Think Tank Webinar - Reach: My Journey with Houdini

[Music] [Music] [Music] [Music] hello everyone hello hello yeah welcome uh my name is scott thompson uh welcome to the excellent adventure of houdini mentorship we have uh some very excellent folks joining us today um i'm the co-owner a co-owner and senior academic advisor at think tank um but i do take a very sort of personal role when it comes to mentorships and specific students uh like the one we're going to meet today i i did take a very active role um it's a beautiful day here in vancouver canada nine degrees sunny outside uh thank you for joining us i know some of you are up very late or very early in the morning so i appreciate your time um and staying up late or getting up early to join us i'm pretty sure you're going to love what we have for you today before i introduce our guests i'd like to say a little bit about mentorship in general and kind of how think tank looks at it um i've been a teacher for more than 20 years and i've noticed a few things about learning and what it takes to be successful in this industry i would say first of all that this industry is not for everybody it's up you know it's tough to get in it's uh hard work once you're in i would say it's tremendously gratifying and exciting which i think is what makes us go um you must be very dedicated and willing to do the work to get there it's it's a tough push i think that with most people it takes about fifty percent of their energy to get ninety percent of the way there and then another fifty percent to get that last ten percent and that ten percent is the part that pushes you from good to great and i think great is where you have to be to make it in this business there are lots of people wanting in business but not everybody can get there so this is where mentorships come in because i i think that last ten percent that final magic piece needs a mentor to get there um a great mentor like the one you're going to meet today is someone that can push you that last part of the journey uh guide you right to the top um please we i know you're gonna have some questions so please keep those till the end if you were to type them in the chat we can get to those later rather than uh expecting us to stop and answer questions in the midst of it we do want to show you the process that we went through here but also demonstrate houdini as well and we'll get to that a little bit but first of all let's meet our guests um the first fellow i met two or three years ago he hails from israel um he arrived in vancouver and i met him uh and we talked about it got to know him a little bit looked at his work he has he has a very solid understanding of cg and started that way i mean he knew a lot he had a couple of pretty big pieces that were very impressive um he also had an amazing attitude uh he always he gave me the feeling that no challenge was too big for him there's nothing that he couldn't do and i have to say i meet hundreds of people looking to get in this business and every once in a while somebody comes along that impresses me with this ability and it's believe me it doesn't happen too much so when i met orion i was cautiously optimistic that he could pretty much do whatever he wanted he kind of expressed an idea it was a little rough still but he had a kernel of an idea that was pretty good uh we talked about it massaged it a little bit um in a lot of cases when someone brings these ideas to me generally i tell them it's too much it's too big these things take a long time they're hard um they underestimate the effort and and the time it's going to take but in orion's case i kind of thought you know this guy he could probably do this you know this is the guy that i think might actually pull this off and uh and so you know we got behind him as a school and we i mean we get behind all of our students but in his case i have to say i i did give him a little extra love just because i saw the potential and i could just see the gleam in his eye i just knew this guy was going to do some great things um that's that's kind of how i felt about him so uh i'll just uh invite orion to say a couple of words brian maybe you can just tell us a little bit about your beginning uh with the mentorship first of all thank you scott for having me here um it's definitely been a ride joining think tank and um i did have a little bit of experience beforehand and very enthusiastic when it comes to cgi and all forms um and i've gotten a lot of support from you and the school in general and um the fact that you supported my uh my crazy idea i guess i i tend to uh go a little bit further than usually what i think i can handle but just a little bit um and having that kind of support having that someone that has your back um it definitely definitely helps you know a person like me to try and push those limits i think that's that's what this industry is all about um so yeah it's it's uh it's been a great ride um it was fun too i mean it was exciting i mean every time you showed me something new you know i i just get excited for you and for the whole process you know um each time i saw you it had changed in a big way and in a good way uh and i know that there were times when you had a little bit of self-doubt um but you were you're taking big bites i mean these were big bites yeah i wouldn't uh i wasn't surprised at all that you felt overwhelmed so uh can you maybe just briefly talk about um just your decision to take a mentorship and why you felt that was kind of the path you wanted to go on yeah definitely so um i basically came to vancouver in 2016. i kind of like scouted the area so what uh options that there are because i knew that i wanted to study abroad for a very long time i studied for a short period in israel and that was the time where i sort of like felt okay i had a certain ceiling um vancouver is definitely the place to do it i scattered around and um the mentorship program and think that definitely what was the highlight because i personally needed someone to guide me i needed a sort of like a mentor basically that is suitable to understand which direction i'm coming from and which direction i want to go to um and that was my major um decision to to make that kind of like move to a you know the other side of the planet and the choice to actually go to think tank um yeah and you you also were taking on not just an ambitious project but you're also getting into a piece of software houdini that is a bit of a bear like it's it's a big beasty program it's it's really just not for everyone it's complicated there's a lot of math coding helps etc so you know you saying i want to go down the houdini road yeah you know i just kind of go okay you know they didn't know that right off the bat i didn't know i was gonna go for houdini yeah so i kind of came from a compositing background and i thought i was gonna be a compositor um i kind of like put that aside because i said to myself you know try something you haven't tried before um i guess that's that's what i'm gonna try and do like practically forever see what i don't know yet and go go there um and you know houdini is scary first time i opened houdini was actually uh before i came to vancouver i was like uh in israel just like okay i'm gonna i'm gonna download it and see what's up opened it started crying closed it and that that was my first introduction into dna and then the decision came to yeah use basically my time in think tank to you know broaden my horizons and we talked about um you know you once you made the decision to do a mentorship we we did a little bit of pre-training with our good friend brandon yeah which was just sort of laid a foundation for you so you weren't an absolute newbie when you got into uh into albert's uh and then you came to me i think i'm not sure if i suggested albert if you brought him to me but uh i i'd known albert before and and so when you brought him up i went well this guy's a rock star so you can't really go wrong with albert you know in fact i would have to say when it comes to albert i have to make sure the student is good enough for albert not the other way around because you can't just give somebody to a very high level mentor and expect uh you know the mentor to really be able to work with them if they haven't got the chops for it because some of it is pretty steep stuff um so on that note let's uh let's let's hear from albert a little bit now just a little bit uh albert's um he's an industry veteran of 20 years worked up around the world for companies like sony image works weta digital industrial light magic to name a few of the the big rock stars of of cg uh currently working at double negative here in vancouver uh he's mentored for us for a while and has always been one of those guys that he just knew your students were in great hands with albert because he really brings his he brings everything he doesn't just bring knowledge he brings emotional support he brings his personality he's got a great sense of humor which sometimes you need when things are going wrong you know you need someone that's gonna sort of help you along when things don't always work out and things don't always work out you know so uh albert shostakovic welcome to our webinar hey hey um yeah thank you thank you for that great introduction as well and um thank you for for having me here uh can you just talk a little bit about when you first met uh orion and just sort of what your first impressions were um um well it was actually funny story