Gears Forums

Gnasher hit registration test - Lead Gameplay Designer responds

I’m sure we have all read about the Gnasher issues that has haunted Gears 4 & 5.

A few days ago I was playing 2v2 Gnashers. I fired a shot at a player which I felt should have done more than it actually did. I recorded it and watched it back later to try and see where I might have gone wrong. Pings were nothing too high so I ruled out lag.

Upon reviewing it I noticed that I landed 10 pellets from my shot. I asked a friend to help me with some testing so that I could try and find some accurate numbers. From doing so I was able to learn the damage of a single pellet at that range and I felt something just didn’t make sense. I feel like this is a clear way to show that the “inconsistencies” are the fault of the game and not the player.

I’d like to claify as I don’t feel I made it clear enough in the video, the shot that I fired was NOT at the range where I am counting pellets, this can be seen at the start of the video. I allowed the clip to play forward a few more frames so that I could count the pellets clearer. Based on this, I feel like the test shot that I fired was perfectly fair.

I wasn’t going to bother posting this here because I have realised that this game has so many defenders, however a viewer has requested I share this so I have done.

Video Link

16 Likes

Good video man.
Concise and to the point with visual evidence of what is obviously something going wrong.
You should tweet this at them/try reddit with it, as I’d love to see them explain this away.
I will say maths has yet to be tc’s strong point. From the incorrect values on the horde card to the completely made up pings, someone seems to make rookie mistakes again and again.

On top of tc’s dodgy maths that end bit is just too funny. I get how the auo-aim/bullet magnetism gets you that kill, but why the hell is it still that strong in a mode where it is supposedly dialled down?
Just how far can you miss in the normal modes and still get a kill?
.

5 Likes

I did Tweet them, it got a fairly good response too but they appear to have not seen it lmao

1 Like

Maybe they did see it but it was not deemed “worthy” of a response. Trying to get a TC post to any sort of topic on either the forums or a response to a Twitter message is almost like finding the needle in the haystack. Certainly been my experience on both platforms as of now.

2 Likes

Sometimes they don’t even have to be pointed at you when people are all blind firing at each other depending on each players ping times and how many packets get lost by each players internet connection. I have noticed that in the playlists with core tuning that the higher a person’s ping time the slower their gun shoots.

The inconsistent damage we keep seeing all the time depends on how much packet loss a player suffers when connected to the game. From my experience playing all of the games online with a slow internet connection something changed with Gears of War 4. It appears that the more packets that get lost the better things will go for those that are losing packets. Players that have a consistent connection receive all of their packets(incoming damage) players that lose some of their packets don’t take full damage at the same range as players that don’t lose packets.

Packet loss can be seen in the game by viewing the scoreboard during a match. Players with a ping that moves up and down by large amounts are losing packets. The Coalition tried to hide this by showing us the wrong ping numbers via the scoreboard when they implemented the low ping matchmaking option. I am only assuming that they did this to avoid complaints about high ping players in low ping matches.

It didn’t work because I have played almost everyday scince the low ping matchmaking option was added, the only reason I came back after launch and almost every match that I play on a central us server has at least one player that has their ping go from 80ms to 120ms then to 300ms then back down to 80ms. One day I sent a person a message that I was going to record his ping jumping up and down and it stopped moving before the match started while we waited for a good two minutes for the game to load the map.

I also tested this while downloading a game in the background(intentionally causing packet loss) in quickplay and it definately gave me an advantage. I could move faster and more efficiently than I could before. My shots worked almost every time I pulled the trigger some times even when they weren’t lined up. I stopped sticking to the wall like when my connection isn’t slowed down. I could run almost straight at lancer fire and not go down. It was like playing a completely different game.

In my opinion it is pretty obvious that some people are intentionally overloading bandwidth to cause packet loss to try and gain an advantage. If they are doing it unintentionally and are unwilling to learn how to fix the problem or unwilling to do what is neccesary to fix the problem they should be restricted to quickplay or at least their own ranked lobbies where all the people that lose packets can shoot their broken guns at each other.

It’s 2020 and we can’t search for games by best connection like we could with the first game they made that had players hosting games. This should not be something that happens with things like dedicated servers and better connection search options. I should almost never play a game on a central us server because anything higher than 60ms is not a low ping connection.

