Discussion on Reevaluating Friendly-Fire/Decons for GrangerPub

Servers GrangerPub Servers
21

No, his solution is simpler: Keep TK 100% for scrims, disable or nerf heavily in public server.

TOP KEK

Bleed =/= TK and reflection would only be effective if it was applied to bleeding and not the TK itself, but you’ll also fuck I can’t even count how many players with globalized reflect.

22

We can try that but expect people joining Club to play and not to scrim just because of that, or at least that’s how I expect it.

Edit: [quote=“MaeJong, post:21, topic:1934”]
Bleed =/= TK and reflection would only be effective if it was applied to bleeding and not the TK itself,
[/quote]

I said that reflect should be applied in both cases. And yes, the players learn and get more careful with time, because teamkilling would be a stupid option if you just kill yourself first.

23

I don’t have a inherent problem against a command like this but, it would have to go further. Smart people can try to get themselves “punished” and gain an advantage by spamming Luci for their team.

IMO the command could work by also forcing them to /share the rest of their credits.

With all due respect Woodbury, whatever you just posted are not my words. I’m saying TK should be off OR modified on GrangerPub and stay on for scrims in GrangerClub. I agree, TK is a great thing to have, but only in competitive environments where you personally know everyone on your team. TK in pub matches are a different beast IMO.

24

No one is gonna get out of base ever again or even shoot because when people suck they just suck.

TK is a great thing to have in general when your team isn’t filled with brazilians, which used to be the case until like 2 years ago. It was already bad then but now it’s like fucking hell you can’t get out of base with over 70 hps. I still love TK and would like to find an option that doesn’t fuck everyone up directly but rather isolates whoever is bleeding too much.

Some QVMs have a vote that allows the players to decide what teamkill (structures included) should be (anywhere between 0 and 100%), that could perhaps be the better option.

25

Sorry @Hendrich, I didn’t mean to upset you, it’s true you didn’t say that. I just experienced what you are suggesting before and ended bad for the server, maybe it wasn’t the way to say it from me. I just repeat one thing, changing the teamdamage behavior will change the gameplay entirely, so grangerpub won’t be grangerpub anymore. Thats the only thing I’m worried about.

26

nah its cool bae i’m just saying its not what i actually said. trying to be 100% clear and nuanced in a online discussion will just force everyone to post 3000 word essays to each other anyways. :sunglasses:

In that case, its probably the better solution to change how the current TK works. Or keep it the same and (based on Maejong’s idea) create a !neuter command players can vote for against a TKer where they can no longer TK/bleed, but are forced to /share all their current credits/evos and continue to do so until a match ends.

27

It could be just a percentage (25-50% is a lot already) rather than throwing them in lateral security position straight away. Or make them share based on the damage they should’ve been doing if their TK was on, so they don’t push people around with lucis and shit like that and it doesn’t become either a clear dis/advantage.

28

I think it would work better this way:

/callvote neuter (idk what a better name for this would be) hendrich being a faggot and spamming nades everytime he spawns (requires 60% approval)

When it passed, it removes TK/bleed capabilities of the player but it forces them to share all their credits/evos AT THAT TIME

every 7 minutes afterwards, 45% (rounded based on credits/evos) of whatever they earned is shared among the team and this continues until the match is over

29

I’m open to test the teamkilling behavior, and I just can share other games experiences, game engine related but not actually trem, what I propose of the 1.x damage reflection is what I have seen work before, and obviouly disabling TK is what I have seen not working before, maybe trem is different is that aspect, test is needed.

In the case of trem, and taking in consideration several factors that make trem unique I think It wouldn’t work a lot right now, because griefers can cause indirect detriment to the team, for example by trying to teamkill a friendly, dying by reflect and respawning, makes it hard for admins to find the griefer and also makes the respawn waiting queue longer for other players. So I’ll trash my own suggestion because trem is more complex than that solution.

