Jump to content

Dynamic levels - in Payday 2 and beyond


Puddy

rKJhPgx.jpg

Payday 2 is a cooperative first person shooter where players band together to commit various crimes in the endless pursuit of wealth, infamy and cool masks to cover their criminal faces with. The game recently celebrated it’s third birthday, yet it still retains a steady player base. How then has the game kept players engaged throughout the years? The many and regular content updates are surely a big part of it. Another draw must be the fleshed out progression systems that offer tons of customization. I would argue that the lifeblood of the game is its dynamic level design; it is what keeps the game replayable and fun. In this article I will discuss what dynamic level design is and how it was used to build “Hoxton Breakout”, one of the game’s most popular missions.

 

DMk7Oar.jpg

Payday 2, Left 4 Dead 2 and even XCOM2 all use some form of dynamic level design.

 

What is dynamic level design?

Dynamic level design is all about creating levels that are as replayable as possible; it is about retaining the challenge and keeping players on their toes. This is achieved by introducing elements that change between playthroughs, things that make the level a bit different each time you play. Dynamic levels are still designed and built by hand, so to speak, which makes them different from procedural levels which are created from automated algorithms.

 

Dynamic levels are useful in games where the developer wants the levels to provide more gameplay than a single playthrough would. This approach has the added benefit of allowing different players to come together and enjoy the same level, irrespective of whether they have played it many times before or not at all. This can make dynamic level design ideal for co-op games and it can be essential for retaining players over longer periods of time, just like Payday 2 has done.

Building a dynamic level

The process of building a dynamic level certainly differs from more traditional single player level design. Instead of crafting a linear experience in meticulous detail, a designer must seek to create a broader structure of what will happen in the level and then design dynamic elements, things that change between playthroughs, within that structure. These dynamic elements need to be designed with care, so that the level actually changes in meaningful ways between playthroughs. The process of making a dynamic level will vary from game to game; it all depends on the game's mechanics, setting and other details. By sharing the design of a Payday 2 level I hope to illustrate what a dynamic level can look like and also showcase the overall possibilities of dynamic level design.

 

 

p1raSzs.jpg
Hoxton in all his glory, featured here in this promotional art. Shortly after his breakout, he leads a daring break-in at the FBI to uncover who ratted him out.

Hoxton Breakout

In the Payday 2 mission “Hoxton Breakout” players are tasked with breaking their old heisting comrade Hoxton out of custody. During the breakout Hoxton shares his suspicion that his capture was caused by an unknown snitch. To uncover the truth, the PAYDAY gang set their sights on the headquarters of the Federal Bureau of Intervention (not to be confused with any real life organization...). This sets the stage for the mission’s second level and the one I will be discussing here.

 

Level Structure

In this level, the players will enter the FBI headquarters together with Hoxton (an NPC). They will fight their way to the “Operations Room”; the place where the FBI servers are kept and where the Payday gang is hoping to find the information which reveals the identity of the snitch. Hoxton will search through the servers and when he has found what they need, the gang will escape. No matter how many times you play the level, the overall structure will stay the same. Instead, it’s the dynamic elements within this structure that change and make it replayable. What are those, you ask? Let’s take a look!

 

hNFGRhk.jpg

Clockwise from top right: The FBI HQ lobby, a central area in the level. The FBI director hides behind his desk. Hoxton and the Payday gang enter the lobby.

The Operations Room

Players will spend a lot of time in the FBI Operations Room. Hoxton will be hard at work searching through the servers, leaving players to defend him from relentless police assaults. The combat space will change in a number of ways between playthroughs.

 
  • Entrances - Most of the entrances to the Operations room are selected dynamically in various combinations, which changes which choke points the player must defend.

  • Windows - The ‘Operations’ room is two floors in height and the second-floor windows overlooking the room are placed in different positions. Players must watch them for enemy fire.

  • Fuse box - The fuse box, which enemies use to cut the power to the servers and pause your progress, can be placed in a few different positions. Players must defend it.

  • Ceiling breaches - SWAT troops can breach the ceiling of the ‘Operations’ room and rappel down right into the thick of it! There are a few places where this can happen (it doesn't always).

 

These dynamic elements will vary and change independently. This can be very desirable, as it will give you a large set of different combinations and improve the replayability of the level. For example, even if the fuse box is in the same location in two separate playthroughs, the positioning of the entrances and windows will change how the players approach the situation, which will help reduce level fatigue.

 

zTq5sA8.jpg
The Operations Room. The Servers are kept in the room under the illuminated FBI logo. 

The Quests

There are four servers Hoxton must search through in the Operations Room. Between the searching of each server, Hoxton will need the player's assistance and send them on a “quest”. There are five different quests, though only three are selected and used in each playthrough. They can be selected in any order and combination. Each quest and its gameplay have been designed to have a slightly different flavor.

