State of Production #15
December 17, 2020
Hello everyone, my name is Tim Perreault and I’m the Production Director at New World Interactive. It’s an exciting time here at NWI; in addition to our latest release for Sandstorm, we just celebrated our 10th Anniversary. From all of us at NWI, thank you for being part of the last ten years and we all look forward to building and gaming with you over the next ten. There are plenty of things to cover so let’s jump in.
Key Points for State of Production #15
- Operation: Cold Blood
- Hotfixes and Known Issues
- New Map: Bab
- Toggleable Optics
Operation: Cold Blood
Our latest update is now live, bringing you a ton of new content. In addition to the content you did see, we were working on a lot of behind the scenes content aimed at facilitating future development. This included an initial push on new Quality Assurance tools, analytics, and a new Dashboard that allows our community team more control over our message of the day and other in-game communication. Our goal with these tools was to provide you with a better game and better experience as we continue developing Insurgency: Sandstorm.
Hotfixes and Known Issues
We’re still looking into the known issues listed at the bottom of our patch notes as well as general feedback on performance since the last update went out. We’re currently looking into visibility problems with STANAG magazines, two loadout issues (loadouts presets not properly activating when selected and equipment from the second and third explosive slot not correctly being available), and an issue where it’s possible to have an incorrect localization partially active.
We’re also closely monitoring feedback about the changes being made to AI. With update 1.9 we made a significant change to the availability of weapons in co-op. This change was rolled back with our latest hotfix. You can read more about that here.
New Map: Bab
In keeping with tradition, I sat down to chat with some of our team to discuss our latest addition to the map lineup: Bab. This time I spoke with Jeroen van Werkhoven, Level Design Director; Adriana Matic, Level Designer; and Brian Birnbaum, FX and Lighting Artist:
Tim Perreault: Where did the idea for Bab come from?
Jeroen van Werkhoven: Bab came from the early days of Sandstorm. We had just started working on the game and we had an idea for a citadel type map and we started working on that first, but realized we needed more space at that time for a mode we were prototyping. We started expanding it and added this whole outside area with a town. Eventually that turned into Bab.
TP: What was the most challenging part of building the map?
JvW: I think the most difficult part was that because the map was in development for such a long time we really needed to figure out what we wanted to do with the gameplay. The map was massive compared to our current largest map, and we just couldn’t make it work. The journey of going from the town to the citadel was such a long run that by the time the player made it to the citadel they were already exhausted. Making it to the Citadel should feel like a reward, but it didn’t feel that way, just more map. It was not a lot of fun either because there was only one pathway through the citadel once you made it there. It felt like we forced 2 push layouts in a single layout. We tried to fix it but decided it just wasn’t working, so instead of trying to add more we decided to cut. To make the gameplay more fun and cohesive we removed the inside of the Citadel.
Brian Birnbaum: I don’t know how many iterations we had to do of the interior of the citadel to get it to work. I just remember so many pathways leading into it and all the tiny corridors and stairs, it took a lot of changes to get to where it is now. I can also remember in the early days when we were just starting to test the map we managed to get Keith Warner our CEO playing with us. The first time he played with us on this map, Mikee was in a technical driving down the hill and he ran over Keith without a second thought. We still joke about that today (Editor’s Note: Mikee refuses to comment on the incident on the grounds that he might incriminate himself).
Keith Warner: No worries about incrimination but I’d like to know where those files are so I can delete them. No CEO should ever get run over by his team and I want to destroy the evidence!
TP: What was the approach to building gameplay around the medium to long range engagement distances we were looking for with this update?
Adriana Matic: For Bab, it was all about cutting specific sightlines. You wanted to keep the map open but also not let snipers be super OP. So it was kind of like checking angles and anticipating where those camping spots could be and limiting their sightlines so they weren’t over 180ᐤ and making good use of soft cover that just obstructs views. There’s also lots of hard cover, but using trees to block the view of a road was the primary approach.
TP: Do any specific areas of the map come to mind that presented a greater challenge?
AM: Yeah, so there’s this construction overlook that’s across from the ticket building and when we were testing players were having a lot of trouble because as they were moving down towards the ticket building they were just getting devastated moving through those lanes. There were two sort of higher vantage points where you could always aim down so you couldn’t really use walls high enough to block that. The solution ended up being using tarps and cloth and more foliage to kind of obscure the sight line to make it more difficult to clock players moving through.
