YT YouTube2Text

YouTube2Text → Transcripts → Designing Radically Non-Linear Single Player Levels

Designing Radically Non-Linear Single Player Levels — Transcript

by GDC Festival of Gaming · 4,847 words · 849 segments · language en · Watch on YouTube

Open in transcriber Summarize with AI Download SRT/VTT

Full transcript

0:02[Music]
0:05all right
0:06hi everyone i'm aubrey sir i've been
0:08doing level design for 10 years
0:10these days i run a small indie game
0:11company called 12 east games where i
0:13still do level design
0:16but i'm better known for my previous
0:18work at wolfire where we organized the
0:20first humble bundle and made games like
0:22overgrowth
0:23i also collaborate with other indies
0:25making art and levels
0:26my focus is on a specific type of level
0:28which is why i'm here to talk to you at
0:30gdc
0:31today i'm going to share with you my
0:32favorite kind of levels to design which
0:34i call
0:35radically non-linear levels my talk has
0:37three
0:38sections first i'll define exactly what
0:41i mean
0:41by radically non-linear levels how
0:44they're based around the idea of free
0:45movement and sandbox play
0:47and activating the space second i'll
0:50describe the basics on how to build them
0:52like
0:52layout options in typical dynamics and
0:55finally i'll explain some techniques i
0:56use to fix common issues that may come
0:58up
0:58in radically non-linear maps like the
1:00map feeling empty
1:02or not having enough structure then i'll
1:04wrap it up with a recap and final
1:06example
1:08but before that i want to mention an
1:09idea i think we as level designers
1:11probably
1:12agree with that idea is that all
1:14conventional levels
1:15feature some openness and some linearity
1:18in other words all levels have places
1:20you can go
1:21and places that are blocked these are
1:23some of our most fundamental level
1:25design tools
1:26paths and walls linear levels start with
1:29making paths
1:30often even when we talk about nonlinear
1:33levels what's described are levels with
1:34branching paths
1:36these kinds of levels can be great but
1:38an alternative kind of level
1:40is based on the opposite idea instead of
1:42building paths
1:43you start with an area of movable space
1:45and place walls to block the player
1:47the team behind hitman calls these types
1:49of levels non-linear sandboxes
1:51a radically non-linear level is a
1:53sandbox level that is designed to
1:55encourage
1:55unpredictable player movement and
1:57exploration of the space
1:59all sorts of single-player levels fit
2:01this paradigm which i'll give some
2:03examples of in a moment
2:04and there's also a lot of stuff in
2:05between even highly linear levels
2:08have open areas in them and there's an
2:10obvious overlap between
2:12this kind of level and most multiplayer
2:14arenas so i think these techniques apply
2:16to any space the player's free to move
2:18in
2:19so i hope some of these ideas can help
2:21you in whatever kind of level you might
2:22be working on
2:25despite that overlap i think this type
2:27of level is a distinct
2:29is distinct and interesting i didn't
2:31always feel that way though
2:32let me tell you about my first
2:33commercial level design job
2:36it was for a game called darkest of days
2:38i was hired as an artist
2:40but was immediately put to work on level
2:42design
2:43outside of a few excitebike tracks i
2:45made when i was eight
2:46i had almost no level design experience
2:49in fact
2:50this was the first major game project
2:51for most of the team
2:53we had our own custom engine and tools
2:55but they were lacking some basic
2:56functions
2:57for example the map editor had no undo
2:59function so in that way it was a lot
3:01like this talk
3:03i had to try to do everything perfectly
3:05the first try
3:09on the one hand it's nice to get paid to
3:11learn on the other hand it's a lot of
3:13pressure to care about your work and not
3:14know how to do it
3:15but don't worry it turned out okay
3:23as we were making these maps our lead
3:24engineer jeff kept
3:26mentioning that he thought his test maps
3:28were more fun than what we were making
3:30we were constantly running up against
3:32the problem we couldn't reconcile
3:34we wanted to make a cinematic game like
3:35call of duty modern warfare
3:37but we also wanted large dynamic battles
3:39and open spaces
3:41trying to solve this problem is how i
3:42became interested in the idea of
3:44radically non-linear maps
3:46worked on a lot of games with them and