because um i i heard about orion um willing to to to have a houdini guidance completely mentorship from multiple people and um then then you scott contacted me and i thought that's that's that's great that's a great opportunity and yeah when i when i met orion i noticed that that he he has this huge project and i was like that's a huge project i was a little bit scared about it but i i noticed that um orion uh was picking up a knowledge like a sponge it was just instant and i knew that okay that that's i think we can do it that's that's doable um because this is always the problem that kind of beginning you when you don't know person you know you don't know you know what what are the limits um you know what what are the possibilities some people you know have some kind of artistic background some people have technical background and you have to figure out what puts your ground and it turns out that that this project is perfect for ryan and and went great well let's be clear about projects in general for mentorships i mean these projects are vehicles these are these are projects that are designed to show off learned skills and so you want to be able to showcase your talent and you need a project that's just right you know you need one that showcases everything properly but also is doable in a reasonable amount of time and i imagine when i first saw the first animatic i was like oh she's you know and especially because we were thinking live action plates a lot too and uh you know there's a there's a whole part b to reach by the way there's a second part that we haven't seen yet but uh you know i i just kind of we talked about it a lot orion and i and it was one of those things where i thought i'm going to let albert make this call because he's the only one that really knows well with reach also when we saw the whole uh whole picture i just realized that it is a great project and he can definitely execute it um but i felt that we might have to divide it into two sections just making sure that we have the first part nailed in and and half um the first you know ground with houdini and then we can move to the second part which which is even more challenging the first part i have to say so so that that would be another ride yeah well fortunately and unfortunately ryan's now at ilm so i don't know whether they're going to give them five seconds rest to do part two i know that that company is uh what's that all right i'm trying to find time to breathe yeah right basics first yeah um i'm going to pass the baton over to my good friend and colleague john biermann john's the guy that looks like he's at the international space station um he's actually not in space although he he appears to be so i'm gonna let john sort of guide us through some of the uh sort of the the more technical stuff as we move forward and we really want to make this uh webinar informational so we really want uh orion and albert to showcase some of the tools they used and some of the techniques and and show us some of the power behind houdini and and what it's what it's capable of so i'm going to pass it over to you john john's a composer he's a videographer he's a camera guy he's a tech guy he's a musician he the guy can do it all so very good speakers so here you go welcome john biermann oh i think you're muted john oh oh he's he's he's he's in my language he's miming all right well we'll work on that that there for a minute and see if we can get john back but in the meantime i'm just going to pass it over to uh to orion and albert and i'll let you guys take it and you can discuss uh the making of this project all right um so hey let's start from the beginning um so so um when i met um orion he already had some idea in his head he he already had a storyboard and i have to say that um orion it you actually have a perfect you know drawing skills um i'm not sure if you can also share with us or something but um this that was our starting point and i remember you were going slowly towards the animatic right you had like a first animatic slides so i remember that most important for me was to to make sure that this project will be not only a personal project but will be also something that would prepare for work uh for for real-life work um therefore we um immediately structured um your project into um like like a project like it would happen in in a studio that means we we have uh created a show with a show code because um in this this big big budget movies usually are secretly coded they have secret codes um and that's what i haven't done before or didn't know about you know and then we named it uh teddy yeah code name hi guys sorry about that can you hear me now hey hey john okay okay perfect do you think troubles what what happened when i was gone um um no i was starting um to to describe how we went into the project the overview of it yeah yeah and you storyboarded this whole thing right yeah you put pens of paper or cintiq to tablet yeah it kind of started where um i i did this uh robotic arm kind of model uh look dev animation uh and at some point i kind of felt lost and i went to scott and said you know maybe you can help me figure things out like what am i supposed to do um and he gave me this advice an idea of just like creating maybe an interaction between a robot and a child and i was like okay i'll develop something on it and i start sketching um a few a few rough sketches um and developing a storyboard for it um and i remember receiving a lot of amazing help from you know students just uh helping me a little bit with composition giving me ideas developing uh you know story aspects sometimes when you have an element you want to know what background is for storytelling's sake um can you throw the animatic up on the screen right now yeah and then we can watch the whole thing like the second part uh yeah that's that it's completely you know together and i broke it down into uh give me some filler guys i'll load it up oh it's really great the second part is great um a little less houdini in the second part i think but yeah the bulk of the stuff really happens uh in the first part in terms of eye candy and um and whatnot but yeah the the robotic arm that i pulled from your art station is that is the kind of prototype idea for for what is it going to emerge from this pod right uh yeah basically um so funny guys uh that the whole button is gone we can you cannot share i mean yeah all right wait wait um uh yeah we cannot we are not allowed uh we we cannot share john i need your help oh dear um i have a share screen option no john i i i see what's happening when the screen goes slower smaller the button disappears okay because i did not have an answer for you there i see it now okay great ryan is that the case for you as well is your screen available to you now indeed yeah um awesome so yeah um this used to be the old name i basically just threw a name in there for you know some dramatic sake and i'm yeah apologizing in advance it's all dramatic um the idea was to basically create a bunch of interesting shots um i didn't really think too much uh regarding to uh going deeply into a full feature film storyline um but the idea was to definitely get a few elements in there that could demo uh a bunch of skills and taking houdini into consideration um so definitely a reentry effect would be a very cool thing to do a lot of smoke um and a dramatic big explosion that uh you know violently uh covers the camera uh and an aftermath to go with it this is basically where the first part kind of stopped uh where the bulk of the effects is um and from this part uh comes a lot more of a character interaction kind of story uh that we see basically something emerging from the pod um and a child from the beginning that you saw very briefly that we don't really know uh anything about so i sat down and basically try to make up a story again a little bit cheesy but maybe think along the lines of interstellar kind of uh world where resources are are kind of lacking and uh humans are kind of trying to come up with new solutions uh descending droids into space and back to get information um so us as of as viewers we don't really understand what's happening and i saw this interaction of this really tall robot looking figure uh sort of like almost intimidating in front of a very small child um but then we see that they have this really you know curious interaction between them um and i didn't really plan on having a plot twist i didn't really know what i was going to do with it but at some point i talked to a few friends and then this whole idea came up that the child is actually a robot and they're transferring information between one another so uh yeah that sums up uh that portion of the storyboard that is definitely cut in between the at the middle uh to um you know create arsenal i think that's beautiful i i didn't actually see that whole animatic before so i didn't even know what the twist was that was the first time i saw it i remember i saw it it looks great and then i thought that okay we have also correct animation and um it's doable but in order to cover all of those topics at the same time um it's a little bit overwhelming so we have to we have to focus first on on the first part where we can you know get into houdini into into your facts and then slowly we can get to the human because um now now we have to deal with a human interaction skin shaders and and so on it's a beast oh the idea was actually to yeah film all this like go out to the field you