Btw, we need to avoid exploitaishon of the !neuter command/vote, since we can have a not lovable player in a team and his own team votes for taking his evo/credits out just for the lolz, since is the last tendency.

I just foresee it…

X player - 69 kills
Y Players - 3 kills
Y player: Hey share something
X player: why should I, you suck
Y player called a vote: !neuter x player
Vote passed (21-1)
X player disconnect.

I thinking about a new one, how about auto-forcespecting someone who reached a teamdamage limit except for lucis and nades, and ill think about those 2 cases later.

30

I agree.

The problem with this is figuring out what that “damage sweespot” threshold would be, based on the weapon used and the player’s behavior during the match. An “automatic” solution is going to require alot of testing.

31

I agree it can be difficult to config, but trusting the people vote is harder. Yeah there are mute and kick votes, but they last 300 secs and 120 secs respectively, the neuter one have a whole game lasting, so It is very dangerous. But I do agree with an admin only command, since responsible human intervention makes it different in that case.

All this suggestion needs testing because trem is so unique that trying to bring other games solutions and make them fit is hard, but all problems have solutions, only thinking a lot solves them. By the way, since 1.3 is the future right now, maybe some karma usage is not out of the question, only players with 0 karma are forcespected if they teambleed/teamkill to the limit.

32

Hi, expert griefer here, 10 years of experience, here to talk about the cons of removing friendly fire for both structures and teammates.

HUGE problems with removing FF damage for structures:

  • When the enemies rush with a dozen Tyrants all at once, a single player can defend the entire base by dropping grenades in their own base. This will not hurt the base (since friendly fire for structures would be turned off) and will output thousands of damage over a matter of seconds.
  • Alien structures which block teammates would not be killable. A griefer (or just any retarded builder) could now build barricades or other buildables in strategical locations to block access to parts of the map. For example, putting the Hovel in a bad spot would prevent any Tyrants from getting to hall on ATCS.
  • Feeder eggs would be invincible. A granger could run to human base, plop an egg in front of the turrets, and leave it there, and no alien could deal with it.
  • Human builders could place armouries all around a telenode, blocking in stage 1 humans. Alien builders could do the same kind of thing in a number of ways on almost every map (barricade or OM in front of door of window room on Niveus, etc.).
  • Defending human base with the highest DPS weapons would suddenly become viable. Flamer does 100 DPS, Prifle does 90 DPS, and Lcannon does up to 265 damage in a single shot. (Psaw does 200 DPS, but that’s beside the point.) These three weapons do a ridiculous amount of damage, and allowing them to freely defend base without any repercussions would make camping and defending against rushes WAY stronger. The game is fairly balanced, and most people complain about camping, so making camping BETTER sounds like a bad idea.

Problems with removing player FF damage:

  • Luci and Prifle spam from camping humans into hall on ATCS becomes a GOOD idea, not a terrible one.
  • Just in general, this lowers the skill ceiling of the game: good players are way less likely to hit teammates with Luci (unless the teammates are really bad and just run in the way) while bad players are frequently found dealing large amounts of damage to teammates, maybe even more than to enemies. This also lowers the skill ceiling of Prifle, which requires accurate prediction and pre-firing along alien movement paths; not needing to worry about hurting teammates would make the weapon way stronger.
  • Teammates could “luci-jump” off of each other’s grenades and luci shots without taking damage.

Problems with “threshold” FF damage (i.e. can’t lower a teammate to below 15 HP):

  • Intentional bleeding becomes way more conspicuous because they will never accidentally teamkill someone they just want to bleed.
  • You can bait aliens onto a teammate’s grenade, survive the explosion, and make a bunch of aliens die.
  • Bleeding a rant down to 15 HP is a lot worse than bleeding a basi down to 15 HP.