3:47ultimately how they became my favorite
3:49kind of map
3:57with that out of the way here's part one
4:00i'll define exactly what i mean by
4:01radically nonlinear levels
4:05when i first started researching this
4:06question i wanted to find a really
4:08fundamental example of what i meant that
4:10everybody would be familiar with and the
4:11oldest
4:12clearest example i found was pac-man but
4:14we're going to talk about miss pac-man
4:16because it's a better game
4:17so why miss pac-man the map is open
4:21enough to offer free-form tactical
4:22choices
4:23like which path to take to avoid ghosts
4:26and large enough for strategic choices
4:28if you're gonna eat a power pellet or go
4:30for the bonus fruit
4:32the designer of miss pac-man mentioned
4:34that each ghost has a different behavior
4:36and in this way each ghost is a unique
4:38risk factor which the player
4:40player needs to consider and there's
4:42some other interesting things about miss
4:43pacman maps
4:45it's small enough so that the whole map
4:46is in play and i'm going to touch on why
4:48this is important later
4:50all the systems are legible in some
4:51sense complex enough to be interesting
4:54but transparent enough to be predictable
4:56there's no exit
4:58this is a common thing for this kind of
4:59map even in uh radically
5:01even if a radically non-linear map has
5:03an exit the player often ignores it
5:04until other objectives are met
5:06in that same way the starting point
5:08isn't very important either
5:10because you don't know how the player is
5:11going to move through the map you have
5:12very few tools for pacing in linear
5:14storytelling
5:16miss pac-man handles tension with
5:17dynamic systems where timing is
5:19important
5:20power pellets ghost responds these kinds
5:22of dynamic systems are
5:24also a reoccurring feature of these maps
5:26at this point you might be thinking
5:28hey aubry do you just mean open world
5:30maps
5:32there are a few big differences between
5:33this style of level and an open world
5:35before i mention the whole map being in
5:37play what if it wasn't
5:39if we imagine a miss pac-man level that
5:41stretches on indefinitely
5:42you can just run from the ghosts you can
5:44never eat all the pellets and the power
5:46pellets become an unlimited resource
5:48in other words it would have no set
5:49challenge and totally different gameplay
5:52another question is are these kinds of
5:54maps just the same as multiplayer
5:56competitive maps
5:57maps played with bots the answer to that
6:00is sort of
6:01they are similar but multiplayer
6:03competitive maps have a lot of
6:04additional
6:05design and gameplay constraints like
6:08symmetrical balance
6:09leg tolerance limited physics that
6:11change how you make the map and so
6:13they're really very different in
6:14practice if you feel
6:16familiar with mda theory the idea that
6:18any game can be categorized by these
6:20qualities i think you can see that this
6:21kind of level occupies a distinct
6:24mechanics dynamics aesthetic niche apart
6:27from multiplayer arenas
6:30miss pac-man was made in 1982 that's a
6:32long time ago
6:34is it even still relevant to modern
6:36level design
6:37here's why i think so if you look at
6:40doom 2016 you see
6:42that many of the levels have linear
6:43sections punctuated by non-linear arenas
6:46and last year's talk embracing push
6:48forward combat in doom
6:49they point out that these spaces were a
6:51byproduct of two of their core gameplay
6:54loop goals constant movement and
6:56improvisation of
6:57in navigation this causes 360 degree
7:01combat and resulted in maps that they
7:03felt
7:03had more in common with multiplayer
7:05levels than traditional linear levels
7:09doom 2016 exam levels are a great
7:12example of a hybrid approach and it's
7:14part of a cycle
7:15away from shooting gallery style design
7:17that peaked around 2008
7:19and a return to less linear design like
7:22the original doom
7:23and i think this trend still has a lot
7:24of room to continue people are
7:26innovating in this area
7:28every year which is one of the reasons i
7:29wanted to give this talk
7:33as i was saying with the doom series and
7:35pac-man
7:36these kinds of levels have a long
7:38history including games like attention
7:40battle zone 98 and there's a lot of
7:41modern examples too like metal gear
7:43solid ground zeroes
7:44or hitman where they tried more linear
7:46levels in absolution and backed away
7:48from them with the more
7:49recent hitman games which we heard a
