CyberPunk — Night City Analysis

Iuliu-Cosmin Oniscu
10 min readMay 5, 2021

I decided to return to Night City.

Building space, in some ways, can lead to confusion if we ignore the fundamentals of orientation.

I think that’s what makes Night City so interesting. It’s committed to breaking all the rules to support the narrative of the game.

I am not going to be the advocate of building a perfect theme park world. I don’t think that’s an accurate way of creating spaces.

It could be, and it most definitely works when it comes to Navigation and orientation, but these days, there seems to be a heavy focus on moving away from that trope.

I think the reason is pretty simple: When you adhere to patterns that restrict your capacity for imprinting identity into your world, you end up with clear, cleanly formulated game spaces that work as such but might not be interesting to justify from a narrative or visual standpoint.

I know environment artist has a hard time with that especially when it comes to justifying spaces (they also have a hard time not ignoring gameplay considerations, I guess it comes with the name tag).

So with that, I would like to deep dive into Night City and explore it a little.

I will start from the starting location represented by Mega Building 10 and focus on the exterior areas, main roads, landmarks, essential nodes, perceived map edges, gameplay pockets and thoughts, and I will try to expand outwards.

The purpose is to build a better mental model/image of how the city from a street-level (since that’s how most people play it) and identify what makes it confusing.

In this article, we will only focus on Little China and its role in the Ward, and we will see where that takes us.

Some of the topics we will talk about:

  • Linear frame by frame perception
  • Assumptions about the space we inhabit
  • How minor deviations in street angles can lead to confusion
  • Major streets, landmarks, nodes, memorable places
  • Cyberpunk world-building design language
  • Gameplay Pockets

We start the journey from what we would call the central hub in this area. Mega Building 10.

The map

Little china sits between a variety of features that constitutes the more prominent macro elements of the map:

  • An extensive canal to the south spills to the sea.
  • Along the side of this canal, we have Palm Avenue.
  • Following Palm Avenue, we can get to California Avenue
  • Moving northward, California Avenue takes the right turn and exists the district.

A slight detour on the right leads us to a street that attempts to separate Little China from a different district called Kabuki.

Inside the streets, we have various streets that weave and out of each other connecting the major arteries: Ellison, Bradbury, Ferguson; and different more minor connector roads that bridge gaps in the road network: Cortez, Cartwright, Ettien, Muren, Brookland, Buron.

The streets:

The biggest issue that I have with this part of town is that it tends to be very confusing when walking around.

There is little foreshadowing regarding where these streets lead.

There is no clear hierarchy or roads, and the streets seem to tie into each other in unexpected and displeasing ways.

A great example of this is Ellison.

Ellison starts a large traffic-driving street that seems to lead somewhere only to fizzle out straight in a side street called Cortez (in front of the medical centre).

It might be that the intention was for this large multi-laned street to lead into the sizeable suspended highway that hovers above the main roads on the fringes of the District (like California Ave).

The lanes connecting these streets are tiny and narrow that they would suggest this was an afterthought.

Ellison and Bradburry collapse into a single two-way multi-lane street that connects to Palm Ave on the south side. However, going in the opposite direction, we get into a weird intersection that I like to call the Bradbury-Goldsmith-Ettien-Cartwhrite circle of madness.

Bradbury fizzles out into Goldsmith, a sidestreet that connects to Ellison, somehow in an underground intersection. It’s hard to figure out where this intersection is in the context of the entire map, but some close investigation revealed that this is precisely under the Mega Building 10. The exit points leading to the Medical centre in one direction and towards the Palm Ave waterfront in the other direction.

I find this very confusing. In addition, the lack of forwarding visibility and foreshadowing tends to lead to a lot of disorientation. If you want to commit to not getting lost, you need to save some severe map-making and orientation since streets don’t lead where you think they show.

If we follow the Subdivision of space logic, we can consider the spaces between the streets as sub-sub districts; however, here, we can perceive them as large monolithic structures that can only be avoided rather than explored.

Some minor exist from place to place, but their directionality is questionable since it tends to move you away from where you want to go.

The best examples are the overhead bridges above highways that would typically serve as crossing points. Here they act as paths that lead to places where you might not want to go.

The Chunks and clusters:

I remember working on a video game once that had a similar problem. You had to sail around an island to get to a specific port. The issue was not the sailing itself but rather the long time it took to circumnavigate that island.

I remember saying to my colleagues that there was a pacing issue there, but since we were doing non-linear, they look at me as if I was crazy.

Here’s the way this works.

That’s why if the island is too large, the tendency is to often give the player more choice in terms of how he would like to engage with it:

  • Do I take the long paths along its side and risk nothing
  • Do I go through the middle to shorten the trip but risk damaging my ship?

