Recalibrating Our Limits: Lighting on 'Ratchet and Clank: Rift Apart'

[Music] hi gdc thanks for coming out today uh we're going to get started on our talk uh this is going to be about the lighting of ratchet and clank rift apart this is a game very near and dear to our hearts and we're really excited to share just how it all came together so today we're going to run through a high-level overview of our artistic process and workflows as well as talking about our biggest challenges and successes on the project we'll start with a general overview of what lighters are responsible for and then we'll start talking about the role that lighting played in pre-production and the work that was done that laid the foundation for the game then we'll dig into the really fun stuff of going through our artistic process and taking everyone through uh standing up one of the coolest levels in our game and then we'll wrap up something uh some insight on how we spent the final weeks of the game really diving into the nittiest of nitpicks to

polish this product but before we do any of that let's do some quick introductions my name is brian mullen i have been with insomniac for six and a half years i started way back on the early days of marvel's spider-man as a lighting artist for the project and i was moved into a lead lighting artist role for ratchet and clank prior to working at insomniac i worked as a cg artist for a tv and film and i worked for a few years as a professional photographer all of which laid the groundwork for a really cool job working on video games hi i'm anna rohn i was a lighting intern at insomniac and got hired after i graduated i've been i've now been here for over two years and unfortunately not all of our incredible lighting team could be here today but it's my honor to introduce mickey mickey has also been an insomniac for over two years and does has done some incredible work on rift apart also she's just a delight and last but not least we have oz oz's

years of work in feature animation helped round out our little lighting team into something fantastic now you may be thinking what is lighting that question comes up all the time you know what it is but what does it mean well it's not just adding lights it's adding fog weather color correction and anything else that makes a picture feel finished at the end of production in games we have to wear a lot of hats compositor color artist sometimes even matte painter because of all this lighting plays a huge role in the final look of the game as one of the final steps in production it's our job to add that last bit of life into a piece whether that's in a level or cinematic we work on enhancing what's already there we make sure our characters are lit in flattering ways we add depth to an already rich world and we call attention to the best parts of a product now a huge responsibility of our job is to determine the mood of a space not

only do we need to make sure that the level is playable and visually appealing but we need to make sure that it tells a story so much of that comes down to atmosphere brightness contrast fog color all of these things make lighting nearly as vital to storytelling as the dialogue which is why lighting can take an image from this to this here we can see an approximation of what rift apart looks like before lighting and after might not seem like much but all these individual tools i've been mentioning come together to impact every pixel on the screen when lighting is done well it's not something most people will ever notice i've always said that you can build the most amazing environments and fill it with the most incredible characters but if letting's not strong everyone's work suffers like we said it's our job to bring an image to the finish line but let's start at the beginning we knew going into this project that we

were going to have a big task out of this uh but as anyone who's been through pre-production knows the only thing we can be certain of in the early days is that everything is going to change so this pre-production phase was focused on the things we knew we could control identifying our pillars our areas of concern and making a game plan let's dive into a bit of what that looked like i am going to start with a bit of a cheat here our biggest hurdle didn't actually happen in pre-production but it's hard to talk about anybody making a game in the past few years without addressing covid uh it's we had a pretty small team and it was forming right when coveted started and we were we were all just starting to work from home and it was a stressful and uncertain time no one really knew how all this was going to work and frankly work wasn't even really anyone's primary concern that said we have an incredible i.t team at insomniac games that did amazing work getting us all going really quickly on our work from home transition

and it's hard to figure out how it could have gone any smoother as for the day-to-day work uh we knew that every day wasn't gonna be the most productive some days were just gonna be tough and that was always going to be okay we knew the priority is everyone had to take care of themselves and we needed to keep our focus on that but i also think i speak for the team when i say that this really colorful silly and super fun game was kind of what we needed it was a nice distraction during a really strange time jumping back to our pre-production hurdles one of the first big ones that we knew we were going to deal with was that this was a pretty sizable game and we were gonna do it with a pretty slim crew small teams can present workload issues and you have to work through the giving people the right combination of uh or you need to have the right combination of folks in place to make sure that you're playing to everybody's strengths it's a challenge i really enjoy and i really like working with a small team it gives everybody a little bit of an