7:52little bit about
7:53earlier today the design techniques i'm
7:56going to talk about
7:57later can also be applied to games that
7:59use pre-designed tiles for randomized
8:01levels like spelunky which i'll show
8:02a little bit in my final example so what
8:06makes radically non-linear maps cool
8:08why do i like building and playing them
8:10there are a few reasons
8:12one of the things you notice about the
8:14games i've mentioned is that they
8:15emphasize gameplay over story
8:17remember how i said our programmer jeff
8:19preferred his test level it's because
8:22a more open map allows the gameplay
8:24systems more room to breathe some types
8:26of gameplay that really benefit from
8:27this are long range
8:28engagements using vehicles and stealth
8:31and gameplay based around controlling
8:33enemy movement
8:34this gives players a lot of choice with
8:37more agency to make
8:38strategic and tactical decisions map
8:41knowledge is important
8:42in this kind of level and so exploration
8:44is naturally rewarded
8:46because there's no need for fail-safe
8:48funneling of the player destructible
8:50terrain or killing any character become
8:52options they also emphasize dynamic and
8:55environmental storytelling
8:57because of fewer artificial barriers
8:58it's easier to feature realistic
9:00locations in architecture
9:02and as i was mentioning before aspects
9:05of radically non-linear design can be
9:06implemented in all sorts of maps
9:09that's a lot of stuff that i think makes
9:10these kinds of maps cool but the most
9:12important thing
9:13isn't it just fun to scout around come
9:15up with a plan and test it
9:16i love the kind of level that lets me
9:18come up with different solutions and
9:19that i can play twice with totally
9:21different outcomes
9:28it's just water
9:32part two here are some techniques on how
9:34to build these kinds of levels
9:37i'm going to borrow and adapt some
9:39theory from randy smith's 2006 gdc talk
9:41level building for stealth gameplay let
9:44me take a moment to explain this diagram
9:45here
9:46this is an overhead view of a map from
9:48thief and we see how smith
9:50used these safe areas and rewards shown
9:52with plus signs and enemies and
9:54exposure shown with negative signs to
9:56predict and control the player's path
9:58i want to use them in the opposite way
9:59so that it becomes impossible to predict
10:01the player's path
10:03this is what makes our level radically
10:05non-linear and also how we activate the
10:07entire space of the level for play
10:10the first thing i do when i start on a
10:12new map is to have a clear idea about
10:13the
10:14layout by picking a strong layout you'll
10:16be able to differentiate
10:17each map you make rather than ending up
10:19with a random hodgepodge of
10:21obstacles enemies and rewards there are
10:24a lot of different possibilities for
10:25layout but here are four solid
10:27solid examples symmetric
10:31we have a low tension center line with a
10:33left right choice
10:35this is an interesting layout because
10:36the player chooses which side to tackle
10:38first
10:39and then halfway through comes back to
10:41the center line
10:42that helps break up the tempo and
10:44creates pacing this is a great layout
10:46when the theme of the map is a major
10:48path like a highway
10:50concentric the outer area of the map is
10:53safer
10:53and rewards extensive scouting before
10:55committing to the center
10:57this layout works well for maps about
10:59assaulting a base or castle
11:02node these are a series of clearings
11:04that interconnect in a lot of ways
11:06this is the kind of this kind of map
11:07keeps up a high tempo
11:09of quickly alternating encounters with
11:11travel
11:12the player needs to resolve challenges
11:14carefully so that they don't bleed into
11:16other adjacent nodes in unpredictable
11:17ways
11:18this layout fits with ruins or slums
11:21field a large area of increasing risk
11:24this layout promotes alternating probing
11:26attacks and retreating
11:27and gives the player a lot of control
11:29over tempo this is a typical layout for
11:31something like a beach assault
11:33again the main point here isn't to know
11:35what the player will do but just to make
11:36sure that whatever they do it isn't
11:38boring
11:39one thing to note here is that it's also
11:41possible to use gameplay systems for the
11:43layout rather than building it into the
11:45map
11:45and i'll have an example of that using a
