Antishyr here, the community manager of Streets of Rogue 2. It's about time we brought you some news on how the development is going. I sat down with the game's solo programmer/designer, Matt Dabrowski, to chat about what he's been up to lately, and here's the result: a miscellaneous bunch of developments that are either happening right now or have recently been completed. Expect this devblog to be slightly shorter than the two previous ones, as this one is less visually rich and is meant to give the community an idea of where SoR 2 is at currently.

Note from Matt: Please bear in mind that anything and everything you read here is subject to change. Some of the specific design elements you'll see here have been planned for a long time, but haven't gotten much (if any) implementation work until fairly recently. Good games are built on strong foundational systems, and it's taken a long time to get those foundations in place, but I think it was worth it! My intention has always been to make my game systems super flexible and easy to mess with. In fact, I'm planning to allow a large number of my "official" design decisions to be totally subverted with the mutators system. It's entirely possible that some of the design elements below just won't work out. But that's what pre-Early Access and Early Access is for!

First things first, I'm fully aware you've been expecting an update for the demo for some time now. I happen to have a quick update on this front for you: Matt's planning on delivering the next version as soon as the demo is, quoting him, "a much better representation of the core gameplay loop that I'm currently working toward for the Early Access release."


❓ What is Design implementation?
Speaking of gameplay, you may've seen a section called "Design Implementation & Balance" on our Development Roadmap[sor2.notion.site]:

You probably understand what balance is, but might ask, "Just what the hell does 'Design Implementation' mean exactly?" In short, it's about writing code for elements of the game's design that are not yet in the game. Example: Matt has designed a whole system of Mayors controlling cities across the world, and the player must take those guys (or gals!) down to gain influence and eventually attempt to topple the President. In order for the whole process to play out, however, you need to have all the elements in place:

• The Mayor's dwelling must have several floors and function as a dungeon.
  
• Mayors will come in different varieties—e.g., a cultist mayor who has a massive following (and does podcasts every now and then) or a sewer-dwelling cannibal mayor whose primary source of protein is another NPC's body.
  
• Each of those city boss types must have their own unique dialogue, types of quests they dole out, etc.
  
• Methods of taking down the Mayor or gaining their favor (and the repercussions for these) are intended to be quite varied. SOR1 offered a glimpse at this, but SOR2 will take the system much further.

...and so on.

It sounds like a lot, but thankfully, the game's infrastructure was built to allow for gameplay systems like these to be implemented quickly and rapidly iterated upon.

⚰️ Death Penalty
"What's the penalty for death?" has been on Matt's mind for a while as one of the most crucial design/balance dilemmas. The goal is to avoid making kicking the bucket too punishing (running halfway across the city to your corpse to retrieve lost items in a dangerous area is unlikely) while keeping the stakes high enough to maintain tension. There are always players who demand the difficulty to pack more of a punch, and those who just want to chill. Mutators that tweak the conditions will be meant to solve this.

💼 Professions
Another outstanding item on the dev agenda at the moment is profession specialization. In the first game, the Hacker, in addition to stealthily fiddling with computers, could actually put up a decent fight if the situation demanded it. With the sequel, the story will be different—picking a fight while playing as a weak character won't be as viable of an option. But, the game will offer more varied ways to improve your system breaching skills, as well as new skills you can acquire via augmentation booths. Essentially, it'll be more important to actually role-play!

Another important aspect that's being fleshed out as we speak is region difficulty tiers. Think of these tiers like a typical RPG system that includes higher-level monsters, weapons and armor in higher-level regions, but applied broadly to a bunch of different types of skills and aspects of the game world that may not necessarily be combat-oriented. To give a general example, if you've picked a Hacker and your skills are currently around tier 2, cracking a tier 3 security system might be achievable but quite tough. Or, a tier 8 police station might flat-out refuse to assign missions to your tier 4 Cop character due to being underqualified.

So, how will region tiers be determined? That's still up in the air; Matt is leaning toward a ruleset similar to one Skyrim utilizes. Each region is given a range of tiers—for example, a region has a range of tiers from 5 to 10. If you enter for the first time when your character is below tier 5, the region locks its level to 5, and you'll probably have a tough time unless you improve your character first. If you enter for the first time when your character is tier 7, the region will lock itself to tier 7 or 8, giving you a tier-appropriate challenge. And if you enter when your character is above tier 10, the region would lock to tier 10, and you'd have an easy time. This is oversimplifying it a bit, and this approach to difficulty may not even pan out to Matt's liking, but we'll see how it goes!

