"VS OP" Why?

I don't think he actually claimed all VS infantry weapons were more accurate. But in any case while it's obviously false on the basis of bloom, it's the same thing as saying all NC guns deal more damage or that all TR guns fire faster. Broadly speaking they have largest numbers of the highest accuracy guns across all weapon classes, and it would seem people get carried away with that.

Yes, and it for that reason I think it is misleading to say "VS weapons are more accurate" since it is only a handful of weapons which actually are more accurate. You could argue that lower RPM increases accuracy (and you'd be right) but that isn't a VS exclusive. All in all though, with the releases over the years faction traits have become faction suggestions. If you're an infantry player it barely makes any difference which faction you choose anymore. Mostly preference at this point.

I'd like to see the heat mechanic for the BG looked at though as it either needs a reduction in mag size to like 35 (overheats on the 36th shot)) OR taking it down to a 143/698 damage model. The BG really do stick out like a sore thumb but that's it.

I mean, technically yes, but when you set that bloom against the increase in horizontal recoil and it starts looking a lot more like a case of "six one, half a dozen the other" than a real advantage.

The Goose sticks out because unlike the GODSAW it's unique trait actually comes in handy once in a while and unlike the Butcher it actually has one.

Well it's more of a general trend really. Broadly they are, even if in "reality" this doesn't mean as much. I would maybe argue lower user RPM and not lower weapon RPM, as in PS2 lower RoF usually doesn't correlate to higher accuracy with automatics. Every time I go through the weapons it surprises me just how similar most of them all are. I agree we're generally approaching suggestions with each and every addition.

Personally I think it'd be better to lower total shots per overheat than the RoF. At that point one might as well just use a Pulsar instead for close to a flat upgrade. At least enough of one to not warrant the chore of unlocking the BG anyway.

Well recoil isn't really a great way to quantify accuracy though, especially if we only use one element of it. The long and short of it is that recoil isn't consistent across weapons, types, damage models, or factions, and even the same values can have different impacts on different guns. How do we consider a strong single-direction horizontal recoil compared to a lesser two-direction recoil?

Bloom is very consistent, such that one can determine the bloom of any given weapon purely by knowing the damage per shot in the maximum damage range. It's easier to interpret how this will impact live gameplay and is completely independent of user skill, unlike recoil. It isn't affected by attachments like grips or compensators, nor magnification sights, and behaves in exactly the same way every single time.

Bloom is easily quantifiable and can be applied to "real life" situations very easily but recoil cannot. Hence why bloom is a more accurate measure of accuracy than recoil.

Consistent, predictable things are much more easily handled than random, unpredictable things.

There is a difference between vertical recoil - which can be easily managed by player skill, and horizontal recoil - which cannot. Having higher horizontal recoil is just as, if not more detrimental to getting shots on target as having higher bloom which can be managed simply by burst firing (incidentally, something the VS are also weak in due to their relatively high FSRM).

In the end you are comparing weapons that have a tighter cone of fire but bounce around a lot with weapons that are easier to control but less mechanically accurate. Which is better/more consistent/more intuitive/etc. is up for debate, but calling one "notably more accurate" based on just that one statistic isn't getting the whole picture.

8 pages in.

Has anyone said "So shut up and play VS then!"

Which is exactly why bloom is a better method of measurement.

You mean to imply horizontal recoil cannot be managed in any way? A lot of guns flat out have a single direction bias and can be pulled in one diagonal line. There is no mouse movement pattern which can affect the bloom of a weapon. The VS are relatively average in most categories of FSM, and again this value is so inconsistent it can range from 0.5x to 3x in the same weapon class while bloom has a total variance of 0.02.

It's not even close to an objective qualification of accuracy on the basis of recoil. Take the Reaper DMR and A-Tross ARs. Both are very similar. Bloom and CoF numbers give them identical accuracy, yet recoil wise the Reaper has a higher vertical recoil and FSM with no directional preference while the A-Tross has a higher horizontal tolerance and a single-side bias, but a higher min and max angle. Which is better? The recoil patterns are going to be random within these parameters but the bloom is not and can be calculated to exact values at any point.

Ever further still, let's compare this to the SAW. The SAW has a higher vertical recoil than both but has a lower horizontal tolerance than both with no directional bias and a min/max angle of 0 degrees with a higher FSM. Does the lower horizontal recoil make it more accurate, or does the higher vertical recoil make it less accurate?

Meanwhile bloom and CoF says it's objectively less accurate. A difference of 0.01 bloom per shot is quite notable. Do I need to bring up some more CoF math again?