11:47field layout at the end of the talk
11:50as well as layout i like to have a clear
11:52grasp of how the gameplay is going to be
11:53unique for the level
11:55one method is to try to find an aspect
11:57of gameplay that's underutilized and
11:58focus on it
11:59this is hard to generalize since every
12:01game is different but here are some
12:02concrete examples
12:04from some maps i worked on this is
12:06brigadore
12:07an isometric shooter that was designed
12:09from the start to allow players to use a
12:10wide range of vehicles
12:12because of this it had some interesting
12:14design constraints for example within
12:16the interior of the map
12:17all the terrain is destructible in this
12:20game certain aspects are randomized
12:21which is one of the easiest ways to
12:23ensure players try different strategies
12:25on replated
12:25repeated playthroughs here the enemy
12:28spawns are hand placed
12:30and whether an enemy actually appears
12:32depends on several factors
12:34similarly the player will randomly enter
12:36the map from one of these two locations
12:38i think this technique of using hand
12:40play spots for enemies and rewards that
12:42are then controlled by systemic factors
12:44is really useful
12:46the gameplay idea for this map is that
12:48the center area is a trap
12:49it's tempting but if you go in there
12:52before picking off most of the enemies
12:53you'll probably fail
12:55as you can see from the main travel
12:56lines this map is roughly concentric
12:59the main hazards and rewards are
13:01concentrated in the middle to draw the
13:02players into the trap
13:03it's possible to clear the map by
13:05defeating enough enemies around the
13:06border so you don't need to go in the
13:08center
13:09but it's a lot faster to beat the map if
13:11you complete these central objectives
13:13in that way the concept of the map that
13:15it's a trap to lure the player in
13:17and the layout of the map that it's
13:19concentric relate to and support each
13:21other
13:22here's another example from brigadore
13:24the game has soft cover that blocks line
13:26of sight but doesn't offer much
13:27protection
13:28in the existing maps it was mostly
13:30combined with hardcover to add variety
13:32but i wanted to replace the hard cover
13:34with mud pits to give a really
13:36dangerous exposed feeling to this map
13:39this layout is one i haven't shown yet
13:41where the objectives are in the four
13:42corners
13:43this means the player has to cover a lot
13:45of distance which emphasizes even more
13:47how exposed the player is
13:49i think this shows how it's important to
13:50pick a concept and a layout
13:52that work well together here's a
13:55different game i worked on called raven
13:56field a single player game where you
13:58capture control points and shoot enemies
14:00in an effort to outscore the opposing
14:02army
14:03unlike similar multiplayer games the
14:05levels in raven field don't actually
14:07have to be balanced
14:08or fair on this map i was motivated by
14:12my frustration of the ai
14:14allies taking all the airplanes on the
14:16other maps so i made a map with
14:18a lot of airplanes in it during
14:20production i bumped the starting bases
14:22out
14:22out to the corners of the map to
14:24maximize playable space for the
14:25fast-moving airplanes
14:27but kept the roughly symmetrical layout
14:30unlike my previous example of this kind
14:32of layout this is an inversion where the
14:34center line is more risky and these side
14:36areas are a bit safer
14:38i think this works because in the
14:40aircraft you have to keep moving and
14:41naturally end up flying over the center
14:43line
14:44all of these examples benefited a lot
14:46from being for functional games with
14:48several existing maps
14:49in cases like this it's pretty easy to
14:51play with existing systems
14:53make tests and figure out something that
14:55works if however
14:57your game isn't very finished you may
14:59notice your map has serious problems and
15:01not know why
15:10okay welcome to the third and final part
15:12of this talk how to fix common issues or
15:14why is your map all busted
15:17the number one issue is that the map is
15:19too linear it's very tempting to want to
15:21guide the player and reflexively fall
15:23back on that skill set
15:24but if you're going to design a linear
15:25map or have linear map sections
15:27it's important to go all in on good
15:29pacing and have all of those existing
15:31techniques that make that style of map
