State of Production #20
Hello everyone, my name is Tim Perreault and I’m the Production Director at New World Interactive. We’re excited to have just released Operation: Exodus, our latest free update. Let’s dig right in!
Key Points for State of Production #20
[h3]Spring 2021 Player Feedback Survey[/h3]
Every so often, we send out a survey to the community to help us better understand how our community feels about the current state of the game. You can fill out the survey here. It will be open until the end of the day on June 4, 2021, so please take this opportunity to let us know what you think.
[h3]Operation: Exodus[/h3]
Our latest update is available to play now. We’ve added new guns, new cosmetics, a new map, and a new game mode available on our new map and many of our existing maps. When you win your first match of Survival Mode, you’ll be rewarded with a set of character cosmetics for both Insurgents and Security that reflect the battering you take fighting your way through to an extraction point. For full details on the update, check out the patch notes here.
[h3]Survival Mode[/h3]
I’ve invited three of the key developers of this game mode: Adriana Matic, Level Designer; David Poeung, Gameplay Programmer; and Jason Campbell, AI Programmer to talk about the experience of creating Survival Mode.
Thanks to Adriana, David, and Jason for sharing that little behind the scenes look at what it took to bring this new mode together.
[h3]New Map: Citadel[/h3]
Citadel is a map long in the making, which has been worked on by many of our developers. I’ve invited our Level Design Director, Jeroen van Werkhoven to speak about the map and what it took to finally get it ready for release.
Like Jeroen said, every level designer on the team has had a hand in this map, so big shoutout to all of them and our art team for bringing this map to life!
It’s been a long SOP, but they usually are around releases. Until next time, keep your stick on the ice.
Key Points for State of Production #20
- Spring 2021 Player Feedback Survey
- Operation: Exodus
- Survival Mode
- New Map: Citadel
[h3]Spring 2021 Player Feedback Survey[/h3]
Every so often, we send out a survey to the community to help us better understand how our community feels about the current state of the game. You can fill out the survey here. It will be open until the end of the day on June 4, 2021, so please take this opportunity to let us know what you think.
[h3]Operation: Exodus[/h3]
Our latest update is available to play now. We’ve added new guns, new cosmetics, a new map, and a new game mode available on our new map and many of our existing maps. When you win your first match of Survival Mode, you’ll be rewarded with a set of character cosmetics for both Insurgents and Security that reflect the battering you take fighting your way through to an extraction point. For full details on the update, check out the patch notes here.
[h3]Survival Mode[/h3]
I’ve invited three of the key developers of this game mode: Adriana Matic, Level Designer; David Poeung, Gameplay Programmer; and Jason Campbell, AI Programmer to talk about the experience of creating Survival Mode.
Tim Perreault: What was the biggest challenge developing AI for the new mode?
Jason Campbell: For the AI in this mode, while it’s still an objective based mode where you’re capturing objectives, unlike others, there’s not a fixed objective ahead of time where the AI will be assaulting or defending from one side. The players can approach the objective from any side and we wanted to have them feel like they’re constantly being chased in the streets, not just assaulting a fortified objective. We needed to have our AI’s approach to the objective be entirely different which led to developing a new spawning system. Instead of using fixed spawn points, either on the objective when defending it or around it when assaulting it, we wanted AI to be able to dynamically spawn anywhere in the map based on where the players were and where the current objective was. While it may sound simple to just spawn enemies nearby but out of sight, this system actually had a lot of complications. One of the big complications was performance. There are so many possible places for the AI to spawn that when we tested where they should spawn, it resulted in a lot of calculations and permutations that really slowed things down. We needed to build a way for the AI to make better assumptions about where they could go and this is where we started building out our new spatial awareness system. Right now it’s primarily used to make sure the bots don’t spawn within line-of-sight of the player, but it’s also something we plan on using in the future for enabling additional tactics. Getting a working version of that wasn’t terribly difficult, but making it fast and making it good took a lot of iteration. There were multiple times during early playtests where you’d see bots spawning on a blocking volume in the sky or on top of a billboard or who knows where. It took a lot of iteration to get the system working.
