[Suggestion] Is it too late to move Sniper Rifles from Infil. to Engi.?

This may indeed provide a little relief for the 'victim' . It's going to come at a pretty heavy cost to the shooter though. I prefer to avoid "messing with the player". That includes concussion of the victim. Messing with the shooter via CoF or a recoil-ish type motion you describe here is a bit more acceptable but still breaks that chain of stimulus->action->reward.

Rather than making it harder for the infiltrator I would prefer we tackle the part that is most painful to the 'victim'. They have no stimulus->action->reward loop at all or ->consequence variant. It's just pure frustration as their death has nothing to do with their actions. (yes, I'm discounting the common sense actions of don't stand still in a window facing a distant mountain)

It's the cloak.

Even with really clean network and the best reaction time, there just isn't anything a player can do. They are going to get killed by something invisible and cannot do anything about it.. It's a sucky experience. One of the worst I've seen in a game.

I run infravision a lot so get keenly aware of it for two reasons. One the decloak is super obvious to me. A cloaker is utterly invisible before, but I'll be the first one to spot the decloak starting. Two, infravision gives a sneak peek into when you die. When the screen turns from black and white to color, you're dead. It happens before any other indicator. Paying close attention you'll see this occur even why you are free to move and fire.

When you combine these an interesting phenomenon occurs. If you are killed you will not see that really obvious decloak occur highlighted by infravision that you are used to. Instead you'll see a normal decloak in full color. Why? Because you're dead while the shooter is still invisible.

It sucks.

I have pretty fantastic network to my server, 50ms is a bad day. But I'd say about 1/2 the time the network of the other guy has tipped it over the edge. I die to the invisible.

Fix I'd do:
Drastically INCREASE the power of the cloak. Make it last longer, make sprinting drain it less.
But put it on a newly added weapon slot. Make the cloak be a device you have to switch to and from.

Or - drastically alter the decloak, or decloak to fire time.

200ms is an example number, it's pretty close to what I'd expect in a game like planetside 2, in a mid size 24-48 fight. I feel you're forgetting to factor in timestamps, player reaction times, and the other intricate issues related to online gaming, specifically through planetside. Even a 100ms advantage over someone with low ttk weapons is a massive advantage. As I've explained, 1hk requires 0ms. Leaves no room for player reaction time, much less for the lag difference.

A simple example without even using a 1hk weapon would be pre-firing. Assume I see an enemy dot heading towards me on the minimap while I'm guarding a door. Being an average player, it takes me about 400ms to react to that information, decide I'm going to prefire him, and set up my position where I think his heads going to appear. I fire as he's about to walk through the door on my minimap. Although he's technically got a clientside advantage over me, say 200ms. He's still got to process I'm in the room. We'll assume he's better than me, with a quick 250ms reaction time. Although I started firing without seeing him, my hits won't count until he's crossed the door on my screen (200ms) during that 200ms window, he's spends 250ms processing the fact I am there, aiming his weapon and firing. That's already 50ms I've been firing at him without him able to react. Most weapons have about 400ms ttk (factoring nanoweave, excluding head/leg shots). Unless I miss all my shots due to rng, or that the guy somehow quad headshots me with a weapon that has much faster ttk than mine, I win the engagement thanks to the 50ms advantage I got from prefiring. It's not a lot in this example, because I'm using the latency I'd expect in a decent sized fight, I'm using an average weapon without landing my prefired headshots, and that the opponent has above average reaction times. In a small fight however, where the connection delay is slower, there's less enemies to focus on, perhaps more average opponents...this prefire advantage can get quite large, and has fueled many hackusations in the past.

In short, even weapons that have above average ttk can get quite massive advantages over opponents in certain scenarios, thanks to reaction times and delay between clients. Now imagine the 1hk shtogun ambusher flying through the floor, or cqc bolters, cyclone heavy or literally anything that's 1hk or with extreme low ttk in the game. Not really fun to fight against something you physically are impossible to counter. As mentioned, squads of extremely skilled players have redeployed to other bases simply because of an above average bolter. Keep in mind due to extrapolation, the enemy sees you moving in linear trajectories, amplified with lag. This is why laggy bolters are especially difficult to deal with, since they clientside you, and you're a predictable target on their screen, even if you try to jump, crouch, zig-zag erratically. Not to mention snipers are a 2hk anyways, so if they land a body shot and you react you're basically dead to the follow-up. Not the most enjoyable gameplay when you're on the receiving end.

They're old vids, but I highly advise you check them out because they are really well done, and not much has changed, or in fact it's gotten worse.



