Subliminal
Star Citizen Live

Star Citizen Live Gamedev: Sense of Sight

19 March 202101:01:071,820 linesWatch on YouTube ↗

Transcript — every row opens the video at that moment

  1. 00:00:03

    hello everybody welcome to another episode of star citizen live game dev i think it's our first game dev episode of the of the year actually uh i'm your host jared huckabee and we are joined by members of our weapon content team today to explore a little bit about their process now if you've never seen star citizen live game dev before uh it's a special variant of our friday show where we take about an hour and we hang out it's not the usual kind of q a that we normally do you can still submit your questions live in the chat we'll you know if if they're applicable you know i'll pull them out and i'll send them to the group but we're gonna we're basically gonna sit here we're gonna watch we're gonna watch jake jake do his work and then we're gonna we're gonna we're gonna just rip it up while it's happening it's just it's not gonna be good it's

  2. 00:00:51

    not gonna be uh let's take a let's take a minute and introduce folks here uh jake since i already called you out we'll start with you who are you and what do you do for star citizen um i'm jake taylor and i'm the one of the weapons artists for cloud imperium and what uh what kind of weapons have you had do you usually work on is it just exclusively ship weapons is it handheld weapons uh it's primarily handheld weapons that's my priority i think i've made one ship weapon i do bugging with them as well but outside definitely fps weapons and gadgets and things like that all right uh and corey uh no stranger to the short uh to the to the shore you're going straight through the show hey we're still getting going here it's early in the morning

  3. 00:01:37

    uh who are you and what do you do for star citizen uh i'm cory banford i'm lead prop artist i also kind of do lead stuff for the weapons team so i look after those guys as well to the shores of scl yep they're making fun of me in the chat now uh that's how you know it's live there's always somebody that's like i think this is pre-taped i'm like no it's life it's life well maybe we do this on purpose to make it look like it's not think of how much better it would be if it were pre-taped all right so all right so uh like i said today is a game dev episode uh that's where we're gonna explore our process we'll explore a little bit of our process see how our how our uh developers work so jake uh i actually don't know much about what we're doing today so why don't you tell us what you introduce us what are

  4. 00:02:25

    we doing today sure uh is my screen shirt currently uh somewhere there there it is there it is cool so today i thought we could go through making this scope it's kind of like something small we could probably rush for it in like an hour maybe um and then i'll jump into engine and we'll start firing a pistol with it so i'll go through like the modeling process block out probably mostly and then apply some materials and then shoot some stuff with it well some targets right now this will be an abbreviated process as as you mentioned we only have an hour for these shows and it's very few things in video games can be built from start to finish in an hour but we'll explore some of the process um we'll jump ahead of

  5. 00:03:14

    here and there we might not cover every nitty gritty detail uh but yeah so uh what is this we're in maya right now yep we're in mia i've got some material boxes just set up here because i thought it looked pretty so i'm just doing that and this thing is what i'm going to be working on so this is kind of the um the standard lock that attaches to weapons they're all the same so i've already got that in engine um in maya and i'm just going to start blocking out this so normally what i do that's what i think everyone should do is just figure out those those uh large forms so i'm kind of seeing this kind of rectangle here this weird front bumper looking thing and like the site so i'm gonna start there and then just work

  6. 00:04:02

    my way in okay now not to bury the lead here because you know not everybody is uh super familiar with every aspect of our game this looks like some kind of uh weapon site a holographic site probably yeah yeah yeah this is just um times one uh yeah so you can see the manufacturer on it as well so vault so we have different manufacturers for our weapon attachments generally have different styles um i don't actually know how many volt uh attachments we have in the game at the moment i don't think there are any or any bolt attachments we just added our first the the was the lightning gun vault uh no that's uh the next the next lightning gun is for the next level the yes the the vault parallax is the i think we might have shown that off um

  7. 00:04:51

    yeah i think i think that comes from the scope so that we'll that will be going in i believe yeah that's that's currently blockout i looked at that today actually yeah briefly so you're just going in on the big blocks at the moment yeah so i've got that square in this weird front bumper and now i'm just figuring out what kind of angle i want this to be uh for folks uh watching at home who are mentioning the the frame rate uh screen sharing or resume has two wonderful options one where it optimizes for frame rate but makes everything blurry in order to get the frame rate up or one where it presents everything crisp and clean and nice so you can see everything but then sacrifices a little a couple of the frames and since we're not like you know running and gunning through

  8. 00:05:39

    a maya level right now uh i opted for the uh let's let them actually see what they're looking at option yeah uh should i go slower no no no you're fine cool all right i think coming around the viewport is something i just do like ferociously oh yeah we're doing this like constantly whenever whenever you whenever we're doing art reviews and stuff and you're like okay show us your work and then you're over teams and especially like covered at the moment everyone working remotely that shows you're working oh yeah i'm just working on this and then it's just like literally just motion blur because they're just panning around yeah let's slow down slow down show me what you're working on yeah because because we see it like all the time we're working on it for hours we understand like every nook and cranny so we're like yeah we get it and it's also important

  9. 00:06:28

    to yeah to see it from different angles to make sure it's reading well like silhouette wise yeah so so you're starting with you know some very basic you know your fundamental shapes you're just trying to you you're basically recreating uh the silhouette here yeah i would say do you always work in that manner or do you ever just like drill in on one small part and build that and then build another small part and then build another small part uh i can't like not this early no but there is a time where like let's say like all this is blocked out but then i'll just hone in on the detail here because i think okay if i hone in on this this isn't really gonna change it'll help me inform other areas so i'm