TP: Speaking of iteration, what was your process like?
JC: Yeah, it’s interesting because it felt really fun right at the beginning but it also felt really broken, like I mentioned with the AI spawning in the sky, plus we needed to spawn tons of bots to create enough pressure, and even then they would sometimes just wander away and get lost. The process basically was about looking at what was going wrong and then investigating and fixing, for example why did the bot spawn in the sky? What could cause it to do that and how do we eliminate the behaviour? How far from the players should the bots spawn? How can we hint to the AI where they should be without giving away the players’ location? A lot of the difficulty of the mode is also about how many bots are spawning and when they’re spawning so we needed to add a lot of configuration and rules around spawning to make it easy to adjust by the designers. Once those tools were in place, it was a lot of playing and tweaking and trying again, building out something new if we found another gap.
TP: When development first started on the game mode, where did you start and at what point during development did it start to feel like “Hey, this is actually really fun!”?
David Poeung: So at the beginning, the original pitch for the game mode was that we wanted something that fully utilized the entire map, which none of the other modes did, that was really difficult and made you feel like you were always in danger with constant tension and pressure. The idea for the mode was that there was an overwhelming amount of bots attacking you at all times while you were running around trying to capture objectives spread all across the map, the constant danger really encapsulating the idea of “survival”. So that was the basis that I started with. In the initial iteration there were 32 active bots at all times which was just way too oppressive. As we started playtesting, we began toning it down, trying to find a balance where it was both difficult and fun, so we dropped the number of bots and made the spawning more intelligent and added variability by randomizing objectives and removing the linearity. I would say the moment things started coming together was probably about two or three playtests in when we were having a lot of people on the team, who had never seen the mode, try it for the first time and it clicked for everyone. The feedback was extremely positive and everyone agreed that the mode was instantly fun despite being very rough. So we had a really solid foundation early on and it just took a lot of iteration and refinement over time along with many quality of life additions to strike a better balance on where we wanted the difficulty and tension to land.
JC: One of the things we needed to iterate a lot was balancing the bot pressure from after you captured a point and were trying to resupply from the weapon crates. Early on, the bots would continue attacking you right after you captured the objective so you could capture, respawn your team and, then very shortly after, get wiped by a suicide bomber that had gotten into the room. There were also points that if you were the last man standing, you just couldn’t clear out the point quickly enough because the bots would keep coming.
Adriana Matic: Interestingly enough that really impacted the objective design. A lot of the survival objectives are pretty small because of that, it was an attempt to make it a smaller area to clear so it couldn’t be contested as easily.
DP: One of the other things that had a huge impact on the design was how different Survival mode was compared to other modes due to the high degree of randomness. It made for a lot of new considerations when testing. With how static the other modes were, things could be handcrafted and chosen manually. But with randomness, there were a lot of considerations around the mode’s flow based on the limits of the randomization. Just the range of possible distances between objectives was its own conversation. We had a lot of discussions around how far each objective could be from the others, whether objectives with a lot of obstacles between them should be closer, if the last couple of objectives should be further apart to make the end more intense, etc. We also spent a lot of time trying to figure out which combination of weapons and attachments felt good to pick up, fiddling with the spawn rates of specific weapons and upgrades to ensure that you didn’t always get the exact same weapons every time you opened a crate or a series of bad combinations. Same with the enemy types, how often melee and bombers should spawn was very contentious for a long time and we nearly cut them from the mode entirely at several points in development.
TP: What were some of the challenges with building the layouts for Survival Mode?
AM: So there were probably three main things that were interesting challenges. First off, Survival initially had two objectives activating at once, but the distances between those objectives would be extreme. We also had a linear layout like the kind you would see in Checkpoint, and unfortunately it just wasn’t very fun to play so we ended up with clustered objectives around the map. The other challenge was that in Sandstorm, layout design is kind of based around buildings, but in Survival the actual building doesn’t matter so much as the path that players are taking through the level. It kind of forced a change in thinking about how to design these layouts. We needed to ensure the player wasn’t just going straight down the side of the map for example or off to the middle and just stuck there all the time.