It already takes longer than 2 minutes to get into a match with every other lobby being dissolved because someone quit searching so just leave players that have high ping to farther servers on their nearest server and then bring players with faster ping times to the players with slower ping times.

That was supposed to be the point of having dedicated servers for games. To put players on as equal a playing field as possible. They should have just left it peer to peer with host advantage because they have essentially turned it back into that anyway with ping advantage.

8 Likes

They dont care about any of this they only want money and to make every Gears original fans life a living hell…

2 Likes

A couple years ago, I posted still frames, clips, etc that were clear examples of hit registration issues. The best was a hammerburst bursting a full magazine on an enemy’s head (an enemy who didn’t know I was behind him) and he didn’t drop. I’ve never had a TC response.

The best we can do is be aware, share info, and be vocal (in a mature way). The pressure may cause them to eventually address it.

So many people complain about hit registration when they miss their shots that it’s tough for a dev to know what’s real and what’s rage, and I understand that. Plus, it’s online gaming so it’s not going to be perfect. But, Gears definitely has an reputation for feeling “off” at times and could use some work in that regard.

1 Like

Here is what i have found out.
I recorded a death in ffa to try and figure out why I messed up or why my character didnt do wht i wanted him to.
So while it was fresh in my head
I reviewed it.
Wht was recorded WAS NOT WHAT HAPPENDED WHEN I WAS PLAYING. I am 100%positive on this. Because it was so fresh in my head.
I try to up A a opponent but get stuck on the wall but still was able to up A but my shot missed completely and i said to my self ( tht wasnt right)so i record it and watch what happened.
THIS IS WHAT IT SHOWED IN MY RECORDED VIDEO.
It shows my character miss the corner, im right before the edge of the corner. I sit there and shoot against the wall instead of up A ing and the opponent is soooo much closer to me.
But when i was playing i take the corner and up A and the opponent is too far away. After i up A and miss. He walks up Gibs me

BUT AGAIN. THT ISNT WHAT HAPPENED WHEN PLAYING AND WAS COMPLETELY DIFFERENT WHEN WATCHING WHAT WAS RECORDED.
So when i get a chance. Im going to record it live with my phone and if i see something sketchy. I will record it also with my xbox and review them both.

SO BASICALLY. THT EXPLAINS why when you think you got screwed on something and record it and review it. It will show you how you messed up. But thts the problem. Thts not what even happened in the first place.
Really hope tht made sense.
If not. It will make sense after i record it will my phone and xbox and you will see the difference of real time and wht the game shows you what shouldnt have happened in the first place.

P.S. I have been seeing alot of GHOST BULLETS. ( GUn FIRING but no pellets coming out) it not just me. Ive watched it happended to my friend also. Literslly back 2 back gnasher shots. He didnt even realize it ( guessing because it was in the heat of the moment )

2 Likes

I don’t want to defend or put accusations on anyone, I just explain something that was posted on lancer and Gnasher tunings blog article, plus my own explanation based on my experience with unreal engine scripting.

But first, very nice breakdown on that clip. Some things I want to add before I start, values that I have provided are taken out of TC articles:

  1. Those damage were made on short distance, it implicates that damage reduction is applied on every unreal unit. Short distance starts at 230uu (1uu is 1 cm).
  2. You hit 10 out of 12 pellets, every pellet damage is calculated separately - so every pellet hit on model has different distance from barrel to target.
  3. I’m assuming that you and enemy had similar ping.
  4. TC is making a moving predictions when you fire shot, it’s About 2 frames which is around 32 milliseconds if we assume that game server runs on 60hz tick.
  5. Player health is set to 600 points. If Gnasher deals 108% damage, this means that it is ~650 damage points in full 12 pellets hit. It indicates that 1 pellet do ~55 full damage.

So starting my explanation:

  1. You hit player with 10 pellets, at short range, with possible maximum damage 550 points(90%), but you did 495 points(81%).
  2. Now why you dealt less damage? Answer is quite simple: Damage reduction on distance is applied, the only one case that dmg reduction isn’t applied is when target is in gib distance (closer than 230uu distance from gnasher barrel to enemy closest hit mark).
  3. Every pellet landed at different place on enemy body, so it has different distances from barrel to body.
  4. 10% of pellet damage is 5 points, so it’s possible that you could deal 45 damage more (statistically 9 of 10 pellets could have do separately ~5 damage less due to distance) if your enemy would be 50uu closer (simply one step closer).