know there are a lot of nice farming fields here in canada i could have gone anywhere i had multiple chances with scots got offered and i remember i talked to you albert as well and a bunch of friends also offered but um at some point i think i just realized i don't want to go out i want to stay home and it's uh it's easier to do it in cg for me yeah we're thinking about even drone to make some kind of a photogrammetry exactly but yeah the um at some point uh when when we were still designing your your project um going from that part um into into layout i'm not sure if you have layouts uh yeah definitely so did that was the moment where we started designing it a little bit deeper and and further and uh it's sort of kind of a transformed into the um place where we realize that that is actually we can do it fully in 3d uh because uh combining that in a real life uh footage uh is yet another um problem to solve oh that's it yeah that's the first layout yeah because there were many you know many elements first of all it's outdoors um and doing an environment that's convincing outdoors is very challenging i've never done that before um that was my definitely like my first take on that um as you can see some portions are like chopped i remember albert the first time we actually met in sessions the first thing we did was uh see how we could improve the layout as much as we could we tighten it up uh initially it was a little bit more lengthy like you saw on the uh and the storyboard there was a lot of drama to build up uh but we kind of tighten that up and make sure that um you know it's appealing and you know more flowing like a dynamic timing yeah exactly it looks to me like you were pretty much locked with the camera angles and the moves right from the get-go like you and that did not change um yeah this is this is where actually i remember um ryan um and i have to um highlight that fact as well it's a a usage of references how references were very important uh because um you know you you can you can be very um talented um uh dop or director um you know and with with time it gets this kind of experience but if you if you're not a person who creates a cameras on an everyday basis uh it's it's it's something that definitely you need some kind of references and i know um orion you had a nice set of references over places and even do uh this and so not only are you referencing you know the look of a texture or the design of an asset but you're also actually referencing camera angles do i have that right like and hands and moves and chases and that too this is actually something that um if i'm thinking about a lot of movies that i watched as you know as a kid um there are so many references things that we've seen before that um you know we can always mimic because i you know i i trust myself in certain things but you know reinventing the wheel for me is not uh not really something that i feel comfortable doing um so a lot of the stuff that i i'm gonna share the other screen just for uh uref is just uh kind of still weird for me so i'm not oh my god yeah yeah in pure f um i basically gathered a bunch of elements um and mainly shot compositions so you can see that i really did not invent the wheel um a lot of the things that i drew inspiration from men of steel um gravity life um there are all all of the stuff even small though right there all this stuff that you could probably relate in you know shot composition to my storyboard comes directly from that so i would say it's a huge advantage to you know look just observe movies observe as much as you can they did it so right so we can only learn from that and that's exactly what i did i reverse engineered most of the shots that i have ever seen in movies just to put a lot of ideas in my head and this is kind of how this came up um and then you know you can always like re iterate on top of that that's exactly what albert and i did uh because you know you get a reference but sometimes you want something to work a little bit in more more in your favor uh sun direction like lighting in general general composition once your effects come into the picture things kind of shift balance so you always have to keep on top of that regarding the pod itself which is really the hero asset i mean obviously it's the hero acid of this part of the movie it is the character as such like where did you where did you look for inspiration on how to design that uh so yeah that's a good question so in this storyboard i think i took yeah i took from uh just an image of that i found of star wars just uh an escape pod kind of thing um then i actually like looked uh on art station for just a bunch of designs um i didn't really want to clutter clutter it up with something too complex i wanted to acknowledge my skills as a 3d artist i am not the most amazing modeler there is uh there are things that i still have to learn and you know art surface is a challenge um but i found i think uh the concept uh that i felt was uh the most appropriate for me to use uh it's made by fong and um you could find it on art station henry fong um and i think in some ways it's very stylized it has something you know of an element that could look a little bit toyish uh but i really really liked how you know it's it's elegant it's designed in an elegant way that i could also utilize my effects on top of it i wanted to look for something that you know has that kind of shape at the bottom so it wouldn't interfere too much with my you know my effects because everything has to be in a symbiotic relationship at the end and this is this is also the moment where i want to highlight the importance of references uh really it's nothing wrong with having inspiration like that from other great artists great directors it's it's something that that you can harvest and create something new out of it and as a supervisor on everyday basis when when we create any type of effects or or even when we when we do lighting we always gather references and we always discuss with the references so whenever it's a a real-life reference or uh even you know like um animated movie reference or other movie reference that's something that we can relate to so so this is great yeah and albert in our in our talks earlier one of the things that we kind of talked about was how it may seem obvious but what are effects exactly like what does it mean when we talk about effects well that's a that's a good question and effects is actually a very broad topic i i think i i settled down with a description that if there is something that you cannot register using camera and cannot hand animate it's probably a fx um so it's not only explosions not only water simulation but it could be a lot of other stuff sometimes you have to create maybe animated forest or maybe some magical stuff in harry potter um anything that cannot be really honey animated so so sometimes there are some kind of how we call it procedural effects which we can get deeper into that um it can be a lot of stuff i have to say and in the case of reach it's the sparks flying off the um off the pod is this entering for example yes a a sparks a fire a um even like lucky animated um foliage and that was that there was a lot of kind of i would say invisible effects that that is happening there oh yeah you have that yeah so um definitely stuff that uh we we knew that we could go for simulation like albert said i think one of the strengths like many strengths that albert has is to evaluate a shot and you know go towards the approach that would be the most efficient the most you know reliable in that kind of way because when you're working on a big project you always have to make you know certain decisions and that was a huge part for me to understand that not everything has to be simulated like um for instance foliage and um and uh vegetation that we used um uh just systems of vex to manipulate them manipulate points and um you know you could always create something without having to to go into you know simulation and um detail about that yeah here exactly that the uh yeah yeah so these are probably l systems right oh well actually it is but not necessary it was just for for the presentation purpose okay but the idea is that um you know with with an experience of an effects artist in my personal opinion um the the the more experience you have the less of simulation you want to do because simulation um however great it is it's very hard to to control um and when you you end up getting some kind of a feedback a comment change here change that you realize that uh giving um most of your work to the software gives you a a smaller ability to of a control so this is this is where you as an artist uh you you need to um try to to figure it out how to do stuff um maybe other way something that i it's it's really controllable and this this kind of also gets back to um like first movies like the abyss like when we have this kind of a water character coming out there's no simulation there um so in this very example we had um parts that is hitting the ground and and there is a explosion and with that explosion that um we have foliage that that kind of waves and reacts to that explosion so one of the idea was simply have ability control that with some kind of a geometry we could quickly and interactively animate our foliage something that you know completely do not require any simulation um you can think of that it might be uh simple but um when you look at the final image something like this is visible on the screen