• Please make sure you are familiar with the forum rules. You can find them here: https://forums.tripwireinteractive.com/index.php?threads/forum-rules.2334636/

Culling the final specimens in a wave

Slappy Cromwell

Grizzled Veteran
Jun 16, 2009
441
175
I suggest a system that automatically begins to cull the final few specimens in a wave after a certain time period has passed while playing in multiplayer mode. I believe out-of-view specimens do get culled in solo mode, but multiplayer requires every single specimen to be killed no matter what. That has the tendency to lead to this unfortunate result:


The survivors can't see the last specimen. The last specimen can't see the survivors. Neither party has any possible avenue of reaching or damaging each other, thus the wave is indefinitely prolonged. In effect, the game is quite over and the only options the players have are to vote for another map or to leave. That is all because of one stuck specimen.

Yes, this is technically the fault of the mapper. However, it is incredibly unfair and cruel to the map author and the players (if not poor game design) to allow such a minor and unreproducible oversight to ruin an entire game. More often than not, these incidents are one in a million, hence why thousands of specimens can have gone through the exact same scenario before and never trapped themselves. In this particular case, that hapless Scrake only managed to get stuck there due to a combination of an amazingly small gap in the blocking brush, a barrage of grenades, a nearby Fleshpound, and a peculiar but unfortunately timed jump on his part.

Basically, the problem is that a few specimens at the end of a wave can literally halt the entire game and there's no system, preferably a time limit, that can cull them out. I've pointed out an extreme example where the game cannot possibly continue. However, this problem can have lesser occurrences that don't necessarily require a custom map or a stuck specimen. It could be the last Crawler who decides to take on a 100% welded door by itself on the other side of West London. It could be the last Bloat hobbling along somewhere in Farm. It could be the last Clot just aimlessly wandering around in Foundry like a lost child. All of these instances require quite a bit of time to assess and resolve. This destroys the game's momentum and essentially coerces the players into a game of hide and seek that's guaranteed to be dull (Unless the last couple of specimens happen to be angry Fleshpounds ready to fly down from the sky, but more likely it's just a Clot/Crawler). Also, it's no fun going on these searches when there's the nagging thought that the final specimen may be permanently stuck out of reach, thus forcing you to end your game right there. This has happened to my team more than once on several different levels, and I can tell you it's a fairly unpleasant experience watching everyone's hope slowly wane away as everyone slowly begins to realize there's nothing we can do.

As I said before, I believe solo mode already has a system that culls specimens, but in my experience it has proven to be a bit too stringent for multiplayer. I would suggest using a similar culling method in multiplayer only once there are fewer than ten specimens left and after 30 seconds have passed. If it takes longer than half a minute to kill the final ten specimens, either the problem I described has occurred or someone is taking far too much time kiting. This isn't a perfect solution, so I'll leave the specifics up to discussion. Nonetheless, I believe this to be a considerable issue in the game that needs to be addressed.
 
This has never happened to me
Then you haven't played enough.

it doesn't seem too big a problem to warrant a fix.
Stopping the entire game dead in its tracks isn't a big problem? With that logic, the program may as well just save us the time and crash to the desktop.

A 30 second restart once in a blue moon isn't too problematic.
And what, just pretend we didn't waste all that time making the progress we did? Just restart everything and hope every second that it doesn't happen again and force us back to square one for the nth time?
 
Upvote 0
Heh, believe me when I say I've played enough. I've even played the map in your screenshot many times. I'm not saying it's not a problem, I'm just saying there's other more pressing issues that deserve more attention than this. You aren't really losing progress unless you're shooting for the steam achievements. Even then it would only matter on the stock maps.
 
Upvote 0
Heresy ftw! I loved that map in the mod days!
We had a session where three of us lined up with axes and typed into the chat:
Hold it!
Hold it!
Hold it!
CHAAARGE!!!
and then we stormed into battle.:D

I'm for what is being suggested. I also think it's a rare but, when it occurs, fundamental flaw in the game and I also had moments where this ruined games.

However we might need to lay out more clearly defined rules for when this auto-killing should be called into action. For example, there used to be a map around that had a bunker with bit of a trench-like area in front of it and a huge area where the player couldn't get to around it. The specimens where spawned out of sight and approached the trench over an open field then. In the later waves players would get pushed back into the bunker and it made for some fun fights BUT. The map had a huge problem with specimens getting stuck outside so they wouldn't even come into view! This wasn't just a lone clot that got stuck in some freak-accident either! this was common place! The auto-kill mechanism maybe shouldn't act here. Of course the player isn't at fault in this case either, but killing off more than half a wave worth of specimens seems a little extreme to me.