opportunity to handle different variety of tasks but it is a challenge and one that we knew we needed to take serious also the camaraderie and communication seems to be great when you have a small team hurdle number two was a pretty big one and it was one that i think anybody working in a launch title window can understand we weren't really sure how best to push forward on the ps5 because ps5 was still being worked on when we'd started so we needed to be wise about how we set our lofty visual goals we knew we were going to have caps we knew we were going to be able to push things but we weren't sure exactly where our new baseline was going to be we weren't sure what kinds of things we were going to be able to afford now and we knew that we were going to have to establish where our limits were but honestly one of the biggest hurdles that we dealt with really early in production was understanding that we needed to make a clear generational leap in quality over the last ratchet and clank game on ps4 and that game was

stunning it was lauded at the time for its amazing visual fidelity and there was a real concern that we needed to figure out exactly how we were going to improve on an already outstanding foundation in a way that shows the promise of next-gen lighting so where did we start we we needed to start identifying exactly which areas we knew we could improve on first off we wanted to make artistic advances to make the game feel more cinematic we wanted the player to feel like they were living inside of a movie instead of just playing a game second we wanted to showcase technical improvements in both reflections and indirect lighting to make the world feel more realistic and last we we wanted to expand on our post-processing to match that of feature animation levels of polish further refining and implementing bloom chromatic aberration color correction all that fun post stuff just really to dial in the polish and while we had some ambitious goals here luckily we weren't starting completely from scratch

insomniac was just coming off of the huge success of marvel's spider-man and with it came a lot of knowledge about best practices and workflows as i mentioned before we're still learning the limits of what a ps5 can do but we know what a ps4 is capable of so we started with that knowing that we would be able to push things more as production went on we started from a foundation of making sure that we followed a pbr pipeline that had a foundation and lighting that could be measured in the real world we would make sure that our contrast ratio of direct and indirect light matched real world measurements and a major lesson that we learned on marble spiderman was that the act of lighting a space tended to have stronger results when we kept things simple which we'll explain more in a bit so i've been talking about the pipeline what specifically does that mean for the insomniac lighting team we'll get into the details in a moment but essentially this is our roadmap for clean simple predictable lighting practices for our games by using these steps we're able to pace

ourselves keeping polish for the last steps in the process and instead focus on getting the fundamentals right up front the first place that we always start when we're lighting a level is exposure exposure is really the foundation of everything that we do it affects everything that we do afterwards and we need to get it locked down early not setting this correctly at the beginning can be tricky because the impact is not immediately obvious and can really start to become more obvious as production rolls on an improper exposure can cause inconsistency in glowing materials and visual effects and in the intensity of placed light values throughout the game it can also lead to situations where the player will walk into a really bright area and the camera doesn't adjust or we walk into a dark area and it doesn't get brighter essentially we need to treat this step less like an artistic or an aesthetic step and more like a technical first step that is consistent with values that we've set across the game what we're aiming for in our first pass of lighting is predictability

let's take a look at an example of the impact of different exposure settings can make on a scene in a way that isn't entirely obvious to the lighter when they're starting out we have a really simple hallway here the results in the right and left if you're only looking at the lighting are not really very different but there is a very noticeable difference in how bright this green cube is getting the reason that exposure has such an impact on materials is because essentially we're brightening and darkening the entire scene and it's bringing things like a missive materials along with along with it on the ride think of a good comparison or a good example to think of is like a cell phone screen if you open your phone up you have uh auto brightness turned off if you open this up in a dark room the movie theater everybody's going to look at you because you've got this really bright glowing screen in front of your face if you take the same phone outside it's going to be really dim we have to think about those sorts of things when we're trying to set our exposure

many of the assets that are going to go into our lighting may not come in for weeks or months so we're not going to immediately see these sorts of things until later down the road which is why we need to be predictable and the lighting that we set up we not only want our lighting to be predictable for us we want it to be predictable for everybody in the across the game once we've established our exposure we're going to set our key light intensity again this is not a value that's set for aesthetics but a value that is coordinated with the value of the exposure throughout the work on the level we like to treat these two settings as locked variables that the rest of the lighting work is balanced around whether it be light intensity fog brightness god rays you name it everything's impacted by these first two things this ensures that we're working on the same baseline across for the whole lighting team across the game and making sure we have consistency the next big step for us is to get a good ratio between direct and indirect

