"VS OP" Why?

Discussion in 'PlanetSide 2 Gameplay Discussion' started by Vanguard540, Jan 14, 2020.

  1. pnkdth

    No one react per bullet and, like I've said over and over, it depends on the rest of the characteristics of the weapon. Some weapons have buttery smooth recoil, others bounce around, and everything in between.
  2. Campagne

    And this has little to no impact on actual weapon accuracy, which can be very easily, clearly, and succinctly defined as the size of growth of the cone of fire. :cool:
  3. csvfr

    Recoil patterns are NOT random. From wikipedia "In the common parlance, randomness is the apparent lack of pattern or predictability in events." (https://en.wikipedia.org/wiki/Randomness). Recoil follows a clear pattern with well defined probabilities for each successive shot. The rule is: equal probability of displacing the distance defined by to the recoil value to either side, unless at the edge defined by the recoil tolerance, in which case a kick to the opposing side is certain.
    The image I've shown lays out the probabilities for the next 4 shots given a dead centre starting point. The calculations can also be done for a starting point at the left side, which gives different probabilities:
    Basically when the recoil tolerance edge is hit, the location of the aimpoint 3 shots thereafter can be determined with 75% chance. This is also about the time where a player may react to the fact that recoil has thrown the aim off and commence correction by moving the mouse in the opposite direction. A headhunter who knows the recoil pattern may not simply correct but also exploit knowledge of the location with 75% probability to score a headshot. In my previous post I said that is generally worth the risk, because some factors can cause it to be better aim for the body:
    • Health left in the target, if it takes just 1 more bodyshot to kill.
    • If the target is far away then attempting a headshot might increase the chance of missing.
    • Accumulated bloom.
    No but lower FPS only decreases RPM giving even more time to react.
    I'm not saying that no, even a dog can catch a frisbee without creating a model describing the flight trajectory.
  4. Campagne

    They are. There may be a restriction in the flow of the choices but we simply can't ever predict exactly where and how it will go. For example, let's look at the middle 50%. There's a 75% chance of going left and a 25% of going right, but the direction it ultimately goes in is random. We can never truly know the outcome until it is decided, there is no method of prediction with 100% accurate results. It is truly random, even though it's under a restriction.

    If a headhunter does this and is able to do so reliably and with desired results, would this not A: invalidate recoil as a method of measuring accuracy when even horizontal recoil can be controlled reliably, and B: apply more or less equally to all other weapons? How would this manifest in a real way, as far as accuracy statistics?

    Well even a 500RPM gun has a fire rate of 8333.33 rounds per millisecond, which is still faster than a reaction time of ~200 milliseconds. Unless my math is wrong at least.

    But there's a huge difference between visually tracking a projectile and accurately maintaining the position of a reticle over a small fast-moving point while simultaneously tracking the number of directional kicks at a rate faster than human reaction and deciding precisely when to compensate for a very near future event. This is just not realistic.
  5. csvfr

    I don't think it makes sense to measure accuracy using one metric in isolation. All sources of inaccuracy must be combined and seen together to define it. The relevant factors are
    • Flinch
    • Starting stance penality
    • Distance
    • Burst-length
    • Bloom
    • Recoil
    The best accuracy, as in "capability to put bullets where desired", is obtained when all these values are as small as possible. But they do all interplay in a manner such that even if one value is high the accuracy is still good. For example: at shorter ranges longer bursts are possible, with lower recoil comes greater effective range, less bloom allows longer bursts. etc.

    If we, for the sake of illustration, only consider the interplay between recoil and bloom, then the example with the headhunter going for a 75% headshot chance can be continued. Assume further that the accumulated bloom (both from starting stance and burst length) is so high that it covers an area twice the size of the head, meaning a 50% chance to hit when seen in isolation. If combining the inaccuracy from bloom and recoil prediction we get a hit probability of 37.5%. Should the shooter still go for the head, even though bloom does not yet influence his chance to hit the body? This depends on the static recoil value, if it low enough such that by dead-centering the aim a 100% hit probability is obtained then going for a bodyshot is the best course of action.
    • Up x 1
  6. Campagne

    Only two of these actually describe the area in which a bullet can be placed though. Cone of fire/starting stance penalty and bloom, which describes the rate of growth of the CoF. Distance has no affect on weapon accuracy, flinch as far as I'm aware is a literal universal constant for all weapons, burst length doesn't actually effect accuracy directly and instead only resets the formula. Recoil as we've been saying simply moves the position of the reticle around, while the CoF is the only thing which actually controls the placement of bullets.

    Any given player's ability to use a gun isn't the same thing as a gun's inherent accuracy. If a player is not very good with a bolt-action, the distance is long and there's lots of things shaking his scope, so he gets a bodyshot accuracy of only 25% while stationary, is the rifle not an accurate weapon?