For one final example, take the Promise. Let's for a moment assume CoF didn't exist and a weapon's recoil was the only means of measuring mechanical accuracy given a perfect statistical accuracy. For the sake of an argument let's even assume the recoil is static. The stats seem nice, but it has a horizontal recoil tolerance of 5.0, which is virtually unheard of compared to every other infantry gun in the game to my knowledge. I don't think there's even another gun with a value above 1.0 off the top of my head.

This tolerance of 5 means the Promise's accuracy given a pinpoint CoF is entirely random, as the weapon can swing through varying degrees based entirely upon RNG. If the gun shoots 15 times and never happens to swing wildly, is the Promise accurate? What if it flipped angles every shot for 15 shots, is it suddenly not accurate? In terms of mathematics the accuracy literally doesn't exist because the same value cannot be in two places. Mathematically, a constant like CoF or bloom is the only appropriate way to measure accuracy.

If x = 5, it cannot equal both +5 and -5. When x is measured it must equal the exact same value or ya' get a DNE. With bloom, b = 0.05. Fire 15 shots and measure a thousand times in a row, b will still equal 0.05 and will result in an identical final CoF every single time. Fire 15 shots and measure the difference between starting location of the reticle and final location of the reticle and we will not get the same result every single time. AKA when measured recoil is not a constant value and cannot be used to measure accuracy because at any given point it can potentially equal more than one value. $h!t don't work fam.

Not if your recoil angle is zero (which incidentally for most VS guns, it is) in which case you are just as likely to kick left as right. This tends to have a rather negative impact on one's ability to remain on target.

No, but letting off the trigger for a split-second resets your CoF which is why firing in bursts is so important.

Since we were just talking about the Goose and since it's one of if not the most picked weapon type in the game, for the moment I'm sticking to the LMGs, where the VS's average FSRM is higher than either the TR or the NC.

In any case, we're losing the plot. Semantically, yes the VS guns with lower bloom are more accurate. If all we were concerned about was mechanical accuracy this would be as far as the discussion went. However, when it comes to the game your ability to hit a target is a function of both the mechanical accuracy of the weapon as well as your ability to keep it on target. Something that recoil is very much a factor of.

Again: In my opinion, both from a mathematical standpoint based on the statistics of the popular VS, TR and NC weapon choices as well as personal experience the generally lower and more predictable recoil (especially horizontal recoil) of the TR and NC weapons is a perfectly acceptable tradeoff for the generally lower CoF of the VS weapons as far as hitting a target is concerned.

MSW-R
Anchor
Goose
Orion

Looking at the average accuracy to me it looks like my original assertion is vindicated.

Well looking at the HSR in the stats you quoted seems to suggest otherwise, as do these stats which show the Betlegeuse having the highest headshot % out of all LMGs in the game:
https://voidwell.com/ps2/oracle?sta...04218&startDate=2020-09-15&endDate=2020-11-03

The capability to hit the head is of course why high accuracy is needed in the first place, and the Betlegeuse having average statistical accuracy but supreme headshot rate does indicate that it is a more controllable and accurate weapon. The statistical accuracy, aka "shots which hit the enemy" also takes into account shots which were not aimed at a target, such as supressive fire, and counts it as a miss. Headshot rate on the other hand, considers the ratio between confirmed headshots to body- and legshots, which can only be attained through intentional aiming.

The Orion/BG has a vertical recoil + horizontal deviation but mitigate this it has reduced bloom compared to other weapons of their type. The horizontal recoil does put a limiter on the weapon but since both work within an area where most battles take place they both see heavy use and in the case of the BG performs well above the rest (too good imo).

You always need to consider all factors (bloom, recoil, weapon type, rate of fire, etc) when determining how accurate/deadly a weapon is. Otherwise you cannot understand its role and the data we get from sites like voidwell and dasanfall. Furthermore, you can't omit certain values/factors because it gets more difficult or inconvenient to calculate, you have to account for that as well and accept there will be a variance.

I feel like VS is seen as OP in the same way that the Russians are seen as OP in Heros and Generals. The better players play Russian for shorter Queues. Many shooters pick VS for more targets, so it might be the people that it attracts?

Yes, this is bad. But no, this isn't only true for the VS nor do the VS have the most equal recoil biases. In fact, the VS actually has the least equal recoil biases and consequentially the most single-angle biases. (NC & TR are tied with LMGs, TR has the most double biases for ARs, NC has the most double biases for carbines. VS is universally more single-angled.)