lighting almost all of ripped apart is taking place in large outdoor areas and in every level we're starting with a wide majestic vista shot and of course these are all alien worlds so matching earth lighting ratios may not seem necessary but i felt like after doing a lot of testing early on where we were just kind of eyeballing things and sort of trying to get in the ballpark things weren't really clicking starting from a strong place where lighting matched what our eyes are familiar with and gave us better results quicker to get it right we want to know what the real world ratio is and this is a rather simple process if you have the equipment you'll need a digital slr camera and a gray calibration card you set the camera to a fully manual mode so that it'll override anything the camera is trying to do for you including shutter speed f-stop iso all that good stuff you can refer to a photography guide for best settings to use for the time of day that you're shooting in but essentially once you've set your f-stop and shutter

speed you don't change it again so you're going to go outside with your camera and you're going to snap a picture of your gray card in direct sunlight and then you're going to turn away from the sun and snap another picture you can also just go into a shadow area like i did in the image here and there you go take these pictures into a photo editing software desaturate it and get an rgb reading and figure out how many times brighter this sunlight card is than the card in indirect light and there you have it you have a direct to indirect lighting contrast ratio for a sunny los angeles afternoon we ended up with roughly a three to one ratio here the next step is to just recreate this exact process in engine we're looking for the same three to one ratio in our direct versus indirect if key light and adaptation are locked this process should run rather smoothly if it's not it might be a good idea to either brighten or darken your sky and re-bake until you reach the expected ratio the next steps of the python is actually

placing our lights prefabs and probes the big goal for this project was to try to work simple and realistic every light needed to have a motivated light source and we didn't want to have any phantom lights on ratchet or manipulate the bounce or strength of our bounce light anyway we want it to be simple and pbr in this example of rivets hideout a majority of this room is lit by just a few very large lights while generally seen as a negative to have overlap at this extent we found that this actually caused less performance issues to deal with and it kept our setup simple after all it's easier to dial three to five lights than 10 to 15. and aesthetically we got more realistic bounce light and more natural adaptation in our exposure from then on it's an adjustments game tweaking and re-baking until you get something that you're happy with it's important that you rebate continuously the game is in constant flux and other departments are always adding things to levels and you don't

want any surprises midway through production so after you get it technically correct and in the ballpark artistically it's time for the exciting part post process you don't want to do this step too early because anything you're doing to add visual flourish now can actually complicate your initial setup and lead to technical mistakes later on but all that being said the post-process step in production is where a project can really find its voice in rift apart we were going for a more futuristic colorful look we wanted lens flares that would make star trek jealous and a color scheme that would make players excited to see more that meant adjusting the color correction to saturate the scene and pushed the contrast for a more cinematic feel it also meant adjusting the fog to soften the overall look of the piece to match the quality of feature animation we mentioned earlier that lighters are the finishers of an image and this is exactly what we mean these final steps can in the process can dramatically change the tone style and

feel of a game completely and that's why it's so important that everything is done correctly trying to polish lighting on work that wasn't built on a solid foundation is an immensely more difficult task one that takes substantially more time and is likely to never hit the quality bar that we're looking for we have an example of exactly that this is a train station in nefarious city that we've re-lit using a series of common workarounds and inefficient workflows to start out there's entirely too many lights in here and it's starting to flatten everything because we have so many lights i had to start making several of them skip shadow casting because we were running into performance issues even that didn't help with performance so i had to start having some of the lights only contribute to bounce light and then we still have performance problems so i reduced the falloff of the lights but because those lights were falling off so quickly now i'm getting less bounce light so then i'm adding more lights to get more fill into the corners

if this is starting to become our sound a little complicated and messy it's because it is uh this is a weird foundation to start with and what we're doing is we're chasing our tails trying to get a look that wasn't built on a strong foundation so at first glance you may not be able to see anything truly wrong with this scene you can see the color tone and the mood we're going for and most people wouldn't notice right away some of the issues but compared against a scene lit using the pipeline that we've talked about the difference can be pretty stark when we light something correctly it's just obvious this is the same train station lit using the pipeline we've been outlining this has fewer lights somewhere around five lights i think are taking up most of the room better performance and easier to adjust for performance if we needed to oh better performance but if we need to dial those lights we can narrow cones on four or five lights a lot quicker we have got better player pathing and