for split of the second nobody is going to think how it was created as an artist it's important what's the final output of it so um just to quickly show you what's happening here i simply created a few points on a grid and we have our control geometry with with a color i use the color because this is very easy way of controlling um aria in this case i'm transferring color from one object to another and with a very simple math of literally mixing one vector onto another which tells you whenever um our uh copied foliage is going to be upward on just laying down i'm controlling that mix literally using color it's it's so simple but yet again so effective and in the thing is is is not about even simplicity of of this effect it's about being able to distinguish when and how you can make a decision do you really need to seem to simulate that um um i think your choice to use red in this context is especially the whole it illustrates beautifully the power of houdini because red of course is the first of the three vectors in color and red is on x so effectively when you're writing red to a points attributes yours and then you're you're basically saying well x go in x direction which is effectively what you've written and this is also how i do my effects i always try to simplify things like you know we always think uh like they are matrices quaternions and you can go very crazy about it and try to simplify your problem to the zero and one it's it's really math between zero and one is way simpler um second example is a little bit more complex when it comes to to to um to vax for some but the same idea just maybe ability to not only to control where that object is but maybe we want to control how it actually reacts so so maybe it reacts and everything goes down uh with certain distance that we can also say how how wide so this is this is what i was trying to um go through with orion just making sure that we can create effects not necessary with a simulation but mostly understanding um the final output of it this was actually applied for more than one thing that was applied for instance for the vegetation but i used it also for um just eliminating the problem of what am i going to do with the ground like the ground is supposed to explode into a million pieces but there was no need so i see a lot of projects that people do for effects usually are very complex they can collapse a building down and obviously that is a very complex thing to do but at the very end um what happens when a building collapses is that you get a ton of smoke a ton of dust and it's covering up the camera um technically speaking that is definitely you know an amazing way to learn how to do multiple stuff and also show it and attack demo reel but in my case i think what i really wanted to do is go you know go through the shot and see exactly what i need and in this case i didn't even simulate the ground i just created blend shapes and in 3d studio max exported it and used it as my base um and it needed no more than that uh from this direction all you see for five frames is basically the ground elevating a little bit and that's it yeah i know that a lot of people will start thinking about uh grains and uh and and some kind of a now you could do stuff like that which you know it takes time and eventually it will look amazing depends on your angle of camera uh what you're planning it to show etc like stuff like that so yeah i i can also relate to to a few projects i i did in the past in my life um on one harry potter movie for example that there is an effect of characters kind of teleporting so it's kind of a very fast effect of kind of rotation and characters disappearing up here and i remember we got this kind of a task in houdini to to create and we tried with transferring a plate on the points and maybe deforming points with pop maybe with some kind of a forces and then one day a colleague of us ours who is a matte painter he paints usually backgrounds and and he he literally uh took that shot and and painted three frames and and done and that was that was the final effect and this this what realized to me is the application effect yeah it it's it's about a final image it's not about how you get there it's about how you how what you get out of it yeah and yeah that blend shape would just work great because literally that was like what um three frames yeah exactly yeah i mean you really are forced into figuring out what is the most efficient path to to the the final result that you need i think that you know the biggest problem is when you when you get comments because when you simulate something like that and and you have to change something sometimes it's an hour of waiting of you know getting the simulation back yeah so you're constantly thinking about how do i how do i make this as art directable as possible um also i wanted to mention here that um we tried to structure the whole project how um professional vfx project would be structured that's why you see we went through storyboarding and then layout then on top of that when we have effects and then lighting we try to to do really in a very structured way um you know this way i wanted to make sure that when when orion is getting into into the project that it will be uh simple and easy to to to get for example right now in ilm and it's the same type of working yeah we did a structure almost right off the bat i think the the first few sessions we immediately created templates um basically as you can see i got comp laid out right at the top uh slapped right underneath it lighting layout previz and the storyboard which would have gotten here in a different version but it helped me keep track of basically everything if i create a new shot i go through each of the departments export my movies and basically i have nesting layers so i can go into layout check out my layout here or obviously compare it from the previz let's see what changed from that point and if i'm making changes sometimes like for instance with this shot you could see sometimes the angle changes a little bit sometimes timing change changes um or the animation itself so i went through multiple revisions uh per shot which otherwise couldn't could have been like very you know hellish to track it um and especially in like a shot like this which was hyper complex um and i went through so many revisions just trying to improve it and i remember each time i was talking to albert about every shot i wanted to show him like every version and compare it back and forth so that was the probably one of the most important things i've learned the track versions actually when we went through um through that shot that that uh the pot is coming uh free uh for the air uh which one i just remember about the um about the effect of of the fire that that you have created that's also something that a lot of people are asking remember oh yeah so i can definitely even demo it so basically i can show you the basics of my approach um we can even go um like that uh so like we said uh we took inspiration from uh basically gravity life and a lot of other pieces that um you know and sort of introduced a certain level of re-entry fire uh i realized after a lot of research that it's not really fire it's a lot of sparks um that are basically conjoined together to make it seem like it's fire uh because fire can't really sustain itself in space so um we were definitely going for that kind of look um i would just add that the reason why we also took references from the movies because well you can't find real reference of uh of anything and that this little artistic imagination here 71 we found one it was like super blurry and everything was shaking and you know yeah you can't really film it um so yeah we have to rely on that and kind of like the work that studios did they they did it sometimes mathematically so i went and learned from whatever i could from that um and definitely created passes we broke it down to elements and try to investigate when we're seeing that kind of effect how do we break it down into what kind of elements are we seeing so we went kind of like broadly we definitely see the smoke right here we definitely see that there is a high amount of volume of sparks that are kind of like dense together and then there there are the individual sparks that are kind of like stretched out longer so we know there are three uh major layers to that and obviously it's up to the user the the artist uh to add more than that and i went ahead and just gave it a little bit more um you know language of my own um so i definitely had something along those lines i had sparks that are separate and two layers of smoke uh that is one is more dense and volume that kind of trail is behind and another um that is kind of like the strips that are coming from solo edges um kind of like to complement the shape of the pond um so i have multiple revisions of that i'm also going to add here that one of the reasons why we have divided that into different elements is to make sure that we can focus not only on one particular problem at the same time but also um when one of those elements do not work or we have to improve it you don't have to render everything at the same time you already have every other element you just have to put it together so it was easier to have them separate and then put them together in nuke and we'll definitely get to that part with uh using deeps and yeah that you also introduced me to um in this setup uh i basically went uh with a kind of like