Letting off the trigger makes accuracy a null value, as bullets are no longer being fired. They can't have an accuracy value or statistic if they don't exist.

Well I'm sorry to say that's not true either. The average FSM for the ES LMGs is as follows:

1. TR = 1.96875x
2. VS = 1.9625x
3. NC = 1.84375x

The difference is very small but regardless the VS is once again not the worst off, factually demonstrating the falseness of your claim "the VS's average FSRM is higher than either the TR or the NC."

There are no semantics in saying guns with lower bloom are more accurate. I've shown this through the numbers many times already. Recoil is a factor in user accuracy, but not in weapon accuracy. Regardless, the numbers simply don't support the claim that VS have more recoil, in fact it would argue the TR's LMGs have more uncontrollable recoil instead.

Sure the weapons are comparable, but no one is saying they weren't. User stats don't mean anything here, we have no control or information on who's using what gun. The Orion is default and obviously suffers because of it, yet the near identical Betelgeuse has the higher performance across the board. The BG which doesn't have access to a forward grip I might add.

Well in general a lower variance means a higher accuracy, both in literal situations and in statistics.

If our method of measuring accuracy is extremely random with no definable pattern of bullet placement for one of the two primary factors we're only going to care about the one we can define.

At best, if this were a real situation where the numbers were very important to know, we could create a plot with box & whiskers plots placed one it, where the box would contain the known specific CoF values at a given number of shots and the whiskers would show the minimum and maximum possible recoil variance from the origin point. However needless to say the boxes would be a substantially smaller point while the whiskers would be so long they'd extend into meaningless areas. Most points within the whiskers would not be a the minimum or maximum values anyways.

Can you imagine now why this is a very bad and ineffective way to measure statistical accuracy? Well you don't really even have to. I'll make a simplified plot of just the Orion to demonstrate this.

No, because we have the values and parameters for that randomness (curtesy of the wiki and other content creators) and failing to account for that means your calculations are meaningless, outside of making it easier and more neat for you to calculate. So the process of calculating things would not be random, we would just have to accept that you need a certain sample size in order to get a satisfactory answer.

Unless you want a simplified and wrong answer, of course.

I said I was only going to do the Orion but I got curious and wanted to see what things would look like when the horizontal tolerance was effectively uncapped, with which the Promise was more than happy to oblige.

Anyways, the whiskers represent the continuous set of possible distances the stock gun has travelled from an origin point given no user correction. The box represents the effective cone of fire in which the next bullet will spawn in irregardless of the weapon's distance travelled from the origin point. It obviously wasn't drawn professionally.

Wrong and wrong. Is an Orion with a randomly generated recoil pattern more accurate than an Orion with a slightly different randomly generated recoil pattern? No, the accuracy is identical. Yet somehow a constant set of values is producing random results. You're trying to argue one set of random numbers is more accurate than another set of random numbers and giving the false pretense this cannot me controlled or mitigated to any degree while suggesting an exact and known value is wrong.

You're arguing the potential for RNG matters only because doing so would yield the result you want. Mathematically it makes no sense.

The argument that recoil should be considered in terms of weapon accuracy is just not viable. As above with the box and whisker plots, the boxes represent exact and know values while the whiskers represent variance. When the statistical variance can be anything between practically zero and over triple the known values it simply doesn't represent reality. This is especially true with the Promise, wherein with the recoil scaling in effect good RNG can mean next to no real recoil. Does this mean it's more accurate than the Orion, which has an objectively smaller CoF? What of when the Promise gets bad RNG and gets the same pull in the same direction for several shots in a row? What if it gets a few then flips? It's not predictable and the user isn't going to be landing many shots at range. It's both extremely accurate and not accurate at all. The accuracy is both positive and negative simultaneously until it is fired and therefore mathematically does not exist and cannot be represented objectively.

But that doesn't tow the line that the VS have abnormally accurate guns, so you'll just ignore the hole in the logic and say something literally objective is incorrect.

This made me remember an old picture I made from an accuracy & recoil debate we had 2 years ago (https://forums.daybreakgames.com/ps...ons-better-than-tr.251099/page-4#post-3523963)

Looking at the picture, representing the typical recoil pattern of straight angle guns, there are several takeaways and possible analyses as to the controllability of horizontal recoil. RPM and human reaction time also plays a role, with a 500 RPM gun a human may have calculated and reacted (with unconsious reflexes using 240ms) to being in a particular spot on the second shot thereafter, and with 750 RPM on the 3rd shot.