the materials all read better and bonus it also just looks prettier let's take a look at the details the image on the left is missing shadows from this platform in some ways it's hard to tell because that there's even a platform present because the contrast is missing and it's making the image harder to parse one of the primary reasons we're losing contrast here is that the light above the platform isn't casting any shadows it's a common technique to move to in the ages long battle with making sure a game is performant and in previous generations of hardware it made a certain amount of sense to do this in the past we've all had to cut corners to keep the cost of lighting from impacting playability sacrificing visual fidelity for performance but with the power of the ps5 we were determined to find other ways to get that performance back and not resort to tactics such as removing shadows from lights we lit scenes with the idea that anything that a light is touching is casting a shadow and anything else and anything less and we're losing immersion the other issue we're running into here

is player pathing the blue of this light is supposed to act as a beacon and pop so that the player knows where they're supposed to exit the image on the left does have this to some extent but the colors aren't contrasting enough and the path to get there is less clear in the image on the right we have a clear sense of not just where we need to go but we can see the platforms leading me to it and what about when we focus on our lombax friend here by over optimizing a setup we have a character that is not doing the ps5 or the amazing character team justice ratchet has lost his highlights he looks extremely evenly lit when the cinematics begin to roll in we really want to make sure that the level lighting was done correctly or the cinematics will become substantially more difficult to handle the glowing mouth on ratchet on the left is of major importance as soon as he starts opening his mouth in a cinematic it's going to look like his mouth is glowing and there's not a lot that we can do in lighting to sort of bring that back

we're all trying to figure out how to balance aesthetics and performance and all too often we find ourselves making tough decisions with the amount of lights the falloff of lights etc making sure that a pretty picture isn't creating a game that frankly no one can play but what we found is following the pipeline that we generally put ourselves in a good position to get a strong foundation of lighting that didn't have to make too many sacrifices our simpler setups were easier to tune for performance we didn't need extraneous fill lights because our bounce light was done correctly in the first place and if you look at the images above the image on the left isn't awful but it didn't quite hit the mark for a game that we wanted to show the promise of next-gen lighting with it's missing specular highlights and shadows and our characters and environment are suffering as a result with all that said we're still making a video game here and it's impossible not to run into performance problems to keep the game running smoothly smoothly each department has allotted a certain percentage of power that we can use to affect the scene in lighting's

case that's about eight milliseconds now that's not a hard and fast rule as different scenarios require different departments to take up more of that budget but it's a strong place to start taking a look at a differently incorrectly literally seen we can see the lighting is actually at 25 milliseconds here that's three times the size of our budget the reason lighting is so high here is because the scene was flooded with smaller unoptimized lights that were inefficiently placed and covering mistakes that could have been easily fixed with proper care that's why it's extremely important for us to tackle performance early and often because it affects the entire game but of course even when we do our job correctly not everything is as simple as lighting a static room if everything would have just sat still our game would have been a heck of a lot easier to light but it's not and a big reason for that is because we were making a complicated game one of the main features of the game is the ability to open a portal and enter a

completely different level that requires both levels to be open at the same time and if we aren't lighting efficiently that's too inefficiently lit levels at the same time we built a game featured around fast loading and lighting is one of the most expensive things in the game that's why it's ultimately so important that we are responsible for our own work and then look for creative solutions afterward if you'd like to learn more about the incredible portal work in the game you can check out peter cowe's gdc talk shifts and riffs dimensional tech on ratchet and clank rift apart here's a specific problem that we run into for uh with performance a lot and it doesn't have an immediate answer we did all the things we just said we kept our lighting simple we followed the pipeline there's only one real driver of light in this scene and it's the sun and we need the sun so if we've not gone crazy and placed a lot of lights why is lighting double our budget what happened here

well there's a few reasons first off the power of the ps5 isn't just going to the lighting department everything that lights are hitting in this game is getting more expensive to light ripped apart is known for its stunning environments huge crowds amazing characters and active vibrant scenes and we love them but they come at a cost in this particular scene what we are seeing is the cost of lighting a complex and dense environment without a lot in the way of optimization available for us on the lighting team but we did accomplish our goal here of getting this very lovely shot performant and let's dig into how first off our core team made incredible looking fur but fur that somehow didn't tank our frame rate for lighting we skip rendering shadows for the first shells and only render the base geo for shadow maps which makes the shadow map rendering much cheaper and unintentionally but neatly also adds a sort of sub-surface scattering effect to the fur in the deferred lighting pass we combine shadow maps with screen space contact shadows to pick fine shadows from the

