ILM Artist Webinar: Advanced Water FX in Houdini with Juri Bryan

welcome welcome ladies and gentlemen to yet another rub away webinar my name is urban bradesco and i'll be your host for this evening with us today is the awesome yuri bryant currently the senior fx cd at ilm london previously head of effects at important looking pirates in stockholm yuri has been working in the industry for 10 years as an fxtd pretty much all of the his time spent was working with water so you guys know he's an expert with water effects uh spending five of those 10 years working as an fx supervisor across movies and television yuri welcome to the show thank you very much it's a pleasure being here and it's really nice seeing so many people here willing to spend a few minutes with us awesome well i'm i i think it's going to be worth uh worth their while i'm pretty sure let's hope so uh so right so let's uh we're going to be jumping into what yuri has to show for us today right after we showcase the advanced water effects promo so let's roll it [Music] [Music] [Applause] [Music] [Music] [Music] yes yes i think that's one of the best looking promos uh we ever did did awesome job on everything i mean uh thank you to saber and the team at rebelway for actually making it look cool i just did the sims so yeah well it's a team effort it's always a team effort there's there's there's no other way very very true yeah with all of that being said uh yuri uh go ahead so it's all you man cool uh so i guess i'll share my screen so i thought it's always nice to get to know each other a little bit so let's start with actually talking a bit about my career while you guys see my show reel um i've started my journey in visual facts quite young i would say uh when i was listing my family in the states and my cousin was animating i forgot how the website is called but it's one of those where you can animate a paw for a little guy on a sled and he's sliding around and it's flash animation stuff and we got really into that and did more and more flash animation together and um cut forward two years he's starting down his path in the animation world and doing 3d animation and now game animation and i started with visual effects at 18 i decided that high school was a bit boring and i know what i want to do with life so i dropped out of high school got accepted into vancouver film school and moved to vancouver uh did vancouver film school which was very nice because you know you learn a lot and you meet amazing people but i always knew i want to do effects and i was a 19 year old quite cocky guy who says hey i can do water effects and nobody hired me i couldn't get a job because i couldn't show up show reel right back then that's like what 10 years ago now um god there wasn't that much training out there especially for the more high-end stuff that you can really market to a studio and say hey give me a job now after a few months of trying to find a job i got contacts with a studio in vietnam through a friend and started working for them started bidding for them commercials movies all of that and we won a project and that project was quite big so the company owners had come over and run the project for us so a week later i moved to ho chi minh city and stayed there for three to four months and worked on that project sadly financial crisis was a thing back then so all the investors kind of pulled out and the studio shut down right it's crazy and the crazy thing is i had to move back and move in with my parents and all of that and try to do any job and anything i could in the meantime just trying to learn more polish up my skills until in december i got an email from ilp in stockholm and i interviewed with them and a few minutes later well an hour like it was like 30 minutes interview and they asked me when can you start and i'm like yeah monday and that was on a wednesday the interview and they just kind of laughed and i find a place first all of that so okay i'll find a place and i'll let you know when i can be in stockholm and i on thursday evening i called themselves like so monday what time and i started monday right away and started working on commercials with them not on water but uh doing a like super bowl commercial for fox which those are always crazy but we did that and uh i was struggling right young artists trying to do some cool stuff with an amazing team but just working working a lot but over time i got the chance to work on some water shots showed that i can do some cool stuff work on a majority of shows and after being at lp for about a year i got asked to lead the team and back then that was well two people and so i was leading myself and another artist and over the next like three four years the team slowly grew from like two to four and then five and we were taking on more and more projects that were all bigger scopes right so the shot count just went up and up and up and up and after about five years i kinda ran the department supervised and worked on all the shows and all of that and uh eventually once the company was big enough to have heads of departments and all of that i became the head of effects and effect supervisor and worked a lot on that kind of stuff right just management while doing shots while managing the team and trying to build things and make things better all of that but earlier this year i kind of decided that hey it's time to do something new i'm turning 29 in here for a while and i want to learn more right grow more as an artist myself so i decided to send out some applications and happily for me ilm wanted me so i moved to london earlier uh in september and i'm working here now just for almost yeah just over two months now as an effects td for the london team and having a blast so far awesome that's pretty much my career it's just a bunch of hard work and having an absolute blast working with some of the best artists in the world who thank god are all amazing lighters um because i'm not so they can make my stuff look actually really nice and really cool um that's my career love it dude i love it it's i have a question because you i feel like you kind of did both and now you're working on movies what is like for people learning you know or not maybe not just like beginners but just people that are kind of getting into the industry is it better to start working at a company that does commercials or movies like what is better for learning that's a that's a tough one so if you work in a movie company in my opinion you're exposed to a lot more higher end stuff right so the the quality they push you to is very very high obviously and you have a lot more time to actually achieve that quality but you also have a lot of very senior people to learn from which is always great and you have a lot of amazing tools that have been written for you right so a lot of those easy things those easy mistakes they're all taking out of the equation just because well a big studio can't afford that if you're working on 500 shots simultaneously you can't do the little mistakes right um i always thought when i was hiring for lp that i want somebody with big picture experience who learned and came up from the commercial world and that is just because you don't have these safety nets you have four weeks to do a commercial of two minutes full cg yeah right and it's just crazy and taking somebody from that environment and then putting them into the tv environment which is crazy but way less crazy than commercial they're gonna be fine right all of a sudden it's like no you don't have two days to figure out that shot you have two weeks yeah and they're gonna be happy right they're gonna do cool stuff and they're gonna be amazing but at the same time if you take somebody who's used to having this whole support network and is used to having pretty much a lot of time to deal with it and like a lot of people to lean on and a lot of software and a lot of pipeline stuff and all of that if you then take them into an environment that's basically here's the software bare bones there's import and export that's all you have from the pipe and yes you have one senior artist as your mentor but they also have 20 shots to deliver so it's difficult it's it's harder for people to step in and at lp we always set like coming in as a senior from a movie