Problems with “percentage blocked” FF damage (i.e. only take 50% damage from friendly fire):

  • Human armor would affect this amount, and so nude humans would still die in one hit to Luci/nades.
  • Not a lot else. I like this idea the most. CS:GO has this:

    To explain, those default values are for the casual game mode, which has no friendly fire (except by your own grenades). Competitive has them somewhere above 0.5 but below 0.85 I think (so FF deals 50% to 85% of what it would deal to an enemy). Even in the pro scene, you don’t do 100% damage to teammates. You can ignore the first command since bullets in Tremulous never go through enemies or teammates or walls, and the bottom command is primarily (or maybe exclusively) for controlling damage to teammates with the knife (also not relevant in Tremulous).

Problems with “backfire” FF damage (i.e. taking some amount of damage when you hurt a teammate):

  • Doing even a tiny amount of damage to teammates would put a helmeted human low enough to die in two hits to goon. (Currently, taking even a tiny amount of damage does this, since two goon chomps is 96 damage to a helmet.)
  • Grenades would be detrimental to buy, since a griefer/troll/anyone who hates you could just go stand on your grenade and make you die. The same thing could happen with Luci.
  • A teammate walking in front of your Adv. Goon barb suddenly cuts your health by a massive amount (at 50%, you take 55 damage in return; at 120%, you take 132 damage in return, more than half of the Adv. Goon’s health).
  • TKing a Basi already makes you lose an evo; you shouldn’t also lose 40-100 health for doing it.

Problems with !neuter command or vote:

  • Vote used against people who are hogging evos or camping in base with chainsuit or whatever and not canceled by an admin. (See Woodbury’s example here.)
  • Half of the player base doesn’t even know how (or just doesn’t care enough) to vote on team votes.
  • Abusable by players who want to luci and make enough money to not care about losing some; they could use it to never worry about TKing again.

Problems with ANY “automatic” solution:

  • As long as griefers and trolls can figure out the details of the system, they can easily abuse it to its maximum limits (if 250 damage in 30 seconds triggers it, they make a teammate do that much damage in that length of time, or they do only 240 damage then wait 30 seconds).
  • It will always be possible for it to hit “innocent” players (bad players who bleed a bit too much, good players with bad teammates who get unlucky with Luci, etc.).
  • There aren’t enough griefers and trolls who are intentionally bleeding to warrant adding a system to prevent intentional bleeding which also causes problems for people who unintentionally bleed.

So, I know that’s a lot to read, but all-in-all, what it boils down to is there are inherent problems with almost every plan, but the best course of action is probably just making friendly fire only do 50%-80% damage (and it can be modified as necessary). I know I would LOVE if Rants only did 50 damage to my poor little Basilisk, and Luci shouldn’t (I think) kill light armour humans with halved damage. (I’m not 100% sure how much damage armor protects from since I try to avoid hitting anything that isn’t the enemy’s head). I definitely disagree with removing buildable friendly fire, for the many vital reasons I listed in my first list of cons in this post.

It is possible that removing friendly fire from certain weapons would end up alright (rifle, shotgun, MD, and painsaw to name a few), but I feel like that is way more in depth than necessary for such a simple and negligible problem. I honestly don’t think friendly fire is a problem at all currently, although it can be annoying at times, especially if the perpetrator refuses to refund you (RIP baslick).

Note that this post is entirely based upon current 1.1 gameplay, and has nothing to do with Tremulous 1.3. These are reasons why various solutions would be detrimental to the current gameplay on GrangerPub.

4 Likes
33

Really a bigger issue is maliciously deconning your own team’s base. While it can be annoying, most of the time tking/bleeding doesn’t have a major impact on if your team loses a match, but maliciously deconning can result in you losing the match immediately. Where tk of buildables come in with that case is when the malicious deconners get denybuild and resort to tking their team buildables, however, buildable tking should still not be disabled for the reasons Menace gave above.