 
  • Security Office - The next server happens to be heavily encrypted. You need to break into the Security Office, download the encryption keys and get them back to Hoxton.

  • IT Department - The next server is missing and the log states it was taken to the IT Department for maintenance. You must locate the IT Department, find the missing server and bring it back to Hoxton.

  • Archives - Hoxton finds a reference to some physical files kept in the archives. You need to go down to the basement, search through the archives and bring the paper files to Hoxton.

  • Forensics - Hoxton learns that the FBI has evidence related to the traitor. Players need to break into the evidence locker, find the right piece of evidence and then scan it in the nearby laboratory for clues before returning to Hoxton.

  • Director’s Office - Hoxton encounters some files on the next server that can only be accessed by gaining direct approval from the FBI Director. You must head to the director’s office and use his computer to approve all of Hoxton’s security clearance requests.

 

What this means is that players won’t know exactly which “quests” they will tackle each time they play the mission, or in which order they will face them. As the difficulty slowly ramps up during the mission and the players’ supplies generally are lower towards the end, completing the same quest as either your first or last one can become quite a different experience, even though the quest itself doesn’t change that much. Allowing the quests to be arranged in any order and combination simply gives the mission a slightly different flow each time.

 

IU2RlIY.jpg
The five quests, clockwise from top right: IT Department, Security Office, Archives, Forensics. Center: Director's Office

The Combat

Now, it’s about time we talked about the combat. It is essential for the replayability of a level that the combat isn’t static and that encounters vary between playthroughs. To solve this, Payday 2 has a spawning system that serves up dynamic enemy encounters. The system unburdens individual level designers and creates a consistent and tweakable way for the game to spawn enemies in all levels. For those of you who have played the Left 4 Dead games this may sound very familiar. The system isn’t completely automated and the level designer can control a few variables.

 
  • Difficulty - The player selects the overall difficulty of a level before starting, but a designer can tweak the difficulty to a factor between 0 and 1. This can be adjusted at any point during the mission and can be tied to certain events.

  • Spawn locations - A designer designates spawn locations manually. The designer can toggle spawn locations on and off, change how often they can be used to spawn enemies and which kind of enemies are allowed to spawn from them.

  • Enemy Wave Mode - Police assaults occur regularly and this is generally handled by the system, but a designer can force a police assault or a complete break from them.

  • Snipers/Harassers - The placement of snipers and so called harassers, regular SWAT troops who harass players from vantage points, is done manually. It is up to the designer to place them in challenging, but fair, positions and script logic which decides when and if they appear.

 

What this all means is that while the spawning system does the heavy lifting and creates varied combat encounters, a designer can fine-tune the experience and still direct the combat somewhat. For example, in Hoxton Breakout the difficulty is slowly ramped up after each completed server, the spawn locations are continuously tweaked throughout the mission to make fights fair and when it is time to escape an endless police assault is forcefully triggered to increase the stakes!


bZmmdn2.jpg

A dynamically spawned enemy squad moves towards the Payday gang.

The keycard economy

In Payday 2, keycards are single-use items that are occasionally used to open certain doors. In order to add depth and strategy to the level, I added something to this level which I like to call “the keycard economy”. In every playthrough, players can find 3-4 keycards which can be used, i.e. “spent”, on a variety of options like overriding doors to seal them off from enemies, unlocking rooms that contain precious resources or opening doors that lead to objectives. The value of the different options can change between playthroughs, depending on dynamic variables and which loadouts the players have. Since players can’t have all the options, they must choose wisely. This allows players to refine their strategy over the course of multiple playthroughs, adding to the level’s replayability.

 
The little things

We’ve discussed all the major dynamic elements of the level at this point, but it is worth mentioning that replayability also arises from smaller dynamic elements too. These smaller surprises can throw players off and force them to adapt accordingly. A good example can be found in the Security Office, where the police sometimes pumps in tear gas when players are trying to complete the objective inside. This forces players to leave the relative safety of the room and charge head first into the police forces which are surely waiting outside. Part of making a dynamic level should be to identify and implement these little game changers!

 

Kvyixap.jpg

Clockwise from top right: The Security Room fills with gas. A keycard has been used to seal a security door. An innocent keycard. A SWAT team rushing to thwart the payday gang.

Discussion

To summarize, the level we’ve looked at is about defending a location and completing short “quests”, with both activities changing in different ways between playthroughs. In addition to this variety, enemies are dynamically spawned, occasional surprises appear and players are able to learn and master the keycard economy over the course of multiple playthroughs. These dynamic elements, this variety between playthroughs, is what turns the level into a dynamic one.

