulation, it is impossible for a shot to miss its intended target and hit someone else. When projectiles are simulated, it is possible for an inaccurate weapon to have different effectiveness simply by the quantity and density of enemies.

    One of the main beneficial effects of introducing inaccuracy is changing how the inaccurate weapon scales when used in large groups, and when used against large groups. Inaccuracy also introduces predictable variation which makes the outcome of a single shot semi-random, but making the effect strategically predictable in aggregate.

    By contrast, perfectly accurate (or just extremely accurate) weapons create a very hard effectiveness curve as force size increases. Large forces of extremely accurate units have been a problem in previous games. TA; Samson/Slasher, Hawk/Vamp. SupCom; ASF, and others. With inaccurate weapons, "real" DPS is also dependent on a variety of factors which reduce its DPS below its nominal maximum value, and does no in a complex, variable way dependent on range, target density, projectile speed, obstacles, and other factors.


    Rapid Fire Weapons

    Rate of fire and inaccuracy are interlinked, because inaccuracy creates a probability distribution. More shots, consequently, has the effect of increasing the effective accuracy of the group, but the curve is much softer than more shots from highly accurate shooters.

    Rapid fire weapons are naturally suited for engaging a large number of small targets, and against a group of such targets their inaccuracy is not a problem. Against just a few small enemies, inaccuracy does result in reduced efficiency.

    As demonstrative examples, consider the Warrior riot unit, and Stardust turret, both from Zero-K. Both weapons kill quickly, but are significantly inaccurate and miss frequently.


    Artillery

    Artillery should have a target region within which the shell will fall. For many types of artillery, this region should actually be quite large, to the extent that you cannot expect it to actually deal damage to any particular single enemy target.

    Precision artillery is a special type of asset, notable by having both the unmatched range of artillery and also enough precision to hit a single target reliably. Most types of artillery in PA currently are sufficiently accurate that they qualify as precision artillery in my opinion. I think these units are in the realm of reason for cost and effectiveness, but this type of artillery really should not be the primary type of artillery used.

    Inaccurate artillery can be much, much cheaper, and might even have superior range than even more expensive precision guns. And they might even have a very considerable amount of splash. Massed fire of cheaper guns creates a continuum of barrage strength based on the quantity of guns, which can be purchased individually, instead of a large, hard distinction between zero and one precision gun.

    And artillery barrages with inaccurate guns create a much more dynamic, mobile volley/response interaction using scouting than picking off individual radar blips one at a time. Bombardments using groups of inaccurate artillery that move around the map, and which have to select their targets more carefully than just firing at every isolated blip one at a time, create more interaction for both sides.

    Currently artillery will hit its intended target, unless it moves. This is not a good situation. Inaccurate artillery will instead strike somewhere within a target circle, and the target moving about really doesn't help its chances much. It might move away from a shot, it might move directly into the next one.


    Rockets & Missiles

    A large volley of unguided rockets is another possible method of quickly issuing splash to a target region. However, unlike a single large explosion, that region can have a variable shape, such as for focusing around a moving target, by firing each rocket at the same target with a small inaccuracy around it.

    Rocket artillery, for example, is a different type of artillery which quickly fires hundreds of rockets to carpet-bomb a target area over the space of a few seconds. Multiple MLRS systems firing together can deliver a truly incredible barrage of firepower over a short period, instead of gradually over time like conventional artillery.

    Missiles are guided, and as a result it makes more sense for them to more closely parallel precision weapons which are more expensive for the same range. However, multiple-fire guided missiles with the expectation that not all the missiles will hit is also a viable weapon type.


    Conclusion

    Very precise weapons have their place, but carefully designed inaccuracy should be a major design consideration for most weapons, with high accuracy carrying a cost premium. Rapid fire close combat weapons, medium range missiles and rockets, and long range artillery all benefit from carefully measured, but significant amounts of inaccuracy rather than very reliable accuracy on every shot.