studio is rough the first six months are rough just because of the amount of stuff you're responsible for um yeah and i feel most commercial studios most tv studios are a lot more that way than movie studios so yeah the pace is way different but uh just from my experience i feel like if you want to learn like to be fast and good like doing commercials can be very very educational and then obviously if you like like you said transitioning to movies at uh ilm they give you some more space to breathe and then but you you already have the knowledge to work fast so you can do faster iterations so you know it's always a good skill to have exactly and i feel also like the biggest thing i feel in visual effects is communication right we always say oh it's a team sport it's a team sport it needs everybody to work together to do something great yeah well you can't do something great if everybody's on a different page yes right and in the movie world because it's so big the communication is a lot more bottlenecked in a lot of studios right it's streamlined through certain people which means as a junior effects artist the amount of chances you have to actually talk to a camper about the issues they're having comping your effects is very limited uh some studios don't allow you to talk across exact departments whereas in the commercial world you're sitting next to your lighter you're sitting next to your comfort zone and they're going to tell you hey fix your stuff yeah and then you fix it you know what's up and that communication skill is something that takes years to really perfect right and get it down to nuance of like hey i know this is going to be an issue for the next department so i'm going to fix it in five minutes without having to talk to them because i know it will be a thing yeah and that skill a lot of juniors have when they come out of a commercial shop but from a movie shop they kind of know like well if my animation doesn't work i talk to my lead and that's like that's not how smaller studios work that's not how commercials work that's not how tv works yeah well cool man i have a bunch of more questions but we'll do them later after your uh i think it's best if we just jump into the the meat the meat and potatoes i thought it would be fun to start this whole thing off with actually showing the first water clip i've ever done in a production yeah it was for ilp um let me just meet that it's actually funnily enough for the swedish military the swedish military does a lot of advertisements that are very elaborate every year not as crazy as the us marine one but you know if you gotta advertise for your navy why not flood an office kind of showing like it's not just the usual office job uh so this is actually the first water shot i've ever worked on a bunch of old school houdini flip stuff i think houdini 12 just came out when we did this and it's all rendered in v-ray and most of this played like the boxes are fake and the water's like yeah and you know it's just a fun little shot to show where it started still looks good it kind of holds up ish um and from that i think we should go to one of the projects i think most people kind of know my work from and that's really lost in space um lost in space was a crazy project um we did water wise we did a lot of work a lot of volumetric work and a lot of crazy detail fire and all of that but relevant for this is this waterfall that we did and the thing is it was a weird brief from the client they basically came to us before they had the script written so they came to us and said like hey here's the first few drafts of the script we want to do a water planet and have some crazy stuff in there we're kind of thinking maybe having a massive waterfall and the supervisor at lp nicholas jacobson sat down with our concept artist and they came up with a bunch of crazy ideas one of them being this massive uh so so like just an equator of waterfalls yeah right so that is just nuts and the thing is because we didn't know what we're gonna do yet what are the shots layout any of this wasn't anywhere decided we actually decided very early on that we would build an environment where we can just plug in a camera and it renders and no matter where we put the camera it'll hold up the resolution will be good enough and then we can just film it and what we ended up doing then is build a one kilometer waterfall uh which is actually not as long as it seems when you have to cover an equator yeah but the thing is we made the environment loopable so the environment could just be stuck next to each other and all the rocks all on the water environment everything will line up perfectly which means you can line up the water perfectly right so a shot like this right all of this water is just repeating and if you really look for it you can kind of see the tiling but we basically simulated one kilometer and it starts here and then goes all the way to the edge that's one simulation that's all one sim done at once we used distributed simulations for that which were crazy um those sims alone took like 750 hours on two 96 core machines and took 750 gigs of ram just because we didn't know where to put a camera right so we needed some resolution in this yeah i think i simulated one kilometer at 0.02 uh pixel uh particle separation which is like everything that's like all the particles yeah it's it was a lot it was uh i think in total one version of the waterfall with all the secondaries and everything came up to about uh 600 terabytes of storage and that is compressed so how do you how do you work with that in the viewport like you don't you don't um yeah funnily enough because we had well i come from a very luxurious place at lp where we have unbelievably good lighters um so i actually got in a very bad habit of never looking at simulations i had too much to do so i just simmed stuff check that every system worked on a five meter square and then hit export and three days later we saw the renders and um they normally worked out so this waterfall for example the top layer so the crazy sim i'm pretty sure this is version two of the set um and that's basically just laying out a bunch of mega scan rocks and then running a sim now when you look very closely on the static ones here you can kind of see a transition line right here right where it's like just a bit brighter and that is because we have actually four simulations in each one of these we have the top layer which you can kind of see over here but this stuff falling down is actually not in that simulation because that would be way too heavy so what we did is just simulate the top chunk of the waterfall until it falls down and then you just cut away the edge make that the emitter for a new simulation at which point you don't care about all the tiles on the left and right over this and you can just emit water from it run an airfield simulation and then render all of that as white water so this stuff here the white water is actually the same flipsim as these mesh chunks here which is we isolated them so we get a little bit of meshing falling over the edge right yeah yeah yeah and that way you actually end up with a top sim that's crazy heavy but that crazy heavy sim actually dictates all the other simulations so once you have that and that looks cool you can just say well throw it into the waterfall let it drop add some foam to it and then send it over to light now because this is crazy heavy uh we could obviously not render a kilometer of this like there's just not enough ram to actually fit all of this high-res into the system onto the machines so we had to export this whole waterfall three times just so we could have an auto loading system that just switches these guys out based on the camera and as we do a fly over they slowly switch in the decks uh distance and that way we can kinda get some detail now that was crazy and like these kind of shots are all nuts and all of that but one of the craziest in my opinion is actually this guy i call it the most expensive background ever that's blurred at least lp yeah uh yeah so yeah that's 500 terabytes of storage yeah about the yeah a month of sim for that background but it looks crazy looks good that's nice um yeah do you think it would be cheaper to just go shoot a river like shoot a plate put