15:32work
15:34otherwise open up more paths and space
15:36to move
15:38similar to the above but with gameplay
15:40you have one idea in mind with how the
15:42how to win the map this can stifle
15:44organic and robust gameplay systems
15:46it's important to avoid trying to
15:48restrict the gameplay too much
15:49trust that the core game plays fun on
15:51its own another possible issue is
15:54the map has no structure the map is too
15:55uniform so you end up doing the same
15:57things over and over
15:59hopefully by having a unique layout and
16:01concept you avoid this issue
16:04in a game with very dynamic systems
16:06sometimes the player can feel helpless
16:09the level should be balanced so it's
16:10reactive to the player and they act as
16:12the tipping point for the gameplay
16:14systems
16:15the gameplay can feel random or
16:17non-existent the player isn't able to
16:19see the cause and effect of the systems
16:20enough to understand why things happen
16:23for example using an invisible trigger
16:25to spawn enemies on the map is a lot
16:27worse than an alarm state in a stealth
16:29game
16:30alarm state gameplay still has triggers
16:32and still spawns enemies
16:34but gives the player a lot more
16:35information if you want to learn more
16:37about making systems legible to new
16:38players i very much suggest watching the
16:40gdc 2016 talk titled level design and
16:43hitman
16:44guiding players in a non-linear sandbox
16:47these previous issues
16:48can usually be a fixed assuming you have
16:51mechanical
16:51complexity to work with but if you don't
16:54then it may be something to bring up
16:55with your design team
16:57the sorts of systems these games usually
16:59have are things like stealth ai
17:00companions inventory looting
17:02destructible terrain
17:03dynamic enemy movement player tools for
17:06controlling the enemy
17:07sub-objectives like capturing areas or
17:10destroying enemy resources
17:12they have pacing systems and various
17:15movement modes like vehicles etc
17:17when discussing how these systems work
17:19having a map that feels like it's
17:21missing something
17:22to show the rest of the team can be very
17:23useful
17:26on the other hand if what seems to be
17:27missing is grand cinematic moments
17:29type pacing are very well defined
17:30challenges i'm not aware of any way to
17:32fix this
17:33yet it's part of the trade-off that
17:35comes with this kind of map
17:37which may not be the best choice for
17:39your project it isn't that dramatic
17:41scenes or clear challenges can't happen
17:43but they tend to result from the
17:44existing systems
17:46and so they're hard to predict i don't
17:49think this is an impossible problem
17:50though because there's been a lot of
17:51exciting progress in dynamic
17:53storytelling lately
18:05before i end my talk i just want to
18:07recap with an example that brings it all
18:08together
18:09and shows how a lot of this theory can
18:10work in a more linear game
18:13the game i'll use for this final example
18:15is one that i was in charge of level
18:16design on called
18:18receiver it has some aspects of linear
18:20design
18:21but still allows free and unpredictable
18:23movement
18:24the level was made from tiles each tile
18:27is roughly the size of a small office
18:29building and has three
18:30floors with doors on each side these act
18:33as lanes
18:35when these tiles are put end to end the
18:36lanes can combine or switch positions
18:39if i had just stuck these together i
18:40would say it was similar to a
18:42multi-path level kind of like this x
18:46but we did something a little bit
18:47different instead the tiles are randomly
18:50placed so the player often has multiple
18:51choices about how to progress
18:54in addition to the strategic choice of
18:56which lane to take
18:57the tactical choice of the best path
18:59through the tile also isn't
19:00predetermined either
19:02because the enemies and objectives are
19:04spawned randomly from hand placed spawn
19:06points
19:07this is how the space is activated even
19:10though almost all this content was
19:12pre-authored and the random elements are
19:14fairly minor
19:15it's still enough to change the ideal
19:16strategy and to incentivize exploring
19:19the space
19:20which makes the level radically
19:21non-linear
19:23because of these systems it made it
19:24difficult to support a standard
19:26narrative
19:27instead we made the objective to find
19:29these audio logs which can play in any
19:31order