That’s my explanation what happend. Sorry for my English if something isn’t very well written. It can be hard to understand when you don’t know how Unreal engine counts distance and trajectory, but if you know how it works, those mysterious situations loses it mystery factor.

But, that’s only my assumptions, it can be different as I haven’t seen code, but it covers most of things that happens on screen in logical manner.

4 Likes

it could also be distance and the part of body that you shoot at probably. i know a lot of games have this but obviously made clear whereas in gears its just a total mystery from the devs. i dont get it, like why cant they just explain to us simply how the damage mechanic works? like distance, area of body shot, etc

2 Likes

Good write up. it is good to see people with unreal experience chiming in, you may know things us laymen do not.

According to what you have said, this should be entirely testable and would just require tha swarm guy to back up towards the post and test at each step back. We should then see a logical drop in damage as the swarm guy moves back.

I think all this will reveal though is that damage was calculated on a player position that does not line up with what is shown on livo’s screen. Leaving us with the confusion of why it would not calculate the damage from where he is on screen? And instead calculate it from where he was on his screen?

Annoyingly we already know that this is what it does, but it doesn’t make it a logical looking system. It is not something a player would be able to parse from visual feedback alone though. They would need to have followed the devs streams to learn this, excuse. that tc bring out.

That brings me to the only dodgy part of your post :
covers most of things that happens on screen in logical manner.

None of what I see on screen seems logical , in fact to a player is just appears broken . TC shouldn’t rely on players like you explaining what their system are doing.

@Livo Sadly that is no supirse.
I guess it at least means he couldn’t explain it away with a pithy PR response easily and this possibly indicates it is a problem they can’t solve. They tend to be quiet in those situations.
Keep fighting the good fight though dude.

I don’t explain what their system is doing, I did not saw their code implementation, I am only basing on what I see on screen in relation to what my experience tells me.

It is testable, if you want to test it, just do it. But take into account that you need to ensure 100% good connections for you and other guys that test it.

Damage is calculated on place that pellet landed. All hits are calculated on player side not servers side ( it explains why player with 100 ping still hits you, even after you have hid behind cover half seconda earlier), it means that what you shoot, shot is calculated on your side, and then game adds 2 frames prediction movement(32 ms) and then calculated hit damage.

We can trying to explain and predict what happen when we take a shot, but there are too many variables that are counted in trajectory calculation and damage calculation to fully break down what happens afterward.

Also I agree that during gameplay you wont catch what happened on screen, because game is super dynamic. I’m not saying that game hasn’t any issues. It has it flaws. But what i saw on Livo clip just made me sure enough of my theory that I wanted to share it with all of you.

What seems for casual player illogical, for a programmer can be some Simple mechanics. All is related to what knowledge the viewer possess.

1 Like

There is no doubt issues with this game, however the one thing that people seem to continuously dismiss is the factor that distance will play.

The Gnasher is programmed to have a damage curve where it is impossible to gib an enemy outside a certain range. Then beyond this gib range the curve for damage slopes downward for every additional ‘unreal unit’ of distance. So, effectively as a test you could consistently land all 12 pellets on an enemy moving 6 inches (about 15 UU’s) away each time and your percentage of damage would drop respectively. Now, while this seems straight forward enough, there is an additional issue to exacerbate the confusion. Because each pellet is it’s own damage packet, the distance to where it engages with your enemies hit box could also vary. This variance can even be applied to two character standing still AFK, as the breathing animation can technically change your distance to the enemy.

Now take a look at the average Gnasher fight and the amount of wall bouncing that happens, Your character is twisted in all sorts of positions and this is where a huge amount of the inconsistency can come from when you land all pellets but don’t gib or only do like 87% damage.

1 Like

Thank you!

My goodness, there are so many people in this community who try to dismiss the basic math going on here, it’s such a breath of fresh air to see someone else who has done their homework.

Yes, there is most certainly issues with connection, netcode, magnetism and its effective range, and general latency. However, at it’s core the Gnasher is actually very simple to understand. Nearly every clip that someone shares complaining about it can easily be explained by their lack of understanding of the gun.

I cannot emphasize this enough! In a game where you are basically doing constant gymnastics, calculating distance for 12 pellets can make things a nightmare.