  10. 00:07:16

    just going to hone it in sometimes that's beneficial sometimes you realize that you wasted a bit of time but that's the nature of it really yeah because obviously like the more detail i go in now without doing those kind of broad strokes like you're painting if i need to go back it's more work so like a painting really you go through those broad strokes and then slowly hone it in yeah the more geometry you throw in as well the harder it becomes to make like large changes to the asset so it's really important to start with the big forms really primitive uh we got a question we got a question here from bad panda uh now this is this might be a question for the weapon feature team as opposed to the weapon content team but maybe you guys know

  11. 00:08:03

    um has there been any talk of scopes with thermal vision thermal vision i'm gonna say night vision i know it's on a list i don't know when and how and and where yeah i think it would be really cool to be honest uh whether it's actually gonna be done anytime soon unfortunately i'm not too sure yeah basically um there's a table for all of the the scopes and it contains like what features those scripts will have some of those features haven't been implemented yet um you know some of the scopes are fixed zoom some of them are variable zoom some of them feature um you know different types of optics uh rangefinding that kind of stuff so you know as those features come online

  12. 00:08:50

    from like an art perspective it doesn't really it doesn't really affect the weapon art team because that's that's pretty much all you know the picture-in-picture stuff or the the shading and all that like you know currently shade wise it's quite simple we use um a very simple shader just to get the like um the actual like um reticule it's kind of just at the moment it's just a plane that sits inside like the scope tube um in the future i would personally love to see some like parallax you know if you've ever if you've ever used the real telescopic site they have um a level of eye relief so you have to be looking at it from a certain distance or else you can't see right because the way it's refracting the light um

  13. 00:09:37

    you know currently it's we haven't really modeled any of that in the in the shade of it's it's just like basically a sticker um but hopefully in the future we kind of you know can improve that i feel like thermal vision or or that kind of stuff or infrared or whatever that would be more like a like a helmet kind of thing like like something that that that you'd overlay on your entire vision through it through the visor and the helmet as opposed to maybe as opposed to a scope but i'm saying that as a creative person on the video team and not as somebody who works on the helmet or weapon team so i understand what makes sense to me i think to be honest it's a question really that you you'd want to speak to the graphics department about um because it's not like you know you need a strategy for

  14. 00:10:26

    um you know thrusters on ships admitting he you know it kind of thermal thermal vision would affect like a lot of stuff right it's not it's not a single department it's not like the weapon it's not something like the weapons well the weapon speech team could be like we want this feature but you need graphics programmers you probably need some art support from you know other departments to make that possible like um i guess you'd be deciding what things grow and what don't doesn't because we don't currently have like a thing that decides what's hot and what's not yeah exactly yeah yeah so um i want to take a sec i want to take a few seconds here and talk about the difference between a content team and a feature team uh in its simplest form a feature team are the people who are is the team that determines

  15. 00:11:15

    how a thing will work so a feature team would be would be the person would be the team that says yes we should we should get a thermal vision helmet in the game and then they would work out how the thermal vision would work what you can see what you can't see how you know what what the field of vision is they would work out all those things and how that would function then the content team would build the physical helmet that would go on to your character for that so when we're talking to content teams like we are today questions about how things work in the game how things function in the game are generally more for feature teams uh content teams are about how things are made how things are

  16. 00:12:03

    constructed uh and stuff like that so just as a general guidance if you're wondering why i'm picking some questions and not other questions it's because we have the content team today so we're looking for questions more about process more about how things are created in 3d how things are how things are added to the game physically and less about how things work which is a feat which is a question for feature teams and that's actually i love the i love the transform tools in maya so much yeah the fact you can you can align the pivot to edges like that and then just drag something um relative to like a set of a set of edges it's so nice in 3ds max the pivot tools are so different it's um for those who i am sure people we've

  17. 00:12:51

    spoken about it before but um some of the content some of the um 3d artists at the studio will work in 3ds max um others work in maya it depends on the department largely teams that work closely with animation work in maya because the animation department used maya so characters weapons fps weapons other content teams like uh sorry otherwise teams like invite the environment team uh props ships working 3ds max mostly because it's it's got more custom tools for modelling um right but yeah that's all sets have different um features basically now uh aaron blade aaron blade says cool name uh uh where where is it scrolled up i lost it i'll i lost you aaron blade

  18. 00:13:41

    oh he said this is why the game isn't out because they are making stupid stuff like this instead of actually working on game breaking problems actually that's why i just disc discussed the difference between content teams and feature teams aaron blade the feature teams are the teams that work on fixing bugs and stuff like that and they are not here today they are in fact working on the content team who makes things like helmets and weapon sights would not be very useful going through and fixing game breaking bugs so they can be with us here today yeah i don't think you'd want like an artist working on um eye cash it probably wouldn't go very well yeah very true very true okay so right now i'm getting pretty close

  19. 00:14:31

    at least to the forms i think i'm going to try oh god they're going to try and get this kind of black border around the edges and then we'll see there's an interesting shape at the top of the um at the top of the the frame it's kind of like um the transition between like a like that edge there is like plain iron and it's like a little triangle and then it's almost like it kind of sticks out on the side of the frame if that makes sense yeah i think come on to resolve i think i already have an idea that's the thing about um modeling is it's um a lot of people have a different approach to some of these problems yeah something like that that's like super quick you can just like fix that um but when you're looking at