19:32obviously this doesn't result in the
19:33most traditional story but i still think
19:35it added to the experience
19:37that said the story wasn't the strongest
19:39aspect of this game
19:40that this was part of the trade-off we
19:42decided to make
19:45another gameplay system at work here is
19:46that each new tile is harder than the
19:48last one
19:49this gives the player an incentive to
19:51check all the lanes in a given tile
19:53which may require some backtracking as i
19:56mentioned when i was talking about miss
19:57pacman
19:58a systems-based approach to tension is
20:00an important pacing tool
20:02and i think this could have been boring
20:04in some games
20:05but because receiver is extremely high
20:07tension back tracking serves as a much
20:10needed break from clearing new areas
20:12this increasing difficulty as you move
20:15forward is
20:16also provides a layout structure i
20:17mentioned earlier called
20:19field because these levels don't have a
20:22lot of scripted events
20:23or built in gameplay they may seem a bit
20:26simplistic
20:27but part of making a map in this style
20:29is to not obstruct the core gameplay
20:31in receiver most of the gameplay is
20:33built around a real realistic simulation
20:35of a gl
20:37of a gun including all the working parts
20:39and buttons
20:40we surfaced all this gameplay for novice
20:43players with a context-sensitive help
20:45system this makes it clear to the
20:47players that every bullet is precious
20:49and that every enemy encounter is
20:50difficult simply because firing your gun
20:52is
20:53is difficult this relates directly to
20:56the levels because you can scavenge
20:57bullets
20:58and obviously cover and enemy positions
21:00matter more
21:01when the situations are so lethal i
21:04think these levels do a good job of
21:05giving the player interesting choices
21:07while staying out of the way
21:08enough to let the core gameplay shine
21:11by working closely with your gameplay
21:14coder and because the core gameplay was
21:16so deep my job was made a lot easier and
21:18i was able to avoid any really tricky
21:20problems
21:21part of that was making trade-offs early
21:23in development to avoid those kinds of
21:25problems
21:26in the end focusing on interesting
21:28dynamics that shift
21:30rewards and hazards keeping layout and
21:32pacing in mind
21:33and giving the gameplay room to breathe
21:35really helped receivers
21:37radically non-linear levels come
21:38together
21:40finally i would like to say i feel like
21:42this is a complicated topic
21:44and because these maps integrate so
21:46closely to the gameplay
21:48they're a bit slippery to nail down but
21:50i think they really are
21:51a distinct and interesting style of map
21:54with their own history and design trends
21:56hopefully i've provided some concrete
21:58ideas on building a map like this
22:00and how to approach certain problems
22:02whether your future levels
22:04are radically non-linear or not thank
22:06you for coming to my talk
22:15and uh it looks like we have time for
22:18plenty of
22:18questions uh if you have any questions
22:22you can
22:22go to the mics
22:28hello and thank you for the talk oh
22:30thanks so you mentioned that
22:32it is a constant challenge being able to
22:34nail the level design just right
22:36before shipping the game now the
22:38industry has changed over the years and
22:40now we have tools to measure all sorts
22:43of things what's your approach to
22:45nailing down the the interesting part
22:48uh up to now what i've mostly done is
22:51just have
22:52play tests do early builds and just get
22:54player feedback
22:56watch people stream the game i i
22:58personally like
22:59more than trying to collect metrics
23:01through the game i like to
23:03um watch live play tests where i can see
23:05the player's face
23:07so you can get a kind of a more
23:09immediate read on what they're
23:11reacting to thank you
23:17hey what do you think of the dangers and
23:20benefits of using procedural generation
23:22with these kind of levels uh i think
23:26it should be kept to a minimum like it's
23:28a very useful tool i know that there's
23:30been recently a lot of like kind of
23:34negativity towards procedural generation
23:38and i don't think that's really makes
23:40much sense
23:42but at the same time
23:45sometimes you just want the level to
23:46play a certain way and
23:49by introducing just a tiny bit of random
23:52elements you can
23:54massively expand the playability of a