individual strands this is all to say that often when we're lighting the most expensive thing that we're dealing with are the characters and it wasn't a problem that we ran into very often on ratchet and clank and it's a really lovely game it's great seeing the characters look this good and not cost a ton for our second item it's worth highlighting the the gargantuan effort of our core team in exploring and implementing more and more refinements optimizations and improvements to our ray tracing throughout production ray tracing was always going to be a major feature of this game clank is after all a walking chrome ball and you're seeing everything on him at all times but that it was done this impressively with this minimal of an impact on performance is the reason that this team was able to win best technology award at this year's dice awards and also a tech award last night at the gdc awards a major tip of the cap to this team and then finally we can't stress enough the importance of the hard work that the

environment and character teams did not only creating some of the most stunning environments and characters imaginable but by putting in the work to optimize and ensure that lods were in place circling back to this scene lighting dropped almost nine milliseconds in a matter of a couple of weeks uh due in large part to the work of people that i mentioned in the previous slides i didn't really do anything here lighting dropped nine milliseconds with minimal impact or minimal input on lighting because lighting is affected by everybody else we could have reduced shadow quality of our key light to make this work but performance was really a team effort not all solutions are going to be immediately apparent and we don't want to try to bring out the big guns too early so by working together we're able to keep that incredible ps5 quality graphics look and run in frame at the same time so now let's dive into the fun stuff making the art

following these rules and more allowed us to start with a good foundation for our lighting but our next step was taking on what we've learned and putting it into production and bringing it into our levels let's take a look at some of what we had to do it was a lot this was a massive linear game with expansive worlds and on top of that we had over one hour of cinematics taking place on 13 distinct planets that all needed to have their own unique look some of them requiring multiple lighting setups for return visits or alternate dimensions they each had to express a different mood and story while still feeling like it lived in the same universe with the same level of polish developing a look can be a daunting process that takes a village from initial ideation all the way to the finished project you're constantly working to hammer out the identity of a space to explain how this process works let's take a look at one of the levels that exemplifies not only the process but the pivots we had to take during production this is blizar

the level that underwent some of the most narrative and design changes throughout the game's lifetime at the beginning of production lisa's narrative concept revolved heavily around those green crystals placed around the environment because of that they wanted them heavily contrasted against a red monochromatic look it also took place underground meaning that the majority of the light had to come from the glow of the orange lava underneath most of our concept art was based heavily around mood and theme of a space and it's up to us to interpret that mood and make it something usable for practical level lighting early concept art while helpful in finding a visual tone doesn't always take into account player pathing that's the need to guide a player throughout a level easily and naturally and that's where things can get really tricky after we tried to match the concept art in the game we found we were contending with multiple issues here we wanted the players to be impressed with the beautiful open vistas of the level but we also needed them to know

where to go because of the monochromatic color scheme it was hard for the player to understand where the points of interest were and without another source of light it was difficult to point the player toward that next objective while still looking natural through exploration we quickly determined that we needed to add another color to break up the monotony of the lava we needed another source of light but in an underground space that can be really hard to come by so after talking to narrative and environment we determined that changing this location of the space was necessary to get the look we wanted so we talked to our ad and concept team about possible artistic avenues we went back and forth trying and experimenting with color and theme and eventually it was decided to knock out that back wall and get the key light and the sky contribution into the space to contrast nicely against all that orange after they decided on this general concept look it was our turn to bring it into the game

like we said before interpreting concept art can be really tricky even though we had solved some of those issues with color scheme we were still having those same player pathing issues the overall complexity of the environment and just the size of the space led to the player getting easily lost and frustrated above or to the side are two examples of brines and eyes interpretations on first glance they seem really similar after all we were following the same concept art we were trying to address the same issues with player pathing readability and mood but this was one of the first times in productions where we had differing opinions on how to achieve it like we said we talked a lot on this project from favorite types of pizza pizza to level direction we were constantly in communication we talked back and forth with each other about how we should tackle this issue and more and there's countless slack messages between us discussing every problem and every solution

we can't really stress how important this was both because we made a majority of this game from home and because it made our work stronger when we work together so this was my first time leading a team and i knew going in that i wanted to extend the same level of trust that i'd always had with our awesome art director grant hollis approaching feedback to this team was always a little bit less of a this is how i need you to fix this and a little bit more of us identifying issues and starting a conversation about how best to resolve in this example of blizzard it was very clear that we all understood what the problems were and we're trying to um solve it with lighting and we know we know what the the issues are this is a perfect example and a great opportunity to let someone in the team follow through on the vision that they wanted to do and it made all the difference not just because this space turned out tremendous utilizing our fog to create a strong sense of foreground middle ground