  20. 00:15:21

    especially when you're looking at like a 2d vendor um you know it's a very interesting challenge having to like resolve that in 3d space especially when you get a concept that's been done in 2d um because you know some of the forms can be really like wishy-washy and they look really great but then when you come to resolve it in 3d you have to really think about how those you know how those different faces interact um you know it's it's a really kind of that's the interesting thing about like production art is that you're kind of translating 2d art into 3d um or sometimes you're even translating 3d art from a um a concept artist who works in a 3d program and like you're basically recreating that you know as a game asset you're having to like be really you have to um really kind of consider you know

  21. 00:16:12

    how you can add all the things they've added but make it you know affordable basically that's the big challenge uh for uh folks in the chat are discussing bug smashers and i'll go ahead and say uh i don't know what you guys have heard on the internet but we were very clear that the reason we uh we stopped making bug smashers is because mark aben got promoted mark aben is now in charge of an entire you know gameplay feature team uh in the us and as such has much more important things to be doing than taking time to make video videos every two weeks so that's why we stopped the show because mark got promoted which happens sometimes when you do good work like mark aven does so he'd absolutely love to do the show but it's just like i said priorities we we keep the feature guys

  22. 00:17:00

    where they need to be so not gonna lie to do this bug smashes i i i bucks was one of the favorite my favorite things we ever did it it take it it took a lot of work though i mean setting up every episode of bug smashers was you had to go through every single bug that was that you know that existed in the game at this at that moment then find the ones that were fixed and then basically he had to reverse engineer the fix so he had to break it again and then so he could fix it live on the on the thing it was actually quite a bit of work quite a bit of work yeah i mean sometimes you sometimes you get a bug and you spend ages investigating it and then you're like oh that's for a different department and you just literally just reassigned the jira to a producer for another team you're like bye

  23. 00:17:47

    i know it was a lot of fun and what's great about when that happens specifically is this mix this mixed feeling of of oh good it's for another team and then and then but also like but i spent so much time on it yeah it's it's this very real uh pyrrhic thing where it's like you feel good and you feel bad uh somebody in the chat asked if you have to model the transparent part or if there's some like uh systemic game system that will take care of that for you um we just put a plane in so there is um we have a system called building blocks which is how we create our 2d ui it's basically like a kind of node based um system for creating kind of like 2d sort like flash i guess but it's not flash it's some

  24. 00:18:34

    other technology i don't really know that much about the back end of it um basically what jake will need to do is create a plane by literally just like a you know four points in space you know two triangles connecting them um and then we in the uv space which is essentially how um pixels are mapped to those faces we put that basically over the whole lot so the entire entire uv space um we have some magical things that calculate the aspect ratio of that plane and then we use that in building blocks to like apply the ui we have this thing called uh renaissance texture again not a graphics programmer so i'm not going to try and explain it too much but essentially it will like cast the 2d graphics onto that plane um so if we have like animations or you

  25. 00:19:26

    know things like ammo accounts going up and down we can we can put those in the building blocks canvas and then kind of you know magic those onto this plane that jake creates um he will probably add um a plane with a glass shader on it as well just to get the um the tinting as well as like a subtle amount of um reflection you know i mean that's kind of a i know that's a contentious topic um glass and uh shiny bits it looks good huh it looks good though doesn't it it does so that matters the rule of call baby wow well it's like 99 it's not all that matters it's just most uh what matters as well as people being able to see stuff i

  26. 00:20:15

    think but yeah uh here's a question here's a question uh i really like questions like this uh from from oracle about the chat stop moving uh all right the chat keeps going i know i know what you said basically though oracle asked uh how you find your your motivation when you work especially at times when you're when you're working on something that maybe isn't so interesting to you uh that is a good question um i think when you're kind of working and it's your job whether you have motivation or not it's kind of dumb doesn't matter and that's the that's the brutal answer but the more the best answer really is you just take a few more breaks that day like sometimes if you're really enjoying

  27. 00:21:05

    something the time will make a way melt away if it's something new and challenging um and a lot of time if it's bugs or just something that you're really struggling with and you can't really get your head around it i normally just take breaks like every hour for 10 minutes and that just helps refocus your mind yeah yeah i think as well like um the main thing i found is a varied a very diet of tasks is is always good if you're doing the same thing every day it can get you know even if it's something that you found compelling um it can get a bit tedious so like having like breaking up your day with different types of tasks um wait you know where you can i think that's really important um yeah like jake says as well just sometimes you just need to take a step back and and relax

  28. 00:21:52

    like you know some not every day are you gonna have the same like creative energy sometimes you know you might just not feel that you know in the zone um so i think with all kind of endeavors that require a certain amount like creativity or attention you just kind of you know you have good days and bad days but you just got to try and make the most of it really yeah the variety is a good point though i think like i've kind of realized especially recently i think that's the thing that i enjoy the most about my job currently is that like every week is something new and like if i'm not enjoying something one week i know that like it's not gonna long lap it's not gonna be long lasting there'll be something better down the road i'm very lucky to be in that kind of position you guys you guys touch on a much larger