This way, you get more opportunities for:

  • Organic gameplay
  • You get to decide the pacing of the action
  • You can decide to opt-in and out of the danger zone
  • It leads to better zoning because now you can attribute the exterior of the area to safety and the interior to danger, or you can do it the other way around; either way, you get more options, and the entire space becomes more attractive to explore

The good side is that it’s not just something you get to do in games with ships. The concept of islands happens in every open-world match to some extent, and as long as it translates into good zoning (easy to understand from a player perspective), that’s a success.

Here’s how other video games do it:

First, they fracture it and build something inside it:

It isn’t too different from how a location in the actual work would be composed.

Also, this opens the possibility of introducing more believable spaces in your game that serves an actual function associated with the part of the building you are showcasing.

When I was working on Watch Dogs Legion, one of the things I kept telling my self was:

  • What function does this space service in the context of the game? What happens there? What’s the world logic?

Because of the game logic we know, that the player is going to use it as a shortcut, or we are going put a collectable there or maybe even an objective) but what is the actual point of this space?

When a player looks at it, what would he call this area? What’s the identity of such a space, and how would he communicate this identity to other players?

I think Watch Dogs Legion and Mafia 3 do this exceptionally well. Building meaning in back alleys and side courtyards creates a clear delineation between public and private spaces. It becomes apparent when you are not supposed to be somewhere, and not just because the UI tells you so.

In the absence of direction that would aim at having the player interact with such a place, what can you do?

It all burns down to a pacing issue. What happens every 30 seconds?

If you don’t want to commit to breaking out the chunks, you can engage in facilitating minor indentations in these chunks that would host engagement elements.

Night city does this a lot, and it works. You get away with not building an outpost, and you get to populate your spaces with more narrative contextual elements.

So how do we fix it? What should we do to improve orientation, player engagement and overall exploration?

First of all, we need to look at the roads and establish where the main issues are?

First, we draw in the main arteries:

Then we do the analysis:

And here are the results:

It seems like Root nodes: 9, 10, 11, 13 and 14 get the higher values.

Is it interesting because of these side streets?

Why would this happen?

The Syntax Analysis aims to identify the number of connections and sub connections between paths to improve flow.

In a nutshell, the streets with the higher value would get a higher amount of traffic because they tend to connect to other avenues and getting from one note to another in 3 steps should be easier this way.

In our case, the issue is that there is no distinction between primary, secondary and tertiary roads. As I said before, the space separation method is not applied there, for navigation becomes a chore, and orientation is complicated because there is no way for affordances to be appropriately generated.

Putting it simply, it becomes challenging for anyone walking down these streets to formulate an accurate interpretation of how the city works.

It looks pretty, but it’s not functional.

So, How do we fix it?

Well, the first step is to make sure that the main arteries are correctly functional and act as you would expect.

In our case, the main arteries are 1,2,3,4,5 and 6, and the target is to ensure that the values associated with them get bumped up. It is because higher importance streets need more extensive usage and therefore more significant numbers.

Root node 7,8, and 9 also need to become more prominent as they seem to fall to the bottom of the list due to poor connectivity.

Maybe something along these lines?

In this proposal, we can see a few distinctive features:

  • We now have better parcels for zoning
  • Intersections are now a little more varied, and we get rid of the confusing three ways paths that were very present previously.
  • Roads now have a better directional function since they connect the sides of the district without sacrificing the original intentions.
  • Some of the paths have been collapsed in one direction.

The new district form mirrors the original one but offers better orientation, leading to fewer chunky buildings. Still significant but easier to grasp.

In my previous entry about Cyberpunk, I mentioned that Night City mirrors, to some extent, cities such as Tokyo, Osaka, Hong Kong, etc. Due to their confusing set of road networks and lack of actual city planning rigours.

However, despite their apparent absence, such functions still exist.

The recipe continues to be:

  • Large roads that carry traffic
  • Secondary arteries that split the bigger chunks into smaller parcels
  • Tertiary roads
  • Alley and paths were leading into people homes.

Even if these things are not immediately apparent or are easily dismissed, it’s hard to ignore that they are there, their existence modelled by peoples needs.

But despite all these flaws, Night City fascinates me, and I have the same feeling coming out of other game, like Sleeping Dogs, where these issues are also present.

As designers, I feel that in our attempts to capture the spirit of these places, we should also strive towards fixing the things that don’t work in the real world because games are about escapism beyond any else.

I don’t think turning every virtual city into a theme park is always the best solution.

We need to be willing to take advantage of the theme park design knowledge we have available.



Iuliu-Cosmin Oniscu

As a level designer I am not a creator, I am a facilitator. Senior Open World Designer. #Leveldesign #Open World #Design @notimetoulose.