It's worth adding that the game world is planned to be dynamic—as the player, you'll shape it by your actions. Random events occurring every now and then will impact factions and mayoral standings. All of this could potentially impact a region's difficulty regardless of its initially assigned difficulty tiers.

High RAM usage has been one of Matt's primary concerns when it comes to SoR 2's performance. The higher volume of art used in the sequel compared to the first game gradually resulted in the game using up to a crazy 11 GB of RAM (Thank you to all those who reported this issue). It needed to be addressed ASAP, and thankfully Matt has been able to reduce this to about half its previous size by improving asset handling in the Unity engine. Further improvements will likely come at a later point, but for now we think it's down to an acceptable level.

Everpresent. Don't get us wrong here, we don't want to fix every single bug before shipping the Early Access version. But the most likely candidates for fixing will fall into two categories: It breaks the game in a not-fun way, or it's super noticeable and ridiculous. NPCs getting out of bed and immediately getting stuck between the bed and their wall would fall into this category. If that's ever happened to you in your personal life, you're probably aware that it's a terrible way to start your day. In addition to that, Matt's addressed a pesky problem where the game wouldn't start properly sometimes, but instead would load up one of your save files with dozens of MB of garbage data. Not a good call, video game.

---

As always, we appreciate your support! Share your thoughts on the devblog below and wishlist/follow Streets of Rogue 2 to stay up to date with the news:

https://steamhost.cn/app/2165810/Streets_of_Rogue_2/
Read previous dev blogs:
Devblog I: Vehicles
Devblog II: Chunks

Watch video devlogs:
Devlog 1: How I Approach World-Building? 🌎
Devlog 2: Creating an Open World Solo 🗺️
Devlog 3: Games That Inspired SoR 2 💡
Devlog 4: New Release ETA, Roadmap 🗺️

---

Join the community on Discord:
🔗DISCORD[discord.gg]

Follow Streets of Rogue 2 on social media:
🔗TWITTER (X)
🔗TIKTOK[www.tiktok.com]
🔗BLUESKY[bsky.app]
🔗YOUTUBE
🔗REDDIT

Antishyr here, the community manager of Streets of Rogue 2. In this devblog, I'd like to offer you a peek at dungeons, show you how we handcraft interiors for the game, and talk about how Matt, the dev, hired a player to help him construct "chunks." Who is that guy? And wtf's a chunk? Let's find out.

SoR 2 is an ambitious open-world game, which means everybody expects its environments to be immersive and diverse. Matt Dabrowski, the sole programmer/designer of the game, created a system that generates a whole island country with cities/towns, rivers, and forests when you start a new save. However, for every piece of the game world, large and small, there's always a looming question: "Is it worth spending time writing a system to generate this procedurally, or would the results be better (and, potentially, more efficient) if I just made this myself?" Generally, Streets of Rogue 2 follows a similar formula to the original game: Procedurally generate the maps, while handcrafting the individual pieces of those maps. Those "pieces" are classified as "chunks".
Here's an example of an interior office chunk:

In the open world of SoR 2, regions are the largest form of measurement. Then there are chunks, and the smallest form is a tile. The smallest regions are comprised of 4x4 normal-sized chunks, while the largest are 16x16. A normal-sized chunk has 16x16 tiles, while the smallest has 8x8 tiles, and the largest has 64x64 tiles, making it the size of the smallest region. A greatly expanded version of the level editor from SoR 1 is used to create chunks.

By the way, if you want to learn even more about how chunks are developed, check out this video devlog by Matt.

Compared to Streets 1, SoR 2 will have a much bigger variety of chunk sizes. The inclusion of smaller chunks comes in handy when Matt fills the world with small filler locations:

Initially, Matt wanted to create all the chunks by hand before realizing it was an enormous amount of work that would, before long, drive him crazy if added to an already extensive solodev list of various tasks.