thought okay there is definitely a vector field happening uh with a wind tunnel and i wanted to sort of visualize it to understand what is happening there it's kind of like an onion shape depending on the shape of the object obviously but uh creating this from a kind of a low-res um smokesome that if i go in you can see that it's not really high-res it's um taking a collision from the object itself after i vdb it um and basically simulating just with um just an uh which i'll show you in a bit and uh basically i think yeah i'm using the valve from the wind tunnel the wind tunnel you could see right here which i dictate based on the direction of the object now um in this setup i'm not really utilizing the optimization that we did but essentially actually before before you go further ryan we kind of did because first optimization that you have done here is like um if you if you guys notice everything is is positioned in an origin in a zero zero zero right well while your your part is actually flying through the space so using at least houdini in back then there was things 17 or 16.5 i think 17. it does not utilize a vdb sparse volumes therefore simulating everything under the angle would generate a huge containers of the volume so we had to normalize it in into the space that would be in an origin and we inverted all the animation so it's along the x-axis therefore we could really focus on a specific section of the animation rather than dealing with something that would be under the angle on top of that we also had to take in consideration that we have to inverse rotations of of the motion of the part um we have to compensate for that motion uh so pod is not moving in this case but air is moving so we reverse the logic instead of pot flying through we did a a word flying uh instead of the part and the gravity force also had to be inverted so that instead of being just down in this case is a little bit angled um because the whole world was rotated so that was those one of your optimizations i remember that i came to you and i said like listen sensei this object is moving at like you know uh i don't know what was the echo speed mach 1 i normalized it to mach 1. it's way more than that the speed of re-entry is insane and i had to normalize it down but also trying to simulate an object moving at that kind of speed is basically going to require someone to create multiple grids if they want to simulate it um but then again the camera always follows the object in this case at least there are there are always stuff that you have to think about where is the camera you know where can i where can i take you know measures to try and utilize better optimization and this is exactly what we did here we by localizing it i didn't have to move my grid my grid could have stayed static completely static um and just creating a wind tunnel that is based on the object of the speed of the object um i could create basically just all the physical attributes uh while still remaining in world space um that allows for a lot more control if you want to do something uh and save time and you know otherwise it would have taken a lot of a lot of effort so i'm definitely creating the wind tunnel that is taking the vector from the object it doesn't really matter what what object it is but uh the new method that i'm doing is i'm taking a low res sim and i'm basically only taking the um the information of the uh vector so i only use it for basically uh was what's it called um advection sorry so advecting the particles through it uh will allow me to get basically the same shape that you saw in the vector field so that's how i'm creating and it's not really a simulation per se i would say i mean there are ways to simulate it directly in a network in one go but again there are so many ways that that's one thing i learned from albert there are practically infinite amount of ways to do at least one thing so i kept even reiterating on top of any effect that i did try to find the right solution for it um and that's kind of how i'm separating three different elements one into sparks and i have another one that it's gonna cook or probably yeah it's pretty good um so it creates that on the fly and the thing that i kind of wanted to introduce is that i can basically i don't know if it will work that way but yeah i can introduce like any object that i want it's going to cook for a little bit so for instance like a cube and i'm just gonna you know give it a little bit of emotion uh directly from just up just the cube i didn't do anything to it um and it will basically take any information that it has if it has any vector information it will take its speed into account and we'll just generate it on top of that um and i can definitely show you like a final result that it's just a cube but you could still generate it from anything so that kind of a procedural way of thinking changed it completely for me because uh before i actually wanted to deanie i used 3d studio max and you remember albert i kept telling you like but in 3d studio max they used to do like this and that but they don't really give you access to any information that happens behind the scenes you don't really get any vector information you you know if you use uh fumefx which i used in the past and it's amazing but if you really want to get into it and make it procedural and work you know as a tool uh in a pipeline that is definitely the way uh the way to go about it um so yeah and i personally learned a lot from that so that's specifically just taking the pig head the famous pig head and applying the forces on it uh just from the basic smoke sim uh the low res one from this i basically advect particles and that that could get definitely give you the gist of um particle friction if you want to do a reentry at any point and so are we sort of skinning these particles as they fly away like how are you making these street clits oops um uh how i'm making the particles themselves so basically um i want to create uh the vector field this is where i visualize that sorry this is visualizing the wind but that wind is a reverse of the motion of the of the pot that i wanted to mention because it is very important to match that speed because otherwise it will look kind of a weird that your simulation could be either too slow for its motion or too fast yeah oh that's another thing that i learned from all the things we're going to share here basically albert teaching me all kinds of like you know they're subtleties but so basic that once you understand them and know them they apply for everything so reach physical correctness at a certain level uh which is important in terms of speed in terms of you know impact velocity uh lighting it doesn't really matter lighting direction but once you have that at least you have a certain base that you know works and you can bend it to your shot if you want to artistically direct it a little bit more you can you know match your lighting a little bit better the next shot if it needs to so that's definitely definitely related to that um once i basically create you can see it's very very simple and i don't think really heavy um i basically feed it into a dot map work that just feeds particles and ejects directly from the vowel so in a practical sense that would be the fastest way at least that i think to create you know um kind of like a system that utilizes particles into a believable level of density that could give you something that looks like sparks and fire um in this case i'm using motion blur in render time um so you could see a little bit drafty but you could see that there are multiple layers uh within it so definitely the fine particles that are getting motion blur directly from the render and the sparks that are a little bit more particle refined lines that you could see that i'm using uh simple tray lines uh so you trail the points and by id you basically just connect them so you can see that we've had a few people asking about rendering i suppose maybe a decent time to mention you are working with mantra here yeah and is it slow yeah uh well anything could be slow anything really depends on what you do so i remember a lot of times they came to albert and again came down with like i used to be right all this time and you know it's a different kind of approach you have to understand a lot of things i'll give you tips right ahead like um i converted all my textures into rap format houdini does this in mantra right right off the bat it does that in render time but you could save that time because it will try to do it again and again and again if you have exrs so i would suggest take your exrs at the very end just convert them into rat um and you could do that from uh the command line uh there there's a little command line you could google it and just convert them and it will be speeding up your render quite significantly um and there are so many other ways you could optimize um if i go into my render settings right now this wouldn't be the best example because i'm rendering on a black background and there are no uh things affecting anything really but every setting that you have here is very vital yeah to basically what you're trying to accomplish if you have uh you know in the limits reflection limit at like 10 it will try to calculate in your environment 10 10 bounces if i'm not mistaken yeah for all your objects and there's no real need for that sometimes you just want to lower it into two just max it like that cap it um and you'll speed up your render um