In particular, on the shots with 2 possible locations, one can calculate which one is more likely if noticing being off centre 2 or 3 shots ago, or that they are equally likely if being dead centre 3 shots before. Since a headshot gives a 2x (2.5x vs nanoweave) damage multiplier it is generally worth the risk of jerking the aim into on of these locations to hit the head on these shots, unless the accumulated bloom says otherwise. I believe this to be the reason the Betelgeuse outperforms in headshot % given its advantage in bloom.

[redacted] (hit reply rather than edit)

I wrote, "The Orion/BG has a vertical recoil + horizontal deviation but mitigate this it has reduced bloom compared to other weapons of their type." Before that I listed both the Orion and BG as examples of having less bloom, a list you then complemented with the LSW and three other weapons.

Where things does start to become interesting with recoil is how trivial it is to control the recoil on the Promise, i.e. you are able to put boxes in the same spot as previous boxes. The Promise also have a lower RPM which makes this process easier still. That is the context if which I argue recoil matters. With other weapons who have much more deviation and might even have directional recoil it becomes more problematic to stay on target. You may define this as user accuracy but I think that's wrong since the user has to overcome set values on each weapon, some of which will favour accuracy while other will favour DPS or <insert gimmick and/or specialist role here>.

Huh, so it's slightly more accurate while also having that OP heat mechanic? Yes it's OP, I die a dozen times a night reloading or after being ressed. VS using this stupidly overpowered gun never have to worry about any of that. Luckily it's not like the main objective is to guard entryways while sitting on point. ....oh......

Vanu wins most of the time if the numbers are balanced, someone must have recorded this statistic somewhere

The really big issue with this is the same as I've been saying so far, in that recoil patterns are all random. Random patterns are inferior to set values when attempting to determine a specific quality of a weapon. The image you've shown has a net average of 0 recoil given equal events occur on both sides of the centre. Wouldn't this flat out invalidate the recoil consideration?

RPM as we know isn't ever constant since it relies on FPS, neither is the human reaction time. Statistically speaking, the more different averages are lumped together the less accurate the overall grouping becomes. Modelling a chair for example, based off average measurements creates a chair that isn't well suited to anyone. The fewer things a stat relies on the more accurate it is. Lumping on recoil, and the probability of each direction, and FPS, and human reaction time, it becomes less accurate than the flat statistic of CoF: The set value describing an area in which bullets are placed around the reticle.

Perhaps so, but consider this: The larger the bloom of a weapon during fire, the larger the cone of fire will be. The larger the CoF is, the lower the probability of hitting a smaller target such as a player's head. Therefore, lower bloom directly equates to a higher headshot rate. Occam's razor (which is not so coincidentally the practice of removing extra steps or probabilities to minimize error) supports this over analyzing the recoil pattern mid-engagement and making the probability-based deciding to compensate for this under the assumption it will occur.

Quite frankly you've gotta be kidding yourself if you think players are actively chaining headshots in an ADAD fight because they're analyzing probabilities while also tracking the opponent and correcting existing recoil.

Yes? Am I missing your point? My direct response to that post was with the second paragraph. Sorry, I should have cut the quote down to reflect that, if this is where any confusion is borne from.

The problem with the Promise's recoil is that it isn't trivial at all until a few shots are fired. In my experience with it I found it tends to shake a lot when bursting in 2-3 shots, which kinda sucks at range when the CoF becomes way too big to hit when the recoil finally smooths out. But we're not really here to talk about the Promise so I'll try not to derail this thread with it.

Why would a lower RoF make recoil easier to deal with? At 698 rounds per minute or 11.63 rounds per second at max fire rate it's still shooting faster than a player is able to react with each shot. If anything a higher RoF is easier to work with because the overall change in recoil tends to be smoother, compared to the heavy "steps" of a SAW for example. (Shoot, gun goes up, pull down, shoot, gun goes back up, pull down, repeat. Can pull down faster than the gun shoots with a 1x sight so it becomes a rough ride on the eyes.)

Well another issue I tend to have with the recoil as a measurement of accuracy argument is most guns can have attachments which reduce recoil. Is this synonymous with improving accuracy? Some players are better at controlling recoil than others. Is a stock Orion more accurate in the hands of a veteran player than it is in a new player's hands? If one would argue so, wouldn't this be a difference in user accuracy? As I've said there is just too many random factors to maintain accuracy in our measurements of, ehm, accuracy.