    Inaccuracy smooths the effectiveness curve of increasing quantity, and makes dense groupings of large forces less efficient because inaccurate weapons are likely to hit something, if not their intended target. Increased rate of fire with reduced accuracy also has the effect of making ranged attacks very visually impressive, which is a nice bonus.
  2. brianpurkiss

    brianpurkiss Post Master General

    Messages:
    7,879
    Likes Received:
    7,438
    I agree with you. Inaccuracy is one of the things I like about SupCom and don't like how units in early C&C games are perfectly accurate.

    If I remember correctly, Uber has stated they'll be adding inaccuracy as well as leading capabilities.

    Don't have any citation though.
  3. igncom1

    igncom1 Post Master General

    Messages:
    7,961
    Likes Received:
    3,132
    Inaccuracy should come from how the projectiles work and are unsuited for particular targets.
    MrTBSC and beer4blood like this.
  4. tatsujb

    tatsujb Post Master General

    Messages:
    12,902
    Likes Received:
    5,385
    just a nitpick if I may :p you probably know this already and agree anyways, the terminology here is : they didn't have simulated projectiles. hence when a unit met an enemy unit it would simply calculate distance unitl it's firering range was met.

    ...by the time you saw the "projectile" (just special effect) fly, the unit had already deduced damage from the enemy unit, with a slight delay corresponding to the travel time of said projectile between them.

    in some games the delay was even too short and it could be noticed that the HP was deduced before the graphical 'projectile' ever hit.
  5. brianpurkiss

    brianpurkiss Post Master General

    Messages:
    7,879
    Likes Received:
    7,438
    I totally understand technology limitations. Still didn't like that mechanic.
  6. tatsujb

    tatsujb Post Master General

    Messages:
    12,902
    Likes Received:
    5,385
    :p and neither did I
    brianpurkiss likes this.
  7. Dementiurge

    Dementiurge Post Master General

    Messages:
    1,094
    Likes Received:
    693
    The first C&C was interesting. As I recall, all attacks were area attacks, and most were somewhat inaccurate. If you had several units pile in the same area, such as infantry clustering together as they swapped grid locations, a bullet could hit all of them at once. Explosions could be on the 'edge' of a unit and do less damage. Ultimately projectiles were still 'flat', stuck on the same 2d plane that all other entities existed on, but at least their interaction wasn't completely boring.

    Compare with Starcraft 2, where I'm not even sure attack ground is possible to produce in their almighty map editor.
  8. veta

    veta Active Member

    Messages:
    1,256
    Likes Received:
    11
    This was a nice quick read. OP is right.
  9. LeadfootSlim

    LeadfootSlim Well-Known Member

    Messages:
    576
    Likes Received:
    349
    For artillery, having inaccuracy apply to radar-only shots, but not visual-confirmed shots, would be a great compromise in my opinion, and would increase the usefulness of scout units.
    MrTBSC likes this.
  10. Culverin

    Culverin Post Master General

    Messages:
    1,069
    Likes Received:
    582
    TheLittleOne used to play FA competitively.

    We was unhappy when things got a bit more accurate.
    Less room for micro.

    I tend to agree with him, there is some good micro in the game that should stay.

    a 12v12 of the same unit with 1 side being micro'd should win.
    Just barely.
    But it makes it much more interesting... For the player to eke out every list bit of efficiency out of his units.
  11. ledarsi

    ledarsi Post Master General

    Messages:
    1,381
    Likes Received:
    935
    I disagree. I think introducing radar-wobble for radar blips is a better solution. Imprecise artillery doesn't care, and precise weapons benefit from scouting.
  12. Gerfand

    Gerfand Active Member

    Messages:
    575
    Likes Received:
    147
    me too, when an arty shot, winds make the projectile balance...
  13. sovietpride

    sovietpride Member

    Messages:
    65
    Likes Received:
    21
    With regards to accuracy and micro, can people please be aware of how different weapons work.
    Again referring to SC:FA, It was harder to micro vs a mantis than say another factions t1 tank because they were more accurate.