and that goes for a lot of solutions for you know maybe you're rendering a shot that is a little bit noisy sometimes you could you know he's even pokes sometimes that's a good solution and you don't have to crank up you know samples right here or anywhere else um it's about picking your battles i i had a lot of them with mantra for sure i i i can only add here that um it's this is the point where it's important to understand your software rather than know your tool because tools can be different you can jump from one renderer to another from one software to another but you have to understand an underlying idea of it in case of mantra we have been using mantra till today almost every day for for certain uh stuff for for high budget movies and nobody complains you have to understand how to adjust it some people can say that uh maybe gpu renders uh uh are fast yep they they can be but uh for example gpu render might be limited to the amount of memory of your credit credit card graphics card or credit card as well yeah big time all reloaded together and i understand that actually gpus are maybe not so efficient at rendering certain things like volumes perhaps or not is not as efficient or not as not as fully featured not as powerful as what a cpu renderer on mantra can do i mean you know each one of them have a pros and cons see each i i think um every software is going towards gpu even mantra is leaning towards that direction as as i believe um but you know you have to know you have to know your your your software uh your tool set you have to know how to use it and you can definitely utilize it in the best way guys i'd like to just interject here for a moment and to call attention to time i think we should go to q a at about quarter past which gives us a little over 10 minutes i thought it might be a good idea to mention something about compositing or at least um talk a little bit about that if you feel uh ready to talk comp yeah for sure um albert you could definitely give an introduction to sure yes sum up nuke in uh six minutes you have six minutes um so in terms of compositing um of course uh same approaches as in any other visual effects uh studio we would make sure that we are splitting everything into so-called aovs that means all the necessary bits are uh rendered separately so we don't have to re-render them over and over again when we change one thing let's say we have a a broad view of of the shot and there are a few trees and we we want to change just the explosion there's no point of re-rendering trees so we have divided all of this and we have utilized a deep compositing especially um for for your shots because there's a lot of volumes uh in this case so i wanted to make sure that we can render volume or like explosion or maybe just kind of a low hanging fog in in in your white shot on the ground and not necessarily needed to to re-render it all the time so with with deep compositing which is um in a summary it is a way of storing a 3d pixel or rather pixel in a 3d space and you can also store a color in in 3d space therefore you don't have to cut out your renders map out your renders with other elements you can just do it on the fly in the nuke exactly and i would like to first of all apologies for this i am really sorry i remember having a few bad days with this with this shot um make it more organized if you're doing new comps okay but uh websites a scripture it's also good to show what not to do we can do it a lot more organized that's for sure always i have your new script going down um and i'm pretty sure i have a friend over there doing like that um so yeah basically the one suggestion that i have for deeps use them because like albert said uh they could save you time uh but use them for hold outs so basically uh deeps are very very heavy you got a point cloud in a certain way coming from the camera's direction and it could be very heavy once you have all of these together if i combine combine them right now we'll be setting a very long time for that to calculate so i basically make a right note for each single holdout so you could see that i'm basically exporting um holdouts of alpha so i can use in my very messy comp later on but it makes it so much faster if you just um you know lay that out first and then operate with it because you could just re-render only the elements you want just like albert said and really save a ton of time and make your workflow just a lot better so yeah and you you sort of specifying how uh how far back in z space these deeps allow you to see right there they are technically already there like the info is already there based on your camera you can import your camera which uh god forgive me where the hell is it um it should be somewhere here um and once you have the camera and once you have your deeps um again they're in a certain relationship that every point is exactly where it needs to be um to occlude each other um and once you merge your deeps together uh that's that's basically the way to do it because yeah otherwise they get very very heavy yet they do and and they take up all your hard hard drive space so is this something that is is encoded into the exr image sequence or is this a separate file system or how does that once you check uh once you check deeps yeah so it's written into the axr well yeah so um at the beginning um dips that were invented at weta digital they had their own file format and i think ilm who as far as i remember is one of the creators of the xr format with the xr version 2 they implemented the deep data in the xr version 2. so for for last i think 10 years we have deeps inside xr's and the ability to actually keep um by default i think when you put mantra it it will try to uh store deep sing to rat format but you can always change that and just keep it as a xr yeah exactly and yeah many people don't like them because yes i i agree they tend to be very heavy but at the same time it is always a question um depending on your workflow what's faster re-render a very heavy um you know simulation explosion or or some kind of a volume or just do a pre-comp i it's it's it's a tough uh yeah it's some cases in some cases it's um you have to pick your battles and uh some cases when the simulation you know and the rendering of this kind of a volume can take up to hours let's say 10 hours and and doing a pre-comp could take half an hour but yeah it can be sometimes the situation where precomp can take longer than simulation itself not simulation sorry rendering yeah so it's a picking your battles definitely um well i would say we could potentially get into the questions now unless there's anything that you felt you wanted to say um before we before we potentially take a different direction um i i think i can i can um close with um one also another another tip when i was talking about uh being optimized i i had quickly prepared a very simple example here let me do just as another example of why you want to understand your software rather than learning where the tools are in this case i have a a very simple um text and what is happening is we are generating uh this effect of the light of the neon light ourself as a color data that means everything is interactive it's it's not a light it's it's literally a data um just just to show you how it looks i'm i'm giving a set of points and has another input is just a grid nothing else and we are computing what light would compute for some can this can be a little bit terrifying at the beginning and yes i understand it's it's actually not that complex it's it's a very simple uh a code that's just uh goes through every point and checks what's the distance that to the point we we are calculating the quadratic decay of of a point light and also what would be the lamborghini response on the geometry and based on that we are calculating color um but what i want to show you here it's it's about the understanding how computer graphics works per se um because this kind of a knowledge can help you as an effects artist because as an effects artist most the time when you work on on the new type of effects or maybe some kind of a magical type of effect it's it's this is the knowledge that you have to utilize to come up with something new original and um and and and utilize this this this kind of a technical knowledge and artistic knowledge to to uh for for creation um and even we can you know put this kind of a effect into the volume and this is really interactive no lighting needed no no rendering needed and this is physically correct as normally light would uh behave because we are using exactly the same mathematical properties of any physical light that is calculated by mantra or any other renderer that's in the viewport people yeah it's amazing and you're simply just telling each and every point or in this case voxel have this color yeah you're just saying be this color exactly it's we are just literally coloring every point it's it's no magic here it's it's a it's a very simple literally a three lines of of a very simple code but it's not about the code itself it's not about you know grabbing some somebody's scene file it's not about watching a tutorial and and just recreating that it's about understanding because when you as an artist you get the task to create something let's say magical let's say you have a new harry potter movie this is where you have to utilize your imagination combining with your knowledge and this is where where you can combine all of that to to create something new and beautiful