Another example where the map isn't flawed: BioticsLab. What if you barricade yourself into the room with the two sliding doors and the trader room to one side? If enough people help welding the doors they might be able to lock the specimens out long enough to call the auto-killing into action and thus avoid fighting them altogether.

I'm definitely on your side on this issue, but maybe we can define this more clearly instead of leaving all the fine tuning to TWI.
 
Upvote 0
I swear, there USED TO be something that did this in the game. But people were taking advantage of it. This was the pre-gameplay-revamp patch, mind you, so Fleshpounds didn't rage if kited too much, so... Folks would dodge all the fleshpounds and kill the other specimens until it was just fleshpounds.

Then, as when only the fleshpounds remained, since there were less than 10 specimens, the specimens would slowly die, so they'd continue to kite the fleshies until they died. This auto-kill feature seems to have been removed at the same time they gave fleshpounds the anti-kiting ability to random rage, so I'd guess it's removal period was an anti-kiting method.

That said, something does need to be done. On Farm, we had a Stalker stuck on a fence somewhere. So we had to spend upwards of 10 minutes combing the map to find her...it really kills the feeling of immersion and breaks the flow of the game...
 
Upvote 0
but killing off more than half a wave worth of specimens seems a little extreme to me.
Which is why I suggest that culling only begins once there are fewer than ten specimens left. Hell, that number can go lower, but I have seen more than one specimen get stuck before.

Realistically speaking, the very final specimens aren't likely to make a huge difference unless they're powerful and damaging, namely Scrakes or Fleshpounds. Fleshpounds will rage on their own at some point, but Scrakes will just continue to tread along at a snail's pace unless the players actually decide to confront him. That's a flaw with the Scrake, in my opinion, and not really a problem relevant to a culling system.

What if you barricade yourself into the room with the two sliding doors and the trader room to one side? If enough people help welding the doors they might be able to lock the specimens out long enough to call the auto-killing into action and thus avoid fighting them altogether.
They'd simply be collateral damage, unfortunately. If a culling system were implemented in multiplayer, it should be foolproof (Otherwise, what's the point?), but it will likely result in a few false positives here and there. However, that's a small price to pay in the long run. But again, what I'm proposing would only result in ten or fewer specimens at most getting removed.

On sort of a side not, I would also consider the door welding thing to be another issue in of itself: strong specimens should be able to destroy a door faster than players can weld, but that's simply not true. Any decently leveled support can weld one door to infinity and only a truly collective effort on the specimens' part could manage to overcome his repairing.

I'm pretty sure the game already does that to some extend
If it does, it's clearly not working.

and there is no need to fix what isn't broken.
If it's not working, that would mean it's broken. More probably, though, is that it's simply not there.

What you showed us is a problem with a specific map, not the game itself.
It's certainly a problem with both. If one specimen has the power to hold up, if not outright ruin, an entire game, then you have yourself a massive flaw in the game design.
 
Last edited:
Upvote 0
Which is why I suggest that culling only begins once there are fewer than ten specimens left. Hell, that number can go lower, but I have seen more than one specimen get stuck before.
As have I, but if you play on beginner the 1st wave consists of no more than 14 specimens! Kill four and then weld yourself into a room to have the rest killed automatically?
And what if you play on a full server on suicidal? More specimens means more specimens have a chance to get stuck. Maybe ten isn't even enough then?
It's just not as easy.

UT2004 has a handy function for bots that get stuck. They have an aneurysm, ie commit suicide so they can respawn. An extended form of this could help KF in more than one way. Read this suggestion:
http://forums.tripwireinteractive.com/showthread.php?t=34674&highlight=respawn

Not only would it work as suggested over there, it would also solve the problem of specimens getting stuck.
 
Upvote 0
I've never seen more than 1 or 2 specimens get stuck on the default maps, although the custom map 'escape to heaven' has a fun quirk where a full 30 or so specimens get stuck in the trader room (makes for a fun couple of nades). I propose set the max kill limit to 3 and kill the specimens if they haven't moved for 10 seconds.

An alternative would be to run a random-move script that activates if the specimen doesn't move for 10 seconds. If zombies are bunched up behind a welded door, they won't be able to move, so gameplay won't be affected in that regard. If they get stuck on the map somewhere for 10 seconds, the script could cause them to turn around and move 10 feet in the opposite direction for example, and then resume their normal human targeting AI. This would help most zombies get unstuck, but there might be the occasional fluke whereby the zombie auto-kill could take place at the end. OR, if a specimen is stuck for 30 seconds (i.e. fails to get unstuck after a couple random-move attempts), then auto-kill or respawn. This solution would virtually eliminate zombie hunting altogether.
 
Upvote 0