What I’m thinking should be done for tking/bleeding and malicious deconning is a semiautomatic approach that includes karma. When you die have a menu popup asking if you forgive specific teammates that contributed towards your death. When certain thresholds are reached to indicate the likelihood of griefing, and have a vote automatically called asking if a specific player has been griefing, depending on the situation either just put the player in question in a “timeout” during the vote, or even pause the game (if it is a malicous decon) during the vote, and if that vote passes, forcespec that player and reduce their karma (also if it is a malicous decon revert).

But I think that %FF is a good idea to try out.

34

I 100% agree. Basenading is much less common than simply deconning since you have to build up money to do it and there has to be an open spot on the human team and they have to be S2 before you can do it. Deconning can be done at any point in time during a game, without any prerequisites. I think the conversation flew far off track into the topic of friendly fire when deconning is the more major problem.

One possible option would be give low-level admins (who otherwise could not revert things; maybe even L2 admins) a special version of !decon which, instead of automatically doing the anti-decon commands, starts a vote which:

  1. Pauses the game.
  2. Upon passing, does the anti-decon commands (I’m not 100% sure what !decon does in the first place).
  3. Unpauses the game.

An automated anti-decon structure that can be prompted by votes would work well when admins are not around or are AFK or are too busy playing on the other team to handle it in time. This way, a player who is trusted with a command which can pause the game could potentially react much sooner and prevent a decon from ending the game.

My problem with this is it slows the game down for that person. When I die, I’m usually already spamming my “spawn as Dretch” bind to minimize downtime and get in the spawn queue as fast as possible, and this would force me to slow down and deal with a menu popping up. As it is, the only time I EVER use an in-game menu in Tremulous is when I’m building (or, rarely, when joining a team as a spectator, but I usually use console for that). Furthermore, players who frequently run back to base may end up having taken over 100 damage total split between half of their teammates over the course of one life. What if you took a lot of damage from a teammate that made you use your medkit, then died because you couldn’t use your medkit to stop the poison? Just in general, I don’t think you can make a good automated system for malicious vs. accidental bleeding.

1 Like
35

Since gpp at least, the damage from each player to anything that can take damage is tracked in the code, but resets if that entity gets back full health (or if it dies).

The current way the menus work by conventional mouse cursor point and click is inadequate for a fast past game such as trem, there are much better ways to handle input for User Interfaces that needs to be implemented. We already have a slight improvement with the menus on test7341 where you can use the scroll wheel to change highlighted selections, but still more improvements are needed. The goal is that you should be able to use the menus at least almost as quickly and easily as using binds.

36

This is me bringing this thread back from recent basenade reports.

Can we all atleast concede that grenade structure FF shouldn’t be around at this point?

:green_apple: :granger: :green_apple: :granger: :green_apple: :granger: :green_apple:

Still happens tho.

37

Just use vstrs and make yourself some binds niqqa

https://knightsofreason.net/forums/viewtopic.php?f=114&t=2249

38

Then nade spam at the base would be a viable strategy when rushes occur. Would it not be better to not allow nades to be dropped when you are within X distance from friendly structures (at least not without certain permissions such as a minimum trusted level) ?

39

long time ago, probly on AU, for short time there was a mod that warned you if you tk and kicked you if you continue. you couldn’t kill om because it kicked you before. i think it counted dps, bleeding time and target and its hp. sometimes it was annoying because you couldn’t move base by killing it alone, but it could be disabled for lvl2+

basically some autokick for unknown players can work. it will make learning curve even steeper for new players but to some degree will prevent desalas from ruining the game

3 Likes
40

Considering how right now the meta allows good players to use their excess evos to go in as Kamikaze Tyrants, I wouldn’t see a problem with this unless we have 12+ players all constantly buying grenades and dumping them every rush. TK seems to get harder to balance when more players are on each team (especially at around the 32 player limit).

That could be a solution, but then it would have to be low enough so that it doesn’t overlap with aliens building close to human bases.

If only QVMs were shared more proactively beforehand. :frowning: