TF2 megatask: making capture the flag popular

PHISH

L1: Registered
Sep 14, 2012
41
21
First, thanks for having me at your forum.

As a long time TF player and once upon a time TF mod project lead I’ve never been satisfied with the way CTF (capture the flag) plays out in TF2. Even in prior fortress games CTF has always suffered from some drawbacks in public play.

Current CTF issues:
  • Lack of clarity for roles:
    With dual goals of protecting your flag while also needing to get the enemy’s flag a typical public game will devolve into 2 or 3 people going for the enemy flag (or none at all), with the remaining either defending the base or wandering aimlessly. An individual player’s role is undefined, and so they often play “undefined”.
  • Spawn and travel times:
    Players going for the enemy intelligence already have to cope with significant travel time just to reach the enemy base, while defenders have spawn’s a few steps away from key defensive areas. In prior games the increased movement and concussion grenades could negate much of the defense’s proximity advantages. TF2 lacking concussion grenades and with increased spawn times shared between offense players and defenders tilts the game even farther against “offense”. Noone wants to wait 15 seconds to respawn only to have to travel 15 seconds to reach the enemy base against 10 defenders, die by yourself and repeat the process. Together with the lack of “Rambo” capability for most classes this results in a lot more frustration for the few players trying to work against a 10+ man defense.
How to fix it?
  • Clarify roles:
    Similar to gravelpit, one team should be offense, one team should be defense at a time. Every player on each team is aligned to the same goal for the round. Offense cap as many flags as possible, defense protect the flag at all costs. At the end of a 15 minute round teams would switch sides. Winner would be the team with the most captures after both 15 minute rounds.
  • Adjust spawn times and locations:
    With roles defined per team you can tweak offensive spawns (blue) and defensive spawns (red) individually. Offensive spawn times should be near instant, defensive spawn times could be slightly longer depending on spawn locations. This can improve balance and increase the pace of action. You also remove the need for mirrored maps, saving on map sizes and resource budgets. Existing CTF maps could be modified to work with this mode with minimal pain (I hope).
How to implement it?
That’s where you guys come in :p.
  • Game Logic:
    I’ve been diving into the logic system and it seems like there a few hurdles to overcome. (With my limited understanding, anyway)

    The CTF game mode entity defaults to 3 captures per round and for whatever reason valve hasn’t included a way to adjust that per map without sending a server command (potentially messing up any subsequent maps in the rotation). So as of now I only know that you could set tf_flag_caps_per_round 0 to turn off the auto win on X caps.

    Second would be to add team_round_timer and game_round_win? I haven’t figured out exactly how to set it up yet to properly trigger the switch after the time ends and preserve the caps.

    Finally the victor would need to be based on the total captures from each team at the end of the round. Seems like this gets to be a problem when you factor in the teamswitch, but I’m not sure.
---
Anyway I really appreciate any feedback or help on this. I’m hoping there’s some TF vets out there with an interest in this idea. It basically mirrors the OvD format that became common in TFC for pickup games in later years.

Tl;DR version: Idea is to implement a modified CTF version which is played offense vs defense for 15 mins, then alternating rounds, adding up the offensive captures at the end to determine the winner. Trying to nail down the proper game logic/ determine if this is feasible using current entities.
 

tyler

aa
Sep 11, 2013
5,102
4,621
Have you ever played invade CTF (neutral flag, deliver to enemy base) or A/D CTF in TF2 (like A/D CP, but with a flag)? Both of those, I think, are better than what you propose, though A/D CTF is kind of dumb.
 

Pocket

Half a Lambert is better than one.
aa
Nov 14, 2009
4,694
2,579
Have we ever actually played A/D CTF that wasn't invade?
 

tyler

aa
Sep 11, 2013
5,102
4,621
A/D CTF is invade by definition, but I didn't know what to call the other one. Football, I guess.
 