it in the background no i don't i actually don't think so just because they always want changes oh that's right and as soon as they make changes it's kind of broken and i i really like the way we at rp had a chance to work with the guys from austin space because they trust us so much um if they would have if they wouldn't have trusted us to actually be able to pull this off they might have gone and shot footage for this stuff and then you know a month before finals they might have changed their mind and then we had to build this bunch of problems here yeah the way it was is like we knew what we're getting into they knew what they're getting into we had a very open conversation with them it's like hey iterations take ages so let's just figure this out we got some help from side effects to really optimize the stuff and they you know suggested linking ram buses to the cpu course and all of that so that we could actually send this efficiently and uh yeah i think we got a pretty cool sim out of it in the end and a very happy client yeah and also a vest nomination uh which sadly went to uh stranger things yeah i i do think this is one of the best looking water sims i've ever saw and in theory you're gonna be showing how to do this right in the workshop yes so basically what you guys gonna learn the last two weeks with me is waterfalls and you're gonna learn every single cheat every single trick everything i've used to do these sims um lighting and all of that i'm not going to cover because again i'm the worst slider you can imagine but you're gonna learn how to do like this shot like you can set that up and run it uh yourself yeah afterwards and you'll know every trick everything that we use for this and the nice part with all of this stuff is that actually it's not insane like there is nothing like the complicated stuff of doing these kind of things in my opinion is always the concept like crossing the initial idea of how can we get to that detail level uh reliably and then once you have that the systems and the actual houdini side of things is actually it's feasible it's not where you're like oh i need a physics degree for this and we need to code stuff and we need to write our own solvers to make this possible it's just using the tools that are there in an efficient way and driving simulations in a very controllable way so that you can really go like well if i place a rock here right like if you look at the sim like this they clients will want to art direct where white water goes so well running the sim will take a month you can't do that but if you build systems if you build everything correctly you can just place a rock and you know what will happen you can't say hey they want a karma area here okay i move my rocks around and i know i'll get a comma area here and that kind of predictionary thinking that's what really makes this stuff complicated but building it um everybody will hopefully afterwards go like oh okay that's actually not crazy and i can use this stuff in my daily work um relatively straightforward yeah because like i see a few people are asking how much ram is this taking and then somebody was saying oh we don't have 500 terabytes of storage like well like yuri's gonna show you guys how to do this on on a budget exactly and you always got to think you probably don't have to do a kilometer of this yeah right like if you want to do a kilometer of this like okay you know you might you might have to go to the shops um but if you want to do a waterfall you'll be fine right it needs a lot less ram than 500 also if you know what you're like as always with effects right stuff gets really expensive if you don't know what your camera is going to be exactly because then you have to do everything for any and everything right oh you might be super close i need all the resolution but if you know what your camera is going to be and if you kind of keep yourself limited in just not going the most epic scale ever um and rather focus on showing off good technique showing off good you know shapes good motion getting all of that nailed down um it's going to be way more beneficial for getting hired and learning stuff than saying like well here's a massive sim that took me a month to sim it's like that's cool but you know can we see some details of it's like no look how big it is nobody's going to be impressed by that right exactly i think what's impressive about this is just that not the fact that it's long what's impressive is that we can't put the camera anywhere and that these shots work without having to tweak a lot on the simulations and yeah so ram and all that should be okay as long as you're like 32 gigs of ram um you should be okay yeah exactly cool so from lost in space let's actually go over to another fun project which was one of the first few projects where i was like full-on lead of a bigger team than two which is the shallows uh shallows is this a little surfer shark horror movie um came out a few years ago and we did a bunch of shark interactions for it and um they were all really my first approach of like hey we need to get stuff into an ocean that is not just a boat in the ocean but like actual full-on ocean interaction and all of that close to the camera and quite high-res and we did this with a team of uh three p three additional artists and i think it looks really good in my opinion this is still some of the best white water we've ever done um the reason for that is uh the lighting soup uh bobo skipper is an insanely good artist and um for example for this shot which i think is the best whitewater i've ever produced yeah um it's all mesh we just meshed it in like an insane resolution and v-ray is amazing so uh we render this with sss and v-ray it's so good it looks like it looks like a real play yeah and honestly that's the thing like bobo just doing his magic is unbelievable and it it works and it's it just looks good and the nice part is like you know motion wise and all of that we could kind of match the plate uh really cool stuff that we actually did for this is um the client filmed this tank right so this is just a massive tank that maybe expands five more meters on the sides here and they put a wave tank in it right so you have like the walls on the sides moving and building all of these waves and we kind of wanted to match those waves so that we can actually match them you know on our shark and all the interaction on now the thing is if you just put down an ocean descriptor for this you can kind of dial it in but this modularity of like just the waves being hit from the side constantly is a really hard thing to match so we had uh one of our match move guys uh actually rodeoed um and drew lines curves and then animated them by frame to kinda match the overall motion of the water uh you know like oh this wave coming in here like that timing is really hard to do with an ocean what the yeah so we just drew a curve and animated that guy coming in and then i use those curves to inject velocities and to drive and um to wrap up oceans to then actually get all the motion to relatively line up right it's not perfect yeah but it's a hell of a lot closer than if we would have just said well there's no ocean or here's a random spectrum that works on a few frames right yes so here it's actually moving with the actual plate that they shot in the face so yes it helps a lot it helps a lot in the same thing here right if we had like big waves coming in we tried to match those a little bit and uh then it's just a lot of resolution back then we didn't have those crazy simulation machines at lp so we just cranked the res up as far as we could and render this guy and the nice part is here like for example i think the shark looks fantastic in the water yeah and that's mainly again the lighters being fantastic and doing really nice little tricks that are super underutilized like here we took the shark that is model and looking cool and then we duplicate the mesh displaced it and just put a water shader on it and instead of shading your shark and tweaking specularities and tweaking like your diffuse maps and all of that together like oh it looks wet now well if you just refract it through water it'll look like it's in water and all the specularities will behave correctly all the roughnesses will behave