TP: What did we do to optimize the performance of the level?
BB: So we did our usual optimization pass where we merge and instance as much as we can in the level but we were finding that performance wasn’t where we wanted it, partially because Bab is such a big and open map. We had to go through and limit as much that was rendering on screen as possible. We paired down all the similar textures in the level to reduce the load but we ran into a problem replacing textures. Unreal actually doesn’t have any easy way to replace a lot of textures and materials at once, especially instanced actors. There isn’t any way to replace all the materials on instanced meshes. So we ended up creating a tool specifically for Bab to try and lower Material and texture usage to lower the memory count.
JvW: We worked with the object view distances a lot too, because the game has to calculate when the player can no longer see an object and then stop rendering it which takes up CPU cycles. By manually setting all those distances we offload some of the computing, so we could look at all the props and say “Okay, after 50 meters, we want this to fade out”. By making that decision we could free up the CPU cycles to allow the game to focus on other things and ultimately keep frames higher. We’re still working on more optimization as we’re seeing feedback from our players using a broader range of hardware and figuring out the pain points.
TP: How did you build the unique feeling and atmosphere on Bab compared to our previous Sandstorm maps?
BB: So originally, we didn’t have too much of an idea of what we wanted the level to look like until we sat down to discuss it and got a paintover of the visual look. That visual look was a very cold and blue sky day.
We decided to give things nice visual cues and colors with the kites. Originally it was a very overcast day. When the concept art was created we didn’t think that we would be able to make the level look like the concept art. But it actually ended up looking a lot like the concept art in the end. I had to create a new blue cold skydome texture. It took several iterations to make the clouds and the blue in the sky to look right. I went through the level to add snow blowing off the trees and rocks, and even just on the ground because those little touches add some real visual interest. There are a lot of little snow effects to make the work seem alive. The hardest thing to create was actually the kites or at least it seemed hard at first. But interesting fact, they are Niagara Particles. Not blueprint or anything else. In Niagara you can simulate wind and basically all I had to do was do what a kite does in real life and tie a string to it.
TP: Bab is an Arabic word for door or gate, so it begs the question: what’s inside the gate?
BB: … Giant snowman.
Maybe one day we’ll find out about this Giant Snowman, but on that ominous note, we’ll move on.
Operation: Cold Blood introduced the long awaited addition of toggleable optics to the game. I sat down with Lead Game Designer Michael “Mikee” Tsarouhas in the Cold Blood Tavern to talk about the effort that went on behind the scenes to make that happen.
Tim Perreault: Toggleable optics have been in the works for a while. Can you tell me a little bit about the history of toggleable optics within the game, where the idea came from, when we finally started working on them, and how we were able to finally release them?
Michael Tsarouhas: Yeah, this was one of the more interesting features in Sandstorm I think. It’s one of the things we originally wanted to do when we were first developing the game, but for production reasons we couldn’t. As people know, on release we had 2x and 4x optics that physically speaking would be able to be switched between different configurations. For a 2x you could move the magnifier away, and for some 4x optics you could just use the iron sights on top. We knew we had all the capability, but production-wise we just couldn’t really do it. We didn’t have the resources or the time to get it ready for release in 2018. After the release a lot of players were requesting it, which we kind of expected. We tried to get it in an update in 2019, we put it in our roadmap. But it was just so challenging with a lot of the other stuff that needed to take priority and a lot of the other ways that the game needed to grow. We tried recently to get toggleable optics into our last release, Operation: Breakaway, but it really wasn’t ready. It needed a lot more attention and time to make sure we did it right. Ultimately, we added nine new toggleable optics and updated eight existing optics. So that’s 17 new toggleable optics that came in Operation: Cold Blood, and there was a lot of time and design and technical limitations and iterations that had to happen to get to that point. It was no small task.
TP: What were the biggest challenges with implementing toggleable optics?
