[GUIDE] Fun, Fast and Dynamic CTF design

Discussion in 'Tutorials & Resources' started by Tapp, Jan 20, 2010.

  1. Tapp

    Tapp L10: Glamorous Member

    Messages:
    776
    Positive Ratings:
    211
    Many of you reading this right now must be thinking about the competitive ctf competition right now, and that is largely what this is about. When writing this guide I set out to address the issues that exist in ctf maps, the issues in having small teams, the good sides of what you are provided and how to (hopefully) increase your chances of succeeding in the current competition. I am not a very successful mapper, and have not had a successful map yet, or even a completed ctf map and as such my advise should be taken with a pinch of salt.
    Clarifications
    First of all, we must address what tools we have, as mappers, to control the flow of gameplay.
    To aid a team's progression into enemy territory, we have the following:
    • Adding ramps in the team's favor, to provide height advantage and prevent the enemy from advancing
    • Health kits placed on their side of cover
    • Ammo packs and alcoves, to provide engineers with cover and metal
    • Adequate cover
    • The ability to move un-noticed
    • Flank routes
    • One way paths
    • Straight-foward entity advantages (team-specific health and ammo, one-way doors, team-specific paths and death traps)
    To allow the defending team to defend, we use the following:
    • Sparse health kits
    • Choke points
    • Ammo supplies
    • Height advantage
    • Short respawn times
    • Difficult of slow routes
    And to prevent sentry farms becoming a nuisance, we use the following:
    • Cover specifically for medics
    • Nobuild areas
    • Distances >1024
    • Glass walls
    • Head-high walls (for demos to lob stickies over, double effective if they can be seen through)
    • Spy flank routes and dark areas
    • Alternate routes around the sentry spots (this is a big one)
    As you can see, mappers have a strong control over how the flow of a game operates (although I'm guessing most of you know this, otherwise you would not have become a mapper on the quest to create the most perfectly balanced map ever created). A mapper's control over gameplay is extremely important, and can prevent the game descending into a chore. Which brings me to my next list, what makes a game fun.
    Design elements which prohibit fun, encourage grinding, griefing and camping, allow the game to stagnate and generally destroy the enjoyment of the game:
    • Lack of cover
    • Tight, unavoidable choke points
    • Indestructable sentry locations
    • Impasses
    • Overpowered defensive positions
    • Architecture which punishes offence
    One of the key indicators of bad design is a game which can last for more than an hour because of a stalemate. When neither team can progress, frustration ensues. And since a lot of competitive teams are at a similar skill level, gameplay should balance on a knife-edge.
    Factors which encourage fun, quick and dynamic play
    • Concentration of a team's efforts
    • Constantly moving battle-lines
    • Multiple routes and paths
    • Dynamic defence
    So, we have identified ways of controlling the flow of the battle, and predicting which factors are negative and which are negative.
    Analysis of the CTF gametype and the 6v6 competitive audience
    Ctf has had a few issues throughout it's use in tf2. First and foremost is its tendancy to split up teams and stagnate into deathmatch. When playing ctf a player is forced to choose between offence and defence. Right from the start the players have been split into 2 seperate groups, and the game hasn't even started yet. To make matters worse, defence in tf2 is generally easier and more lucrative than in other FPS games. Between the demoman, heavy and particularly the engineer, classes are designed to hold an area or field, with clearly defined battle lines and forces of conflict.

    Being easier to defend, the players defending often become the MVPs, while the offensive players get domination after domination stacked against them. Compounding this is that the score temptation leads more players to defend than to attack. With more people defending than attacking, any progress is unlikely to be made by either team, since it is in fact in the interests of both teams to stagnate play to the point where people stop attacking altogether. I don't need to tell you that this is bad for gameplay, particularly in 6v6 where both teams are playing to win.

    Of course, on the flip-side of this is the dreaded scout rush, fear of many gamers. While a scout rush does prevent stagnation, with so many intruders it becomes impossible to block off any intrusion, and both teams go on total offence. With both teams deliberately avoiding each other combat is removed from the equation, which is, once again, even more of a problem with such a small team.

    Fixing the problems

    So the key issue is how to prevent both of these extremes from occurring, and the best method is by presenting the teams with roles as attackers or defenders, frequently alternating such that captures are frequent but so is combat. I tried a similar attempt with my 5cp map I am currently working on, in which the attacking and defending team is decided by the centre point, and if no captures are made in 5 minutes the roles reversed. This was actually as simple as locking a point for 5 minutes after it was captured, but did wonders for gameplay. Logically one could adapt that setup for ctf, though it would turn out a bit clumsy and difficult to understand.