correctly and then you put a little bit of displacement on it so all this water here on the nose right it looks like it's running off that's just a displacement that's procedural and shark looks wet just like that right it's relatively cheap well cheap artist time-wise and um you get some really cool stuff from it so that's the shallows which is uh a fun project was a tough one but uh yeah we pulled it off i think and here is also where the inspiration for the shark shot comes from right so the shark shot that we're doing in the um week five uh week four yeah week four of uh the workshop is this guy it's inspired by that and you're going to learn all the methods everything that i used to do this shot so all all of this stuff you'll be able to do the same kind of work right same kind of methods and then it's just dialing it in and getting cool animation to make this stuff work after this guy let's talk about another project that we're going to do and that is one of them this is actually going to be in week two and i'm going to uh talk a little bit about custom ways of doing flip simulation custom sourcing making flip stick without having to use a lot of viscosity and stuff like that and what we're doing is a very very simple effect on screen but a rather complicated effect when you work on it which is we have a character and it's raining like hell yeah so we can't just put a dinosaur in here we need to have all the rain interact with this dinosaur right so we need sprays coming off the dinosaur we need streams of water falling down we need streams on the skin that like move around and collect and do all of the nice things and you barely see it in a lot of those shots which is kind of sad yeah but if it's not there right like these little droplets falling off all of that all of these little impacts that halo of rain around him you'll never ever buy the shot yeah right uh so we're gonna spend like a little bit on like how to set something like this up and how to build something like this and i think just to under for understanding water and different flip issues and all of this um it's gonna be quite nice yeah now these shots are just kind of cool and kind of epic yeah but they're kind of simple it's just cool animation and a big flip sim what we will cover is how to make like white water with this fluffy and this much white water um we'll cover that when we talk about beach breaks in week five and it's that's just a fun little little thing to do yeah i'm so good yeah the dinosaur like with the and the raindrops it's exactly what you said if they're almost invisible but most the best effects usually are right yeah which is which is kind of sad but at the same time if nobody notices your work it's kind of the highest compliment yes if people say like oh wait you did something on that shot it's like yes yeah yes give me an oscar right now it's like just just give it like i had this weird obsession in uh school in ancient film school of trying to talk my professor into letting me do a shot of a single raindrop running down uh played footage like just filming somebody and then making a raindrop just for sure like look it's it's fully integrated you don't see it it didn't let me which was [Laughter] too bad too bad you know it's it's a really really big shame and uh lastly from projects before we jump into like some more of the stuff you learn in the course uh one thing i'd like to show is just the latest water project i did at rp myself uh effects supervised did a few shots on it which is troopers jupiter legacy netflix superhero movie and we were given a storm sequence in the beginning so open oceans and a lot about interactions in storms and this stuff is always fun and it's always super challenging to do um just because the resolution is crazy and you have like dialogue shots i think there's like 60 dialogue shots of them just standing on deck like this yeah and now populate this and make it look good and have it like continuity travel through and all of that stuff that's just always insane and then like doing shots like this is just always fun and dealing with wakes and all of that is yeah that's always a nice challenge so just wanted to show off some of them crazy shots and a lot of this water work is actually done by oh well now my old team at rp by a bunch of young fx guys who joined me straight out of school or like right after their first job and uh learned from me and worked with me for a while and they do shots like this and they look great it's amazing do you guys is the ship animated or is it actually floating on the water so we have a com like lp has a completely custom open ocean tool set for like building the surfaces and with that we also have houdini setups that take animation and then just put it on the boats but you always have to animate right like if you would just do a linear animation of a boat moving you won't get the drops you won't get that hang time at the lip of a wave and all of that stuff so there's definitely additional animation in here um but yeah just a micro wobbles and all of that kind of comes with the toolset and then there's a lot of animation on top obviously but yeah yeah like some of the stuff will cover like some of the approaches to this kind of stuff um we'll cover when we talk about the mail storm and gravity warping and all of that for example none of this is using guided simulations none of this is using the pre-made white water tools like this is basically if you have houdini 12 you can do these shots they'll be a bit more expensive because you don't have narrow band but besides that this is you know as most of the work at lp for example this white water here that's a custom white water solver that i'll teach you how to write in week three there is nothing here that's really like oh side effects presets or side effects tools um which i think makes these shots look better and they make them feel less generic i feel like a lot of the white water stuff nowadays you can really tell which version of the white water solver was used for that so that is really some of the main projects i wanted to share with you guys um that i've done and worked on and supervised over the last few years and some of the crazy amazing work done by the lighters and other effects artists at lp it's um just fantastic people and yeah great work from them yeah yeah dude it looks so good i love it i all like everything you showed us here is like top quality for sure thank you um shall we talk a little bit about one of the cooler things in my opinion that will teach in the course let's do it um that is actually airfield simulations now airfield sins are not really in houdini as of now but obviously you need to do them sometimes so let's say you have a sim like this and yeah that's the sim right kind of high-res and just water but the supervisor kind of says like eh it doesn't really feel like like a nice splash uh it's very stringy right all of that stuff how can we actually turn this into a cool sim without well spending five months tweaking velocities and all that so what we can do with this is turn it into this which gives you a lot more of the soft air feeling to everything it makes it feel a lot more natural as it breaks right you get these hanging particles right away that a lot of people would associate sorry associated with a white water sem or miss simulation yeah the thing is this is all 100 flip there's not one additional sim in here nothing and this is for example the same approach the same ideas as are used for the waterfalls it's just how do i get in right water is kind of simulated in this vacuum of like oh there's no wind in here there's no air friction it's just gravity and then your water simulation yeah this allows you to just say turn on a button and turn on this air friction so that you get this really nice peeling effect of your particles getting pulled away and like bigger masses of water they just stick together they hold together and the smaller they get the more they frail out the more they break up and they just hang in the air nicely and fall at a different speed and the big masses of water so that's a really neat little thing to just have in your arsenal and i thought it would be kind of nice to show you guys the houdini setup for this yeah so this is