This one should complete the first one a bit, and is more recent:



Everyone already runs nanoweave or symbiote + flak armor. It's meta. In an infantry oriented fps, obviously everyone's going to go for the infantry damage reduction in their suit/implant slot, even though it's useless against headshots. Other options are costly and situational, far beyond the use of the average player. They're stuff for situational loadouts that weren't gonna rely on nanoweave/symbiote anyways, and won't change even if nanoweave is buffed to reduce headshot damage. Doesn't have to be a flat bonus, it could be modifiers like 10% less to the head, 20% to the body. It's not going to change the use of nanoweave, it's just going to make players a bit more survivable on their main loadouts. In any case, making something "less meta" really shouldn't be an argument for change. It's like deleting TI or biolabs because that's what a majority of the playerbase finds fun so generate stale fights.

As for quick scoping, you're right. It entirely nullifies the only real disadvantage snipers have. Combined with critical chain and the issue I mentioned above, it becomes something that can not be countered. A delay would be a good idea, and the sort of nanginata mechanic you mention would definitely have it's place. However I maintain the fact we should not add a concussion effect to snipers, and that their body shot damage should be drastically reduced.

Let's be honest, landing a headshot is easy. Landing a body shot is even easier. What's difficult is to chain headshots when factored with randomized recoil, bloom, etc while in a target rich environment. It's why snipers which can get pinpoint accuracy through hold breath and the opportunity to recon their enemies, and hide from them/dictate their engagements easily are so broken. Not only do they leave no opportunity for the victim to react, but you'd have to be a very bad player to miss a shot. Either make it more difficult to aim, and reduce bodyshot damage, or simply reduce headshot and bodyshot damage. We could also add a specific effective range that doesn't include the entirety of realistic engagement scenarios, and a few layers of rng, although it wouldn't be very practical I admit. Best solution I'd say is to force a headshot + bodyshot combo through damage resistance/output tweaks, 2 bodyshots with nanoweave.

Sure, snipers on engineers would make them a bit easier to counter for newer players, and hopefully reduce the number of hackusations, because you'll actually be able to see a non-cloaked dude as you get clientsided. But unless snipers get weaked to be a bit more skill-dependent and allow for counters, I'd rather we give the engineer access to a wider arsenals of weapons, such as ars. Engineer could also get heavy weapons imo, since it is a more suppressive class after all. Ofc most players will ditch the heavy weps for better weapons as they progress, but it might be useful for newer players.

Yeah it seems it's a problem that's been augmented with the DX11 patch, I don't recall it occurring as much before that. It's frustrating because you can notice the guy while he's still cloaked, and die to him instantly while he's still cloaked and you're trying to duck/evade. You don't lose the engagement because the other player is skilled, or because you messed up; it's something that is beyond your control, as you mentioned. It's also confusing for newer players who usually attribute this to "hacks" or suspicious behavior.

Personally, I'd prefer we try to remove or add significant drawbacks to 1hk weapons, and also tone down the hs multiplier a bit for all weapons, even if it's a minor tweak like 1.85x just so you don't get your head torn off by a cyclone or armistices before you can even react. I still don't understand why they switched the snipers which felt mostly balanced in beta to something that's 1hk. Also the stalker cloak and 1hk knives were a stupid idea, but that's off topic as well.

You solution would take us more to a planetside 1 experience. Simply remove the cloak from snipers. Increasing the cloak/decloak times would be another viable solution, but I'd prefer the first one simply because it would be a bit more uniform, imo.

I would argue that this is actually already the case for all weapons except the snipers. The reason we hipfire in CQC is because doing ADS is a disadvantage most of the time, slowing you down and making aiming for a nearby target harder.
Snipers avoid this because they only need one shot, and that first shot can be aimed using the hipfire ease-of-use and with a quick scope you can then benefit from the ADS accuracy, defeating the entire stimulus->action->reward that is supposed to be in place for snipers. By adding a small randomization in a slightly longer COF or offset crosshair when going ADS you correct this.
A longer duration COF forces the sniper to either risk missing or wait a bit longer before taking the shot, which is a perfect stimulus->action->reward loop for snipers.
The alternative forces the sniper to not aim using his hipfire, but with his ADS. He'll be encouraged to ADS before he aims over the target, fine-tuning his aim after he aims down his sight. This is also perfectly in line with the sniper's design and with a stimulus->action->reward as now the sniper actually has an action and reward for the exact thing he was supposed to do in the first place.

This doesn't seem to make any sense to me.

What you are describing here is that the visual cues of the infiltrator decloaking and taking a shot take longer to arrive than the message that the player is hit and dead. Significantly longer even, since you have the time to effectively detect the difference.

So let's say that there is a 100ms latency. If the Infiltrator comes around the corner and takes a shot. According to you the message that you are hit and dead arrives in less than 100ms, because you can see the infiltrator decloak after you are dead.