GPuzzle

L9: Fashionable Member
Feb 27, 2012
638
414
And someone has already done that, it is called ctf_haarp. And guess what? It is preety good.
But CTF primary goal is to allow people to keep an eye at both intels. It is to go offensive, and be defensive, just like 5CP. Why do most CTF maps suck in terms of layout and gameplay (Turbine's layout is so good it is used in competitive TF2) ?
There is one reason. In most maps, teams have to go through the enemy spawn to get the intel. Turbine has three ways in to the intel, and one that is safe, one that is risky, and one that is perfect for an action scene.
 

Dr. Spud

Grossly Incandescent
aa
Mar 23, 2009
880
854
CTF Landfall was all about fixing these very problems.

-It has the spawns behind the flag (mimicking CP maps) so that players always keep running forward, and they both defend their flag and attack the enemy without needing to commit to one or the other. There is no "lack of clarity for roles", because everyone can push forward, like a CP map.

The flags should be the center of attention as much as possible, like the objectives in PL and CP are, so:

-The distances between the flags are shorter and the map is wider, so that flags are picked up more often.

-The flags aren't in easily-locked-down rooms.

-The layout is made to keep a battle at mid at all times, which the couriers will have to run through, so most players in the map see the flag before it can be captured, and the return trip is just as difficult as the initial pick-up for the courier.


Whether or not you think Landfall was ultimately a good map, the point is that these problems can be fixed through better layouts. You don't need a wacky new gamemode. CTF isn't broken. The official CTF maps just suck compared to the CP and PL maps.
 
Last edited:

fubarFX

The "raw" in "nodraw"
aa
Jun 1, 2009
1,720
1,978
ctf_cesspool was a very good attempt at clarifying those goals but it never quite came about.

it was mixed with some koth logic, if you had more flags, you could just defend and wait for the enemy timer to run out or take the risk and get more flags. spawn times were also affected and all that good stuff but hey, sometimes maps don't work out.
 
Last edited:

PHISH

L1: Registered
Sep 14, 2012
41
21
A/D w/ flags (original dustbowl) is not remotely similar to traditional CTF gameplay so it's a mistake to equate the two.

Sadly it doesn't sound like you guys ever have experienced the classic TF CTF gameplay in organized teams, so I think you are underestimating the many elements that made that game unique and exciting (and completely different) than say dustbowl A/D.

TF2 punishes death highly, and due to the mechanics of the medic and uber it drastically changes how CTF plays at the top level. Instead of layered defense and a fast paced offensive attack it becomes a very conservative push and hold game with one ginormous blob. If it's a push and hold game it might as well be CP, because that's what the mode is suited for and that's how it plays.

What the mode I'm proposing would do is open up the door for more aggressive play on offense (because they don't have to worry about the other team steamrolling to their flag if they fail) as well as bring back some of the CTF excitement of flag pulls...most importantly it helps improve the pub game experience.

There are other elements that could be improved because a lot of the CTF maps in TF2 are in cramped low ceiling flag rooms leaving little room for finesse flag grab attempts, but those could come with time.
 

Dr. Spud

Grossly Incandescent
aa
Mar 23, 2009
880
854
I get your intent behind this new gamemode, but it's got some flaws. I don't think the mode you're proposing really benefits from involving flags in the first place. It's simply taking turns attacking and defending, which plays better in a non-CTF map with CPs.

In your proposed mode, if one team is only on defense that means they need to turtle and let the attackers come to them. So if the attackers do get the flag, their return trip is meaningless - they'll just run back to their base through an empty map. Meanwhile everyone else is twiddling their thumbs with no objective to complete. That's a big problem, and one that is easily fixed by replacing the flag with a CP to capture.

This would play better in a map specifically designed for A/D, rather than putting this into a CTF map. Take your ideas to the natural conclusion and you've just got an A/D CP map.
 
Last edited:

PHISH

L1: Registered
Sep 14, 2012
41
21
I get your intent behind this new gamemode, but it's got some flaws. I don't think the mode you're proposing really benefits from involving flags in the first place. It's simply taking turns attacking and defending, which plays better in a non-CTF map with CPs.

In your proposed mode, if one team is only on defense that means they need to turtle and let the attackers come to them. So if the attackers do get the flag, their return trip is meaningless - they'll just run back to their base through an empty map. Meanwhile everyone else is twiddling their thumbs with no objective to complete. That's a big problem, and one that is easily fixed by replacing the flag with a CP to capture.