  29. 00:22:40

    point that that i that i hear from every team in the company i'm one of the only employees that like regularly is talking to every single team in the company i hear a version of that from almost every team that's why sometimes when you when when you look at the order that things are made sometimes when you look at the why this ship comes after this ship comes after this ship or why this weapon comes after this weapon or why this location or space station or whatever it comes after some of those a lot most of those times those things are predicated on priorities you know the thing that's just needed for the game most but there there is always this other this the sense of we need to keep our artists and our designers engaged and interested and because they do better work when

  30. 00:23:28

    they are so if you have done if you have done say nine aegis ships in a row for 2018 or whatever it's like you go all right we're just we're just going to put a little moratorium on aegis ships for a little bit our our guys need to work on something else so yeah and sometimes that like that can have you know effects that ripple all through all through stuff so it's it's it's like 90 priorities you know it's it priorities are always the the dominant uh deciding factor in these things but there's always just this sprinkle of of what do our guys want to work on as well what would be better for for them to stay engaged and better for them to provide better work and stuff like that and then at the yeah

  31. 00:24:17

    productivity can be affected by that so it's definitely something like as a manager it's something to consider if you're putting someone on like you know fixing folks their whole the whole like quarter they're probably gonna have you know they're not gonna have a great time they're not gonna be they're not gonna be on their a game um you know i think it's really important jake can you talk to us about booleans because i noticed you've been using them a bit and i i would love to see some more maya boolean stuff some maya brilliance uh such a good word it's such a good word talk to me about booleans um okay so there's no pressure by the way i didn't mean it i didn't mean to take you out of your flow though i was i was so in the i don't think i've got any other boolean stuff left but i'm going to bring in

  32. 00:25:05

    this little shape here which i would normally do with pom but for a demo i'll do that go online yeah we should probably explain what pom is as well but um if you show us a boolean operation now so i could do palm layer yeah so um there's a lot of different techniques for polygon modeling um there's you know there's like box modeling where you're basically you have a box and you're extruding parts of it out you're like pulling faces out pushing faces in in setting faces um what you can also do is subtract and add shapes together different shapes and it's a really really useful tool for creating interesting forms like that um so you see here you can like you unionize two shapes or you can intersect them or you can subtract um and it basically you know it's just

  33. 00:25:53

    another it's just another um tool on the like art belt really like there's so many different techniques um and there's so many different approaches to the same problem like i was kind of saying earlier different eyes prefer different workflows for that reason you know there is one thing that i wish i knew sooner and this is kind of a different technique but if there's anyone who is into 3d art and they're doing stuff and they want some really like smooth like edges so you've got like a right angle here and you want to like smooth this out if you do it with like a chamfer this you've got like a this weird kind of like curve yeah you need the radius of your curve to have a point

  34. 00:26:40

    and it needs to be like parallel right so it's kind of like you're drawing a circle around and then the radius is getting smaller as it gets closer yeah so you kind of want something like that right um yeah like a uniform a uniform thickness the whole way around and it's just something that's pleasing to the eye a lot of the time you know when you see that weird like boldness bit you're like that change in volume yeah so what i would usually do and this is disgusting i would let's say i want these here and i want to like bridge to turn that face to that face um if i grab this there is a youtube tutorial that i found this on by the way it's not

  35. 00:27:28

    like anything fantastic um if i step this edge to here so this edge is kind of going to be the radius that's the yeah that's the center of your um your your bevel essentially right so i'm going to select these two the face i want to extrude and my kind of like work or what uh select multi select the face hold shift to select the edge and then you can go to wedge and then you've got like an angle here oh it's like a lathe but with faces that's really cool and you could do that so nice yeah i mean maybe that's really boring it's like why the hell is this guy showing me this yeah you see this blue man beautiful consistent volume around that curve it's just it's just yeah so nice geometry now it's like us this is like it's sexy goddammit

  36. 00:28:16

    yeah yeah exactly i was actually we we haven't we had a new guest on he showed me like a really he showed me a thing in 3ds max i didn't know existed that was just beautiful and there's there is stuff like that there's just like um when you get to a certain level of proficiency with a piece of software there's like so many little tips and tricks that you can discover so that's the difference yeah look at weird uneven beautiful perfect radius i'm i'm i'm on reddit right now looking for the star citizen devs learn from youtube thread and then and the inevitable article on you're not good enough no one's too good to look at youtube every now and again to be honest you are always learning yeah yeah every day's a school day if you're not learning then you need to

  37. 00:29:04

    try some new stuff i think you know yeah i mean you need to get a new job if you're not really learning every day it's just it's just the nature of it really like uh somebody somebody had a i don't have my normal person here picking out questions for me so i'm trying to keep up with chat while paying attention to what you're doing do you guys do you folks have experience in real life that you feel has helped you in your digital art um i mean there's not direct experience um but i do think some basic understanding of light fabrication industrial design obviously when you're creating 3d art you're normally creating something that's analogous to something in the real world so if you can understand how that thing is uh put together you're going to make a

  38. 00:29:52

    more convincing version of it um you know so i think really like reference and inspiration you know it can come from anywhere in the real world like you know just looking at how stuff is made i think is really really important um so yeah i would say i would say that definitely helps um being you know being around lots of different objects i guess looking at them i can really help i think for me materials definitely like um especially when you get into the texturing side of like you've just google images isn't really going to do it justice you have to like see it and like move it see how it moves to the light so you can yeah decide roughness value and then um what there's a program called um marvellous