    Accuracy is measured by a single metric in isolation because it's a plain and simple exact value of accuracy. As above half of those aren't even relevant. CoF and bloom literally describe the placement of a bullet where the gun is pointing, which is quite plainly the literal definition of accuracy in this context, is it not?

    Bloom allows for a longer period of fire to remain more accurate. This is of course what the CoF math always demonstrates, that starting CoF and bloom can be used to measure exact values. They very often show lower bloom with a similar starting CoF is more accurate after a small number of shots with an incremental increase beyond the overtake point.

    If the CoF is perfectly placed over a head exactly one half the size of the cone, (let's just assume the body is not in frame some how), then yes there would be a 50% chance of hitting the head. BUT, there is a 100% chance of the bullet landing within that space. When you stick in a recoil pattern we lose the certainty of this. It might be a 37.5% chance. What if the recoil kicks in the other direction? What if the player compensates for the recoil? There is no certainty in our exact values beyond the perfect accuracy of CoF.

    The whole point of the recoil argument is artificially claim some weapons are more or less accurate than they really are. For example, take the Gladius and SAW. Both are very similar in terms of maximum damage models with the same bloom and close to the same strafing accuracy. However, the Gladius has a single-direction bias for recoil and the SAW does not. We could try to figure out the probability of a probability to determine which of the two is more likely to be accurate, or we could just measure their respective cones of fire and get exact values.

    The SAW is objectively more accurate when fired from any stance other than standing and moving and the Gladius is objectively more accurate when standing and moving while being considerably less accurate when not. Recoil can be controlled fairly well on both, especially within the range where the Gladius is going to actually do anything. Suffice it all too say, the recoil is irrelevant, random, and relies on several "what if"s. And even if you agree that in this instance recoil isn't an appropriate method of qualifying the accuracy, one would have to admit that recoil was not an effective tool because it can't measure every weapon to the same degree.

    Recoil as a method of measuring accuracy has about as much validity as saying "only the pros play VS." It relies on way too much random factors with probability stacked upon probability.
  7. csvfr

    No because if not knowing where the gun is pointing, the bullet placement relative to the gun direction is irrelevant as far as one's ability to hit targets is concerned. In particular some guns have extra RNG elements in the recoil pattern itself which creates the same effect as additional bloom for the shooter. The VS's SVA-88 and Pulsar LSW are prime examples of this.

    SVA-88: https://planetside.fandom.com/wiki/SVA-88
    Pulsar LSW: https://planetside.fandom.com/wiki/Pulsar_LSW
    Headshot %: https://voidwell.com/ps2/oracle?sta...,7276&startDate=2020-10-05&endDate=2020-11-05

    With identical damage profile and magsize, identical horizontal recoil and tolerance, overall the guns seem very similar. Yet there are some slight differences. Pulsar has 0.04 bloom, less vertical recoil and FSM, and a recoil which angles to the side with a 3 degree min/max difference (<- extra RNG in recoil here). SVA-88 has 0.05 bloom, more vertical recoil and FSM, and a straight angled no-extra-RNG recoil. Which one is more accurate? It turns out that the SVA-88 has better headshot % notwhitstanding 0.01 more bloom. This demonstrates how declaring one weapon as more accurate based on just CoF is wrong.
    • Up x 1
  8. pnkdth

    Precisely, if bloom and CoF tells us where within the box our bullets will travel then the recoil (plus other factors) will tell us how likely it is for us to place more bullets within "the same" box. Both, in tandem, affect our accuracy with a weapon in PS2.
    • Up x 1
  9. Campagne

    The RNG of recoil isn't anything remotely similar to bloom. Bloom is static and applies per shot, as in a rate of [initial cone of fire] + [bloom] = [final cone of fire]. It has exactly one value possible, but recoil can kick in more than one direction and the direction cannot be known prior to the kick.

    There are a few flaws with this comparison, and not only because you're trying to equal player performance with weapon stats.

    Firstly, the Pulsar costs a third of the SVA-88, so already newer players are more likely to go for it instead. Secondly, the SVA-88 has a tighter hipfire in every stance, which also isn't impacted by recoil regardless, and thirdly there is nothing to suggest how each weapon is used or at what ranges. Both are very comparable in their effective ranges and in these ranges recoil isn't going to be an issue either way.

    However, if we were to consider recoil as a factor on weapon accuracy, the SVA-88 has access to a compensator which the Pulsar does not while the Pulsar has access to an extended magazine which the SVA-88 does not. Combine this with the worse hipfire and the single-angle recoil bias and there is less reason to use a grip on the Pulsar compared to the SVA-88. The data doesn't tell us user loadouts either.

    This is not an appropriate method of measuring accuracy, whether by recoil or player performance stats.

