CTF haters, why do you hate CTF??

[Shmitz]

CTF has its problems for sure, but I'm not sure how many of the problems are tied to the game mode and how many are tied to the maps, or lack thereof. The reason I mention this is that there is only one version of CTF: you penetrate the enemy base, grab their intel, and bring it back. Rinse and repeat until point limit is reached. The more popular CP mode has 3 flavors: Dustbowl, Well/Granary/Badlands, and Gravelpit. Also Hydro, if you'd care to argue that TC is simply a variant on CP. CTF currently only has one version, and tbh I have a hard time envisioning any other method of implementing it that would still "feel" like CTF, as opposed to CP with extra strings attatched.

My apologies if this post doesn't make sense, it's late where I am.

I have to agree that there are a lot of different ways CTF can play out, and even custom maps have barely scratched the surface of the possibilities.

The biggest problem with changing the way CTF plays in TF2 is it's much harder to communicate to players what they need to do differently. With CP mechanics, this tends to be a little bit simpler, particularly in A/D maps. Players just have to find an unlocked CP to stand on for a little bit, and usually arrow signs suffice for direction. In CTF, players tend to get all manner of confused when the capture zone is in a different location than the intel.

 

[Username]

The biggest problem with CTF in TF2 is that TF2's best style of play is with a front line that moves through the map, with one team advancing, taking ground, then with their engis and snipers, holding that ground; then advancing again. Dustbowl has this, Gravelpit has this, payload maps rather explicitly have this (the advancing payload is the rear of the front line); even the 5-point linear CP maps behave similarly.

With the typical TF2 CTF map, you don't get this: the area you're trying to defend is not your newly-taken ground, it's way back at the farthest point of your base. And you can't--particularly in 2fort--form a front line, and take ground, as your enemies are spawning around you, or even behind you if you're in the enemy base.

The other big problem with CTF in TF2 is that both teams have exactly the same goals: Each team is simultaneously defending and attacking, and you don't get the different dynamic on each team that attack/defend maps of any kind give.

One of my map ideas was a CTF map that would try to take the best of TF2's battle style and adapt CTF to fit. I'd planned a map with a single flag; one team's job is to defend the flag, the other's is to infiltrate the base, pick up the flag, and return it once, for a single capture; the flag would have no timeout, so that it would progress steadily through the map.

That's exactly why I've always wanted to see a map where you push your intel into the opponent's spawn. Then you've got a frontline, where your intel is, and a defensive line, trying to stop your opponent's intel. And camping your room to prevent them from taking it would be stupid, as it would just stay right up at your final lines.

 

[Ross_C]

Yes, camping and turtling are the main reasons why CTF isn't liked that much. I do believe there were some good ideas that were already said in the thread that I may bring up again.


Obviously we have to combat massing engineers into a intel room, but we don't want to cripple engineers either. We don't want Engineers to feel like they aren't useful.

Turtling isn't just boring for the attackers either, as a engineer, I find that being forced to turtle to win is boring for me, but its keeping my team from losing. I think I have a solution to this.

First off, the Intel Room needs to be sentry proof. They need to have multiple entrances or even be somewhat open. We need to have engineers focusing on other choke points in the map to place their sentry. Now there has to be multiple choke points, so one lone engineer can't triumph over against a entire team.

There is a problem with this though, as what happens if the defenders all go engineer and cover all their choke points? To add to this, what if all the choke points are sentried up, and the defenders decide to be even more lame and build a sentry within the intel room so even IF a attacker got passed a sentry chokepoint they'd still have to deal with a intel room sentry.

The way to fix this is to make it so the chokepoints can make good sentry points, but it must be flawed. Think about it. In other game modes such as cp or payload the engineer needs to rely on his team to make sure his sentry is safe from spies or attackers. If a sentry is in a area that it can greatly exploit the engineer doesn't need his team to help him.

So Sentry choke points to the intel room need to be somewhat flawed. Have one wall protecting the sentry, have multiple areas where the attackers can flank the sentry. The only way the sentry will be able to survive is if the defenders actually DEFEND their team sentry.

Not only does this solve the problem of troubling sentries, but the engineer isn't bored out of his mind. As annoying as spies and demos are, it's alot more fun for a engineer to feel threatened than for him to be overly powered. I find it much more fun fighting for my Sentry's survival.