background and adding visible fixtures to add lighting that contrasted against the organic environment but also because trust had been established and it carried through the rest of the project we both talked later on after the game had launched about how cool this moment was on the project and how much it helped us grow as a team sorry about that uh with felizar really starting to take shape at a broad strokes level it was time to start digging into making each area within the level feel distinct and exciting the different districts the honeycomb mission the underground lava cave the grand vista they all work together to create to create a cohesive vision but one with enough variety in the color palette in contrast that individual spaces were able to stand on their own and that leads us to one of the more exciting and challenging opportunities on this project figuring out how to compare and contrast not just the levels with each other but the with levels that have multiple dimensions ratchet and clank is all about two

things blowing stuff up with crazy weapons and insane levels that encourage the player to explore each new world you go to is a chance to see something spectacular and that meant it stabbed after establishing one planet we needed to make sure that it felt entirely different from the rest each level had to have its own unique color scheme and identity in cases like blizzard we even had to compare some levels against their own interdimensional counterpart doing these constant checks helped us make sure that each space felt like an entirely new alien world as well as making sure that our lighting felt consistent throughout the game something really special about rift apart was how much time we had to accomplish real polish on this project as many of you know that doesn't always happen in production whether that's a triple-a game or a midterm you're often running right up to the wire but because of our faithfulness to the pipeline our constant checking of the game and the dedication of the team we had a little bit of extra time to make things as good as we could that meant we were able to go back to each balanced area and really dial in

and explore what we wanted to do it was through careful planning and many slack threads that we were able to take something from good to great when you're deep in production it's easy to get overwhelmed with design player pathing and tech technical work it's so much to keep track of that it all makes art take a back seat in your brain while the previous image was it was fine it wasn't where we wanted it artistically with this added time we were able to dial in on exactly what our key areas of focus were and put in the work to fix them sometimes that meant doing more post-process work tweaking the color schemes of specific areas to make them feel more unique other times it was fog adjustments to really dial in the mood of an area in this specific spot we needed to make a more cohesive color scheme which involves changing the emissive signs and the light they gave off for a more focused polished look later in production many of our biggest

cinematics started coming in and it became a huge undertaking that needed a lot of quick turnaround and fantastic results it was at this point i decided i was going to have azamiki two incredibly skilled cinematic lighters begin focusing on taking these major parts of the project to final they took on the significant role of not only lighting these cinematics but like we kind of mentioned before functioning almost as quality control for the final image this meant tackling a lot of things that aren't really associated with lighting like making material adjustments making fur that looked wet when it was raining animating the fur to blow around if the characters were up high or in a storm and always always always making sure that our two long boxes had great eye spec meanwhile brian and myself turned our attention to wrapping up our level work we would regularly play the game creating extremely long pages of notes for us to address as well as items of feedback for other teams then we'd split that work up knock

everything out we could and do it all over again we did this repeatedly for the final month or so of production and i think i can speak for brian when i say that not only was this some of the most fulfilling polish work i've ever done on a project but it was some of the most fun nowhere is there a better example of collaboration really bringing a space to the finish line than nefarious city one of the first levels that the player sees in the game and a level we just needed to knock out of the park this was one of the biggest challenges in the game because it had to feel moody and atmospheric but also full of life that's a lot of hats to wear for a level not to mention that the player needed to be able to navigate through all of this visual noise one of the unique things about this project was the amount of collaboration that was done we spent months passing this level back and forth between us building on each other's strengths getting ideas from one another and lending a hand too many times to count closing in on the end of uh production i handed this level off fully to anna to bring home and she absolutely nailed the space it went from looking like a great

level to just an absolutely incredible environment nefario city is one of the things i'm most proudest of in the game not just because it's a great level but because because of the challenge and collaboration it took to get there i started this project as an intern and after i finished college i was given the opportunity to come back and finish what i had started i learned more about lighting teamwork and the struggles of game development on this game in this very level than i ever had in a classroom to me rift apart isn't just a great game but it's a sign of the personal and professional growth it took for me to get right here on the stage today and it's something i couldn't be more happy to share with you so in the end through a lot of communication and hard work we were able to make something memorable together in challenging circumstances it took incredible buy-in from everyone to accept a task like this and to collaborate like we did this is for me the most pride i've had

