What are your top tips for new mission makers?

[Pr3sario]

Hi guys, I'm pretty new to mission making so tell me your top tips on making a compelling, fun mission?



EDIT: Of course not just me, but anyone else new to it who happens to be reading this post...

[Dogface]

From my perspective, framerate is king. There are certain missions that run so poorly on my (and others) machines that it's impossible for me to enjoy them. So whatever you do, don't spam AI all over the map, even if your PC is good enough that you don't notice the framerate drop yourself. Many people in FA are skirting the low end of the system requirements; please be considerate and bear this in mind.

Use dynamic spawning and AI caching if at all possible, or use other means to limit the number of AI. Avoid scripts that use a lot of CPU cycles, either through poor design (large triggers, too-frequent checks), or bugs. The mission 'Swept' (EDIT: I mean Skyline) runs extremely quickly because the AI are only placed along the ridgeline. Red Snow and Arrival are examples of missions that are usually near-unplayably slow for me. Adversarial missions with no AI are always fine, unless there's a script error.

On a related note, I personally think that missions that are winnable are better than meatgrinder ones -- especially if they're made 'difficult' just by placing loads of AI, because that also affects the performance as above. There can be exceptions; Finale is fun because it has a lot of atmosphere. But as far as I'm concerned a few patrols and light defenses still provide plenty of enjoyment, but a map that overwhelms the ability of soldiers and leaders to react and overcome it is not so fun to me.

Be wary of making things too complicated. It can work but it can also not work. So if you're thinking about making a mission with three teams, that splits the team into several parts with different objectives, has a complicated winning condition, features a complicated insertion/extraction procedure, or has lots of things happening across the map that the players aren't directly involved in, think about whether the mission could work as well or better without them. That's not to say those things should be avoided entirely because sometimes they work great. Just be cautious about it.

I look forward to seeing your stuff! Really should have another go at it myself some time...

[Pr3sario]

I personally think that missions that are winnable are better than meatgrinder ones

Couldn't agree more - finding a nice balance between enough AI to challenge and not result in TDK is a priority - but one that is hard to judge while sitting in the editor on your own. I'm often pushed to rage quit on missions where most of the team and I are dead before we can pinpoint the attack... Not fun.

[wolfenswan]

Most of the stuff here applies to A2 as well.

• Rule of thumb No.1 : 3 AI rifle-infantry per player (expected player count that is, not per playable unit)
• Rule of thumb No.2 : Players will find a way to fuck up and completely ruin whatever way of the mission to be played you had in mind. Thus avoid making a mission that's only fun for a very narrow/specific style of play (e.g. "Dark Business").
• Unless there's a good reason, transports with dedicated slots should never be necessary/required slotting (esp. "glorified taxi" helos)
• Unlocked AI vehicles upset mission balance. Locking only their driver seat (this lockDriver true) is a good compromise between fun and balance.

On Perfomance:

• Use the body removal tool in F2. Deleting non-used objects (dead bodies) will improve perfomance.
• It's not the total number of AI that affects perfomance, it's the number of AI being active (engaging, having spotted the player etc.). A2 does a fairly good job in keeping calculations low for "dormant" AI. That said, AI in the hundreds will benefit from caching.
• Triggers hardly affect perfomance client-side unless they'd trigger every frame or you have several hundreds of them. If they happen only server-side they should not affect client perfomance at all, unless they cause a significant drop in server fps.

[Black Mamba]

I second that "meatgrinder" statement as well.

Keep in my mind that making that "good" mission is also a matter of luck. As you said, testing all that, by yourself or with a couple volunteers, can't really give you any idea about how this will turn out during a session. Don't be afraid of having to make adjustments after the first playthrough, and the second, and so on.

Regarding Dogface's point about complicated things: while that's entirely true with coops, I'd still encourage any new mission maker to try and think out of the box a bit, especially when it comes to adversarials. After a fair number of years playing Arma, we've beaten the good old attack/defend to death. Obviously V+ and A3 bring a bit of diversity when it comes to locations or specific gameplay mechanics, but to be fair, once you have something like random engagements, you basically have the possibility to organize an attack defend anywhere on any map. Cratesistance, Goose Chase, for example, are innovating missions bringing new forms of gameplay, at that is part of why they are interesting to me (note that they remain quite simple in their premise).

Brings me to my next point: when you make a mission, you will have an associated gameplay in mind. That's natural, and as we all have distinct preferences when it comes to gameplay, you will imagine the things you would do. Just try and make sure at some point to get back to reality and imagine what it is actually gonna look like during a session. I know that is something I struggle a lot with.
In any case, though, and I think this is an important point, try to have a good idea of how you want things to play out. And craft your mission towards that ideal.

As for the workflow itself, we all have different methods. Me, I'll usually have an idea (e.g "I want to force a player convoy by having supplies to transport" or "let's have fun with non-lethal bullets") and I'll explore it to death: I'll make five or six missions with it. Then I'll release one, maybe two, probably zero.
It's far from perfect though, as I'm pretty sure the more missions you release, the more feedback you get, and the better you get at it.

