Monastero

Monastero (de_monastero) is a community-made map for Counter-Strike 2. It was released on the Steam Workshop in August 2024.

The map is designed for the competitive 5v5 game mode.

Plot: The Phoenix Cooperation, a notorious criminal organization, has set their sights on Lake Como, aiming to disrupt the peace of this idyllic Italian locale. Their target remains uncertain, with intelligence suggesting they could strike either the strategic boat dock used by government officials and wealthy elites or the renowned natural thermal springs. Both locations are critical, the dock being a covert hub for transferring sensitive information and illicit goods, while the springs are a major tourist attraction and local economic lifeline. To cripple their enemies’ operations and send a powerful message, Phoenix Cooperation is planning a high-stakes mission. The local authorities and special forces must act swiftly to prevent a potential catastrophe that could devastate the region’s landscape and security.

Links: Steam Workshop | Mapcore
Team: Just me
Development: 4 months
Platform: PC
Game: Counter-Strike 2
Technology: Source 2 Engine, Counter-Strike 2 Workshop Tools

Making Of

As soon as FACEIT x Mapcore announced their Big Adventures CS2 Mapping Contest, I knew that I wanted to participate in it. It was a challenging undertaking, as Counter-Strike: Global Offensive had transitioned to the Source 2 Engine and became Counter-Strike 2 only five months earlier.

The change of the engine brought a big paradigm shift with it. The Hammer Editor had been around since the original Quake engine (Quake, 1996), then used in the GoldSrc engine (Half-Life, 1998), and finally in the Source Engine (Half-Life 2, 2004). For historical optimization reasons, maps were built using BSP brushes. A map had to be fully enclosed without any leaks. Thus, most Source games featured indoor areas, as they could be heavily optimized.

Source 2 brought the mapping experience to modern industry standards and aligned it with other map authoring tools. Maps are no longer built from brushes but individual faces. For obvious reasons, rendering a single face is much more performant than rendering six faces. The new workflow is most comparable to modelling in Blender.

Having only worked on smaller maps before brought a second challenge. The scope of producing a competitive 5v5 map is exponentially higher compared to a 2v2 map. As I was still familiarizing myself with the new tools, the process was much slower, and changes to the layout were much more expensive.

Pre-Production

Already prior to the announcement of the contest, I had the idea to build a map situated at Lake Como. Originally planned as a 2v2 Wingman map, it now became a competitive 5v5 layout. I started the pre-production by collecting interesting reference material and browsing Google Maps, where I also found the map’s name—Monastero.

For the setting, I chose a warm and sunny day in spring with a clear blue sky. I figured it would be the best option to enhance visibility and make player models stand out from the background—crucial for a competitively viable map.

I started my creative process with sketches on paper, mixing and matching interesting elements from different references in a way that would be believable and create a sense of place. The first problem I was confronted with was that most of my references featured open outdoor spaces without any cover.

The first sketch features elements present in the final version of the map. However, the layout was too compact, and I knew I had to start over.

The second sketch is very close to the final version of the map. Although nearly all competitive maps follow the four-square layout, I wanted to experiment with a rather unconventional layout where the attacking side has a lot of map control early on and can enter from various directions.

The advantage of drafting on paper is that it’s cheap to iterate and doesn’t hurt to scrap ideas. The disadvantage is that it never feels the same once you are in the editor.

The problem with this approach is that you are trying to build the final product immediately. The significantly more sustainable creative process is to make guesses and prove them in order to find the final product through exploration and evaluation.
In hindsight, if I were to plan another map at that scale, I would jump into the editor earlier and iterate through greyboxing and player feedback from playtesting.

Production

After the map had been laid out, I began with the blockout. Most of the time here was spent learning the modelling tools to get up to speed. While modelling, I stuck closely to my design sketch and converted it into a playable level. I used models for reference to get the scale of the environment right.

I already used quite a lot of props during the blockout to better convey my idea to the other contestants. However, all the details just got in the way. Iterations got a lot slower, and build times started to increase. In hindsight, I would begin with simple ramps instead of fully modelled stairs. I wouldn’t add any detail to buildings unless relevant for gameplay, and I would use props sparingly only to get the scale right.

Building the final map from the start meant I had a single shot at it, which could be either a hit or miss. Only one day prior to the deadline, the layout was finalized, and not a single player had ever set foot on the map.

Launch

The approach I had taken was risky and didn’t pay off in the end. The map had major design flaws, with some areas being too wide and open, and others too narrow.

Monastero didn’t move on to Stage 2 of the contest. The feedback from the Steam Workshop had been overly positive, and as some players still enjoyed the map, I decided to add bot support, which was missing from the original release. I decided against continuing working on the map, as large parts would have to be scrapped to fix the layout.

Within the first day, the map climbed to the top of the most popular maps on the Steam Workshop and was featured on the front page for several days.