top of page
Z2
H1Z1 Z2

10km x 10km

The Arena

Z2

H1Z1's early access launch map, Z1, was originally designed for a zombie survival game. After the addition of the battle royale game mode that quickly grew in popularity, Z2 was planned to create a balanced map that was appropriate for the twitchy, competitive, last-man-standing gameplay. The massive tracts of empty land and limited road connectivity were part of a slower world that pushed for cautious exploration and scavenging in contrast to battle royale's constant action. Z2 intended to give players more options to navigate through playspaces, as well as giving a generalized distribution of POIs.

​

Development on Z2 started in December 2015/January 2016 while I was finishing work on the the Indar v3 map update for Planetside 2. When I took on lead level design responsibilities for Z2, the world team had gotten together the rough topography and city locations, as well as art leadership having already planned much of the asset list. Starting out on a fresh map and having some of the basic foundation down already was helpful to start defining minimum dimensions, planning layouts, and organizing design information. The heightmap was being iterated on by the world artists at this time, so it was a crucial time to help integrate design requirements as well.

​

Since the editor was undergoing a significant technical refactoring, there was a good cushion of time to iterate on the planning aspect of the map before we 'broke ground' on it. For instance, the Proposals/Revisions shows some of the experimentation with the major arterial types and layouts. This phase also included defining POI types/map positions, the world tasking pipeline, time estimates, and a lot of info organization. Getting incoming world builders caught up quickly will reap rewards on tight schedules, so smaller details like a usable wiki and intuitive filename conventions will help it feel like the team is ramping up instead of scaling a wall.

​

By the time the world team was able to move on to the next phase and begin work on the map itself, we had a robust gameplan together and were able to hit the ground running. We had a lot of implementation resources and building groups (instanced arrangements of props) available at this point, so filling in the world went fairly quick. Once the basics had been blocked out, it was a standard iterate/review cycle until the map's launch in September 2016. I went to work on an unannounced project shortly after its debut, and the Z2 team continued iteration and optimization, as well as adding a few additional POIs in empty field dead space.

          

With the game's popularity steadily increasing the months following the big Sept '16 update, resources were reallocated to H1Z1 and I returned to the team to assist iterating and implementing new features and POIs. From early 2017 until starting on the Z1R update, the team optimized, added and removed POIs, placed new rock/boulder field cover sets, and experimented with new level layouts. Hillside Villas (and later Ponyvale) shows the evolution of breaking out of standard suburb blocks, whereas the Wynn Lake revamp attempts to unite and improve existing game spaces.

​

(Original)

Z1

vs

Z2

- Sporadic POI placement led to spotty loot  density on the macro scale

​

​

- Exaggerated heightmap and low field

cover compounded high ground advantage

​

​

- Long runs of impassable terrain or  partitions box players out of gas & navigation unnecessarily difficult

​

- Tighter spaces originally designed for first person camera before game became de facto third person

​

- Modular, grid based highway pieces that iterate quickly but require significant unnatural terrain adjustments

- Subdivided & generalized distribution of    POIs averages out loot better and keeps a few distant destinations in view at all times

​

- Heightmap value clamps and extra field cover shifts combat dynamics from terrain to object based, reduces verticality

​

- Encourages flow from playspace to playspace with wall breaks, crate stacks, etc

​

​

- Bigger interior dimensions reduce disorientation, getting stuck and camera-related deaths

​

- Organic spline-based highways laid on top of existing terrain topography

  1. Proposal #1

    Here's the basis of what would ultimately be refined down and become Z2's major road/river layout. Black was intended to be a railroad that signified your presence in the NW/W part of the map.

  2. Proposal #2 (Rejected)

    Closest rejected proposal to the chosen Z2, this one went easy on the highways and ideally would've used a railroad across the southern half.

  3. Proposal #3 (Rejected)

    Variant with lots of continued proximity between major artery landmarks. This would've been a good choice for the survival mode, since it concentrates a lot of the flow towards the center with lots of isolated hills and valleys.

  4. Proposal #4 (Rejected)

    This one had the highway 'touch' all the major cities, and then laid the less important river/rail arteries horizontally across the map.

Proposals

  1. Z2 Revision #1

    Honestly, this is nearly identical to the "Proposal #1" image, but I figure it'd be nice to put it down here to help show the iteration. Also, it has less pen scribbles on it.

  2. Z2 Revision #2

    Here the railroad got the axe, and the river slicing into Pleasant Valley got cut. Despite all the map maiming, a segment of highway from Cranberry to around Harris Bluffs was added in.

  3. Revision3.jpg

    Added the county roads connecting the outer parts of the map together. Threw the extra river segment back in since it was only creating an additional 2 bridges.

  4. Revision #4

    The Pleasant Valley river segment is finally put to rest, extra water is added for the survival portion of the game, and an extra county road breaks up the huge empty spot in the northern half of the map. The river now functioned as a single path you could orient yourself along. The far ends of the river that were intended to cut into the mountains ended up turning into terminating lakes.

Revisions

bottom of page