#Fix standing-to-prone hitbox

1 messages · Page 1 of 1 (latest)

wintry yacht
#

To insure you this is not a ping related issue, or packet loss situation. I had my connection to the sever up on another monitor at the time because i wanted to test this exact issue.
This video was on BBC dom server 127v127 on 22 ping.

Anyone can recreate this and its a big issue for every player. I suggest this animation & hitbox be looked at so its not shooter favored because this looks really bad on the proning players perspective. This is something id expect on 200 ping. Not one tenth the amount.

As someone whos often on the move my self im also given free kills like this all the time, and it just does not feel right.

#

I understand one could argue that the top of the screen does not represent the top of a players hitbox model, but if you take it in frame by frame I am hit twice you could even say 1 of those shots was fair because of this reason, but few frames after and still being hit when im in the full prone position is just bonkers to think about being remotely arguable and even with a shot from one of the other players shooting at me missing.

modern temple
#

this would inadvertently make dropshotting even more advantageous but i still favor the hitbox being accurate to the model in any scenario

wintry yacht
#

I agree it would, though as much as id like something like dropshotting fixed I want support players not to die behind their cover they have made when they're going from standing to prone to reload or simply duck to save themselves from a firing squad of players.

This change going to effect all players equally (unless you have never proned while being shot at before lol?) . Making cover more powerful, making reaction times matter more.

The way it is now it feels like you're getting punished for pressing the prone button in any situation. @modern temple

modern temple
#

i do it so often that like 1/3 of my deaths are being shot while fully behind cover

#

so i fully understand

split jasper
#

The problem isn't hitbox, its just movement delay in general. The game isn't updating movement as often as it is updating bullets.
The good old peekers advantage, you run around a corner your camera comes in before your model (or in this case, your body takes longer to prone on their screen than yours).
It's just because it's very obvious with prone because it's fast.
This is easy to notice on running animations to ADS - If someone you're looking at goes directly from sprint to shooting you, you'll see bullet tracers and damage indicators before they finish raising their gun.

compact verge
#

It's happing for me too often and very annoying. Going behide wall and .... dead 🤬

nocturne lantern
#

yes 😔

#

love when that happens

steel wadi
# wintry yacht To insure you this is not a ping related issue, or packet loss situation. I had ...

In this clip you died within ~100ms of being in a position such that they should not have been able to hit you. Taking into account server tick rate this could feasibly happen with as little as 60 ping on the enemy side which is fully possible, and remember there is packet loss on both ends too

If you wanted to show that this is a hitbox issue, you need to record yourself shooting someone else and getting hit markers when you shouldn’t have.

#

What seems likely to me is that you died about the same time as you begun to move into cover, the hits just did not all arrive until you were hidden. As far as I'm aware, hitboxes in this game are literally just what you see is what you get, regardless of animations

tame osprey
#

this happens to me all the time

#

so annoying

regal pebble
#

its your ISP

#

not hitbox

#

get a better ISP

tame osprey
wintry yacht
steel wadi
#

Gather some though, because I've never seen a single one.

wintry yacht
#

You're making up 60ms or something? You can't just pull numbers out of the air for a debate. Please don't do that it makes me not wanna debate the topic at all with you. It makes your argument just invalid. Like I said I had 22ms ping at the time of making that clip.

steel wadi
wintry yacht
#

Rough is not enough

steel wadi
#

Well, the exact value is irrelevant, so I'd say it is

#

unless you died like several seconds later

#

but in that case claiming it's a hitbox issue would be crazy

split jasper
wintry yacht
#

You claim 60ms yet the clip is almost a full second.

steel wadi
#

it definitely wouldn't stop it

split jasper
steel wadi
#

but I said 60ms as an estimate for the enemy ping

steel wadi
wintry yacht
#

It reduces it by its exact number you compare it to. I feel like you don't know what you're talking about, and your refusal to get real evidence is clear you don't care enough to debate it so why are you here.

split jasper
steel wadi
wintry yacht
#

No offense but you're coming off as just a troll

lapis epoch
steel wadi
wintry yacht
#

You can't even claim what you're saying with real numbers???

steel wadi
#

what?

#

I don't know what you mean

#

I don't know how much ping the enemy had, or their packet loss rate, or your packet loss rate, but presumably neither did you.

#

Again, my point is the clip provided is not evidence of the claim you made. It might be, but that seems by far the less likely explanation

#

Considering you also see this exact behaviour from things like walking behind cover

wintry yacht
steel wadi
wintry yacht
#

It's in the very beginning. I was monitoring the packet loss on another monitor.

steel wadi
#

You were tracking packet loss to the game's servers?