    The second such method for assigning roles is using a single neutral flag in the centre of the map, with the team owning it being the attacking team. The capture zones would be in the opponents base, somewhat like football. I have seen someone else either suggest this idea or use it in their contest entry, so I know that is has/will/is being used. This method is great for gameplay, (while being a bit more deliver the flag than capturing it) and is a healthy medium between optimum gameplay and understandability.

    The third method is difficult to employ, but can work out better than the one above. It is to use the normal ctf entity setup (though it can be modified if necessary), but some tricky architecture. The idea is that it becomes very easy to enter the enemy base and collect the flag, but escaping is very difficult. This focuses the gameplay more on the returning of the flag reather than the collecting, and is very experimental. The gameplay follows a specific order:
    1. The team attempts to infiltrate the enemy base. Thanks to a combination of height advantages, well placed healthkits and counter-sentry positions, all sentries are wiped out on the way in.
    2. The flag is taken, and the defending team co-ordinates to block the exits. This should be encouraged by choke points (though they should be avoidable on the way in)
    3. After conflict(s) the flag is either taken back to the attacking team's base or guarded by the defending team until it returns and they go on offence or it is picked up by on attacking team's second attempt.
    This design is my personal favourite, as it mimics the back-and-forth gameplay of competitive 5cp, while preventing play from stagnating. Special precautions should be taken to prevent scout rushes, and so that as the attacking team retrieves the flag they can easily be intercepted by the opposing team. It's difficult to mobilise and entire team before any of their scouts can make it out.

    Like I said before, the above advise should be taken into consideration however you feel necessary, but they are worth considering before you start blocking out your ctf contest entry. Good luck!

    Edit: further notes
    First of all, I would like to clarify what I meant by scout rushes, as I was not talking specifically about scouts. What I mean is that when attacking, it is in your interests to avoid the enemy team. And if both teams are attacking, and there are enough routes, they will do exactly that. If both team are simply running back and forth, avoiding each other, it does not work well for gameplay.

    Golden moments
    Golden moments are a term I use to describe the pinnacle of good game design, when creative and dangerous play is strongly rewarded. Most players will remember these moments for a long time, and they become more frequent among heated battles of high-skilled players. It may have been a plan that went off without a hitch, a particularly skillful environmental kill, or even the appropriate use of bonk (as rare as that is).

    Personally I recall a few golden moments I have experienced. One of which was in left 4 dead, when I was playing on advanced with some friends. It was down to me and one other person, and we managed to run back-to-back, rotating, killing hunters before they got close, not getting touched at all, kiting a tank and just barely making it to the rescue vehicle. Ever since that day, I have attempted to re-create that scenario. Another such case was while playing 2fort. I saw that our intelligence had been taken, and as a soldier I ran to the battlements to intercept the carrier. I rocket-jumped, landing directly on top of the scout and killing him with my shovel as I landed. Golden moments can come in a wide variety of situations, but are all the ultimate goal of game design.

    A map which tends towards golden moments keeps players on their toes, and encourages strategic plays. Intercepting the flag carrier, shutting down an assault or even ninjaneering are golden moments that you can design your map around. If a player knows where a team or player is headed, make sure that they can set up an ambush. If a scout is escaping with the flag, make them take a longer route out, so that a defender can stop them. If there is an area of large conflict, add a health kit so that the victors can press on. Golden moments are the appitamy of strategic play, so try to define them.

    Dynamic defense
    Dynamic defense is a rare case in ctf maps, but designed well it can lead to a golden moment. Imagine that the enemy team has just literally blow away your entire team, and are about to escape with the flag. You co-ordinate your team, and everybody positions themselves in an area the attackers will probably need to pass through on their way out. In a combination of height advantages, a level one sentry and the element of surprise, the enemy assault it shut down in the last second. Another, more common form of dynamic defense is when the flag is dropped. An entire base is formed around that flag, and players guard it with their virtual lives. Dynamic defense can best be accomplished by making it easy to intercept the enemy. Design your map so that it is difficult to detour from the exit route once on it, but give them 3 or 4 options to choose from, so that the defenders cannot shut down all exits. That way it can be identified which exit they will be using, and the defenders have 20 seconds or less in which to organise a defense.
     