    However, vs artillery, the opposite is true - It was possible to micro vs the likes of Fervors because they were accurate (I.e, they fire, you know the shell is going to land at point X) , yet nigh-on impossible to micro vs medusae due to how innaccurate they were ( They fire, you have no way of knowing where the shell will land)

    Just my two cents before people go "Accuracy = good for micro" or vice versa.
    MrTBSC likes this.
  14. zweistein000

    zweistein000 Post Master General

    Messages:
    1,362
    Likes Received:
    727
    I agree with the OP entirely. I also think that this game could use a bit more AOE, specially with non energy weapons (Pelter, all the ships and ground artillery do no AOE have negligible AOE now, which is ok if you are going to have a high precision weapons of such range, but as soon as you get inaccuracy you will need that AOE to balance things out as well as create a more unique behaviour).
  15. KNight

    KNight Post Master General

    Messages:
    7,681
    Likes Received:
    3,268
    Accuracy has nothing to with with the differences between the Mantis, Striker, Aurora and Thaam, it actually comes down to Rate of Fire, the Mantis fires every 3 ticks(10 ticks = 1 second) where as the Striker, Thaam and Aurora fire every 10, 13 and 15 ticks respectively.

    It's not that the Mantis was more accurate, it just that it's much harder to dodge shots coming that quickly, and even if you dodge one or two, you were still begin hit by the rest.

    Mike
  16. sovietpride

    sovietpride Member

    Messages:
    65
    Likes Received:
    21
    And... the point of your post is... what?

    If we measure accuracy as "how many shots, X out of the shots fired, Y, hit the target", then... I don't see how you can exactly argue that the Manti was less accurate than its counter parts when more of its shots would hit home.
    Or, in other terms, The manti fires more, therefore it misses less... which is another way of saying it's more accurate?

    That is also an over-simplified system actually.
    Let's take the example of the trained soldier vs amateur.
    They both have the same weapon, let's say an assault rifle. - It's "accuracy" is the same, as the weapons are identical.
    Tell them to shoot at a target. The soldier will (hopefully, given he's a soldier) hit the target more often than the amateur. Let us assume this is the case.
    Are you going to say that If i were to conclude: "The soldier is more accurate than the amateur", the statement was wrong?

    In your example, the Manti and various tanks have a weapon which would be 100% accurate if all objects were stationary. If they were moving, I don't see how you can argue that the system demonstrates that the manti is capable of maintaining a higher hit%, and is therefore more accurate with its weapon system than the others

    In either circumstance, the result is still irrelevant. Let's assume i used the term "accuracy" incorrectly. My point still stands. Less Accuracy does not always mean that micro can be used as a counter play, neither does more accuracy mean that micro can be used as a counter play either.

    So please, get off your high horse. Attempting to fob someone off with technical details that bear no relevance at all is... unnecessary.
    Last edited: October 20, 2013
    TheAnnihilator likes this.
  17. KNight

    KNight Post Master General

    Messages:
    7,681
    Likes Received:
    3,268
    To explain things plainly. To many people take things at face value and end up misunderstanding things, this is the internet after all.

    Mike
    stuart98 and cwarner7264 like this.
  18. sovietpride

    sovietpride Member

    Messages:
    65
    Likes Received:
    21

    Which is what I was trying to achieve, to
    a) stop people from throwing around sledgehammer comments relating accuracy to micro when it wasn't that straight forward and
    b) so people would utilise terminology and mean the same thing by it.


    *shrug*
    Where (honestly) was I raising a source of confusion?
  19. ledarsi

    ledarsi Post Master General

    Messages:
    1,381
    Likes Received:
    935
    I think it really doesn't matter how inaccuracy affects micro because of the size of groups involved. A group of 50 bots with machine guns spraying bullets isn't exactly the most amenable to precise micromanagement, especially since 50 is actually not that large and the player probably has more important things to do.
  20. sovietpride

    sovietpride Member

    Messages:
    65
    Likes Received:
    21
    You're right in your example. BUT.
    Again using SC as the example:
    Early game, not many units or resources. It's like 1 bot vs 1 bot. Argument is either whether intensive input of commands should yield an advantage over someone who doesn't. Should it always be mutually assured destruction ala 2 submarines going at each other, or like 2 LAB's which could be microd?

    Because that can lead to further repercussions like an engineer being killed and expansion being stopped.

    Of course, no ones really going to bother by the time huge blobs hit the field...

Share This Page