#

though anyway, theirs is also very important

wintry yacht
#

You need to prove with factual evidence that it's not the case in my clip in your next few messages or I'm just going to stop responding all together. I'm tired of this back and forth and don't have the energy to continue.

steel wadi
#

mate you are the one making a claim in the post, burden of providing proof is on you

wintry yacht
#

I provided more that enough proof

steel wadi
#

...where?

#

all you posted is a clip which shows behaviour that looks identical to that caused by latency

#

to prove your statement you need to show a scenario where latency certainly is not a factor

wintry yacht
#

Me dying almost a full second behind cover on 22ms and no packet loss that was being monitored on a 120hz server.

steel wadi
tame osprey
#

imma say something stupid , maybe the camera position is different to the hitbox position or maybe i dont know anything about game dev.

lapis epoch
#

show a video of you killing sb in that situation

steel wadi
wintry yacht
#

You claim that but it supports my argument in another way why are people with packet loss then gaining an advantage over people without any.

steel wadi
#

unless you mean peeker's advantage

#

them having a delay on everything they see is a two sided thing.

#

it also means they don't see you until after you peek and vice versa

#

the only way this can lead to unfairness is if they have a playstyle that deliberately leans into this more than other players.

wintry yacht
tame osprey
#

oh yeah my bad

steel wadi
wintry yacht
#

Peekers advantage is not a factor here because my RTT was not enough to represent the time where I should never have been hit.

steel wadi
#

RTT? not familiar with the acronym

wintry yacht
#

Round trip time. Time it takes from my packet to send a signal, gather information from the other users in the server and be brought back to me.

#

Its IMPOSSIBLE for me to have been hit here if it's not a hotbox or camera related issue.

steel wadi
wintry yacht
#

And if it's a camera related. Then that shit needs to be changed ASAP. Cause my camera has me fucking prone and I'm still dying

wintry yacht
#

They have to send messages to the server for me even to receive them to be shot and hit

steel wadi
steel wadi
wintry yacht
#

If they're losing packets shooting me they miss.

steel wadi
#

if packet loss is too bad the server does deny the hits

#

but usually that is not what happens

#

packets are resent if they do not reach the server

wintry yacht
#

Ya so you're packet loss claim is untrue

steel wadi
#

???

#

it adds delay if packets must be re-sent

wintry yacht
#

I died

steel wadi
#

as that info has to be transmitted back from the server to the client

#

and then back again

wintry yacht
#

If shooter had packet loss I live and shots are denied

steel wadi
#

No, only if they bypass a certain packet loss threshold.

#

Oki even posted clips demonstrating this a little while ago

#

of how hit reg looks with different packet loss ratios

#

it will only deny the hits if the packet loss is too bad for it to be likely those shots were fired fairly

wintry yacht
#

Still I'm behind the wall by almost whole second

#

This would be picked up by that

steel wadi
#

anyway. From the enemy client your position is lagged by yourLatency + theirLatency + yourPacketLoss + theirPacketLoss.

From your client their shots are lagged by yourLatency + theirLatency + yourPacketLoss + theirPacketLoss.

So the total delay in discrepancy between clients is 2*(yourLatency + theirLatency + yourPacketLoss + theirPacketLoss)

steel wadi
wintry yacht
#

We have to remember that this person also has to be shooting me at the exact pixel of free space between the barrier and my head

#

For every single shot

steel wadi
#

wdym by "exact pixel"?

#

Firstly in that hit reg isn't handled based on what the game renders, but also... why?

#

They just have to hit any part of you

wintry yacht
#

Because this assumption means they have the most time to hit me

steel wadi
#

Also consider they are shooting from a greater height than the barrier, so if the top of your head is level with it they can still hit you

wintry yacht
#

But it's incoherent cause the player has recoil unless they're cheating

steel wadi
#

That was very close range. Recoil is a factor right at the last possible opportunity but not much before then

wintry yacht
#

It's safe to assume they're not cheating.

steel wadi
#

not a gun with a lot of recoil either

image.png
steel wadi
#

A clip from the firer's POV would not have this issue as hit reg is handled by the firing client.

wintry yacht
#

It's not though because I've told you my RTT is not coherent with the time I died.

#

So my claim is undisputable

steel wadi
#

that is the time it takes for you to be updated on the info stored in the server, not to be updated with the info on the enemy client.

#

The enemy client is important because that is where hit registration occurs

#

likewise their RTT time is not sufficient to have your up-to-date position

wintry yacht
#

It's either:
a) fucked up camera
b) jank hitbox
And I can do the math to figure out if it's a camera but it's gonna be obvious that it isn't the issue

steel wadi
#

You'd be correct if all processing was done on the server but it looks to me like you are ignoring that it simply is not

wintry yacht
steel wadi
#

No, that only happens with extreme cases of packet loss.