  10. pnkdth

    We do know values for recoil, type of recoil, tolerance, etc. Which means we can assess the probability of the outcome to understand how the weapon will actually perform. It is then trivial to see which weapons will behave in a more/less predictable manner which is a solid way to define accuracy since this will tell us how likely it is for us to be able to place that cone of fire over our target.

    You're fine with using probability on VS pop and what players it attracts but now you suddenly feel compelled to figure out every tiny little deal right down to loadouts? Does this also mean you do not think the BG is accurate anymore since it has, by far, the highest usage in Q4 (which will improve its performance data) and guarantees every user of the weapon has nearly 7k kills with VS LMGs (which will improve the performance data) to even start using it. What's more, we cannot even know if ANY of the accuracy data we have is accurate since it doesn't record accurately missing player targets (your definition is where the gun is pointing, not that you can actually use it to hit anything).

    We cannot know anything... It is all a lie... Not even the cake is re---- [ transmission lost ]
  11. Campagne

    We can assess the probability, exactly. The probability. The chance that our estimations will be correct. Using strict, well-defined values gives us exact parameters, not values within a potential range. No weapons to the best of my knowledge have a recoil pattern which behaves in exactly the same manner every single time, and therefore no weapons have an exact, definable accuracy statistic based on recoil.

    I'm fine with using probability with that because I'm showing when probability does not support the claim, just as I am with this. For VS to be comprised of predominately higher-skilled players, there are several different probabilities stacked on top of each other. The more probabilities that must be used the lower the certainty and accuracy of the claim. For the VS to be comprised of predominately higher-skilled players, there are just too many extremely narrow events which all must occur simultaneously, and each event gives a margin of error each time.

    With recoil, almost every shot gives a new probability statistic to determine in which area the recoil will lead. Therefore, the final position of the crosshair relies on several random events. The final position cannot be known or calculated or measured until the RNG has already decided which path it will take.

    Yeah, that's exactly what I'd say. The user performance stats don't apply to the weapon stats. The BG is accurate because of it's notably lower bloom of 0.04 as opposed to the weapon damage model's standard of 0.05. This can be calculated to demonstrate the results of a smaller CoF, and therefore a higher degree of accuracy.

    Strawman, again. As I've said time and time already, my definition of accuracy is the area surrounding the reticle in which bullets are randomly placed, such that a smaller area is of a higher accuracy. I've stated the opposite of what you try to label as my beliefs, that the position of the gun/location it is pointing to is irrelevant to the weapon's accuracy.

    Perhaps this might have been more obvious if you didn't cut out the bit directly above where I said "there are a few flaws with this comparison, and not only because you're trying to equal player performance with weapon stats." Or perhaps if you didn't cut out the part directly below it where I said "this is not an appropriate method of measuring accuracy, whether by recoil or player performance stats." :rolleyes:
  12. pnkdth

    In both cases we have a set value which depending on where the user is pointing the weapon will yield randomised yet manageable result. In the case of recoil we compensate in attempt to foresee the recoil pattern, in CoF/Bloom we do our best to control the bullets no going all over the place, if we're at range we will try and lead to the best of our abilities, and so on.

    In other words, at best your way of measuring accuracy is a pointless and an entirely theoretical exercise. Even if you're maintaining you only want to measure weapon accuracy you are, in fact, ignoring core mechanics of the weapon itself (the recoil value, velocity, RPM, damage model, is not a part of the player). We don't need to know what loadout, favourite colour, or if the player woke up on the wrong side of the bed that morning to make calculations based on any of those values.

    At this point though, I think it is time to just agree to disagree. I mean, I don't disagree that bloom/CoF can make weapons more/less accurate, I don't see it as a sufficient way of measuring it.
  13. Trebb

    Perception is a huge thing with a 3 faction PvP game, and when 1 faction gets all the fun stuff and a lot of advantages (near hitscan rocket launchers, harder to hit dueling ESFs, no reload heat based weapons, tanks in bullshiat locations etc), it's noticed.

    When I notice at least 6 VS in a squad are all BG using heavies, it infuriates me. Mostly because of the 'its a worse orion!!11!' propaganda they all spout, while it's used more than EVERY DIRECTIVE GUN COMBINED :p Watched one VS zombie get up, kill 3 people before dying, got res grenaded again, and killed another few people. Me? Woulda died reloading ;\

    I call them the deVS faction, and I'm only 20% joking ;) (seriously, look up all the old Wrel youtube vids)
    • Up x 1
  14. Campagne

    This implies the physical skill of the user can influence the accuracy of the gun, which just isn't true.

    At best, measuring accuracy by the listed known accuracy values is the most accurate way of doing so possible. It is a universal constant with no regard for the player's actions beyond holding the trigger down and can be calculated perfectly. It compares to virtually any other weapon or theoretical set of numbers.