at the same time um this so so this this is just an example i we actually use this technique in a few movies that were not released yet so i'm not going to talk about exactly what and how but as as this as a technique it's it's very similar um to to exactly the same what what render would do um yeah that's that's all i wanted to show just um learn how cg works and you can use any kind of a tool you can jump to to blender houdini what not houdini is just a tool that for you to use yeah but it your it allows you to access um cg on an extremely granular level it is all about points all about point points in space and the data that every one of those points has and houdini more than any other software gives you the ability to very interactively very dynamically write information into points based on an endless variety of criteria exactly which makes it just insanely useful for everything yes yes you that that kind of a possibility of a granule editing it's it's great i know that some people were scared of using houdini because it's it is sounds kind of a technical um yeah it does and you know at least 15 years ago when somebody heard houdini that's the same like saying a coding in c plus but today it's completely not um side effects is making a great job to to to make this tool a very useful for anyone and studios are utilizing houdini for modeling for environment for animation for rigging for lighting and it's it's it's a tool that is really growing um out there are studios that are only using uh houdini from from start to finish and some studios are slowly getting into that point and nowadays those tools are so easy to use you don't have to know a single line of coding you don't have to understand even that to to be able to use it sure of course that knowledge will help potentially you in the future to create something unique but it's not necessary to to to do your job um i know a lot of artists uh as senior artists uh which never wrote a single line of code but they generate amazing effects and they are just amazing it's so it's you know this kind of saying that houdini is a technical tool yeah i think that past you can use it like that sure and i i sometimes do i love it but i'm not necessarily you have to well guys um it seems like camera may be down i don't know if you see me or not hopefully you still hear me um let's just move on to a few a few questions and one of the questions that i saw pop up i think it's always an interesting thing to talk about is more the psychology of of learning this stuff um you know he i can't remember what his name was jonathan actually is his name um and he asked you know like how do you stay sane when learning something as deeply complex as this and obviously you know you're hitting some roadblocks and things are getting a little frustrating what what's uh what techniques do you have i think it's the best question to you on i think it was directed at you um how do you stay sane i if anyone will find the answer to that question um that's not me but i can definitely say that you need a purpose you need something to drive you um because the information that's out there is boundless there will never be a point where you say i learned enough i know everything there will never be such a time because information is just so vast um and i often just joke like okay i gotta make room in my brain for this new thing but at some point i think what really helps is having a goal having something like an objective or a target that you just want to reach you just want to set for yourself that will propel you forward whether you like it or not i think that's that's how it works for me um and that kind of brings me to you know every step of the way to try and like break another boundary that's that's all i can get from my my experience yeah too right um regarding uh you know the kind of bigger picture of what you learned throughout this project if you had to do this again and i suppose you're doing part two eventually maybe uh you know what would you do differently or you know how would you approach things differently if you had to do this again so i already went through this question with myself like a lot of times because um usually when i make a shot i finish it i think i finish it it's never finished you can go and look at it and pixel screw it over and over and over again um and i actually for instance like in one of the shots that i showed it went through two different revisions i could probably take it through four other different revisions um it's important to know where to stop but there is so much i would do differently with the knowledge that i have right now um to just say a few even just the setup that i did for the uh friction i definitely want to make it a bit more practical use for um you know uh as a tool not just for that specific shot but maybe for a variety of shots like uh if i'm doing a lot more objects that would be nice to have uh if i'm doing the explosion maybe uh i don't know um i'm always thinking about okay i could have done this angle or that angle but you know it could go in so many different ways um and i'm kind of training myself to let go that that's i'm i'm in that kind of point where i could look at a shot and just butcher it over and over and over again um and i need to train myself to actually let go of something and for part two i think i'm mostly worried about that cut that we will feel okay this was kind of like while studying houdini to as a beginner level and part two will kind of be like okay i know dini a little bit better now so it will probably integrate a lot more complex stuff to you know utilize other things that i don't know so we'll see how that blends in i'm excited for that i'm afraid there's been a couple of questions like somebody asked you know did you model the uh pod in houdini or did you model it in maya and someone else is asked you know um i mean i think people are kind of concerned probably if they've invested in in other 3d suites like do i need to use both or can i do everything in one or what was your workflow between maya and houdini like how how were you able to to break that up okay that's a good question because um i was debating about that i was thinking the first time i'm going to houdini that's a big 3d package that offers basically everything um but i wanted to pick my battles again and just try to focus more on the scene layout and see how much i could create in houdini without going overboard but for the assets i definitely went to 3d studio max you could go into maya just create your assets that's definitely a normal thing in a pipeline working in dna and ilm everybody uses like different programs but you could definitely have your secondary 3d package that you create your models in your uv map you go into substance or mari you texture your objects and then you create your assets and structured folders you import basically everything in olympics format or whatever form that you're comfortable with uh directly into houdini and you know start hooking up your your shaders and you're good to go you can always cross you know 3d packages with no problem uh especially now with usd i think which i haven't explored yet but that's definitely an option uh so yeah definitely feel free i i modeled the pod and i modeled the house and the fence and basically the vegetation is a variety of mega scans and stuff like that but yeah i wanted to add asterisks to that um if you work on your own personal projects uh really does not matter um i would say don't force yourself uh to to to maybe two you don't feel comfortable with you can use blender for that matter but if you think about making a project that would prepare you for for work i would suggest you it's not a bad idea to get familiar with a software that is being used by a company that you wish to to work for because um maybe companies that are utilizing um houdini and maybe if you want to be a modeler and they utilize maya won't appreciate the fact that uh you're you know very fast in free use of your marks because they're not going to to get the software for you they have a pipeline they have a workflow that you have to adjust uh towards to so it's good to to do your homework see what kind of a software is used for in a studio that you wish to work for um but beside that yeah um like ryan said a normal and typical pipeline in uh in visual vfx studios is um people are using uh software that is dedicated for it we don't trap the camera in in nuke we we don't model in houdini we don't do textures uh in houdini there are better software for that you just use what's best for for for the for the task yeah yeah absolutely um another interesting question came up that i think albert you will be well suited to answer um he says instead of using tools to shape knowledge can you recommend a course or reference for 3d computer graphics in general this is coming from a 2d photo background or i guess you know photoshop user but you know general ceo yeah well uh my my personal general uh 3d knowledge was a a set of books especially the older books like renderman books because all that knowledge is identical everywhere every software that that we have today no exceptions work on exactly the same mathematical principles exactly the same physical principles there's no difference so even if you get to to understand how how shading works based on older books that describe it for renderman it's exactly the same everywhere else so the books would be the the best to to understand that