    • Thanks Thanks x 10
    Last edited: Jan 21, 2010
  2. absurdistof

    aa absurdistof

    Messages:
    1,242
    Positive Ratings:
    393
    Wow props for typing all that out! It's a bit late so I didn't read it all (a.k.a. past the bullet points).

    Only comment is that open space, when used well (ctf_supersandvich_a4?) can actually really add a dynamic and non separated feel to a map.
     
  3. Colt Seavers

    Colt Seavers L6: Sharp Member

    Messages:
    290
    Positive Ratings:
    81
    Cheers - interesting read and good to have a coherent reminder of the important stuff.
     
  4. grazr

    aa grazr Old Man Mutant Ninja Turtle

    Messages:
    5,436
    Positive Ratings:
    3,562
    Half of the problem is the scoring system in place. The reward for capturing the intel is improportional to the amount of effort and risk that was put into the capture. Crits were added later, but they are a product of the capture, not capture product. IE people capture the intel and happen to get team crits as a reward, but people don't capture the intel specifically for the crit bonus. Additionally, whether or not your team is capable of utilising this reward effectively is relative to player positions, and the mentality of a player to use it to his advantage, rather than chase people back into their own base for an easy kill, as opposed to using the advantage to capture the intel again.

    That's not to mention the average player mentality in other areas of ctf either. Say there is a relatively good pyro that you are healing whilst on an assault. There is a sentry up ahead that has been pinning your team for the better part of 15 minutes. Your uber fully charges and this is the moment of truth for your team; but the player sees his nemisis run away across the opposite side of the courtyard and uses the invuln to run past the sentry and attempt to kill his nemisis alone. The map stalemates for a further 15 minutes.

    But that also is a result of the emphasise on killing that Valve introduced with dominations and revenges, of which score you an equal amount to that of capturing the intel, which is then boasted publicly on the scoreboard not to mention less likely to wind up killing you than penetrating the enemy fort.

    How ever i did have an idea regarding a ctf cp_steel hybrid, where capturing the flag does have a dynamic effect on the map.

    Capturing the flag could have a variety of effects on the map such as:
    • Turning off enemy health/ammo packs
    • converting team specific health/ammo packs
    • changing the size of health ammo/packs
    • Adding health/ammo packs to the map
    • converting routes to one way
    • opening up locked routes/providing additional routes of attack
    • adding/removing cover to yours or the enemies base

    This would give a much more desirable goal to anyone willing to capture the flag, and is less likely to continue to stalemate with the favour increasing to either side on each capture.

    Of course any of these changes should be carefully monitored because they obviously imbalance the map (purposefully), but should not be too over powering to cause a steam roll.

    P.S. My pet peeve with 2fort is that bloody medium health pack in the sewers. It causes too many people to retreat!! Someone may go in and be fortunate enough to kill 4 people, but lost 50% of his health in the process, so he retreats. By the time he gets back they have respawned and know he's there. Not to mention that the static defence usually resolves around 25% of a team, and that's most of the other than for the likely sentry position.

    The amount of times i've killed those 4 people, rocket jumped up behind the fort entrance and snuck in with 45hp as soldier is rediculous. You're then able to heal up unhindered behind enemy lines.
     
    Last edited: Jul 9, 2010
  5. Nako

    Nako L1: Registered

    Messages:
    8
    Positive Ratings:
    0
    you still need to consider class limits, the so called scout rushes then become less an impact, yes there is still 1 or 2 scouts rushing but there will be atleast 1 soldier oder engie there to stop them (which most often he will).

    Playin in a comp team, I played mainly 2 ctf maps, 2forts and turbine (which I hate).

    They evolve around gaining the control of the mid of the map asap and then def all exits of the other team so noone can leave and get "our" intel, while doin this they build up their single über and then if they got it, they rush in with atleast 4 ppl and try to get the intel, if that fails (cos the entire enemy team is there (4 vs. 6) they try to retreat, (often the medic survives if he's worth his salt) and the process rebegins. Very often an engie builds a sentry in the mid to prevent losing the mid.

    So the deffin team needs to assault the mid in order to get an intel. So they need to do 2 efforts to get the enemy intel, and build up 2 übers (one after another, consdier Medic limit 1!). And that is darn long for a competetive match. This could be hell faster and way more interesting.

    So most often all 6 members of the team attack together or defend together.

    Consider this while mapping my fellow mappers :)

    Also try to make an intel cap so rewarding as possible, matches where every team gets atleast 10 intel caps is not fun.