I would personally focus on increasing the time between decloaking and firing a shot. Cloak should be about getting in a good position before opening fire, not a cheap way to get into someone's face and then blow their face off with a quick decloak, including if you are a SMG infiltrator.

(keeping everything to just 100ms chunks here...)
I would imagine it goes out like... decloak begin packet, some 100ms delay, firing (and a hit!) packet.
On the receiving end I would EXPECT to see... 100ms delay, then deloak arriving, then 100ms, then death.

It doesn't happen that way.
Unsure if there is some server latency introduced on the cloak but not the fire?
It's almost like the client sends a "decloak completed" (not started) and "fire + hit" at the same time after a 100ms delay.
But yeah I totally see the final stages of the decloak in full glorious color with the ripple effects that you *don't* see when in the black and white thermal vision.

I'm literally seeing them decloak as my body falls to the ground. I would need like negative 100ms of time travel reaction to do something.

Same phenomenon happens with or without infra vision. It's just super obvious with the infravision because of the "snap to color" that occurs on death.

Yeah, this cheap one sucks. If you know they are there and about to decloak you can listen to get the faster reaction and juke..
If I don't know ahead of time, or if the network ain't perfect then it's beyond my ability to dodge. I'm f'n old though. /shrug.

Very well explained. It's something that's related to DX11 I guess, because that's when it started happening, or at least became noticeable from what I can remember. Though I wouldn't be able to tell you why.

Something we got used to, but still very cheesy.

Come to think of it, its very plausible.

One of the most common complaints is "I take 7+ shots and they insta-turn fire 2 or 3 shots and I'm dead!". This happens when the first shots happen just after a packet and your client cant predict the shots happening yet, then when it catches up it looks like fewer shots are fired.

The client wont know to decloak your opponent before it gets that packet. If the infiltrator shoots you in that same packet, you get the hit and death notification simultaneously with the decloak notification, causing the client to start the decloak animation as it starts the death animation.

This is exactly why we should mess with the infiltrator. We need to build in more time for players to deal with the latency and reaction times so they dont feel its unearned. Being able to fire after decloak should at minimum take 1 packet. Also we should make work of ways that break a weapon's intended purpose, like using hipfire to aim and then using ADS COF for the shot to defeat the purposefully build in drawbacks of aiming with hipfire.

I have similar network performance and on Connery getting domed by still-invisible Infiltrators has never been my 'norm'. My experience pre-and post-DX11 is largely unchanged: I almost always have time to react to Infils that get close to me.

This may work for Stalkers, but for Hunters this would fall under the "messing with the player" category you mentioned.
Just make the cloak ineffective at < X# of feet. It's already pretty visible if you know what to look for but you might as well just go all the way with it and make non-Stalkers visible if you get close enough. You get to see the sniper that kills you, and we don't add cumbersome baggage to the game mechanics to band-aid a network issue. Problem solved.

^ here comes the guy with 0ms reaction time, 0ms connection, 0ms processing. Sees 0 issues, brings no fixes.

Let's try to deduce what his playstyle is in another thread maybe

TRSpy, I very intentionally have not addressed you in this thread or indeed any threads on this topic because you have proven multiple times over to be incapable of discussing it like an adult. Every time we do this dance it ends with you getting upset, saying naughty things and getting a vacation, so lets skip to the end and ignore each other until we go away. Capisce?

^
again, do not feed the troll

^
again, do not feed the troll

Sorry man, we're pretty far apart on this facet of the topic. Just to be clear, I'm talking about a infil decloaking somewhere on screen. You're saying you have time to react and move to dodge the decloak-on-fire shot? And "almost always"?

In my mind these are the only possible reasons so help me understand..

1. Misunderstanding
2. Superman
3. Playing a different game
4. Making **** up

Idk man, that sounds like whining about others being good and taking something away from them rather than spending the time to get good at it yourself.
lol

Also, pretty sure the game basically does 2 decloak packets: one to ask the server if they can, and another to say that they are de-cloaking on their side.

You can see the first one when the server isn't responding and you get stuck in cloak forever.

The second one kinda makes sense, as you wouldn't want the infil player to be visible on other people's screens before their own. This would explain why despite absolutely relying on a check to uncloak from the server, infil clientside still exists. If it didn't, low ping players would be seeing infils before they even de-cloaked on their own screens.

I think you pulled something while stretching.

There's probably an ACK involved. It might be encapsulated alongside other data, but yeah something back from the server. speculatively..

1. Shooter hits decloak ->
2. Ack <-
3. (decloak shows on local screen)
4. Shooter fires, shooters client determines hit ->
5. Verified (ack), That player shows dead <-

If you get bad packetloss or jitter right before 2 then it could get you stuck for a moment. If disconnected, probably stuck that way for good, yea.