[Black Mamba]

And there was another point that I wanted to make and forgot about: make sure that the experience is compelling for everyone. One easy example is for chopper pilots. I mean, I know chopper insertions look cool, and AI-flown aircrafts are ugly.
But if the chopper crew is basically useless outside of the initial insertion and the final extraction, it's probably not worth having a chopper insertion ("Now go scout... eeerrr... wherever" with a chinook is exactly what I call useless). You'd just be taking a risk of messing the mission up in the first minutes, before the players even get to play.

Another example is the integration of toys. Watching a couple IFVs rain death on the enemy from afar while your element lead does the right thing and wait, is effective and realistic. And boring. Of course you could add a few SPGs here and there, but you have to keep in mind that they are more likely to ruin the infantry's day first.
And that's not to say toys shouldn't be used, but special care is needed when making the mission to ensure they add more to it than they remove.

[harakka]

Simple missions are often the best ones. You don't need to have anything fancy in a mission, just players, enemies and suitable terrain in between.

Find a place on the map that looks interesting to fight in. Figure out where the players start and where they need to go. Create a few enemy groups of 6-12 and copypaste the groups into suitable positions. Add looping patrol waypoints for some of them, and place a few smaller patrols looping around the region that players are most likely to go through. Add a marker for the area players need to control. Add casualty cap for players and have victory occur some time after 90-95% of enemies have been killed. Write a super simple briefing.

It really doesn't need to be any more complex than that. Put down some enemies and let the player at them. Funtimes will ensue.

[fer]

From here:

Is there a particular style of Folk ARPS mission?
In the traditional Folk missions for ArmA 2, the Folk platoon was meant to represent a guerrilla or insurgent force type, operating without high-tech equipment or significant combined arms support. Later, in the (ArmA 2) days of Folk ARPS this evolved, and many mission makers added their own flavor to our missions, from 'Soviets in Afghanistan' to 'rogue UN platoon' and 'USMC thrown into the meatgrinder'. For ArmA 3, we welcome a similarly broad range of mission types, but retain our emphasis on infantry.

Does this mean I can submit a mission where everyone has an APACHE and a M109?
While all kinds of missions are welcome, we ask that you don't focus on long-range engagements by optics-equipped infantry with plenty of tanks and attack helicopters in support. The mission should be a challenge, not a 'brown people shooting simulator'. For example, if you give the players a few IVFs (e.g. Warriors), please give the enemy some SPG-9s.

Some personal tips for new mission makers:

• If you can't explain your idea for the mission in a single, simple sentence, save it for later in your mission making career.
• The core of a mission should be simple - don't be afraid of that, because you can always embellish around the edges later.
• Terrain selection is a far better use of your time than writing new scripts.
• Write the briefing first, with markers (see this guide); this will help in several ways:
  • Issues with the shape or complexity of the mission will become apparent as you write.
  • If you have to make some key re-design decisions, you haven't wasted time on scripting or AI placement yet.
  • If you know what your end goal is, you can avoid the trap of the endlessly evolving mission design.
• Create and test the end conditions (and ending texts) as early as possible:
  • Things like casualty caps and triggers are easier to set up and test early on, when only the core units/objects are placed.
  • Again, if you have to re-think the fundamentals, you won't have wasted time on the embellishments.
• Once you have established your core (briefings, endings etc.), add complexity and embellishments iteratively:
  • Begin with what you think is a good minimum of enemies; you'd be surprised at how few AI can hold up our platoon!
  • Don't throw the kitchen sink of challenges at the players: if the mission is a cake-walk, you can always beef it up later - and you'll have the advantage of seeing how players tackle it, which will better inform the changes you make.
  • Save your iterations/versions in a back-up folder; you'll thank yourself for doing that later.
• Don't beat yourself up about re-playablity:
  • Crateresistance is a fine mission with randomisation, but we've played Swept as many times and people still find new ways to die.
  • You shouldn't be worrying about scripting for randomisation at this stage (about the closest you should get to this is using the 'grouped invisible markers' approach to randomising the starting location of a vehicle or object).
• Stop worrying about making it perfect - so long as the core works, it has proper briefings etc., your mission stands a far better chance of being evolved into something really cool if it gets played, not if it stays hidden on your local drives.

I could go on for pages about stylistic issues (balancing infantry and combined arms etc.), but in truth the best way to learn is to make missions - and play them. If you are starting out, my advice is to follow the F3 coop tutorial pretty much to the letter, but spend your time finding a great location and writing a really tight briefing. You should spend zero minutes on scripting beyond configuring the F3 components. Then play it with people and build up from there. Writing and playing 3-4 simple missions will teach you a lot more than trying to get everything you know or want to explore into your first effort.

Also, if you want to write Dark Business, don't.