Symthic Forum was shut down on January 11th, 2019. You're viewing an archive of this page from 2019-01-08 at 23:17. Thank you all for your support! Please get in touch via the Curse help desk if you need any support using this archive.

Welcome to symthic forums! We would love if you'd register!
You don't have to be expert in bit baking, everyone is more than welcome to join our community.

You are not logged in.

## Horizontal Recoil in BF4

Skynet T100

Posts: 1,614

Date of registration
: Apr 12th 2013

Platform: PC

Location: Guilin Peaks, Finland

Battlelog:

Reputation modifier: 14

Saturday, October 11th 2014, 1:39pm

### Horizontal Recoil in BF4

In BF3, horizontal recoil was created by pseudo-random patterns with a rule of no-three-recoils-to-the-same direction (Horizontal recoil analysis (recoil patterns)). In the absence of evidence (or at least evidence shown here ), BF4 was assumed to have similar recoil generation (Algorithms for the Shooting Mechanics in BF4).

I quickly looked at the H recoil patterns with a memory probing script produced by our revered supreme magician who is in for quite a bit of rep. Have a look at some magdumps with naked M416:

Isolating the 803 H recoil values in this batch and drawing the 200 first

shows us that there are no apparent pseudo-random rules applied.

The distribution of these recoil values is as flat as expected for just random numbers in the range from RecoilLeft to RecoilRight:

Also he number of right vs. left recoils was as expected for M416 (74.47 vs. 25.53 %, theoretical 75 vs. 25 %).

The implications of uniform and random H recoil is that H recoil trajectories will be random walks (analytical solutions possible!) and that simulations (mine at least, dunno about hitrater) of gun accuracy need a slight revision.

Edit: Algorithms for the Shooting Mechanics in BF4 updated.

@Rezal
This is in line with your observations?
"Less is more? How can that be? How could less be more, that's impossible. More is more." Yngwie Malmsten
"Many bullets help." WoopsyYaya
"most rhetorically legitimate ad hominem 2015" ToTheSun!

This post has been edited 2 times, last edit by "3VerstsNorth" (Oct 11th 2014, 2:13pm)

Wizard moderator

Posts: 1,620

Date of registration
: Jun 28th 2012

Platform: PC

Location: RI, USA

Battlelog:

Reputation modifier: 14

Saturday, October 11th 2014, 2:17pm

I like how we get a tool like this and it takes a day to finally settle the H-recoil topic once and for all. There is one thing left I'm curious to test however. Back in January when talking to Frankelstner and asked him about H-recoil using HorizontalRecoilAmplitudeIncPerShotMin -0.3 & HorizontalRecoilAmplitudeIncPerShotMax 0.1(M416 data) as and example and he said this.

"Anyway, horizontal recoil is rotation in degrees. From a mathematical point of view, the correct direction is of course counterclockwise (when looking at the soldier from above). Thus the rotation can be anything between 0.3 degrees in clockwise rotation (i.e. to the right) and 0.1 degrees in counterclockwise direction (i.e. to the left). I'm not sure if Min and Max have any meaning other than giving the possible range of values. I haven't tested outright swapping both values." - Frankelstner

I don't really think flipping the values will change much but it might be something to test if your curious.

Skynet T100

Posts: 1,614

Date of registration
: Apr 12th 2013

Platform: PC

Location: Guilin Peaks, Finland

Battlelog:

Reputation modifier: 14

Saturday, October 11th 2014, 4:00pm

Yeah. It's much like in real science; a vast majority of new discoveries is propelled by new tools. So, all my rep to scriptkiddies . With H recoil and recoil recovery pretty well figured out, any simulations are going to be more relevant.

Re switching the sign, I can do this to a gun with an opposite recoil balance, this would be good to check. So far all the values I looked at appeared as if they where just uniformly randomized in that range but, admittedly, I have not yet checked any higher-order statistics. Or did you think of switching the values in files and replaying the same gun?
"Less is more? How can that be? How could less be more, that's impossible. More is more." Yngwie Malmsten
"Many bullets help." WoopsyYaya
"most rhetorically legitimate ad hominem 2015" ToTheSun!

Rezmer

Posts: 4,259

Date of registration
: Apr 6th 2012

Platform: PC

Location: From the heart of Europe.

Battlelog:

Reputation modifier: 17

Saturday, October 11th 2014, 4:14pm

Yes, this is what I have observed as well. Hitrater has already been updated.
[Aristocrat's Shoes]

### Quoted from "Darktan13"

TLDR -
Teamwork is where players function by themselves, but their effectiveness is multiplied when they work together. Not a checklist of "did we bring a healer so we can start playing?"

This post has been edited 1 times, last edit by "Rezal" (Oct 11th 2014, 4:21pm)

Wizard moderator

Posts: 1,620

Date of registration
: Jun 28th 2012

Platform: PC

Location: RI, USA

Battlelog:

Reputation modifier: 14

Saturday, October 11th 2014, 4:40pm

### Quoted from "3VerstsNorth"

Yeah. It's much like in real science; a vast majority of new discoveries is propelled by new tools. So, all my rep to scriptkiddies . With H recoil and recoil recovery pretty well figured out, any simulations are going to be more relevant.

Re switching the sign, I can do this to a gun with an opposite recoil balance, this would be good to check. So far all the values I looked at appeared as if they where just uniformly randomized in that range but, admittedly, I have not yet checked any higher-order statistics. Or did you think of switching the values in files and replaying the same gun?

Same gun swapped values so then we can compare.