I'm not sure how you came to these conclusions. Why are they taking the flag out through an empty map? There won't be mirrored bases and the capture point ideally would be located in the traditional midfield area a bit away from offensive spawns on something like a raised platform. It's expected the defense is going to contest the flag outside of their fortress all the way to the capture point if they choose to, but if they do not they are resetting their defense. This is how CTF has been played for over a decade and there doesn't exist the issues that you assume.

There is never a time where you are twiddling your thumbs on defense. If one person is capping, the rest of the offensive team is attempting to get a relay cap i.e. they are already in your base ready to pull a second flag the moment it is captured. There's no reason they would arbitrarily stop attacking for even a moment.

The big problem you describe doesn't exist.
 
Last edited:

GPuzzle

L9: Fashionable Member
Feb 27, 2012
638
414
He desinged basically your map.
He knows what he's talking about.
 

Dr. Spud

Grossly Incandescent
aa
Mar 23, 2009
880
854
Uh, that's a whole lot of new information that wasn't in your original post. I did go back and caught the part where you said the maps wouldn't be mirrored, but still.

But the problem is a fundamental one, and the way the map is setup doesn't change it. Defenders are turtling, and the attackers are going to take the flag out of the base with no defenders between them and the capture zone. There's no way the defenders (after a lot of them were just killed, since their flag was just stolen) are going to be able to get outside and stop the courier. The most interesting and meaningful part of the flag mechanic is the fact that a courier has to transport it, and that's the one part that doesn't matter in this proposed mode.

This is why existing one flag CTF maps are always invasion: you've got an enemy team in between the courier and his destination.

edit: I'm not trying to squash your enthusiasm for making a new mode, but I think you'll need to rethink this one before it's worth pursuing.
 
Last edited:

GPuzzle

L9: Fashionable Member
Feb 27, 2012
638
414
Uh, that's a whole lot of new information that wasn't in your original post. I did go back and caught the part where you said the maps wouldn't be mirrored, but still.

But the problem is a fundamental one, and the way the map is setup doesn't change it. Defenders are turtling, and the attackers are going to take the flag out of the base with no defenders between them and the capture zone. There's no way the defenders (after a lot of them were just killed, since their flag was just stolen) are going to be able to get outside and stop the courier. The most interesting and meaningful part of the flag mechanic is the fact that a courier has to transport it, and that's the one part that doesn't matter in this proposed mode.

This is why existing one flag CTF maps are always invasion: you've got an enemy team in between the courier and his destination.

edit: I'm not trying to squash your enthusiasm for making a new mode, but I think you'll need to rethink this one before it's worth pursuing.

Yeah, that's why Haarp is A/D CTF and it doesn't suck.
 

PHISH

L1: Registered
Sep 14, 2012
41
21
Uh, that's a whole lot of new information that wasn't in your original post. I did go back and caught the part where you said the maps wouldn't be mirrored, but still.

But the problem is a fundamental one, and the way the map is setup doesn't change it. Defenders are turtling, and the attackers are going to take the flag out of the base with no defenders between them and the capture zone. There's no way the defenders (after a lot of them were just killed, since their flag was just stolen) are going to be able to get outside and stop the courier. The most interesting and meaningful part of the flag mechanic is the fact that a courier has to transport it, and that's the one part that doesn't matter in this proposed mode.

This is why existing one flag CTF maps are always invasion: you've got an enemy team in between the courier and his destination.

edit: I'm not trying to squash your enthusiasm for making a new mode, but I think you'll need to rethink this one before it's worth pursuing.

I don't want to come off as some snarky bulldog pushing this idea either, but I do want to respond to criticism that I think is a bit off base. Perhaps it will help me explain myself better. The assumptions you are making about optimal CTF defensive strategy and how the defenders would play the map I think are a little off.