  39. 00:30:41

    designer which is like a soft body kind of like sculpting thing and you make that how you would make a bag or a item of clothing in real life so if you know how to do stitching like for clothes well it's designed for it's designed for that industry primarily right it's mostly yeah like fashion and um oh what's the word for it my mind is yeah the first thing the first time i saw marvelous designer was the first time i i like i kind of thought it was cheating i was i was like what yeah would you just look at the breath of what it can do and how it does it and how quickly it's some functions anyway i just i was just like this is i don't know just the genuine first reaction i had was this is cheating yeah back in my day yeah it's so much

  40. 00:31:29

    like that with uh with gamer like it's constantly evolving pipelines are becoming way more like physicalized and simulated um and the tools are just getting a lot more sophisticated you know like for creating materials and stuff there's a lot of like procedural there's a lot like procedural tool sets like substance designer for making materials you know back it back in there what you'd do is you'd um you'd be you know sourcing photo reference and kind of using a program like photoshop to combine photo reference um you know and that would basically be a larger texturing you know it's also or you'd be doing it by hand right you'd be by hand you'd be painting in edges and just like on a 2d you basically you take your uv layout which is like i said it's unfolding your

  41. 00:32:16

    model in 2d space so you can put paint on it basically and you take that uv layout and put it into photoshop and you literally do that just like painting stuff right whereas now we have programs like substance painter where you can take your 3d model and just paint directly on it and that's just like such a game changer you know like the way 3d art has been made has changed so much even in you know even in the like six or whatever years i've been doing it you know it's changed radically and it's just gonna continue changing as we get more like you know more deep learning tools that kind of stuff more physicalized workflows like it's only gonna get crazier you know i think it's so cool uh here's a question chad i really like uh they say art is never finished only abandoned

  42. 00:33:03

    how do you know when it's time to stop working on an item yeah i mean to be honest like you do experience diminishing returns we generally have obviously we estimate all of our tasks so you might give a certain amount of time for a task um that would include you know like what jake's doing right now which is what we'd consider like white box gray box modelling getting like the big forms in with you know take it through to you know without like the chamfers details the the stickers the decals um you know and we have kind of like a set amount of time for each of those you could just noodle forever like you know there's you would never ever like if you if you didn't if you didn't have that kind of you know i have a set amount of time to polish this up i've set my time to do it

  43. 00:33:51

    obviously your stuff needs a bit longer then we'll give it a bit longer because we want it to look great yeah but you know you can you know you can easily work on an asset until you stick a bit and i think that's the main thing is that you do you know you get as diminishing returns partly because you're getting you know it's as it gets more complete that the nuance it gets more nuanced what you're changing but also your brain just starts like you just the fatigue just like becomes real right the the best answer i ever heard once i i won't say who who said it because they said it to me confidence was was that uh making ass making assets is like eating pancakes when you're at when you absolutely hate them you know you're done that's actually a really good analogy i might have to start using that because because pancakes i thought it was

  44. 00:34:38

    brilliant because pancakes are the only food in the world you love them when you start and you hate them when you're finished and you're basically liking that to uh to game art he was like when you hate it when you can't stand to work on it anymore that's when it's yeah that's when it's done yeah and to be honest like if you shelf something when it's incomplete and you come back to it you may notice things that you didn't see before oh yeah you know so and also you know as we discuss like you're learning as you're doing it right like you might try a new technique and you might also when you in retrospect you might look at the asset and say ah actually in retrospect that's a bit too big i could have changed that i could change this you know as an artist i think you're striving for especially when you're striving for realism you're always going to find fault because you're limited by the processes and other technologies

  45. 00:35:25

    to an extent and also your inability um so yeah you could sit there forever basically yeah so we basically like you know you know it's time when you've spent a submission amount of time with it really i guess and that's kind of wishy-washy answer no no no per perspective is yeah you actually touched on a very important answer that that many artists share perspective is so important sometimes you need to stop and step away you know go for a walk you know you take a loop around the block whatever and then come back and you can and you'll see things that you weren't seeing just an hour before and you know your time frame may change whether that's an hour a day a week you might have to step away for a week but that process of stepping away and letting it all

  46. 00:36:12

    drain from your from from your active ram and then coming back to it later and pulling it into the system fresh allows you to see all kinds of things that you didn't before perspective is incredibly important yeah i think to be honest like um yeah a lot of the time you you just like okay i know there are things i can improve on it but you know it's when you start getting to that point where the returns are really diminishing um but then again that ex that like extra like ten percent that can be the difference between like an asset that's pretty good and that's it that really like it's really sold so it's a fine line because you you know if you're kind of it's like exponential right you know you get to a point where it's literally like you need to you know you're not really getting enough that it's worth the time investment i think that's really where

  47. 00:36:59

    the distinction so jake's putting on some decals now so when you run us through that all right i'm putting some decals on this thing thank you that's a wrap you're welcome uh so like palm and decals right um so they're like a separate material with like a bunch of details on them and you're just kind of like floating that detail on the top of things um stickers basically aren't they yeah pretty much and this is good because like this particular pom sheet uh this one uh you can use on like like how many ever assets that you want really you could put on every asset in a game if you if you really wanted to rather than having to have like multiple ones for yeah every material that's a