1 Like

Of course what TC explanation is dose not mean anything. Because they “Hot fix”/ Change it every other day .Without telling anyone what they did!

1 Like

That’s another factor, TC can do some servers settings updated without informing anyone if they think that little tweak does not interfere with gameplay.

Anyway, people will still see this explanation as some heretic trying to confuse then even more than they are currently confused.

Most people does not understand Simple maths, that’s why TC does not go deeper explaining how something works.

I’m glad that TC starter to describe how some things works, its more clearer for some people.

1 Like

A little tweek? More like a Tornado that ruins gameplay! I really think they are tweaking it so hard back and fourth. Collecting data ,and feedback. Beta testing for Gears 6. I think they gave up on this game. Gears 5 is really just a Beta as it stands. It was never finished at launch. And 8 months later not even close to being finished!

1 Like

The Gnasher is a weapon full of legacy with a deeply rooted history – it is what makes it iconic, especially in Versus. It’s behavior quirks and nuances are as much of its DNA as its look and visceral feel.

Unfortunately, the weapon can be misunderstood in how it functions – an issue which has existed since the first Gears. I would love to be able to say that it is a simple as it seems, but that is not the case. The aforementioned legacy has had 15+ years to grow and compound.

We have described the Gnasher’s behavior in detail before, but I believe it is time to be a bit more explicit about it here. So let’s dive in!

First, lets look at the damage graph of the current Gnasher – most of you should be familiar with these graphs by now:

Some details about the elements in the graph:

  • The green line is the damage falloff curve (damage over distance)
  • The red line represents player health – any thing over 600 will kill the player (forcing DBNO or outright death depending on a number of factors)
  • The green box represents the gib range. Any player within this range when shot by the weapon gets special consideration

Next, lets talk numbers:

  • The Gnasher has a base damage of 525 damage in Gears 5 right now (it was 500 in Gears 3 and 4, for reference)
    • This damage is represented by the green line in the graph – you can see that it starts to fall off immediately
    • This means that a full hit of the weapon does 87% of a players full health, but since the damage falls off right away, players will see 80-86% most frequently when just outside of gib range
  • There are 12 pellets to the spread and the damage done is divided by the total pellets:
    • Each pellet does 44.75 damage (just about 8% of the players total health)
    • If you hit 8 of the 12, then you would do 358 damage

Fairly simple so far – but let’s get into the nuances that this weapon has had since Gears 2 and 3: Gib range and pellet round-up

Gib Range

We have talked about this a lot, but the gist of it is that any victim who is within gib range takes bonus damage from the Gnasher

  • Hits within gib range have a 20% bonus to the total weapon damage – that 525 goes right up to 630 damage
  • This means that each pellet does 52.5 damage, or 9% of a players total health
  • Hitting 8 of 12 means you would do 420 damage

This calculation is what usually throws off the user in my experience. In the video, when Livo tested the pellet damage, he was within gib range, so he got the extra 20% bonus. But when the shot in questions was taken, he was outside of gib range which means the maximum damage potential would have been 87% of the players health – more likely 82-83% from the distance.

Pellet Round-Up

This is another system nuance built in the legacy weapon. Specifically, when a player hits ‘most’ of their pellets, the damage done is rounded up to the maximum potential damage it can do.

  • This is tuned to 70% : Hitting at least 9 of the 12 total pellets means you will do the weapon maximum potential damage at that range
  • In effect, hitting 9, 10 or 11 pellets at any range is the same as hitting all 12
  • Landing 8 or less means that your damage is calculated per-pellet based on the ranged modifier

This was added back in Gear 3 to provide a more consistent damage result when attacking enemies at the shorter-ranges usually seen with the Gnasher. It is easy to have a pellet or two miss at these ranges and that would mean that a victim would survive what looked like a perfect point-blank.

So, so summarize, there is more than you would expect going into the Gnasher and how it calculates damage. The best advice is to remember that the gib distance is vital (the red circle below):

Crossing this line means that a victim goes from taking a non-lethal hit (87% or less) to a very lethal (and messy) one.

The Gnasher has behaved this way since Gears 2 and we have not changed it.

I hope this helps!

-Jamie

34 Likes

Gears 5 will last 2-3 years. The longer you blame the game, the longer you hold yourself back from getting better.

2 Likes