actually very different than i than any of the courses i will be holding like any of the workshop tutorials uh i'll just walk you guys through the scene of how i did this effect and slower res but in the workshop itself i never have a scene file where i just go like here's the scene fun let's go through we all the stuff you'll learn all the projects we're building step by step well and we built them together by the way right so it's nothing like oh i'll just copy parameters over or anything like that it's just let's build all the nodes let's test all the things together let's debug things together if something goes wrong so let's say we have the sim right and it's low rest at the moment then we want to turn it into something a bit more natural looking the only thing you have to do and the only thing you have to think about is that variable density in houdini is a thing and to turn that on all you have to go is to volume motion to density and then just check the density by attribute option and that allows houdini now to distinguish between distant different densities right so think about oil floating on top of water that's just because the oil has a lower density than water so it'll always try to come on top now water has a density of one thousand air has a density of zero so if we just add particles around our water with a density of zero it'll look like your water is moving in air pretty simple concept yeah pretty easy to do they actually built this it's a bit more complicated not crazy but you basically do all of that in the soft solver which allows you to have the sub context inside of dot networks so you can just build tools you can build systems that use all the tools you're familiar with without having to deal with all the gas micro solvers and all of that stuff so if we go in here all it is is basically a stream that splits out basically i kill all the water so i split out the uh here's only water that gets turned into a new surface field because obviously as soon as you have air particles in there the flip surface field you can just throw that out of the window that's useless because that will encompass all the water particles but the edge of the water is air right so you need to build a new one we just make this guy bigger fill it with particles and then kill all the particles that we just generated that are inside of the water and this way we get this nice little cocoon yeah of uh water around everything right so if we simulate a few frames here we can see that we get this new shell particles generated right here right that just happens on the first round and those particles they're air particles and that's just done by grabbing all the water and then turning it into the shelf particles now if we had particles all over the place this would obviously not work right because then we had particles underneath here which makes bubbles and we have particles back here which slows down the water all of that so all of the rest of the system is basically cleaning stuff up and like making sure the shell is only in the places where we want it to be and this stuff again like all of the stuff i'm going to teach is crazy simple it's the concept and like the higher level approach to things that really make this stuff work if i would just ask you guys oh just put a shell of air around this thing you're good you know like actually building this stuff is not it's not crazy complicated it's also not the easiest thing in the world and linking things together making sure you know the resolution so whatever you change you get the right amount of voxels filled with air around it or let's say you know when it collides do i need to take into consideration that if my particles collide with right for example here this water comes and collides with the rock now if you're not careful you have air particles here that you don't see in display blast now they will obviously collide with the rock first and also it feels like your rock has a force field around it so they will just you know you need to clean this stuff up or hey these particles hang here in the air and you have one particle all the way up here that now has a box of other particles around it increasing your simulation time and all of that stuff right so it's not just building like the idea of like oh let's drop in variable density we're done bye but how do we clean this up how do we link this stuff up and how do we make this really production ready where we can say hey this works and it's scalable and i can say it worked on a rock so let's drop it into a beach and it works yeah and how can we make this stuff scalable and approachable and nice and that's really all the course is about like high level concepts that allow you to do very low level stuff that's really easy once you break it down yeah right like once you know hey this is this is how i can build airfields you can build airfields and you're done right you can always add an airfield to something but if you don't understand why certain steps are necessary or why certain methods work and certain methods don't um and somebody asked you like hey this splash looks a bit cg looks a bit fake can we add a bit more realism to it your first initial approach won't ever be this technically simple approach my goal as an artist is always i shouldn't take longer than eight hours to have all elements for my supervisor of work and thinking a certain way approaching things as like smaller systems that link together and build something great rather than like i need to build this big massive system breaking them down into smaller easy to build chunks will always lead to the best results and the nice part is once you build them correctly you can change and do whatever you want right if a supervisor comes now and says ooh can we have a second rock in here or can we have a guy walking through this the system doesn't care it's all built and it's all fully acceptant of the fact that anything and everything can change the only variable this cares about is i need a cocoon of air particles around my water particles so it does that and nothing else yeah and you can use this for waterfalls and for other techniques as well right like just going to work this exact same system is what's used for waterfalls uh there's some different like cleanup required right maybe you want an area where the water falls faster than others and slower or you want water to be more concise right less of that breakup at the top and then more break up at the bottom so you add layers of control but the core principle again is pretty much the same thing and the same thing if like oh we need to do beach breaks and we need these really nice roughing waves this but the only difference is now we select where we introduce it right we don't just put it on all the water but we add layers of control of where where do we want that throughout where do we want that breaker same thing like the underwater bubbles in the shark they utilize for example variable density and different approaches of how to control this and to really yeah really be in charge of what the simulation does and it's just all lego bricks in my mind right you just built a system that's just a normal lego brick and then if i use that lego brick today to build a fire station tomorrow to build a star destroyer it doesn't matter the lego bricks the same and it just links into the systems and you get your results from there yeah and i feel like all of these techniques are so important to art to direct your simulations because i think a lot of time people have problems or issues they don't know how to art direct stuff you know like you said they they they create a tank and they put some water in it you get a water sim but then your supervisor comes in and he's like oh i'm not super happy with the water sim and if you're unexperienced you're going to be like yeah but that's how the water like that's how houdini does water right yeah it's not really true like you can when you get to know all of these uh techniques you can start art directing water fire destruction and then magic as well like all of these techniques are like you said it's a brick right it's kind of a tool set that you can then reuse in other uh areas with houdini and your effects yeah like just to emphasize how crazy this is i think 99 of all