 

This level was made for Payday 2 and, as mentioned, dynamic levels will look a bit different depending on the game and its needs. The Left 4 Dead games have less emphasis on objectives and focus more on linear progression through a level, with dynamic enemies, items and minor path changes along the way. The Killing Floor games have arena levels that suit the game’s wave-based horde mode and these levels feature fairly simple dynamic elements: enemy and item spawning as well as the location of the weapon and item shop. The revived XCOM franchise uses levels which have designated areas or “slots” where different buildings and structures can fit in and shift the layout accordingly. The XCOM games also allow different missions to be played on the same level, enabling levels to provide even more gameplay mileage.

 

The dynamic level design approach may fit these games, and others like them, but it is not suitable for all kinds of games and it definitely comes at a price. Since dynamic levels are designed to be replayable, heavily scripted story moments and set pieces may have to be deemphasized or removed outright. Playing through such sections may be thrilling once or twice, but they generally lose their appeal very quickly. Furthermore, some degree of polish is generally lost in the process of making dynamic levels. The fact that you are making an experience that can’t just happen “in one way” means you can’t necessarily polish, and control, every moment of gameplay to an insane standard, like you would expect in an Uncharted game for example. Additionally, an incredibly strong core gameplay loop is almost a requirement for a game with dynamic level design. Since the levels can’t be overly scripted, directed and set-piece heavy, the levels can’t compensate and “lift up” a slightly weaker core gameplay. Finally, one must also consider that creating dynamic elements in a level takes time, time which could be spent polishing or making more non-dynamic levels.

 

These drawbacks must be weighed against the potential benefits. After all, the value of replayability should not be underestimated. As I mentioned in the beginning of the article, dynamic levels seem to be almost ideal for co-op games. Playing games together definitely adds something to the experience and this can help to compensate for some of the potential drawbacks like the lack of set-pieces. Adversarial multiplayer games, i.e. player vs player, don’t necessarily stand much to gain with the dynamic level design approach as the element of human unpredictability and challenge is usually enough to keep players engaged and entertained. By looking at XCOM, we can see that dynamic levels can be used to great effect in a game that isn't a shooter nor a cooperative one. And if we compare them to procedural levels, dynamic levels requires less sophisticated technology to create, but more human labor, and can offer something that feels a bit more handcrafted and unique. Ultimately, game makers need to look at the dynamic level design approach, its pros and cons, and ask themselves: is it the right approach for us?




User Feedback

Recommended Comments

The one of the problems I have found in the industry is a desire from certain people to want to control levels a lot as if that is the true goal of level design, they aren't willing to let elements go to create Sandbox designs. I am finding myself more and more drawn to sandbox design. I'm sure I would still enjoy building a linear experience if I was learning from level designers who are really experienced at making linear levels, its certainly an area I could improve in but I think sandbox design is becoming more and more standard.

Link to comment
Share on other sites

Interesting read! The images turned out beautifully.

On 12/5/2016 at 10:00 AM, Radu said:

Great article! I can't wait to apply some of these concepts in l4d3

haha, thought the same thing. Have a feeling a lot of this will be relevant to that game/mythological unicorn. I like the various ideas for directing people in/out of rooms 

Link to comment
Share on other sites

Great read! 
Generally I think developers should try and elaborate more on this.
It would be interesting if we began to see much bigger gameplay changes and tweaks for a more dynamic level design. Say for example that you could unlock bigger sections in a level like a whole street or building. That would be tougher to balance and it would change a lot regarding gameplay but I think it could be worth a shot. The only thing that is really sad which cannot be integrated in this is the heavily scripted events and/or story set pieces. 

Link to comment
Share on other sites

Well done @Puddy! Thanks for sharing. It's good to have such insights, although I was expecting a bit more detail, or discussion on how you plan around the variables, what might be hurdles, what you struggled with, how do "you" strike balance between how many elements vary.

For e.g. I really liked the concept of the keycard economy, I'm not sure to understand if some areas are still accessible but harder to reach without them (e.g. you open a shortcut to return the files).

In terms of technology, I'm wondering how the dynamic elements are handled, you have scripting tools similar to UE blueprints?

And finally a reflection on how dynamic elements could be used even in SP. I'm thinking this stuff can work to a degree in a SP experience, if instead of creating a rollercoaster like COD, one would want to give replayability, break linear parts up. HL2, the episodes in particular, have some arenas which I see could benefit from random spawning etc. Or you could create a long term objective in a level/chapter, the way you get there could vary, or could allow for different outcomes with different rewards for players.

Edited by blackdog
Link to comment
Share on other sites

Nice article.