  48. 00:37:48

    really important point with with production art like this is that if you can share resources that are in memory then you're gonna it's you know it's good for performance so what i'm going to do now is this is really gross and dirty but i've tried to be as quick as possible um i'm going to put this in the engine and we can see what it looks like and maybe we can apply some materials and stuff to it [Music] export it oh oh no that's using lambeth one instead of tyler is now bombing my chat with pictures of yeah i don't mean to be fair like going full power like this for an hour

  49. 00:38:36

    is it take it out of you right jake yeah okay here we go i've applied it to this pistol i've already got it set up so hey start shooting with it as well all right why don't you switch to the optimize for video uh when you're hearing engine yeah that might be good how do we do i think it's up at the top you move your mouse to the top of the screen share and in the little option window that pops down the top here more or maybe yeah it can be a bit slideshowish in the editor can't it yeah optimize for video clip yeah yeah cool oh wow i see what you mean jared yeah yeah see it makes it blurrier but it gets us our frame rate back this is a function of zoom people it's a

  50. 00:39:23

    function of he's he's the other side of the world i'm going to try and just like slowly move no no you're fine no you keep doing what you're doing we optimize for videos so we're fine now yeah nice okay so now i can actually see this in game which pretty cool yeah yeah and i think as well this is another thing that um because we have this like modular system where these these attachments can work on multiple weapons um it can be interesting like getting it in situ early on is hugely important because that can really change you know like if it sits too high on the model or if it's too big and beefy like you know you need it to feel right so obviously making those changes once you've put in loads and loads of geometry is a really bad idea so like getting it in game i mean i feel like that looks pretty nice

  51. 00:40:12

    maybe a touch on the small side i don't know what do you think jake actually chat was just asking that how do you know you got the scale right normally you just eyeball it for things like this um we do have we do have metrics that we use for these um you know for these assets but it can just be on a case-by-case basis basis like some of the concepts we get they look great um you know they look really lovely but when you put it on a model the way it kind of interfaces with the top of the gun it can feel a little bit too squat or a little bit too tall and skinny and that's just where like there is a little bit you know as the production is there's a little bit of like interpretation you know finessing it to make it feel you know read quite well so corey are you saying this needs to be taller wider

  52. 00:41:00

    what's your preference mate i mean you're making it so i'll leave it up to you but perfect yeah okay it's perfect there we go cory bamford uh normally we just eyeball it yep is that a quote now yep okay i'm waiting for them i'm just waiting for the meme i mean how about how about that that's a tall boy this is what you want this is what we want hell yeah all right there we go everybody knows bigger is automatically better absolutely bigger yeah it's imagining people walking around with this thing so i know what scope they're using you hold it sideways it's like a binocular

  53. 00:41:49

    so you hold you hold the pistol sideways and then you've got two dots yeah that's how it works just pick yourself in the eye with it i think what we could do now is probably re-export it at an appropriate size and probably play around with some materials so it's coming i got the cgf here right um i will set this up at a camera angle like this you want to launch a finished one as well just so we can get like the side by side uh what now you want like a finished one now at some point i might just be nice to see you know this is where we're up to obviously it takes longer than an hour

  54. 00:42:36

    to make one of these guys but yeah we're for those watching at home uh we are skipping ahead at this point like the the 3d modeling would is not done but in order to in order to fit more of the process into our hour-long show we're not skipping ahead to the materials so is this picture quality no uh yeah uh yeah you can turn yeah that's uh for for this one why don't you uh yeah it's fine keep it here yeah because i'm not moving about that much really just the pictures there so what's this thing what's this thing jake that you're you're on now what are we looking at son this is the material editor uh so this is kind of like our texturing thing uh we're not going in by hand and like painting stuff we're going we've got like a material library that will apply to things which

  55. 00:43:24

    is really good for um things like this like i can just throw this in early which sometimes i will do and i'll just throw some materials on it and it'll help inform like how it looks like i don't need to make the entire thing bake out then put it in engine and see how it looks i can kind of do all three at the same time yeah which is really cool i know i know this gets brought up on a lot of our um our art dev kind of shows but um the main difference between um the workflow we use uh on weapons and and other assets in and the more kind of traditional workflow for video game art we don't have a high poly bake um and what that really means is we don't have a really high resolution model and then we don't like translate that

  56. 00:44:12

    detail onto over like a really optimized low poly model what we kind of do is we make like a mid poly it's kind of more expensive triangle wise there's more triangles and vertices than you would have like a low poly but then we don't have to have a unique like bake for each of our assets so essentially we just we rely on this thing called like face weighted normals to create the lovely like smooth curves and really it just means that the texture the texture use we have is a lot lower so it's kind of a trade-off i hope that made sense all i heard was our work is mid yeah i normally have this like onto my second screen so like i'm not looking through this

  57. 00:44:58

    tiny window yeah you can drag it over if you want and we can just see the changes you're making you know yeah uh some of the folks in chat are are asking like if you can do this so fast why aren't there more of these and stuff i want to reiterate that yes this is an abbreviated workflow that that you're that you're seeing here obviously that model is not done um obviously he's he's forced to work on one screen instead of two screens it adds another layer of challenge that you're currently seeing right here but there's far more there's far more work that goes into these and then as far as the multiple customization options and everything uh again that's additional than additional work not just for the content team but for the feature team you know all those different customizations have to be created as a feature set and of course this isn't the only thing