the effects i do if it's water magic smoke uh rigid bodies they all use these methods yeah and not just like oh i take an airfield approach and apply to smoke it's like no they use water sims like i solve 90 of my work with water it's like oh you got to do clouds boom water water that's going to be a water cloud once we had to do for childhood's end we had to do a spaceship coming through clouds yeah that's a water sim that's just a tank in the boat like air spacecraft pushing through the water and then you just convert that into a bdb and use the rest attributes to add noise to it and there you go high-risk clouds that wrap around a ship and look cool whereas a smoke simulation with the spurs or like would go way too swirly right and get way too way far away from the shore all right dude that's crazy water that's crazy to use waters for clouds yeah you want the sand tornado there you go it's a water water yeah and you want magic as well right yeah if you look at all the harry potter magical you know when voldemort and harry are clashing that's just just water like the biggest mistake i see when people show like oh i made a cool flame thrower yeah their source is not water exactly yeah it's gasoline it's kind of spraying yes and 99 of the time it's just a sphere that emits a lot of velocity and that's not what a flamethrower does if you look at any footage it's a line and then things go up and that's your fire the fire doesn't push forward right that's why you'd never get enough detail in it that's why exactly it looks weird and shredded just do it watersome set it on fire yeah and uh you're good it's it's really so easy so what you're saying is that in this workshop people will learn how to do flame throwers clouds water clouds and and magic like legitimately the way i'm emitting water like again i'm not using any of the side effects tools just because of layers of predictability and the way you need to see things in my opinion um that's the same way i do destruction it's this exact same method and i literally explain it on the hands of destruction of like hey let's blow up this chorus and predict where all the pieces go cool let's switch that to a flip sim and there you go we can predict how the first few frames look right like my goal has always been that this industry is crazy with the amount of work we have to do the fun how fast we have to do turnarounds it's just nuts yeah so if you're a good artist you can kick off your shots on a friday afternoon and you can go home and have a weekend and people who aren't in the industry won't really appreciate that because especially with sims if you kick off a simmon that thing takes 30 hours if you're wrong you have 10 people without anything to do on monday yes 10 people without anything to do that's you know a few thousand dollars for the company yeah that probably means that thursday delivery is not going to happen so now you have to explain things to the client side supervisor if that happens too often your studio loses trust with the clients and right that's also a lot of pressure every artist has on their shoulder even at low levels right yeah more so actually at the junior level because we're artists and we're perfectionists most of us and most of the pressure comes from our own mind right so my goal has always been build systems where you know what's going to happen right i never ever want to kick off a white water sim and not know where i will get how many particles i never ever want to kick off a splash sim and not be able to look at the first frame without ever having to simulate a frame like i can just visualize that in the viewport and i will know exactly how that first frame will look and then i hit sim and it adds a bit more detail to it and continues from there right but i can really sculpt all my work i can predict all my work and if you can't do that you're gonna have a lot of rough weekends because you'll have to go in and check your sims right do they look good do they not look good and if you don't do that right either you know your delivery schedule is just going to be kind of messed up and i think we put all a lot of pressure on ourselves through that kind of stuff yeah and the more we know of how to look at things at complicated effects simply and really at a base level the more free weekends we have right yeah that's how it is the more the more times we can actually leave the studio at six pm and not at eight because we had to wait for the first ten frames of the sim and make sure that it works and doesn't run out of ram right um all of those things are super super necessary so a big part like every time we start a new little project we're actually spending the first video like that's like 20 to 30 minutes normally on not doing any effects we're not opening houdini what we do instead is um we're going to like a website with like 3d it's kind of like an open whiteboard and you can just dump stuff in and dump images directly from google you can dump youtube links all of that stuff and we're doing shop breakdowns which means breaking down all the elements we assume as an effects artist the supervisor wants to see and then we're going to order them in the with their dependencies right what do we need first also considering what do other departments need first and then how do we actually talk to a supervisor about these things and get this stuff signed up before you open houdini because it's a million times better to spend two hours of your day getting full approval on your simulations getting that all like that that's all the elements we do and then doing those then it is like hey i'll just hop into houdini and i build my water sim i spent one and a half days on it i show it and now the suit goes like well that's actually not what i had in my mind and now you just waste the time you put more pressure on yourself right you put more layers of oh on on your own shoulders that you just don't have to and that comes back down to the communication right i think a lot of a lot of bigger studios for example they they tell you which parts you have to do as a junior whereas the commercial studios like fill this glass with water and some ice cubes you have two days bye good luck don't screw up don't screw up show us the render tomorrow morning thank you clients expecting a version in six hours um so that's uh that's really what i'm trying to focus on with the course is like how can we make things that look complex that look really high-end that are high-end but make them seem simple if i do my job right you guys will go like oh that wasn't that hard yeah and then you turn around and you do something then you couldn't do before yeah like if if that's the result then i did a good job yeah love it love it so good i feel like you should do another workshop just talking about production and how to because because you have a lot of experience obviously leading supervising projects right i feel like that's a very important skill for juniors mid-level or high level artists to to to have a proper time time management right yes because that's gonna probably save you a lot of those uh crazy late hours uh in the studio if you know that it's not just the artistry but it's the discipline that you have to learn that comes together so i feel like you should do like a bonus but tutorial or something just talking about that i think the workshop's going to be a lot longer than the 25 hours we advertised i think it's it's quite a lot right now just because i i talk a lot about these concepts and those ideas and while building the scenes and showing you guys how to do all of that but i definitely agree like the biggest right that lp i always used to say i promote somebody to a mid-level or senior not depending on what shots they are put right on how much of my attention they need yeah because you can do the best shots ever but if you need two hours of my time a day you're not a mid-level you're you're you know how to do cool stuff but you need a lot of input you need a lot of feedback and technical help and all of that and at the same time you can do pretty cool looking shots but relatively slowly and you're a senior because i can tell you here's a shot and i can trust 100 that i'll get that shot on time yeah and to be able to do that it's not experienced with