Of course, we can have a sentry choke point that is REALLY good, but this needs to be a place close to the attackers and needs to be somewhat in the middle of the map. Not only does it make the defenders want to concentrate on keeping the center of the map, but it also makes action much more fast which in turn makes the map much more fun for both teams.

 

[Ezekel]

That's exactly why I've always wanted to see a map where you push your intel into the opponent's spawn. Then you've got a frontline, where your intel is, and a defensive line, trying to stop your opponent's intel. And camping your room to prevent them from taking it would be stupid, as it would just stay right up at your final lines.

that sounds a lot like the bombing run game mode from UT2k4.
it might work, particularly if you give on screen messages when people pick up the intel, telling them they should take it to the enemy base. the cap zone should be blatantly obvious too, either very well signposted, or an area that gets passed through/by (e.g. on other side of glass window) on the way from spawn to frontlines.


judging what people have said, i'd also wager that a good intel room for gameplay purposes should be large, open, with some cover, but not much, have well spaced entrances (minimum of 3, or 2 very large/wide/tall ones).
the sentrys should be encouraged to gravitate towards the entrances, rather than the centre of the room, or the intel itself. and a sentry should not be able to be placed so that it's within 1024 units (max target lock range of sentry) of all doors, unless the doors are more than 90 degrees seperated from the position of the sentry (i.e. the sentry takes longer to lock on due to rate of turn of turret )
oh and the spawn room should not be on a major path between the map's central areas and the intel room.

as i said before though, this isn't ironclad, as i'm sure there will always be a map surfacing that breaks the conventions and becomes a great success.

 

[samn]

That sounds a lot like the CTF map I'm working on. (Apart from the spawnrooms on a major path part)

 

[StoneFrog]

CTF currently only has one version

One-flag CTF, reverse one-flag CTF, CTF, reverse CTF, "shutdown" CTF (secondary objective such as disabling a shield around the intel room), multi-flag reverse CTF (like flagrun, even if that was technically territorial), multi-flag CTF.

Honestly now. The only problem is that we don't have many ambitious CTF mappers...

 

[ParanoidDrone]

One-flag CTF, reverse one-flag CTF, CTF, reverse CTF, "shutdown" CTF (secondary objective such as disabling a shield around the intel room), multi-flag reverse CTF (like flagrun, even if that was technically territorial), multi-flag CTF.

Honestly now. The only problem is that we don't have many ambitious CTF mappers...

I've never heard of any of these actually implemented is what I meant. If they do exist, then I stand corrected.

(Wait...tc_meridian...dammit. >_<)

 

[Sgt Frag]

There's avanti which is kindof like a ctf/cp. You have to push the flag to checkpoints to the end of the map.

 

[StoneFrog]

I've never heard of any of these actually implemented is what I meant.

Lots of them were implemented in QWTF and TFC. In fact, I'd say 4 or 5 of those were actually in official Valve maps.

 

[TotalMark]

Lots of them were implemented in QWTF and TFC. In fact, I'd say 4 or 5 of those were actually in official Valve maps.

Ctf_push for one.

FF has the flag replaced with a soccer ball in their version.

 

[A Boojum Snark]

Ctf_push for one.

FF has the flag replaced with a soccer ball in their version.

TFC's push had a soccer ball too, which is why FF did it

 

[fisheye]

The main thing stopping mappers being more creative now is the extremely limiting entity set. I'd prepared some models to use in a tf2 map before the game was released. A beer barrel, a big stack of pizza boxes and an old gramaphone. They were to be used in place of 3 flags. The team that captured all the 'flags' would win the round and be teleported to a party room (barrel set up, pizza on the tables, gramaphone playing cheesy music) while any losers that were left alive would be weaponstripped and teleported to a cage in the party room.
I still love the idea of that map, but the stupid intel entity doesn't allow me to choose something other than the briefcase. If I only wanted to use a single custom model, then I could fake it, and pakrat a model into the map that would override the briefcase in everyone's gcf file. But that won't work with three different models. In tfc, it would have been as easy to do as falling over. You can choose any model you like for any prop type, even wrong ones. Hammer allows you to place prop_statics using models that can't even be used as such, and on the other hand it won't allow you to select ANY different model as the intel/flag/whatever. I'm no coder, but there's no way that giving mappers that choice would have been hard to implement. To me, it just smells like valve have withheld this so that no-one gets to trump their next 'exciting new gamemode' hype.
meh.