    A theoretical exercise is saying the probability of a recoil pattern has the highest chance of placing the cone at [Position X,Y]. CoFs can be calculated to exact values as you must be getting tired of "hearing" me say. The only theoretical elements with them is the number of shots taken.

    Velocity, RPM, and damage & degradation don't impact the weapon's ability to place bullets into the world. At best recoil can influence where in the world in some distant sense, but not in the most cleanest and analytical sense. Player loadouts affect the player's recoil.

    So be it! I agree with your disagreement. Well, not really. :p Hold your beliefs, clutch them dearly! Which of us is right or wrong doesn't affect reality either way so it makes no difference beyond the fun of discussion.
  15. VelmaBlackburn

    Coming from a newer players perspective (around 350 ish hours):

    • VS weapons do not have bullet drop
    • VS weapons are generally accurate
    • VS weapons are very deadly in close range
    This leads to some situations where in theory the other factions should have a leg up against VS but are actually still punished severely.

    Situation one: VS HA vs NC HA at range. If the NC HA is skilled enough to reign in their gun their high stopping power should mean they have the advantage in this fight. If the VS HA is competent enough to aim their gun towards the NC HA they are capable of taking the other player out, it might need a couple of bullets more it is generally easier to hit a target at range IF you do not have bullet drop.

    Situation two: VS HA with Orion vs NC HA with Gauss Saw in melee range. This is the engagement range VS are really good at (putting aside NC shotguns). So to a new player, with default weaponry it appears that VS is just great at every range.

    I can tell you that that was certainly how I felt as a TR player, where it feels we kind blow at medium to long range and are only really good in close range against NC.

    And just to give my NC bros some love, yeah you suffer a similar fate, but at least you own TR at range :D
  16. pnkdth

    VS sacrifice DPS for accuracy outside of a handful of weapons (you're expected to hit more to make up for it). No bullet drop does come into play at times but in the vast majority of engagements you will not be affected by it. VS have no particular edge in CQC over anyone else.

    As a TR player you have the Watchman, CARV, MSW-R for closer range and mid range you got the Bull and Rhino (which pretty much are VS weapons) + the TMG-50 which is (which also have a very predictable recoil pattern with a vertical pull). Carbines, well, you got the Kindred. The Jag is hecking solid too. You got the Cougar and T5 for range. TR ARs are also solid but I see why only having burst variants for mid range and up is annoying. The default T1 Cyclar is a fantastic all-arounder and the TAR and TRV means you get the best of whatever world of BRRRRRRRR DAKKA DAKKA. Haven't used the TORQ much though (the TRV is my one and only true love though sweet reliable T1 Cycler comes to visit too). The TR default sidearm is great too.

    If you go through the NC arsenal they have pretty much everything (and more, like the best SMGs) bar TR's super ROF weapons and a 143/750 LMG. It is really quite sad when you look at it all as much of what once was considered a TR, VS, or NC trait is now more of a guideline which can be ignored if the designers feel like it. When once it felt like a decision to choose faction (if you're an infantry player) now it is little more which kind of colour you enjoy.

    If we're talking default weapons though, yepp, why the SAW became the default for the NC is a mystery that's haunted PS2 players since the beginning. Makes about as much as sense as starting out VS with the Ursa and TR with the TMG-50. The VS default LMG is pretty much the only LMG they'll ever need (hence its wide use). I really don't like the BG though since while it won't hit as accurately as the Orion (with a grip) it can still be controlled since it actually is of the weapons which are more accurate in terms of bloom per shot in its class. Add in the heat mechanic and you got a gun which can keep going and going and going. In short, BG needs a nerf. The writing has been on the wall a long time for that.
  17. VelmaBlackburn

    Just to be clear, I said "appear". I wasn't making an argument that they actually are. I think with the exception of NC all factions are generally somewhat balanced, my evidence for that being that both TR and VS are pretty close to winrate of alerts while NC lack behind somewhat, though to be fair that could have other reasons.
  18. pnkdth

    I see. Yeah, I also think it is more useful to look at NC since that's the faction which fell behind. From my perspective, VS without the current BG is in a good spot. A personal pet peeve with recent changes to the NC was their MAX which got hit with a double nerf (both to its main weapons and the complete removal of slugs). It was already a specialised kit to begin with and they made sure you could never reach for than an arm's length. With MBTs I think there's a lot of issues with secondary weapons which doesn't really make sense for the Vanguard.
    • Up x 1
  19. thed1rt

    I think the most OP thing of all is how TR can kill you in 0.3 seconds with a buzzgun and NC can 1 shot you with shotguns that dump what seems like 10 rounds in 2 seconds.
  20. thed1rt

    And its very hard to get kills with a Lasher unless you are already winning. Most people just run from it or blitz you.