Enter Cyan_Light, real name: Clay—a SoR player and a prominent modder who has contributed countless incredible workshop items to the first game's community. He and Matt decided to join forces, and the rest is history: Clay has already crafted thousands (!) of chunks for the sequel. I'm not kidding. Here are some of the examples of his most recent works:

To make Cyan_Light's work less time-consuming, Matt designed the editor in such a way that it attempts to improve efficiency and minimize human errors as much as possible without limiting creativity. For example, if you place a locked door on a wall, the game will determine which direction the door should be facing to ensure that it's locked from the correct side. Or, if you place a generic chair, the game will determine if chair variants exist for the specific type of chunk you're creating (i.e., Police Station or Suburban Home), and display one of those variants. On top of that, you can now randomize other elements like the amount of damage on objects—a table in the President's office would be in mint condition, while one in a drug den... maybe not so much?
When it comes to deciding where to put quest objects in a chunk, Clay drops a marker that the game obediently populates with a randomized item—a chest, a safe, a kidnapped person. It all depends on the requirements of the quest. Some pieces, like vending machines, may be thrown in randomly, but others, like, say, augmentation booths, will abide by common sense: You'll see an aug booth near a hospital. You can also tweak the spawn chance of objects for more variety:

Quality-of-life features like those I listed above help with the pace of pumping out chunks: One takes 10-15 minutes to cook, larger ones—up to an hour.

When putting chunks together for missions, Clay makes sure there are always multiple ways to approach a quest. Here's a vent to put syringes into, and there's a PC to hack, and a bunch of windows to crack and climb through. Maybe a conveniently placed 1500-lb overheated generator in one of the backrooms.

When it comes to choosing how to complete a quest, the material a building is made of is important. You can blast through wood and brick to get inside a building (or even ride through it if your vehicle is souped-up enough). Steel buildings might prove to be more of a challenge. Naturally, you'll see more structures made of sturdier materials in high-level game areas.

A lot of attention is paid to designing dungeons. E.g., mayor's offices are huge chunks that require a meticulous approach from the design standpoint.

...I know you've been waiting for more details on this feature. Work is being put into it as we speak; in addition to wild caves and office dungeons we talked about above, cannibal hideouts and house basements are in the works, as well as other types of dungeon biomes we'll reveal later.

One thing to note: While dungeon levels like the ones you see here are entirely handcrafted, procedurally generated dungeon maps comprised of smaller chunks are also planned. Much of the programming for these has been completed, but Matt has opted to hold off on further development of these until after the Early Access launch because, in Matt's words, "the game should probably ship at some point... yeah, that would be ideal".

Having a swimming pool in your basement is some luxury sh*t, ngl.☝️


Here's a look at a mostly-completed cave dungeon chunk. I say "mostly" because diagonal walls are currently getting an overhaul and thus are missing from this revision of the chunk. It's also missing some extra cave flourishes that were just recently added, notably cave lakes.

"Do I get to do all these fun things in the level editor, too?", you may ask. The short and definitive answer is "Yes." All tools that Cyan_Light has at his disposal should become accessible to the community in the future with the release of the level editor—that one is planned to be rolled out at some time after the early access launch. Players will be able to create new chunks and entire levels for the game, as was possible in the original game.

The level editor is already packed with tiles, wall types, animals, melee weapons, and firearms to play around with, with more coming in the future. Interestingly, adding abundant content to a game is not the biggest grind an indie dev can undertake:

As you can see from this tweet, Matt's currently focused on better floor tile mixing and sorting out diagonal wall issues. As soon as that's in the books, he'll be working primarily on design Implementation and balance tasks. Following that, he'll be completing the local co-op mode. For more info, check out the game's Roadmap[sor2.notion.site]. I'll share more details when I have them!

---

As always, thanks for your support, all. Comment your thoughts on the devblog below and wishlist/follow Streets of Rogue 2 to stay updated:

https://steamhost.cn/app/2165810/Streets_of_Rogue_2/
Read previous dev blogs:
Devblog I: Vehicles

Watch video devlogs:
Devlog 1: How I Approach World-Building? 🌎
Devlog 2: Creating an Open World Solo 🗺️
Devlog 3: Games That Inspired SoR 2 💡
Devlog 4: New Release ETA, Roadmap 🗺️

---

Join the community on Discord:
🔗DISCORD[discord.gg]

Follow Streets of Rogue 2 on social media:
🔗TWITTER (X)
🔗TIKTOK[www.tiktok.com]
🔗YOUTUBE
🔗REDDIT