Secondly, for certain maps, Survival just doesn’t jive very well. So if you have a small map like on Ministry where there’s just not a lot of paths around and it’s very laned or if you want to play on super hard mode play a map like Bab where you can’t take a step outside without getting cracked in the head immediately. So we had to weigh a lot of our maps against that and decide if they would be included or not.Finally, the weapon supply crates. There were 480 crates and they had to be moved… a lot. It was also challenging to find places to fit them once they were no longer toilets, but it was fun.
TP: Toilets… What’s the deal with toilets?
JC: We needed a prop that you could open!
DP: From what I can remember, we wanted to differentiate the weapon crates from the supply crates. We didn’t want players to open one and expect to change their loadout or get confused by the two. It needed to look distinct. The problem was we didn’t have a prop early in development and it was going to take our art team a little while to put something together so we needed a placeholder, and specifically a placeholder that was able to open and close. Aside from the supply crates which we didn’t want to use, the only object we had at the time was a toilet, so our Technical Director coded them in as toilets. It was kind of a sad day when they finally got replaced by the real models, even though the final crates ended up looking fantastic.
TP: What were some moments that stood out to you during the development process?
DP: I have one that I remember very distinctly. I think it was right after we added the MR73 and Desert Eagle but they weren’t properly implemented. They were kind of wonky in a number of ways, but specifically their physics were really off. We were messing around with them in Survival mode playtests and we found that one shot from either gun would send your target flying in the air, so we had a lot of fun just launching enemies across the map.
JC: I remember when we were playing around with objective order early on and we were testing on Outskirts. We had just made it to the penultimate objective and were heading to the extraction point but it was a five or six hundred meter run and everyone on the team died in the first two hundred meters except for our Community Manager fatherHank. He made the rest of the four hundred meter run on his own just booking it down the side of the map over mostly open terrain getting shot at left and right but he survived and made it to the point. We thought maybe he’d had God mode on but he swore it wasn’t and I’m pretty sure he died shortly after actually arriving at the extraction point. It was an epic run!
AM: When we started doing the bigger internal tests, I remember once being on a team with one of our producers, jayvereal, and we kept getting shot and couldn’t figure out where from. We turned around and finally realized that this bot was prone on top of a billboard like across the map. And like, all I see is jayvereal crawling past me prone and she managed to sneak past all the bots. What a champion. She killed the bots for me and I was like, thank you!
TP: Let's talk about the extraction sequence. I know it went through quite a bit of iteration.
DP: Yeah, the extraction sequence, despite how prominent it is now, wasn’t part of the original design for the game mode. It actually came in about halfway through the development process because we felt that the finale to the mode was a bit anticlimactic, aside from more bots and a larger variety of special bots there wasn’t anything memorable about the final objective. The mode was just going from objective to objective, the last one is a little harder and then it ends. We wanted something more exciting and cinematic at the end to pay off all that intense pressure and tension that had been building for the whole round. During one playtest, we had an epic last stand type moment happen on the final objective and we wanted to capture that feeling and reproduce it in every round of Survival. Then someone suggested “What if we added a helicopter extraction to that?” So we made this set piece where you’re pinned down by enemies on the final objective, then a helicopter flies in to assist and once you drive off the attacking bots the helicopter extracts you and your team, rescuing everyone and you survive.
AM: When I first heard the idea, it was presented to me as the helicopter is going to land and it was an interesting challenge because these maps were not built with a helicopter landing in mind. I had to find cinematic locations for a helicopter to land but in the end the helicopter doesn’t actually land! Instead, we have this cool cinematic scene play out. One funny thing that happened during early development of that scene was you’d be loaded into the helicopter and for whatever reason it would shoot you like ten thousand feet off the map. We fixed that.