    I hope I did my best to help you guys.
     
    Last edited: Jan 20, 2010
  6. Shmitz

    aa Shmitz Old Hat

    Messages:
    1,129
    Positive Ratings:
    740
    Nobuild areas are not a tool for preventing sentry farms. Nobuild is a tool to take care of awkward mechanics issues like engineers being able to build teleporters right at the door so enemy spies can teleport into spawn, or so buildings can't be constructed on top of a prop with an unexpectedly buildable collision model.

    If you're ever tempted to use nobuild to prevent sentry farms, then you need to stop and realize there is a serious flaw in your design, and reworking the design to utilize the other items in that list will be a much better long term solution.
     
    • Thanks Thanks x 1
  7. Tapp

    Tapp L10: Glamorous Member

    Messages:
    776
    Positive Ratings:
    211
    I realise that, and put nobuild in there as a last resort if performing any kind of changes to that area would cause the map to implode. Likewise with the weird entity setups such as team-specific healthkits. I also strongly agree with grazr in that capturing the intelligence should provide a better reward, and quite possibly the best way of rewarding someone without breaking the game is to give them revenge on a nemesis, probably the engineer who dominated them on the way in. So ideally one could create a setup whereby capturing the intelligence gave the capturer an ubercharge which lasted for 60 seconds or 1 kill, though I guess that's a bit 'out there' in terms of layout and design. And also, I think you already know this but health and ammo packs are static, and cannot be turned on and off. One could put a box onto them, but that would be the extent of it.
     
  8. grazr

    aa grazr Old Man Mutant Ninja Turtle

    Messages:
    5,436
    Positive Ratings:
    3,562
    I'm pretty sure you can kill them in the I/O. Or just use an entity spawner. I havn't checked on it though.
     
  9. UKCS-Alias

    aa UKCS-Alias Mann vs Machine... or... Mapper vs Meta?

    Messages:
    1,264
    Positive Ratings:
    748
    Backburner + W + M1 comes to mind. Because that pyro got ubered thats most likely the reason they were pinned for 15 minutes. To take out a sentry you more likely go for a demoman or heavy depending on the distance.

    The biggest issue however is that many times the intel is behind the enemy spawn or the route going to it is very close to it and you cant avoid that route. Passing the spawn requires you to beat their defence twice for just 1 cap. And in the meantime they will camp the intel next their spawn so they will be able to farm alot of kills. in 2fort that is actualy the case. There are only 2 effective exits both passing the same area again. Those 2 then go up to the courtyard (very close to their spawn) and next their spawn.

    2 entrances to the intel generaly isnt bad actualy. Even turbine only has 2 effective routes as the one passing the spawn just is impossible to be used unless you were steamrolling already. Its just the placement. They never should require you to beat them as soon as they spawned. They should have a short walk to stop you. That means that once they were taken out there is only a minor chance that they will get a 2nd chance to stop you. They only get that chance when your teammates stopped. Thats what makes turbine better than 2fort.

    To me turbine actualy has better sentry spots then 2fort. In 2fort there are only a few effective ones and as those are allways at the same place people know how to take them out. in turbine they can move them a bit making it alot harder again as you never exactly know where they are standing. This will cost a bit more time. but once the defence in turbine is broken you are allmost guaranteed you will be able to cap the intel.

    To me this is basicly the thing that makes a ctf map possible to win.
     
  10. Paradigm

    Paradigm L1: Registered

    Messages:
    7
    Positive Ratings:
    0
    Damn after reading this I've realised just how much I've overlooked in building the ctf map I'm currently working on. I'm really going to have to think about how to reward attacking players better.

    One idea that occured to me however, was a map where the bases are north and south but instead of the flags being in the bases they are placed east and west so this would encourage teams to try and hold the mid section while one team member captures the flag. This would stop temas having to attack past an enemy spawn point.

    This may have been done before but I'd be interested to hear your thoughts on it as a concept.
     
  11. grazr

    aa grazr Old Man Mutant Ninja Turtle

    Messages:
    5,436
    Positive Ratings:
    3,562
    Many of the contest maps featured a 5 area layout (as in turbine). Players spawn on one side of the base and have to defend the flag on the other, players get the choice of attack through the middle or defence at the intel. This is a good layout choice and allows for better tactical interception of flag carriers, rather than garaunteeing it because players have to pass the enemy spawn points.

    You should avoid letting teams control the central area as it can be prone to allowing steam rolls.