in a project that i've worked on uh and it's because of how well everyone executed across the board environment character animation vfx core design gameplay it's it all comes together in something like this uh i think i speak for the both of us and the entire rift apart crew when i say that this was a career highlight and we hope that everyone has enjoyed experiencing it as much as we enjoyed making it so with that thank you for coming out today and we will open this up for a q a [Applause] hello wonderful talk guys really informative really appreciate it um kind of curious about um you're talking about lighting in the materials how closely did you have to work with the artists on the team to get better reflections or is

it just like thankfully everything worked out like the pbr pipeline um sure yeah that comes up all the time uh we have to work really closely together because as we said this is a team effort so uh oftentimes we'll go in on light a scene and if we see something isn't responding the way we'd expect when we're working with the pipeline we start investigating into the specular the gloss the roughness all of those material channels and if we notice something seems strange we'll reach out and talk to them about if there's an issue yeah thank you thank you first of all thanks for that fantastic talk i thought this is truly generation defining work and it is amazing so hats off that's like once again um the question i have is uh pipeline related but with respect to getting parity between sdr and hdr a lot of times like uh as we are like in our facebook development what we are seeing is that you know there is a push

for hdr from manufacturers but there isn't a lot of hdr devices out there there's a large chunk of sdr devices but if you focus on uh getting the content across following the sdr pipeline and then later magically try to switch it to hdr there can be all kinds of help that can break loose with values and things like that so did you guys run into issues with respect to you know getting your hdr pipeline figured out and getting sdr at the same time to kind of get some sort of parity and you know just be it wouldn't be just like flat flat out washed out or anything like that sure so yeah i don't want to get too much into the technical stuff because i'm afraid i'll put my foot in my mouth with what core does uh but they do incredible work getting our work from sdr to hdr um for us the way that we do it we're all working on hdr monitors um but we have uh at least for myself i have an hdr tv that i'm

launching the game on regularly playing it uh where we normally see most of the differences is in the really bright spots we have more room now on that top end so it's it's really just about validating your work regularly um i mean a big initiative in our studio is constantly playing the game and it's important for us when we're looking at that hdr quality to make sure that we're not losing details or that things aren't looking washed out um it was something that came up a lot on spider-man um so we we kind of had hammered out sort of the the issues of how all that worked going into this game already thank you sure uh hello i'm again really impressed with the talk i love it when you break things down into uh half sort of the creativity of how beautiful the colors can be and the other half is the engineering and then there's a third element of guidance um for people that don't have access to

uh top shelf talent how do we study lighting in a way with you know the the amount of work that's out there on say deviantart or pinterest for inspiration and then learning about how to tweak it in game where's that fine tuning and where's that sort of auto didactic self-training if at all possible that's a great question yeah uh so there's a planet you can get inspiration from anywhere right you can go on pinterest just like you said and just getting into the system whatever you're using and kind of going crazy is a fun place to start uh working together as a team is always wonderful to bounce off ideas of each other just like we mentioned but yeah one of the one of the best things that i've kind of done to really get a feel for lighting is actually going out there with a real camera and just messing around with studio lights and having a fun shoot day with your friends

and your team and just trying different things and exploring different lightings just to get a real feel of what it looks like when you're you're really in there it helps a lot at least for me do you have any yeah i mean for me i think when i was starting out i would look at pretty pictures and i would go i'm going to try to make that look exactly like this and i think that's how a lot of people start i think spider-man's where i learned a lot about trying to slow down and like try to do things realistically with lighting that i could measure in the real world uh because the problem i was finding a lot is i would like a dining room and it looked super cool and then i'd walk to the other corner of the dining room and i'd look back and it didn't look super cool at all um so it was about trying to make sure that i do it right uh get everything sort of set up make it boring essentially and then start trying to look at that stuff like from pinterest or google or whatever and start going okay so how can i start like nudging this closer to that instead of starting with trying to dive right into making it pretty

so then second part is there um is there a point in studying cinematic lighting yeah i love cinematography i've got tons of books sitting behind me and my zoom calls all the time because i like looking at that stuff for me a lot of it is inspiration it's uh we use some of the practices that we have we're a little bit you know we're doing things in real time so we can't do all of the things that they do uh but yeah it's definitely a huge help for me i like seeing how they use color to help with mood and to support narrative and that kind of way yeah absolutely yeah we do that all the time we didn't often do that as much for ratchet i mean we had a lot of things we wanted to pull from sort of inspirationally not so much direct you know artistically but all the time when i was creating skies or trying to set up a general idea of what i wanted we'd look up you know star trek and star wars and all the cool space stuff out there and we'd want to make it as cool as we can you