  58. 00:45:45

    that these guys work on so that's the biggest one is that attachments while you know we prioritize them in some quarters generally like you know it makes up a part of our workflow but we're also creating the weapons which are much more you know much more time consuming um you i mean also if you look at the difference between a you know this this guy on the left here now who's that's like a final asset so you can see like the modeling is a lot further along the material definition is there there's also a lot of technical setup that jake's uh glossing over because it probably you it's not the most exciting stuff we do like i thought sorry yeah i was gonna say like i thought it would be very boring so i did all that behind the scenes like already um in prep for this yeah um yeah so there's quite a difference

  59. 00:46:34

    and i think i agree with you that scope is quite small because this i know this is already quite small yeah yeah i think it does need to be embiggened somewhat um but you can see you can see the palm decals on that one on the left there jake you wanna just show us one of those palm decals so we can kind of yeah it's about bolt there what what how is that turn off the optimization jake we'll lose frame rate but we'll get a crisper picture so you can see the palm details you're talking about uh is it here the stereo high fidelity and then i've got optimize for videos just turn off the optimize for video clip there we go our frame rate will go down but you'll see the we'll see more detail yep cool all right so palm details are kind of like

  60. 00:47:22

    this uh this kind of text here like it's if you've ever heard of normal maps it does the exact same thing but we also use a height displacement which also adds a bit more extra oomph to that normal so if i add one to to the scope that i'm working on now i'm just gonna add one of these but basically what we do is we have a height a height texture so it's like the depth of that bolt um and then we use that depth basically there's a shader that will use that depth to kind of create fake indentation um on the model and basically it's a lot cheaper than modeling it in with with geometry right this is an instance where using texture memory is a lot cheaper because those kind of cylindrical

  61. 00:48:10

    circular forms and stuff that would be really really like you know heavy on triangles so we kind of fake we kind of fake a bit of the modeling with these texture sheets um and it just gives us that extra little bit of detail yeah so you can see there's like a there's like a perceived depth to that bolt right it looks like it's going into the model but it's just a texture jake hasn't actually modeled that at all yeah this this is one of the greatest secrets of star citizen how we get the detail that we're that we shoot for while remaining performative i mean that is a texture that is doing that not an actual you know modeling of the thing that's parallax occlusion mapping for the people who are asking in the chat palm pos so if you want to know what textures you'll need if you want to do this actually can i do that yeah i could do that

  62. 00:48:58

    if you want like a diffuse which acts as like your ao so if i well yeah yeah ao is basically like shadows right so it's when two surfaces close together it gets darker it's like a way of faking that kind of bouncing of light around in crevices yeah yep and then with most games you'll probably recognize a normal map which like kind of fakes that light and then the height map is what really does the magic so i'll get rid of that height map and that's what you see normally so like that normal map does like react to light and that's what most games do but then you add like a height map with some defamation and yeah wow and it's a very iterative workflow because if you want to move that bolt you're not like having to drag ferts around you just it's just a plane it's

  63. 00:49:44

    just a sticker so you can just you can just drag it around wherever you want so like it's a very lovely iterative workflow um using these decals um because it just means that if you want you know if you want to like add these little bits and bobs it's just yeah it's just so much quicker yeah that's how when you when you're seeing ships uh with grading in the floor plates and you can see through the grating and you can see the wires underneath and stuff like wow it's like did they really go through and model all that that's palm yeah which may be modeled but then with that is where we will bake it so we base that onto a texture a texture so we're converting we're converting geometry into pixels um in a way right yeah so you model it once but you just don't have to model it every single time that you want to

  64. 00:50:32

    apply it yeah and you can also create these um these shapes in a 2d package as well um you know it depends like it's just a height it's just this displacement map at the end of the day um but generally like modeling it you know you get a bit more control here we go yeah i mean it's just like baking an asset really like you would anything else but you're just baking it to a plane and then you're copying that plane and sections of it on to other assets yeah so yeah yeah so on props for example we use um like four or five pom shoots covers like 90 of our assets um but then obviously you might need some more specialized ones for weapons i think it's per manufacturer we have like

  65. 00:51:20

    a texture we have texture resources to manufacture so like all the bearing weapons will use the bearing palm sheet and decal sheet um you know which has kind of stylistic elements on it that are very light bearing um you know that's kind of how we do it it's jake's chamfering now so you can kind of see the face where normal stuff a bit i guess right yeah so normally you like oh you get like a high poly and a low poly but we just modeled those things in so when a high poly and hello probably love each other very much yep they connect so yeah you basically select because right now that looks weird you select the faces we have like a custom script that runs through it all and that just tweaks the normal so if i go into

  66. 00:52:07

    a normal display here so all these lines are kind of like those normals where the shading is being directed it's like the orientation of of the face almost right each foot has a each foot has like an orientation and when that's perpendicular to a face it makes that face appear flat so we're basically weighting the normals to the large flat faces and then we're kind of letting all of the angle the difference between the normals is on all the edges and that's what creates that little curve um normally you'd need like a texture resource to kind of fake that um so you know it's a little bit cheaper yeah and like i think like a chamfered edge um like let's say like this and these are soft like that is pretty