houdini like a lot of people learning things through workshops the schools nowadays are amazing you know the technical stuff the stuff people don't get is the approaches how do i communicate how do i break things down how do i work with things that are actually in a correct dependency right like what does my lighter need before the other elements like do i do i need to do a whitewater pass on the first thing and then refine it yeah probably so they can start shading it and comp can start comping it and then you just update the sim and everybody hits render and it looks decent right instead of hugging all your things until it looks cool and then you hit render in the right lighting environment and you see your mesh doesn't work now you start fixing right like all of those kind of approaches all those ideas they are really what distinguishes artists in their levels yeah i would 100 agree uh well cool man we're almost running out of time but i do have a few questions absolutely um so learning and teaching like when you started how did you learn what was because back then there was no rebel way right no we we are making things a lot easier for people by bringing like you know industry uh like senior people like yourself to teach because that knowledge is so valuable and we're trying to give it to the people so they can so they can you know don't have to work super late or have some free weekends so how did you start learning and how do you see uh you know how teaching well learning and education was then versus how it is today well that's the biggest difference right you have actual industry professionals talking about stuff you don't have people who are good at stuff working at home teaching most of the work right you have guys who when girls who work 10 hours a day in production and then go home and are so passionate about what they know that they spend another two three hours teaching people yeah and that is unbelievable i mean jesus like that's right resources that's just i love it it's good yeah um i mean no other industry really does that besides arts right like you wouldn't expect like a chemist construction workers to do a workshop on how to construct even though that would be a cool workshop it would be an amazing workshop i would go in an instant but yeah you know you don't have doctors working in a hospital all day and then going to university and teaching they take the day off and do that um most of the people nowadays don't do that they they just do it as a side gig which is fantastic for all the students the way it was for me back then as vancouver film school didn't teach houdini i tried to convince them a few times they shut that down real quick so i actually spend about six hours a day after school which in vancouver film school that's uh one year of education which means you have 10 hours a day of classes plus homework and then i spend another few hours a day reading documentation and just going through note by note and just from a through set putting down every single note in the side effects documentation and see what the hell it does spend way too much time on art force i'm just asking dumb questions and downloading every setup i could get yeah my smoke systems or i like what i learned about smoke i learned from a guy called bunker on odd force like he has this weird scene file called volcano .01 which shows a custom way of doing a smoke cooling down on an ash cloud it's fantastic everything i know about verticals comes from him so doing that and then the only way of getting industry inside was spending every minute i could every walk from home to school every bike ride every time in the gym listening to effects guide podcast and their videos that they're outputting that they used to output a lot uh which is just a bunch of supervisors talking about the shots and how they did it right like the mail storm for example that we'll learn how to do that method actually allows you not to simulate a male storm you simulate a flat simulation that then gets deformed and just looks like it's a simulation where everything is falling down which means your tank is one meter high compared to 20 meter highs which means you can change the shape like oh we just want to make this like a little bit different cool change it we don't have to reset yeah was that the whole message that's how is that how they did the pirates of the caribbean one as well exactly like that whole method nobody ever showed me how to do it but i re like when i was like oh that would be a cool thing i remembered how um that effects guy podcast or where the soup mentioned that in like a three minute bit it's like oh that's a cool idea let's see if i can do that ends up yeah that works and it's actually really efficient so it's just a lot of a lot of time spent uh listening to every little tidbit you can get from anywhere and everywhere right so it was a lot more cumbersome it was a lot more time intensive and the worst part is you learn horrible habits like if you teach yourself that way like you'll have to spend about half the time you learned learning stuff to unlearn stuff like oh i can do this with smoke and that'll be cool and then you go into production work and they're like don't do that ever yeah it's horribly inefficient it's way better to just simmer five times the resolutions like oh but like i can introduce a lot more rest into my sim using like dual rest fields and doing this cool stuff and then we can comp it's like yeah yeah or it can just send two hours longer and we hit render and it looks good yeah like because you just said uh those two hours of sim are more worth money-wise and time-wise than four hours for your comfort and lighter yeah yeah that that doesn't make sense right so all these habits all those things where you're like obsessing about how cool it is if you're in your own little bubble they're useless so learning directly from a professional is just way better yep every time and learning in production is also quite important i feel so that's what we're trying to do here with web away get people well to educate some senior people as well because not everybody is specialized in water right people are coming from you have senior people that are specialized in pyro and they want to learn water they would take a workshop like this which is also one of the questions i had is it better to be specialized versus being like just a generalist is it is it going to help you in the end well it depends right effects artists you will be type cast that's just if you're good with fire they'll give you fire shots yeah if you're good with water you'll get water shots so you might start out not being specialized but you'll be specialized in a degree right i think i'm way too specialized in a sense like i mainly do water yeah so if somebody gives me magic shots it takes me longer to do than other people just because i'm like oh this is not my traditional water um but as an fx artist nowadays you do everything and anything and then you might have like a heavy focus on one thing right so yeah you might be like the smoke guy for the smaller studio but that doesn't mean when they need destruction when they need magic effects you'll still do them but probably 60 of your shots are gonna be kinda angled to the thing you do best but every manager i've ever worked with they every you have at least one meeting every six months with your managers right and they just go do you feel typecast do you want to do something else what are you interested in learning yeah you know like ooh like i told my guys at rlms i love water if you have some destruction i'm at it but i'd love to do it yeah cool so we'll give you water shots with a little bit of destruction in it right that's cool so i get to learn new stuff yes and that's how most supervisors operate and leads operate it's like hey you're good at this but what do you want to do right because we don't just want to type cost people yeah exactly i feel like in a way so if you are specialized it's going to be easier because you will know exactly you you're going to know everything about water right for instance and that way you will kind of uh work faster you know exactly and so that's that's a and that's a good thing but after a while i feel like you get bored by just doing water so you