MT: Oh god, everything man. Like, toggleable optics sound simple, like, you have two different ways to aim your optic, way one and way two. Sounds easy, but it was very complex. Not only were we doing it post release while we were trying to manage all the other things in these content-rich updates, but there were so many things we had to do to do it right. We had the following conditions: 2x optics that existed in the game already that needed to be converted with new animations. We had new 2x optics that needed to be created like the 2x OKP-7 and then needed animation. We had the optics that existed already with the iron sights on top that needed to be converted, some of which were weapon specific. Then we had the 4x optic with the red dot sights on top, and these needed new animations but different ones from the 2x optics. Then we had the new 6-3x optics that needed to be created with unique animations. So we had all these different types of optics that needed unique animations, and you had conditions where some were already in the game but needed to be updated to be toggleable. With all these different things needing to happen the biggest challenge was the communication and trying to make sure everyone was on the same page. Then there was the challenge of what to call them in-game. What’s an intuitive way to explain how these work through the loadout UI so players immediately understand what they are. The technical implementation was difficult as well. We had bugs where you’d aim an optic and it just wasn’t there, or the field of view was weird, the screen was blurry when it shouldn’t be, or it was just straight up showing the wrong magnification. On top of it all we had the normal challenge of gameplay balance. We had to iterate on these, we couldn’t just put them in and call it a day. We had to make sure they felt like they had utility and that they felt like they were balanced in terms of sight picture, transition times between the options, supply cost, fair trade-off compared to choosing other optics, and availability per weapon. So there were a lot of different factors.
TP: So speaking of iteration, what about the toggleable optics is very different from when they were first testable to what ended up being released?
MT: Some of the biggest differences were how the 6x to 3x ones evolved and how the weapon specific ones evolved. The 6x-3x were originally 7x-4x. We tried this in testing and almost unanimously the testers felt like they were not useful and not really effective as optics. This was because so many of the engagements were so close and you were just switching between two magnification levels that were both designed for further away, so it didn’t feel like there was a lot of value to it. When I first saw the feedback I was a little alarmed and was just thinking man, I wonder if we’re really going to have to rethink these. It’s good feedback and it makes sense because the game, broadly speaking, is mostly close and medium range engagements. But the way we solved it was really simple. We just knocked both magnifications down one level and then ran another playtest where almost all the feedback shifted the other direction. The other piece of this was that as we were developing we realized there were other optics that could be toggleable like the M1 Garand for instance. When we were doing research on the M1C scope we realized you could use the iron sights as well as the optic, so we stretched ourselves a bit in getting it work but we felt like it was important because it makes the weapon unique. It also helps with balancing other weapons that are seen as being cheaper or less effective like the G36K or the AUG A3. It’s great to give them a special upgrade that also gives a unique sight picture, etcetera. There were other ones we would have loved to have done. We would have loved to have done a toggleable optic for the 4x PU Scope but we just didn’t have time for it. We still got those other ones in.
TP: Any funny anecdotes about the development of toggleable optics?
MT: I was really shocked when I saw that we had 17 of them total. I remember when we were writing up the patch notes and being taken aback and even a little confused thinking wait a second, there’s no way we have that many. But I went through the list and realized like, yeah, we really do have that many. I think it’s kind of a testament to how big of a deal this is for the gameplay and how big of a challenge this was. But we did deliver on it. We knew it was something people wanted. And this was no small thing, but our team really wanted this too. We all knew that you could push a magnifier out of the way, I know you know Tim, and so this was an important thing to get to work so we pursued it and we delivered.
TP: Any final thoughts about toggleable optics?
MT: I think this is a good example of something in game development that you need to acknowledge can’t be rushed. You need to think about it from all angles: the gameplay, the technical implementation, the balance. How it might change how people enjoy the game period. All those things needed to be considered and while all those things took time and while we didn’t do it as quickly as we would have liked to, I think it’s saying something about our team and our passion that we put in the effort and we finally got it in. And we did it the right way and we did it in a big way… Again, 17. That number bears repeating.
While it has been an exciting year, the team and I are happy to be winding down. Over the next few weeks a lot of our team will be taking much needed time off. We’ll be coming back fresh and ready to get started on our next major update as well as a special surprise that will be coming out early next year. In our January SoP I’ll be looking back at 2020 as a year in review, things that went well, areas for improvement, and general lessons learned. To close out the year I’d like to wish you all happy and safe holidays and until next time, keep your stick on the ice.