  67. 00:52:58

    much the same price as this because because this is hard edge the engine is going to split them so there's technically like two verts on top of each other yeah on here that isn't happening because they're smooth they're one smoothing group so although you're getting like you've got a much higher poly count that's why like our budget for poly counts much higher your vert count is normally not nowhere near that so yeah yeah it's kind of it swings around about really like um obviously adding more triangles those triangles have to be rendered by the gpu but um you know versus versus having textures for every single asset it's kind of you know it generally wins out um in the case it doesn't win out we use things like pom fake it so it's a really nice workflow for that reason

  68. 00:53:45

    so you're combining a bit of geometry with pom here right yeah so that's like a cheeky little way of getting some extra detail oh it's not in the group i'm in the group mate it's not cool enough to be in the group i was going to say it but i want you to wait for you to figure it out so this kind of looks horrible no i haven't crushed it something wrong with the export oh i've put i've put an e on the export name i do that all the time like i press a button and it stops working i'm like why and therefore you've renamed that you renamed the export oh

  69. 00:54:33

    you've experienced this this is the export path and i accidentally put like a w there and i'm like why is this not working that's hilarious it happens all the time there we go so i've just shrunk that a bit so yeah so i mean when you get when you get your face really in there you can see the the triangulation you know you can see the faceting we call it on the on the bolt yeah you know there's no avoiding that right at the end of the day is there's like a there is a resolution to these things they have they are made out of triangles and points in space so you're going to see them at some point but we try our best to avoid that really yeah go ahead and go ahead and optimize again if you're going to be back in here yeah no i uh official engine break so if i want my engine to work again and start applying attachments i have to close it and reopen it again

  70. 00:55:21

    so yeah i'm going to do that we'll go back to cameras so we're actually almost we're actually almost done we only got five minutes left okay i can just chat for stuff then somebody was uh somebody was asking if 3d modeling is all you do and i don't know what they meant by that but i just i think i think what they mean by that is probably that like oh you're not painting materials or textures and things like that and yeah it's something that i miss like i really um i do miss that i do it in my own like personal work just to keep my skills like up up to scratch um but there are circumstances where we do um like those palm in those decal sheets for example we will if you've got any random like weird like custom assets

  71. 00:56:09

    like i think a while ago and i mean a while ago i don't know if it's even a game i made some like prosthetics and like some hearts in a jar and like organic stuff and that that needed that workflow as well so um in terms of art yeah you're kind of right but yeah um but we also do a bunch of um bug fixing communication with other teams like working with future teams and other departments right that's a huge part of your life as well yeah i mean now um half my job isn't even doing art and just doing technical stuff and like talking to other team members and trying to get things done and working and um yeah i mean that's production art it's you are producing art for a project with restrictions even though our restrictions are fairly liberal by some

  72. 00:56:56

    standards um you know that is the life of production artists really you're kind of it's not just design and it's not just you know making like artwork artwork it's also producing producing work to a standard for a game right that's the difference really it's not you know it's not the same as like um if you're doing like concept art for instance you don't have those same restrictions you know it's purely about the design and the aesthetic and stuff and then production artists make that a reality okay well actually uh we're actually i'll actually wrap up here uh we'll wrap up here i'll give you one final question from the chat because the chat has been asking uh if maya blender and 3d max all enter thunderdome who's coming out alive

  73. 00:57:47

    oh it's maya i mean like i it's definitely not 3ds max i think there goes my partnership yeah blender's the uh blood does the um that's the underdog but yeah um and and maya's just like the the kind of old-timer who like always wins but still wins i see it if you're comparing it to languages like 3ds max and maya are kind of like english in the like they're the dominant they're the dominant kind of program um you know but like if you look at languages like newer languages that have kind of been designed to be better in a way they don't have that same like they don't have that same amount of use i think that's the problem you know what's that language that was like you

  74. 00:58:35

    know invented recently i can't remember what it's called now but like basically that's kind of how i see it like um my 3s max are kind of really entrenched in the industry so a lot of people know how to use them a lot of people and it's like a common language right which blend is kind of quite new it's the tool set's only really developed over the last you know a few years yeah so like if you have to say if you're asking like to pick one as well um to get in the industry just use like just use bender it's free and it's it's really good as well like um a lot of the skills that you'll learn using that are transferable like i started on maya i mean i'm working on may and now but like when i first started here i was moving over to 3ds max but that's why i don't like it because i was forced to come off of it and now i'm back

  75. 00:59:24

    yeah i think your first is always the one that you will um you know your first 3d program is the one you'll feel most comfortable in like never which one of us have more boxes in the background cory you or me all right guys that's it thank you for thanks thank you for taking an hour out of the end of your week to appear on the show uh with us this week as always i i very much appreciate it uh uh feature teams are hard at work uh uh wrapping up alpha 313 uh right now so it's always a good time when we can get uh friends like jake and acquaintances like corey uh to be here on how could you just that are you kidding

  76. 01:00:15

    corey has bailed my ass out of so many things like like like corey uh for corey for jake i'm jared uh thanks so much for joining us this week we'll be back next week with another call with another i almost said calling all dabs now that's in like two weeks there's so many shows every single week until next time everybody take care bye obviously it was time to end the show yep i was just talk about diminishing returns i had some diminishing returns there at the end yeah i think it was a good time you

Transcript from YouTube’s automatic captions — punctuation and Star Citizen jargon are approximate. The video itself is hosted by CIG on the official channel.

Squadron 42

LiveSQ42 Launch Subathon

$2,100of $3,000

Getting Sub to Manchester for the Squadron 42 launch

Next giveawaySubscriber Store pick
$0.48to go