/rant

 

[TotalMark]

TFC's push had a soccer ball too, which is why FF did it

Didn't know, sorry...

A push map has more of a soccer or football dynamic, so it fits. This dynamic might be one we're looking for.

 

[Pink_Panther]

I'm no coder, but there's no way that giving mappers that choice would have been hard to implement.

as someone who codes...its not easy to code in flexibility. normally companies will look for the easy solution to a problem to save costs on development. Because valve never intended for the intel to be anything but intel, they hard coded in the brief case and the change to the backpack carried model with all the fancy pages flying behind it. Coding in flexibility for other models is not an easy task. You have to create code to allow outside items to change the pointer to a model, you have add code to make sure that model exists, you have to make sure the model has both a static component and a carried component, etc. There are lots of bits and peices that go on in the code to make dynamic allocation possible and safe to use.

Now, thats not saying its impossible. Any average coder could do it. Its just that Valve producted a product and delivered it, but never gets another dime from it even though they are adding patches to it. So, they have to go with the cheapest and fastest methods possible.

Static = more profit
dynamic = flexibility valve wont use = less profit for no reason

 

[Ezekel]

however, i recently saw a vid where someone had tied the cap zone to the intel (and made it visible for testing purposes). when the intel was picked up, the cap zone cube was still attached (and rotated correctly) to it.
so couldn't you just parent a dynamic prop to the intel entity in a similar manner?

 

[A Boojum Snark]

but never gets another dime from it even though they are adding patches to it.

Do. not. say. this.
It is entirely wrong. I'll try to find the article I read with Gabe talking about their marketing stuff... and their strategy with the free weekends and 50% off sales when they release big packages makes enormous amounts of money. The only "free" or "they don't get money from it" part is people who already own it don't pay more, but they most certainly still make money.

edit, found it:g4 blog
I hope this doesn't end up derailing the thread

 

[Uriak]

My 2 cents.

For a start (Flubber would confirm) we effectively almost hate ctf gamemode on our servers.

The main reasons have already been stated.

Speed difference and slow pace : running after a scout is boring. Running into 3 sentries as much
2 objectives, and focus on not getting out of flag room : And then we have turtling

Honestly, Turbine is the best at the moment, because with moderate teamplay you can actually take the intel out. But the middle room can become a huge turtling mess.

Conclusion either ctf game mode or ctf map layout basis must be modified, preferably both.

As pointed out, communicating to players new rules is difficult. The best move I've seen so far was with the with a force field around intels that could be removed with a central zone. It shifted most fight outside the intel room, even if it was still camped.

What must be changed at all cost. The main point in CTF is the intel pushing, imho. Getting intel is the same as getting into a CP. Taking the intel back is another thing...

So the concept of remote intelligence room that serves as capture zone at the same time is flawed. You have to turtle it because you'd get scout raped otherwise.

I'd like to see a one sided ctf mode, with decent attack, and level design made to help taking the intel, but not taking it back. The other solutions is making some variant of the secondary objective. Somewhere in the middle 2 zones needed to open the big areas to intel room. Capture zone can be near the spawn rooms, intel room shouldn't.
Another possible variant : each team alternate between stealing team and defending team, with a track between the 2 teams bases. At each "round" the attackers are given 3-5 minutes to get intel. At the end, 20s simili setup, and there you go again with roles switched.

 

[Pink_Panther]

The only "free" or "they don't get money from it" part is people who already own it don't pay more, but they most certainly still make money.

Exactly my point. They just have to coax people into buying it, not keep people playing it. The opposing idea I had in my mind are MMORPG's that have a monthy fee. They patch to keep people playing, not get them to play.

And to get back on topic...I like CTF. I just dont like the stalemates that get caused by teams not working toward a goal. That and having to uber just to move 8 inches forward.

 

[samn]

They also increase sales of future products (like L4D) by frequently updating their older ones. I would have left TF2 less than a year after release if it hadn't been updated and would never have needed to use Steam again.

 

[TotalMark]

however, i recently saw a vid where someone had tied the cap zone to the intel (and made it visible for testing purposes). when the intel was picked up, the cap zone cube was still attached (and rotated correctly) to it.
so couldn't you just parent a dynamic prop to the intel entity in a similar manner?

I wish it was that easy, the model would have to be bigger than the intel case.

I wish you could just change the model too.

BTW: That was my video. On a lighter note, you have to parent the capzone manually (no smartedit), but it still works.