#

and enemy ping still applies, so that would not be a disproof. Merely a reduction in the extent of scenarios in which my argument holds

wintry yacht
#

Packet loss will never be anyone's advantage why are you claiming someone with packet loss is going to kill someone faster

steel wadi
#

what?

#

I never claimed that

wintry yacht
#

You are.

#

You're discussing that he killed me because of his connection

steel wadi
wintry yacht
#

Peekers advantage is based on ping like I said

#

I have 22

steel wadi
steel wadi
#

that determines how far behind each of you sees your opponent's actions assuming 0% packet loss to the server from both clients and infinite tick rate

#

neither of those are the case of course

#

~4ms are added on at each end from limited tick rate in a 120hz server, packet loss not so easy to quantify though :p

#

especially as we don't have a ground truth for the opponent's connection

wintry yacht
#

Do I need to bring out papers discussing how peekers advantage works in video games and then show you how it compares to the situation that I have displayed

#

For you to realize your wrong

steel wadi
#

If you wish.

#

If I'm wrong I need to know about it

wintry yacht
#

First off what's your background to even be making these claims.

#

If it's in the same field of what we're arguing about you should be fired

steel wadi
#

1st year of CS degree, so... basic knowledge?

#

I would think if my assertions here are wrong it should be pretty easy to explain why though

lapis epoch
#

in the current state of the game your ping has barely any effect on wether your shot hits or not

wintry yacht
#

Alright well that's fair then, I don't need to use simple terms

tame osprey
#

aight guys cmon chill now no need to compare professions it seems kinda personal ngl

steel wadi
#

It is weird, but if he can prove my assertions wrong that's good for my knowledge.

#

gonna make lunch rn, will check back in a bit

wintry yacht
#

It's insulting to have to argue with people when I know what's happening backend

wintry yacht
steel wadi
#

I haven’t heard of a game where peeker’s advantage doesn’t exist. How’s that even possible without waiting for server confirmation for all inputs?

wintry yacht
#

you are misinterpreting my message

#

Im claiming peekers advantage impossible to be the cause of my death in these circumstances that ive displayed in my video.

steel wadi
#

Apologies.

#

But I do not see your proof as conclusive, it appears to blatantly ignore that the server does not contain up to date information.

#

(You seem to imply your own RTT is the determining factor in a scenario where the relevant information that needs to be transferred is not stored in the server)

wintry yacht
#

As the aggressor and player that reacted much before the player shot me, this would prove to you that the player here couldn't have killed me due to peekers advantage. Even though in the situation here, im not shooting back.

#

Because this server is 120hz, and I didn't claim that i can understand that this might be an issue if it were 60hz

steel wadi
#

I didn’t claim that was the reason they killed you, those two are just a product of the same thing

wintry yacht
#

this is the players first shot

image.png
#

this hits me because of i guess is player camera

steel wadi
#

I suppose you could refer to it as “cover-taker’s disadvantage” in a scenario like this, but that’s not exactly a commonly used phrase and perhaps for good reason

steel wadi
wintry yacht
#

second shot registered while completly behind cover

steel wadi
wintry yacht
#

3rd shot registerd, dead, behind cover again

image.png
steel wadi
#

Same story

#

Consider you move behind cover. The enemy client has a window of:
delay(you, server) + delay(server, them)
To shoot you before you are behind cover on their client.
You will only see/hear/feel the impact of the shots fired after an ≈ delay after they fire (which is as already described, beginning after you’ve started moving on your client)

wintry yacht
#

now this is in 60fps, 11 frames from the first shot hitting me 16.7ms PER frame that's 183.7ms to kill me. from his first hit. I was behind cover that long. 3 hit markers. Without bullet travel time (because we can call this point blank the numbers are way too small to factor) 700RPM gun, divided by 60 is 11.66 rounds per second. Time to kill in this situation was 183.7ms. so 11.66 rounds/s x 0.1837s is 2.1 rounds shot. Though he shot 3. This makes even less sense now.

#

Unless my math is wrong, this is just flat out impossble

#

To combat this Issue the hitbox needs to be changed in this animation.

#

22 ping, and this shit feels like im getting shot by players with 200, like i said

#

The unknown variable here is the other players ping, which in this case had to be above 200 to even kill me.

steel wadi
#

So that makes sense

#

After the first shot (instantaneous) there were two units of the time delay associated with firing again

wintry yacht
#

yes thats why my caclcuation only starts on the first hit, 0ms

steel wadi
#

He shot 2 throughout the time period

#

But also bear in mind the pattern you see the shots in will not be identical to the pattern they were fired in

#

It is fully possible for them to register out of order, even

wintry yacht
#

no no no i see what you're saying we're starting at 1 shot

#

so 2 shots is right and i didnt factor in that im already lower hp

#

Anyways, 22ping is a disadvantage for me, anything above 0 is, so 183.7ms - 22ms is 161.7ms on his end. To kill me.