PvF 2017 Champion

Posts: 7,304

Date of registration
: Apr 3rd 2012

Platform: PC

Battlelog:

Reputation modifier: 19

Saturday, October 11th 2014, 8:17pm

@3VerstsNorth

Do you think you could offset the values with a middle of the road value (looks like -2.5 in total* don't know how many per shot but it'd probably be -.2) to see what Hrecoil "compensation" looks like?
Data Browser

Passive Spotting is the future!

"Skill" may indeed be the most magical of words. Chant it well enough and any desire can be yours.

Are you a scrub?

### Quoted from "blahdy"

If it flies, it diesÂ.

This post has been edited 2 times, last edit by "NoctyrneSAGA" (Oct 11th 2014, 8:44pm)

Posts: 235

Date of registration
: Aug 28th 2014

Platform: PC

Location: Sweden

Battlelog:

Reputation modifier: 7

Saturday, October 11th 2014, 8:55pm

@elementofprgress

I don't know if it's of use to you, but doesn't the AS VAL have swapped recoil values?
Why I don't use lasers:

### Quoted

The Green Laser Sight you should almost always turn off as it gives away your vacation.

### Quoted from "Rezal"

Double Ass-Ault Tank Lunch - YouTube

Most fun ever in this game mode.

BF4: Competitive Tank Launching

Rated on explosives used, launch, flight trajectory, style/tricks, and landing.

### Quoted from "Pastafarianism"

Shitgun: Whoops did I spell that wrong? Nah, I don't believe so.

### Quoted from "InvisiblePrawn"

Anyone got any ideas about what the UGL Rail on the L86 is for? (13/04 Patch notes.)

XM25 can't mount, can it? Looking forward to my Underslung Ammo box!

### Quoted from "C0llis"

Is there any way for anyone that doesn't have mod powers to know what the IE on 360 thread is? I feel so left out...

Nope. Deleted so only mods can see the contents. It has some... interesting stuff there... *squints @Riesig <.< *

*Nick 30075's Application for Moderatorship follows:*

HURR DURR I IS GOOD MOD GIB IE ON 360

My application would be better.
"Totally not Tapatalk only. Pls gib pwor"

Wizard moderator

Posts: 1,620

Date of registration
: Jun 28th 2012

Platform: PC

Location: RI, USA

Battlelog:

Reputation modifier: 14

Saturday, October 11th 2014, 10:25pm

### Quoted from "Noobspanker"

@elementofprgress

I don't know if it's of use to you, but doesn't the AS VAL have swapped recoil values?

HorizontalRecoilAmplitudeIncPerShotMin 0.1
HorizontalRecoilAmplitudeIncPerShotMax -0.4
and the 416 has
HorizontalRecoilAmplitudeIncPerShotMin -0.3
HorizontalRecoilAmplitudeIncPerShotMax 0.1

So yeah I don't expect there to be any real difference when swapping min and max seeing as both have a clockwise pull in game. None the less it worth testing with 416 so we get a direct comparison. Like I sorta mentioned in my first post, it is more of a curiosity thing.

Skynet T100

Posts: 1,614

Date of registration
: Apr 12th 2013

Platform: PC

Location: Guilin Peaks, Finland

Battlelog:

Reputation modifier: 14

Saturday, October 11th 2014, 10:57pm

@NoctyrneSAGA

Here:

RAW is, well, raw.

PREDICTIVE is the per-doctrine compensation method: practise until you grasp the mean drift and always compensate with that expectation. With BF3-like recoil, this is good even for long bursts because the pseudorandomness prevented recoil from expanding as a function of shots fired. With BF4-random-walk recoil, however, the gaussian governing H recoil expands all the time because by-chance successions of same-side recoils take you far from the expectation value (mean).

REACTIVE is, as far as I see, most ecologically valid compensation method as in being the thing we all do. With pseudo-random, there's little room for reactive compensation. But with the slowly drifting random walks, an adept shooter can continually adjust the aim to keep on target. It takes some time to get the info from eyes through the brain's visual system to the motor system and from there finally into hand muscle control signals. Let's take an optimistic estimate of this reaction time and say that 240 ms (three full M416 recoil intervals) is enough: let's now let the shooter use the information from each shot i as a compensatory signal for each shot i+3. Applying this to the PREDICTIVE recoil patterns prevents the runaway expansion. Having the predictive part is still critical there to control the recoil from the initial three shots that happen so fast that one really can't compensate for them reactively.

Yeah, sounds theoretical but I think this - staying on target - is what we do automatically. The question is how short the lag is. I can now, thanks to magicry, explicitly measure the lag (at least from myself) but not before next weekend coz traveling to France tomorrow for a week.

Another concern of the relevance here is "hey, you're magdumping - we ain't doing dat". Well, because recoil recovery is initially super slow, bursting at spread-decrease optimized tapping rates (as in almost full auto 2-shot microbursting w/LMGs) is essentially full auto in this context and the same need for reactive compensation applies.

3VN signs out until 18.10., cheers mates.

PS. Now with H recoil pretty much settled, I'll redo the T100 (+ DPS!) tables with smarter formatting when I come back.

@elementofprgress

If you send me the logRec.csv for swapped-values-M416 (or tell me how to swap the values without having my ass in PBB master index), I'll pull out that control analysis.

@Miffyli

You're the man!
"Less is more? How can that be? How could less be more, that's impossible. More is more." Yngwie Malmsten
"Many bullets help." WoopsyYaya
"most rhetorically legitimate ad hominem 2015" ToTheSun!

This post has been edited 1 times, last edit by "3VerstsNorth" (Oct 12th 2014, 10:07pm) with the following reason: fixed linguistic derps