do want to jump into other areas occasionally but that's just me yeah and i mean like i always thought and looked at things like if you're the best at what you do at one place you're probably in the wrong place yeah right like we're all too young none of us are 65 and like who i'm going to retire next year i just want to say exactly all the money in the world for the next two years and just be the king of this thing at a place um if you're in a place where you're not learning things you're probably in the wrong place yeah and that means if they only make you do one thing and you're not learning new things where there's not new challenges attached to that and new new other problems that you have to solve around it it's probably time to go and um being typecast is one of those things right like they're gonna give you shots you're good at and if you're good with fire you're gonna get fire shots if you're good with water you're gonna get water shots but they know that people don't want that so being very good at the thing will help you get a job for the studio that has that need right and that that needs to be clear if you show your show reel and it has all smoke on it and you get a job it's not going to be to do water shots you're gonna get that job to do more smoke yeah and once that gig is done and they like you they might ask you hey what do you want to do next um but whatever you show in your show reel you'll get hired by a studio that has that specific need yeah so if you want to have a wide range of stuff that's very high-end your chance of getting more jobs is definitely higher than if you say i'm super specialized on water like that's my that was my show reel super specialized straight out of school took me a year to get a job right i just knew i didn't want to do anything else so i i i was in a position where i could wait but right if you just want a job you better show a lot of different stuff yeah and i feel like if you are going to do that and i hope we're kind of moving into an area about show reels now yeah i feel like if you are going to show a lot of things just have the quality consistent because if you have amazing water but then you put some very shitty pyro stuff or destruction that doesn't look good and it's just from a tutorial it's better to leave it away until it's good enough so it you you put it into a consistent quality because that will just tell people that you don't know you cannot differentiate the qualities and it's not even that like as a recruiter you look at it and go like well okay the water looks cool yeah is that just a tutorial now yes the fire doesn't look good exactly so i don't trust that they can actually do it yeah exactly like if you have a very short show reel showing only amazing stuff if you have a show reel that shows everything grayscale that's a million times better than showing stuff badly lit and badly shaded because now i go like uh no artistic guy yeah like even if it's a sim showreel if it looks bad it looks bad and it all plays together right but if i see a watterson that's grayscale and it's cool water i'll probably go like well at least they know what they're good at yeah and call them in for an interview because they have that judgment they have that i trust them to be able to make that call of saying hey this works or this doesn't work yep exactly there's a few more questions and then then we're going to wrap this up uh there was a good one you were mentioning uh remapping cpu to ram buses uh is is there any freaking knowledge or resource on that you no idea especially the way we did it will not apply to anybody else yeah right it's all custom it all has to do with your render managers your system like like all of that kind of stuff actually scares me like hell like building a computer i'm like oh which parts work with which part and where is the freaking flow sheet it's horrible so no i have no idea i just know uh eric hermaline our then uh cto did that and it worked yeah yeah that's what i know about it yeah because that's a completely different area of expertise usually people handle all that and super technical yeah yeah it's like oh how do we manage the data centers i i don't know i just know i put data out there and they they do it yeah uh when it comes to rendering shading and compositing uh i believe for this workshop sabre is gonna run through that and show you guys how to uh how to actually shade and comp you know all of the shots that you see in the promo so uh yes uh we're gonna be covering that because that's an essential part of uh producing that quality right like yuri was saying a lot of times you are kind of depending on your team to but it's the other it goes both ways you can you have to know what you're delivering to them and you you have to know that they're going to take it to the next level but you know you kind of need to know their process and vice versa for this to work so yes it is a very important process um well i think that's it from me i i think caleb if you're listening i guess we can play the promo one last time and then we're gonna say goodbye [Music] [Music] [Applause] [Music] [Music] [Music] yes we're back looks so good every time it's such a nice promo yeah uh so are people asking where is the volcano zero zero one five uh yeah jeff jeff wagner i think uh posted it the old school chair from side effects he knows where the good stuff is oh yeah i need to i need to see that as well it's it's it's literally just a sphere with a bunch of micro solvers and uh particle scattering systems and it just i don't actually know if it still works you might have to find an old houdini version for some of those stuff stuff to work but yeah it's uh it's really nice wait is jeff actually in the chat yeah jeff's in the chat oh damn and he actually posted the link to the uh odd force topic and uh yeah then you just have to search for a volcano on a score zero zero one dot set and you'll get thank you jeff uh you're hired to be our chat uh chad specialist for now on and he even offers to fix it [Laughter] it's a that's a really nice file for people who want to get into micro solvers for smoke it's a it's a little little sphere rising as a master class that's cool yeah that's yeah i mean we show a bunch of that how to do it in our workshops but yeah because i was meaning to ask you like where is that file i need to open it up and i mean again it's the same thing of like right you see the file and you like oh this is how it works technically but if you don't have those explanations of why it works why those approaches what are the goals with certain systems then actually applying it on a different topic is a lot lot harder to do right it's it's everything we talked about like you can read the forums and learn stuff from it but it doesn't mean you get it and doesn't mean you can apply to other things yeah awesome well uh yuri thank you again for this it's been an amazing chat uh but yeah thank you yuri thank you uh for being with us uh jeff thank you for being in the chat as well super super cool too to see the industry people like of that status listening to us but yeah dude uh let's let's do more webinars it's gonna be let's do it yeah but yeah the workshop is starting soon it's gonna be awesome i can't wait to see it and uh yuri i feel like you're gonna do an amazing job teaching it so yeah i think people are gonna enjoy it and uh yeah i'm really really looking forward to getting to meet all the people in discord and to see the cool stuff you guys do with it awesome well sweet that's that's that's it for us uh it's been we've been talking for like an hour and a half now uh see you soon with a new webinar all right bye everybody you

Channel: Rebelway

Views: 102,134

Keywords: Houdini Tutorial, Houdini, SideFX Houdini, Nuke, Nuke Tutorial, VFX, FX, VFX Industry, Special Effects, Visual Effects, Simulation, VFX Tutorial, VFX Tutorials, Water, Water FX, Liquid, Water Simulation, Houdini VFX, Webinar, Rebelway, Juri Bryan, Industrial Light and Magic, ILM, Important Looking Pirates, Water VFX, River VFX, VFX River, VFX Ocean, VFX Bubbles, Houdini Ocean, Houdini River, VFX Course, VFX Class, VFX Lessons, VFX Webinar, VFX Live Stream, Houdini Webinar

Id: Uagz4BogC5M

Length: 77min 22sec (4642 seconds)

Published: Thu Oct 21 2021