know just try and get anything we could to do yeah fantastic thanks a lot thank you hi so um i was wondering so for something like um an effects that causes a large amount of light like a large explosion um would that be more of your guys department to kind of inter work on that light and how it interacts with the world or to be more of the effects artists or be kind of like a combination collaboration between the two of you uh great question yeah it's both yeah it's a weird domain because obviously you know they want to do their effects they want to make the most incredible amazing explosions which they did uh and then they have to give off a certain amount of light so often times what kind of came out was either a collaboration or we you know just talk for advice or you know if we saw something that that didn't quite match up for what we expected we'd talk about it see why and come up with a compromise yeah it's all it's all just about collaboration but great question

thank you of course morning uh kind of a similar question i was curious about the interplay between like the design team and maybe when they're doing white box or anything do they do lighting that they hand off to you and like this is gameplay critical lighting how does that work at insomniac um i think it kind of depends uh it depends on the designer and the lighting artist but i think what i've seen is when we've moved to work from home is that it seems like in a lot of ways our communication has gotten stronger so i know just early grey box stuff now i'm already talking to designers about what they want to do and if they want to place lights that that kind of works for me generally what my focus has been is starting with an atmosphere that kind of sets the tone and mood of what they're going for so it can kind of help them sell the story that they're trying to tell and then we'll talk about where to place lights if i can do it great if they want to do their pass first that's also great we

also have a system where people can add annotations so there's like little notes in the level so i can i can see they we have a cinematic that will play here so maybe i want to craft some lighting around that or um i need them to go push this big red button over here so i'll kind of try to help guide them in that direction but um it's i'm i i don't want to speak for everybody but i feel like i'm pretty flexible when it comes to uh what how designers want to work with things it's just important for us to make sure we're talking a bunch thank you yeah great talk thank y'all um so i was just kind of curious as lighting artists and the next generation do you see any sort of like emerging trends in lighting that's coming up like dynamic shift more so than static and light mapping and things like that um i don't know about exact trends but i can tell it's becoming more like important i guess to the process and maybe that's the wrong way to put it but

before lighting was something that had to be done to put in and we didn't have the the tech to make it something incredible and now with things becoming a little bit better we can really focus in on its own unique thing and make something really incredible with what we have so i've seen a lot of a big trend people just paying more attention and then trying to get the best details out of what we have but as for specifics do you have any yeah i just i i think the big thing is trying to move more and more towards dynamic shifting things um so that we're not baking quite as much and it's going to take a while to get there i think but we're already seeing some really cool stuff across like with lumen and i think it'll just keep getting more and more refined uh but yeah like anna said i think that um lighting plays a pretty big role in selling sort of the visual component of next next-gen um and yeah i just i don't know it's exciting i like i like doing this thank you yeah thank you

that was a great talk my question is as a liking artist besides deadlines how you decide when to stop that's a really good question because yeah some things can always be better right like does the art director give you a green light this is right or wrong or uh so the art director does it did they give us a green light as og are asking um yeah so what we do is we will we'll get our concept our uh you know our initial sort of this is what we're going for um we'll start lighting the the scene based on that uh we'll have meetings with art director um the way that grant hollis likes to work is he likes to see what we're doing and start doing sort of paint overs of where he thinks this thing could go for final um and we just sort of do that back and forth throughout the process as far as like when we know when to stop like the game has to ship at some point i guess

i mean like anna pointed out in one of the slides we had all these notes and it was uh it was great because we had kind of hit a point in the project where we actually had time near the end and it was sort of like okay we don't have to stop we can we can actually just keep going with this thing and and anna was awesome because you just kept playing the game and writing down all these things and um yeah i mean we kind of worked up to the wire i remember that the day we went gold too we were just working right up to the the time where they were like okay you got to stop now yeah it was awesome okay thank you sure thank you okay okay well if there aren't any more questions we can go ahead and wrap this up i think we're going to be hanging out at the wrap-up room i think it's across the way so if anybody wants to stop by and say hi that'd be cool but yeah again thank you for coming out today

[Music] you