I don't think that prevents outbound packets for #4 though. Are you prevented from firing when stuck disconnected? I guess would be the key.

If not.. On the server side, when 2 is sent it's then sending a decloak on to the other client. If there is server or network side delays of 2, then it's going to arrive at the same time as 4 on the far client.

In most cases I don't think it matters. The bolter isn't on your screen, or are mixed with other priorities so it wouldn't make a difference. They could be decloaking well after but it doesn't affect the outcome.

The scenario where the bolter is in your face, possibly expected is less common. You should be able to dodge if you're quick (and listening). Instead, you watch them decloak from the death screen. Rare. But it's !@#$ memorable.

idk if clientsiding dudes from the comfort of a cloak with perfect accuracy in one shot that leaves no opportunity for reaction is what I would considered skilled play. (And yes, I have played infil daimyo to get my helmet, and have completed infil and sniper directives).

but to each their own I guess.

Here's the thing though, it would make sense to do it the other way around. Because right now the shoot and decloak basically happen at the same time. There's no warning, it looks weird and is another example of poor reflection from the dev's part. If the decloak shows on the other client before the shooter actually decloaks, it gives the room for a potential reaction from the other side (this won't be the case in most scenarios, but at least the kill won't look weird). Realistically, the shot is still going to hit the target, because of how little delay there is between the decloak and the shot (unless the dude has some super ping and/or godlike relfexes, maybe like kane, in which case it wouldn't change anything either). But yeah it won't make a difference to 99.99% of scenarios, just that the animation will look less bizzare. It's just a more polished way of implementing the mechanic.

that would be more of a 1hk stalker with knife or nightmare ambusher shotgun la scenario, but yeah it leaves a bad impression, especially to people who aren't used to the mechanics/rely on the deathscreen to understand what happened.

The "git good" argument can be used for virtually anything, meaning you can use it as a bad excuse to stop any balancing act.

For example you could easily argue that clipping through the terrain and shooting from its safety is a case others being good because they know to use the "mechanics" of how and when you can clip it. If an argument is so universal you can use it to defend a blatant bug abuse, the argument has no value.

I'm pretty sure this is the first time I hear this situation described the way you do. I've seen many a server hamster die over the years but never have I been truly locked in stealth. Sounds more like a desperate attempt to shift the discussion away from the solutions and on to baseless arguing, just like your "git gud" comment above.


So people, how about we keep throwing solutions at this instead?

So far I think good solutions are:
A longer firing delay between decloak and being able to fire. Skilled players shouldnt have a problem with this right? Or are they going to whine about such a change?
Adding ways to keep the sniper in its intended role so you cant use hipfire to aim and then fire using ADS accuracy, such as slowing the COF settling when going ADS of adding a slight offset. After all, a skilled player should be able to handle that without whining about it right?

Yeah.

To de-cloak/cloak, yeah you need a confirmation for sure. You don't even appear on your own screen to cloak/de-cloak.
I can't remember if you can fire while disconnected, but i'm pretty sure you can.

Basically just "Can I de-cloak?" -> "Yeah, sure" -> "Okay, I de-cloaked and did x,y,z" -> "okay other clients, infil de-cloaked and did x,y,z". Which is why infil can still get clientside even with the de-cloaking delay.

Yeah, I know. That was his response to the crouch thread, just using the same logic to poke fun at how dumb it was (even though I agree with most of what he's saying here about Infil).

Unfortunately you are stuck between a rock and an hard place.

IRL a sniper can remain camouflaged for ages. In PS2 you have no vegetation, monochromatic wastelands, no camo and on top of everything, a Q-spot system in which you can "guess" where enemies are. So to make snipers a viable option PS2 designers envisioned the cloak: short duration cloak with long-range weapons, long duration cloak for short-range weapons both with a "audio/video de-cloak effect before firing" built-in balance (which fails, like many PS2 designers choices).

There is no guarantee that the audio/video cue is available before the victim is damaged. You have a race between two events:

Event 1:
decloak takes place on attacker PC (250 ms or 300 ms? I do not remember anymore)
info reaches the defender PC (make it 100 ms, just an example)
audio/video cue is now available according to frame rate and processing delays on defender PC (make it 25 ms)
reaction (make it 200ms? That's a rather good reaction time)
TOTAL = 100+25+200 = 325 ms

Event 2:
decloak takes place on attacker PC (250 ms or 300 ms)
attacker fires, bullet travels and hits (make it 500 m/s speed, target at 100m = 200ms travel time)
TOTAL = 250+200 = 450 ms

Unless the target is far away the attacker will always win... And IF the target is far away guess what? The defender will not have any audio/video cue and get hit unless he/she is always moving in an unpredictable pattern.

There is a reason why the "snapshot snipers at extremely close range" are the deadliest ones, there is zero bullet time.