#

Still means 161.7ms time for shot 2 and 3, and im behind cover on my side for the last 2 shots.

steel wadi
#

My initial guess of 100ms was too low I admit, about 200ms passed between you seeming too far down to hit and you dying

#

so that you ended up with a ping ~100ms longer than my initial estimate checks out

wintry yacht
#

Anyways, with that cleared up. Still i am claming the hitbox for my death.

#

I just cant see my camera poistion really being the factor here. The time is just too long.

steel wadi
#

So, we're back to the start. I think you need to get some footage where there isn't reasonable doubt to suspect the issue is something else.

#

I have consistently noted latency seeming more prominent than it should be in this game. I don't know exactly why this is the case

#

But a lot of people have expressed that it seems greater than it should be

wintry yacht
#

You can just do math to figure out the camera/hitbox situation

#

Then again, even if we're back to the start. People like myself, and at least a few others in here know the extent of the issue. Peoples pings these days to servers are generally under 100 ping, very few players are playing above that. The consistency of this happening and the claims =/= the number of players on high ping.

#

Research would mean noting every high ping player in ur lobby, and recording every death

steel wadi
#

I cannot explain the seemingly inflated latency issues the game experiences. There have been several updates intended to fix it but none have really changed anything

trim yoke
wintry yacht
trim yoke
#

Yeah, I fear people would get scared of dropshotting ngl... but the animation is like, REALLY slow rn

#

And it's just going from standing to prone I think? Does it apply to crouching too?

honest hamlet
#

Wait, is this desync why I keep dying when I prone behind a sandbag after someone was shooting me?

trim yoke
#

No, not dsync. the animation is slower than your camera movement

#

actually lemme see if there's a video of this out of curiousity

honest hamlet
#

Ah, ic

trim yoke
#

But like I notice it a lot when you see teamates either go from sprinting to prone or vice versa

#

they take about a full second give or take to switch between states... but the camera movement is at least half that. It could be placebo, and if someone would like we could test it when I get home in about ~4 - 5 hours here, we could test it

#

(And just to clarify too, desync/ client hitreg definitely is a factor as well, but I really think the sluggish animation vs quick camera movement is a very big part of it too)

trim yoke
#

Anyone here online? I'm down to test this out a bit if someone is

#

Or actually, over here there's a good clip showing how wonky the animations are with it... I'm fairly certain the issues we talked about are there

#

You can even see the animation skipping forward when the switch from prone to standing up, because they're still in the middle of it when they switch confused_af

wintry yacht
#

Bumping this with another instance of this BS, on Official severs, 31 ping with no packet loss, player who killed me was also from Canada.

tulip axle
#

._.

#

Well... Oki make the fix for this look nice... In practious. It will just become cancer.

modern temple
#

a slight improvement is still better than none

trim yoke
#

Didn't even notice... Honestly better than what we've got imo. Accurate camera movement means less jank

#

Good work Oki 👌

modern temple
#

with the netcode update this should greatly reduce dying while fully ducked

tulip axle
#

From what i could tell, it was more consistent, Consistent movement in a near zero TTK game is better for everyone... however...

#

Above video post from the OP here. Just proves how it will becomes cancer. because they become unkillable if they have good reaction and snap.

modern temple
#

i think being rewarded for fast reaction is the basis of pvp

#

plus, nobody likes successfully reacting to fire and dying outside of their control

#

if it becomes problematic in the future, then proning can be slowed a bit

#

which i would approve over any amount of hurtbox jank 👍

willow flicker
frail swan
#

symbol

steel wadi
#

Looks like it was an animation issue, then

vast jolt
#

The next step is to fix deaths in 1 frameBBClown

steel wadi
#

That one isn’t entirely possible

#

Might be possible to reduce it

#

I guess you could fake it to make it look like it was “fixed”, but then the user would see something more different to what actually happened which isn’t a good thing

wintry yacht
#

Animation issue in combination with hitbox was not = what the players camera saw

fierce hinge
wintry yacht
#

Please use this post for the related topic.

steel wadi
#

Related topic is kinda exhausted now

#

Since the issue has been fixed

vast jolt
# fierce hinge Proper armor rework would give people the choice of tankier builds where 1frames...

Man, I started this thread in the wrong discussion thread. I apologize. But even tank builds are not immune to this. When there are support missions and when I play on support, I too catch deaths in 1 frame. On medic I tried playing in heavy armor. I didn't feel the difference.

It doesn't matter how thick you are if the enemy, on their monitor, saw you earlier and managed to stuff you with lead in that time and you didn't even know it.