I've always been a fan of dynamic level design. I personally like it even in MP. The only Source level I released to the public had this for CS:S. My dynamic design for MP was to cut off some paths until different number of players were on the server. Thus the same level can accommodate different crowd sizes. So I don't think that MP should be ruled out of the discussion. A common example of levels that are kind of dynamic in MP are Battlefield levels that have different boundaries or points depending on the player size (albeit set at server setup time).

The main challenge when doing interactively dynamic levels in MP is that many players are not very "aware" of the environment. So my attempts at educating with signs that changed based on available routes was not successful because I would hear players exclaim "Why isn't this path open anymore?" or another scenario of "I didn't know they could get behind me here!" So it takes some educating the players--especially the loud and self-important ones that get angry at things that surprise them. When playing and I pointed out the signs, players might respond with either "STFU!" or "There are signs?". :)

But for me, these concepts are things that I find interesting both as a player and as a designer. I personally enjoy the stress of unknowns in playing, so I like that kind of thing.

Link to comment
Share on other sites

Thanks for the comments and feedback fellas! 

On 2016-12-10 at 0:49 PM, blackdog said:

Well done @Puddy! Thanks for sharing. It's good to have such insights, although I was expecting a bit more detail, or discussion on how you plan around the variables, what might be hurdles, what you struggled with, how do "you" strike balance between how many elements vary.

In a way I only scratched the surface, but I wanted to avoid a too long article.

On 2016-12-10 at 0:49 PM, blackdog said:

In terms of technology, I'm wondering how the dynamic elements are handled, you have scripting tools similar to UE blueprints?

Hmmm. Sort of. Not nearly as powerful tho.

On 2016-12-15 at 7:14 PM, shawnolson said:

Nice article.

I've always been a fan of dynamic level design. I personally like it even in MP. The only Source level I released to the public had this for CS:S. My dynamic design for MP was to cut off some paths until different number of players were on the server. Thus the same level can accommodate different crowd sizes. So I don't think that MP should be ruled out of the discussion. A common example of levels that are kind of dynamic in MP are Battlefield levels that have different boundaries or points depending on the player size (albeit set at server setup time).

The main challenge when doing interactively dynamic levels in MP is that many players are not very "aware" of the environment. So my attempts at educating with signs that changed based on available routes was not successful because I would hear players exclaim "Why isn't this path open anymore?" or another scenario of "I didn't know they could get behind me here!" So it takes some educating the players--especially the loud and self-important ones that get angry at things that surprise them. When playing and I pointed out the signs, players might respond with either "STFU!" or "There are signs?". :)

But for me, these concepts are things that I find interesting both as a player and as a designer. I personally enjoy the stress of unknowns in playing, so I like that kind of thing.

It could definitely be interesting to explore in MP too! But it needs to be implemented more cautiously I think, just like you said it can be tricky to "sell" it.

Link to comment
Share on other sites

Great stuff :)

Payday 2 certainly deserves more props for it's amazing missions!

It has certainly been a reference for me in terms of dynamic level design (I call it procedural but whatever :P)... So much that I have actually used this technique on a project last year

What's great is that they're not just changing insignificant stuff, but instead critical gameplay elements (location of the camera room, keycards, windows that give NPCs more LOS, additional door to break through after opening the vault door, etc.)

...and players need to adapt!

It makes levels replayable, and gives a lot life to the game overall... that combined with the frequent updates, and this game is still popular 3+ years after it's release

 

BULLDOZER!!!! (famous last word)

Link to comment
Share on other sites



Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Add a comment...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.


  • Mapcore Discord

  • Mapcore Supporters

    Our incredible Patreon supporters keep Mapcore's lights on. If you'd like to donate, check out our Patreon announcement.

    Note: This is brand new! The format will be tweaked and rolled out to more pages soon.

  • Our picks

    • [CS:GO] Shore
      One of the projects I worked on after I made de_austria.

      Yeah, I know the timing is really bad (CS2!), but can't change it... 🙁

      In general a lot went wrong during this project. Had a lot of problems with the layout, but I think I found a good compromise in the end. Also learned many new things. Maybe I would even be ready to create a really good map now after all, but I think that it's gonna be the final one. I really like making maps, but it has become too time consuming over the years for just being a hobby. Unfortunately I never managed to get a job in the games industry. So that's it. I hope it's a worthy ending of 20 years of making maps for me. 🙂

      Shore is a bomb defusal map for CS:GO that takes place on Lofoten Islands, Norway. (Also have a Hostage version of this that works well imo)

      Workshop link: https://steamcommunity.com/sharedfiles/filedetails/?id=2955084102

      I took some inspiration from this place in Norway:


      Layout:


      Some pictures:
      • 18 replies
×
×
  • Create New...