Assuming the defending team is going to cluster deep inside their base and fail to meet the offensive team at the periphery of their base is unnatural. A layered defense with forward holds at choke points is typical to CTF. A layered defense means that supposing the outside layer is passed by some offensive players, they then are dealing with flag room defenders. The outer defenders will respawn before it is possible for the offensive players to clear out the flagroom defenses and exit with the flag. If the defense is choosing to pack everyone in the actual flag room then they are playing suboptimally and they deserve to lose the flag in one fell swoop. That is a rare occurance to get a "coast to coast" and I'm not sure why you think it would suddenly be the norm. It would be even less common once you take all those people that mindlessly dm in the midfield and have geared to protecting their flag.

People naturally try to seek out the other team so I'm not sure why this would suddenly go against all player tendencies and cause everyone to "turtle" away as you suggest. I can't think of a single example of this ever happening in any ctf game I've ever played, many of which mirror this mode.
 

Fruity Snacks

Creator of blackholes & memes. Destroyer of forums
aa
Sep 5, 2010
6,394
5,571
People naturally try to seek out the other team so I'm not sure why this would suddenly go against all player tendencies and cause everyone to "turtle" away as you suggest. I can't think of a single example of this ever happening in any ctf game I've ever played, many of which mirror this mode.

Rule 1 (or something. none-the-less, it is a rule) of level design, never ever assume players will do something that you think is natural. What is "natural" is determined by the map.

Spud brings up fair points. If you have it set up so that Red can basically just camp the flag, it will be in reds great interest to just stay there, where it is safe and spike anyone who comes in.

But theory-crafting like this is pointless, I highly recommend you continue with your idea and implement it into a map, then watch how the game comes out, talk to players and then work from there.
 

BrokenTripod

L5: Dapper Member
May 11, 2009
248
65
I hope I've read everything right so far. I'm going to try equating your game mode to A/D CP.

If you look at any A/D CP map (Dustbowl, Gorge, etc.) on the last capture point of the stage with a teleporter, you've basically got Blu attacking Red's capture point without much travel time and Red defending the CP.

It seems to me that your gamemode is exactly this but instead of having a CP, you'd have the intel where the point is. In these A/D CP maps, it's not very difficult for someone on Blu to get to the point while staying alive. The hard part is getting there and staying there for the amount of time it takes to capture the point. I can easily see Blu grabbing the flag pretty easily. I don't know if you intend on making the return timer really short, but with the normal resert timer it would be really easy for Blu to keep pushing up to the flag and bringing it back slightly closer with a scout or a spy.

Additionally, in many of these A/D CP maps, most of the Red team will be dead when Blu takes the final point, so it seems like you would have the problem of the "empty base" that Dr. Spud mentioned.

And I don't know if you've tried chasing many scouts with the intel, but it's incredibly difficult to catch them if they make it past most classes. The chase part, when Blu has the intel and is running away, sounds like it wouldn't really end up happening due to the slow movement speeds of most of the high-damage classes.

Basically, bringing the intel back through the base probably won't be very hard for Blu unless Red happens to spawn between Blu and the Intel, but that would just encourage Blu to spawn camp red. I haven't played many A/D CP games where Red is back at the first point while Blu captures the second point (you know, other than spy caps)
 

tyler

aa
Sep 11, 2013
5,102
4,621
Hey, so, we might be wrong about classic TFC gameplay or whatever? But I don't think we're wrong about TF2 gameplay. You said yourself that it's slower; that's a game design thing, not an eminent play style thing. Remember that.
 

PHISH

L1: Registered
Sep 14, 2012
41
21
Thanks for the feedback guys.

I think a lot of those dustbowl comparison concerns would be resolved by the CTF map design itself, spawn locations and not having immense spawn delays for defenders. The spawn camping aspect is something that could be a potential issue, but with highly reduced spawn times in general it may be less effective. That's something I'd have to see play out in testing. Spawn camping is present in current CTF as well so that's probably not a unique problem.


Entity question
I've got a lot of the logic stuff working now the way I want it but the main issue i'm running against is handling the teamswitch at "intermission". It seems like the only way to switch the teams is via a victory condition being triggered through team_round_win. I'd prefer not to declare a winner after the first team has played offense because a winner isn't supposed to be determined until both teams have had their attacking round. Is there any other way to trigger the team switch besides that entity? Or failing that, is it possible to edit the victory message displayed? It kind of seems like it is hardcoded.