Thanks to Adriana, David, and Jason for sharing that little behind the scenes look at what it took to bring this new mode together.
[h3]New Map: Citadel[/h3]
Citadel is a map long in the making, which has been worked on by many of our developers. I’ve invited our Level Design Director, Jeroen van Werkhoven to speak about the map and what it took to finally get it ready for release.
Tim Perreault: When did we first start working on Citadel?
Jeroen van Werkhoven: Citadel started very early on in the development of Sandstorm. It was part of the first batch of levels we created.
TP: What were some of the challenges you and the team ran into building this map?
JvW: One of the biggest challenges was figuring out the gameplay and linking it to an ancient city. We knew that the answer was closer to a CQB experience, but we still wanted to offer enough opportunities for longer-range weapons. We found a balance for the first part of the level by using vertical gameplay. The ground level is more shotgun or SMGs oriented, and the rooftops (high-ground) are more open, long lines of sight where rifles and snipers are a must! Also, machine guns are great to use on the rooftop to provide cover fire to the players advancing in the alleyways.
The second part of the level is more trench-warfare but uses a similar approach. The players in trenches are better off with close quarter weapons, and the players that are going for the high-ground have different spots to snipe from to protect their buddies in the trenches.
TP: Why has it taken so long to get this map to the point where we’re releasing it?
JvW: It was in production for a long time because we wanted to create something unique that feels like an ancient citadel. It went through so many iterations and hands; every level designer worked on it in some capacity. The map started more as a symmetrical layout; both sides were identical in feel and gameplay, very close to an arena type of map that you would find in the Quake or UT games.
After many playtests, we concluded that it didn't feel like a citadel and wasn't very realistic compared to an actual Citadel in the real world. We put the level in the 'freezer' for a while to focus on the other maps first that we knew we could deliver.
Once we picked it up again, we knew we wanted to capture that feel of an ancient city inside a fortress and made the map a lot less symmetrical, more organic. Later we highlighted the history of an ancient town by turning it into a museum, which also adds to the storytelling of the environment.
We also had to figure out the art direction for this level, how can we capture that ancient feel, and the answer was more or less the new survival mode. We made the atmosphere in the level more dramatic. Started fires, literally by barricading the exit routes of the level and setting the barricades on fire. We added more barricades and fires throughout the level to bring that gritty, warzone feeling more forward.
TP: How was the collaboration with the art team different from when the map was started to now?
JvW: When we started on Citadel it was very much a level design only project, figuring out the gameplay. We did an early art pass, but the art direction was very different and minimal compared to our levels today.
When we picked up the level again, we thought about the art direction much more thoroughly, and the survival feel. The level design and art team had to work closely together to avoid breaking the gameplay, while adding more atmosphere. We didn't want to make it too dark, but at the same time, it should feel different. Also, the map was initially going to be during a heavy winter. But we found it much more interesting to change the setting to late winter, early spring; the gritty look, overcast sky, audio all add to the survival feel.
TP: What is your favourite part of the map?
JvW: The amphitheater, because it's such an iconic element, both for art and gameplay. And it's the perfect location to extract with a helicopter.
TP: Are there any moments from development or testing on Citadel that stand out to you?
JvW: The first time we tested the very first blockout of Citadel many years ago. I knew we were on to something special. It just took us a long time to figure out how to make it fun.
Another moment was when we tried the first version with the ancient town in place. All the dots started to connect, which led to the idea for the final area of the level. The Excavation site and trenches as well: playing that area of the level makes it feel like you're taking part in a war movie.
Like Jeroen said, every level designer on the team has had a hand in this map, so big shoutout to all of them and our art team for bringing this map to life!
It’s been a long SOP, but they usually are around releases. Until next time, keep your stick on the ice.
Although it's (sadly) been fixed, I have to shout out the dude who set the entire first act of Shrek as his in-game name.
I just love this shot. Shout out to