News | Forum | People | FAQ | Links | Search | Register | Log in
The Door Problem Of Combat Design
This is a problem I regularly deal with!

Article: https://gamasutra.com/blogs/AndrewYoder/20190808/348237/The_Door_Problem_of_Combat_Design.php

Discussion: https://news.ycombinator.com/item?id=20645914

(copied from GA thread. Thanks mfx.)
Uh... 
... hasn't Nintendo solved the problem by locking the door behind you (Ocarina of Time)? 
 
I don't understand how this is a "problem"... here are some really simple ideas to force the player to stay in the room/arena...

1-close the door behind him and open it again when the last enemy is dead (to allow the player to go back and explore the map)

2-open the corridor's floor and put lava bellow it so the player can't go back (close the lava pit when the last enemy is dead)

3- activate a spike shooter in the corridor, this way the player can't stay there and have to go to the room (deactivate the spike shooter when the last enemy is dead)

4- spawn a shambler in the corridor 
 
well, if you think of it as "player is making a bad choice" your solution can be to take away choices from the player (Tribal/Izhido are suggesting) or improve/increase the choices available to the player (as the article seems to be suggesting.) 
 
Yep, what metlslime said.

Rather than thinking of it as the player making a bad choice (and punishing them for their conservative play style), I'd prefer to invite them into the arena and give them good reasons to play that way.

I think the locked-door solution works for some games, like 3D Zeldas, but with first person shooters I think that technique is best saved for the most intense encounters, like the final room before a level exit. If you lock the player in for every significant combat room, it can turn repetitive. Even Doom 2016 broke up their pace with platforming and exploration sections between arenas. 
 
I don't like closing the door behind the player and locking them in, it's such a cheap solution most of the time. I mean it works well for a climactic fight but not for some regular fight, and must be used sporadically. 
 
Yeah, getting locked into EVERY combat encounter gets pretty stale, especially if the game is in large part about exploration and traversal. And I actually think the "one-way path" solution is not a big improvement.

I like it when the designer minimizes the number of times that "enter door and immediately encounter enemy" happens, especially when the player is seeing a new space for the first time. If it's more like "enter door, go forward, learn about space, encounter enemies" then the player may at that point have ideas in mind about how to deal with the encounter which seem more fun than retreating way back to a door. (That also means that the environment is going to be more interesting than just hall-door-room stuff.)

I guess that overlaps with some of the other solutions the article mentions but it doesn't feel exactly the same. Rather than explicitly rewarding the player for pushing ahead, instead make them feel like they can handle the situation and that they have plans for a fun attack bubbling in their head.

It's harder to make spaces and routes like that in a game like Quake that encourages enclosed spaces with abrupt, sightline-blocking transitions between them. Not impossible though. :-) 
 
i figured it out

don't make the room

put the enemies in the corridor

done 
Well 
"Don't make the room" is very good advice, and is also often relevant for improved visuals. 
Backspawn Rotweilers. 
Nuff said. 
Agreed 
#9 FTW 
 
Anything involving Rottweilers is good enough for me. Can you explain what is “backspawn”, though? 
And 
when people get tired of rottweilers spawning on their backs? Which is the backup plan? 
Frontspawn Rotweilers And Rotate Player 180d Degrees. 
 
Jaja 
 
Haha 
 
The Bullet-Proof Solution ... 
... is obviously to separate the play spaces with a giant rotating door constructed entirely out of Rottweilers all rotating around a central point.

To implement Hard skill, you make the door out of Shamblers instead.

(In seriousness, I haven't finished reading the article yet, but I will, because it's good.) 
#16 End Of Thread GGWP 
 
Lava! 
One option the article declines to suggest is a big pit of lava. No, hear me out!

What if you put some kind of mild hazard in the area prior to your arena, puddles of lava or slime to navigate around, trap shooters, a long drop, something like that. That would make blind back-peddling too risky, and even a more careful retreat might backfire - if the monsters follow you then you have to juggle environmental hazards as well as combat. Better to fight in the arena and only face one threat.

The Underearth has quite a few examples of this kind of thing - the rooms with slime and the rotating spike shooter all drive you to take the next combat on without retreating. There's also a nicely placed ogre just after a pipe, whose grenades tend to bounce into said pipe and hit you if you try and run. If you run into the open, he's a much fairer fight.

I've had a more gimmicky idea that might be good for a bit arena fight. The plan was to have multiple plinths with powerups/armour/etc on them, each with a button below that causes it to descend slowly. That way, to get all the goodies for the fight, you'd need to visit each plinth twice - once to trigger it, and again a bit later once it dropped. If they're spaced far enough apart, there's a extra tactical decision about which button to press first - do I want armour more that the quad? 
Now Or Never 
I'm intrigued by the idea of constructing a time-limited honeypot. Some sort of reward on a platform that, for example, slowly descends into the lava during the fight. Advance decisively, or loose the goodies forever. 
 
While I think this article is great for a general introductory ideal and spans many different games, I feel like Sock had some more specific and honestly more relevant (to Quake) thoughts on the matter.

https://twitter.com/SimsOCallaghan/status/1161643807760367616

Many of these things are easily observed in Socks best levels, especially Grendel's Blade, how supplies and weapons are breadcrumbed in a way to push the player towards moving forward and getting into the thick of it, and the difficulty never makes encounters overwhelming at least to begin with.
Even the most conservative players like Tens and Arrcee while playing the map seem to bound forward into situations they normally would not, grabbing new weapons, items and upgrades as counters and timed triggers slowly trickle targets into his view.

You can read more about pushing the player towards a more active playstyle here on Simon's website. http://www.simonoc.com/pages/design/sp/ad_grendel.htm
I really need to get to making a full analysis of Grendel's Blade in particular. Many things it does are ingenious, and after many attempts I end up tripping over myself talking over gameplay since things are that frequent. There's a lot of intricacy in design there I feel like a lot of people passively overlook. 
@fairweather 
Yeah Sock's twitter thread is much better suited to the specifics of Quake. (And he's also way more knowledgeable in good Quake encounter design than I am.) With the article, I was trying for a tricky balance between a specific context and general appeal, and ended up favoring something general.

Some of the techniques I listed for the door, like the one-way paths of drop downs and locking doors, are really heavy handed for Quake design but totally the norm in other games. I think that's the spot where my more general approach created the most disagreement, but I'm glad folks have still found it useful :) 
 
I'm not a fan of breadcrumbing. There's a whole GDC talk about pipes and lines and what not that boils down to "draw a path for player to follow" which I find pretty cheap. There are ppl who use it effectively, but this rule when abused (which is often) leads to pretty poor experience. 
But... 
There are certainly gated, one-way sections in Grendel's Blade (or at least sections that stop the action a bit and lock you into a certain path.) sock even says so himself in the article fw linked above:

The solution (though not ideal) was to force the player to drop down for gated doors and then re-use the central area several times.

Andrew's article wasn't presented as "the end-all be-all" solution to the door problem - the solutions were presented as possibilities.

And I hope some of the techniques I outlined will prove useful in overcoming those problems. 
 
To me Andrew's article points in the direction of the encounter design found in Halo and Crysis series: hubs and paths. Paths exist to connect hubs. Hubs host encounters that are composed of 1) points from which one can observe and plan, 2) interesting setups for the firefight, 3) unknowns that spice the encounter and cannot be forseen through point 1 above. There are multiple ways to approach this kind of design and, to me at least, invitig arenas are the way to go. Andrew's article touches on the idea of an inviting arena but doesn't exploit it to its fullest. 
You must be logged in to post in this thread.
Website copyright © 2002-2024 John Fitzgibbons. All posts are copyright their respective authors.