23:55map
23:57so yeah i try to use it sparingly um
24:04so as an educator i'm brady university
24:07in the netherlands um
24:08we have a lot of room to do multiple
24:11level design projects in our curriculum
24:13but one of the things i have to kind of
24:15figure out is uh
24:16how we build complexity for the students
24:19so like where we start simple and then
24:21build up
24:22um and some parts of that are easy like
24:25single player platform game level design
24:28is
24:28a nice starting point but do you think
24:31the uh linearity
24:32and non-linearity is that more difficult
24:35or
24:36i think that especially now making a
24:40really compelling
24:42linear experience is like extremely
24:45difficult
24:45so um and it's mostly different
24:48disciplines like
24:49more about storytelling environmental
24:52narrative
24:54my last game i made was totally linear
24:56totally totally linear
24:57totally linear like adventure game so
25:00i'm well acquainted with those problems
25:02and yeah i i think they're both really
25:04difficult just different
25:06so uh non-linearity is more like a
25:08sideways shape yeah just like a
25:10different
25:11a different set of skills okay cool
25:14hello so thank you for the talk um
25:17i wanted to know you mentioned at the
25:20start that
25:21the making of the levels is like making
25:24sure
25:24what space is blocked and making when
25:27what space is open
25:28but in this kind of non-linear level
25:31where you opened up a lot of this space
25:35what let's say one problem maybe if the
25:38player has a lot of
25:40freedom of movement no there's not
25:42necessarily that they can fly but
25:44let's say you know i know they can climb
25:45walls or something like that
25:47how can you make let's say obstacles or
25:51walls that actually are obstacles
25:54without
25:55impeding the core gameplay in that way
25:57so uh yeah i
25:59one of the games that i didn't mention
26:00because i had a lot of trouble with this
26:02exact problem was overgrowth
26:04because your character is this rabbit
26:06and he can jump like
26:0830 feet in the air and run 40 miles per
26:11hour
26:12and you're you're just like flying past
26:14everything at top speed and going
26:16anywhere you want
26:17and yeah this is a really difficult
26:19problem and i i would
26:21you know i think ultimately you have to
26:23have those kinds of
26:24movement options cost something at least
26:27a little bit of something so that you
26:29can
26:30like have some gradient in the player
26:33behavior
26:34or else they are just going to be lying
26:36to the objective and jump over
26:37everything because their character is
26:38superman
26:40okay thank you thanks
26:44caicho escudigasco thank you for the
26:46talk i recognize this maybe a little bit
26:48outside of your field of expertise but
26:49i'm curious about
26:51the overlaps with vr projects in terms
26:53of linear and nonlinear
26:54design yeah uh there is a
26:58vr version of um receiver
27:02and i've but i've only played it a
27:03little bit
27:05um and i'm not sure
27:08i think it's it it basically works like
27:11i don't think it's like massively
27:12different uh it's
27:13maybe a little bit more confusing um
27:17when you can't like whip your head
27:18around and get for some people maybe
27:20less confusing for others i don't
27:22i'm not sure but i haven't done
27:24extensive player testing with that kind
27:26of thing so
27:28yeah thank you anyone else
27:34thanks for coming to my talk

About this transcript

This page contains the full transcript of Designing Radically Non-Linear Single Player Levels by GDC Festival of Gaming, generated from the public captions YouTube serves with the video. The transcript has 4,847 words across 849 segments, with the original timestamps preserved so you can click any line to jump to that moment in the embedded player.

What you can do with it

Use the transcript to take notes, quote the speaker, build a study guide, generate a summary with ChatGPT or Claude via the YouTube Summary tool, or export it as a timed subtitle file with YouTube to SRT. You can also re-open it in the transcriber to translate the transcript into 100+ languages.

Free YouTube transcript tool

YouTube2Text is a free YouTube transcript generator — no signup, no daily limit. Paste any YouTube link and get the full transcript instantly, with timestamps, click-to-jump, translation to 100+ languages, AI prompts for ChatGPT, Claude, and Gemini, and exports to TXT, SRT, VTT, or Markdown.