and and this is my way of teaching as well making sure that you understand it not rather than know where to put we press which button or what kind of a code to write it's it's not that the case you as an artist you have to be creative you have to come up with your own techniques your own style but first you have to understand it and yes i can probably suggest some some uh some books there i can put some recommendations so um i feel like we maybe have five minutes left we can go longer if you want i mean you know how are we feeling how is everyone feeling all right um is the chat feeling good for another five minutes at least i yeah somebody was asking you know what do you mean is that render man is that a book i think we should maybe try and type out a little list in the chat of useful yeah of useful links and you know one question that i know is on a lot of people's minds is okay i want to learn houdini where should i start who which which youtube tutorial series should i watch which blog should i sign up to and my my entry was really entagma um and then i found guys like cg forge and there's a whole there the amazing thing about houdini and 3d in general and especially houdini is that the more complicated and difficult it is to get into this the more people there are out there trying to help you learn it it's actually kind of it's it's an inverse relationship uh it's you're actually drowning in excellent um tutorial series uh to to learn to to get into houdini i actually want to add to that albert i i want to like explain exactly the reason why i think i did something that a lot of people ask me like what tutorial did you follow and there are no tutorials for uh re-entry as far as i know i look um and there won't ever be tutorials for that exact thing you're looking for um because in the past i i made this portal for instance and um it was based on npc's work on x-men and there were no tutorials on that um and i had to reverse engineer it and understand it in that kind of fashion but here in this manner uh it's like albert told me at the very beginning you could look at tutorials and see what kind of tools they're using and try to like make an average of like all the tutorials you see but these days i actually i barely watch tutorials i can tap into something very specific in a tutorial but at the very end it's like albert said learning the fundamentals and the why of something how something works it will basically give you that ability to assess what it is that you want to do and all the tools that you use before and understood them will kind of like line up for you in a logical sense that that sort of i clicked for me personally because i learned like through albert that you know he has that kind of understanding and obviously there are a lot of really amazing tutorials out there that for sure that could definitely help you that's kind of like for me that's kind of the difference in in that kind of way i just wanted to mention to everyone in the chat room everyone here in the webinar that um albert has just sent me a link um to a amazon list of books which i now can't paste i'm trying to paste this link into the chat and i can't um hopefully someone will do that for me i think it's already there yeah i think so so just just quickly just um just uh one of the the you know series of books um i i own and i went through um there's definitely a lot more and uh you know some of them might be easier some of them might be harder like definitely i always on some kind of a book regarding math uh it's with effects is useful um it's useful to understand some mathematical concepts especially when you're getting into into them so um that's a suggestion the the art of behi and art and science of digital compositing uh this is amazing book for any cg artist this is something that you definitely should go through it explains a lot of concepts uh when it when it goes to computer graphics and the renderman is is more in-depth uh into technical things of how shading works and how how uh lighting will work uh it's it's it's very very good to know there are a few other books that i can i can suggest in general about the lighting maybe um not all of them i personally read but i found that they might be useful i i want you to get into them but yeah just just uh check it out fantastic i think that's probably an excellent place to stop that's a solid uh 90 minutes that we've just had here i think i have the one pretty well hope you guys feel the same way um yes okay yeah maybe i'm not sure if we can continue even in answering on the chart um because i see that there's a lot of questions i don't mind answering them i mean if you guys want to stick stick around for a little longer i suppose we can extend this out it'd be good to get a feel from the chat if uh people are still keen on answering questions um did was there something that you identified uh already from the top i see that somebody's asking if you develop great tracer compute shadows as well yes it did uh um and it's all described so so just to um um answer that question a long time ago i i did say a brief tutorial on my my personal website uh i'll also add it here um i wanted to to explain how rendering works uh how how ray tracing works and i went through a few of those things like explaining how shadow is calculating how how few aspects of rendering are working and it's not the point that um you should know it and you should be able to write your own renderer i mean sure if you're interested be my guest but sometimes when you work on some unique effects you may find yourself that knowledge that you gain here it might be compatible with something that you have to create here and and something will spark in mind that hey maybe i can utilize the knowledge how to uh um ray trace something and maybe i can do projection onto my geometry and therefore i can do some kind of effect this is this is the idea of understanding how how computer graphics works can spark some ideas of creating original effects cool there's definitely a few people asking about learning vex which i'm encouraged to see and i was a definitely a sort of person who thought there's no way i could ever learn to code there's no way i could do this and actually once i started learning houdini um i thought you know what maybe i can and i really should and i think people should if if you're going to do computer graphics it's always going to be useful to you to have a little bit um so you know where do you where's a great way to learn vegs and i think and again in tagma's vex course was excellent and there's also this really great guy junichiro horukawa who is currently doing a vex uh series and it's free totally free he's excellent yeah but there's just so many um my suggestion uh because you know i i came from a bit different background starting over 20 years ago i didn't have any tutorials there was no youtube even but it's vex is just a language computer graphics language like c like c plus plus um if if you get to any kind of a computer computer language coding language not computer graphic language just coding language you find a lot of similarities uh so just understanding any type of language would help you um take the same way of thinking into a new spot whenever it's it's python c plus plus java whatever you do it's it's about way of thinking uh that's that's what's most important yeah and actually with regards to vex um houdini in a way makes learning to code vex and therefore see even easier by providing vops because vox is an amazing way to bridge um to help you understand the syntax and the way that you would need to write your code in order for it to have a successful operation um and it sort of it makes it a little easier to get into coding by giving you giving you the option of laying it out in nodes and box also i just can quickly show you guys that even if you do something involves if you're not familiar with x let's say i'm adding one thing to another you can always right click and say hey i want to see that node that voltage of x and it will give you a bunch of additional information that not necessarily is needed but you get the general idea of how this code works um this is sometimes i actually get into that back in a days where i was not sure how to approach it and but i knew how to do it in notes so i just quickly connected with the notes and i uh checked how the uh how it looks for houdini in the vex side of you and okay that that makes sense um because houdini is um working in vex um every any every vops are actually vex they are translated into vex on the fly so you can always check that out that's one of the way of learning i guess yeah and one important thing to know is that actually working in vex is a much more efficient use of your computer cpu do i have it right like it's better multi-threading i mean with vops it's no difference vex and vox will be the same because it is translated into uh into vex um i would say maybe vex could be a little bit cleaner um but um vex overall is super efficient sometimes it's comparable to to c plus because it's already multi-threaded with c plus plus you have to generate codes that would be multi-threaded python is not multi-threaded so if you compare python to the x always specs will be faster it's just that multithreading that that kills it it's uh but between uh vex and vox will be should be no difference of the speed it's it's the same thing it is translated at the end of x when houdini executes it cool unless you see any other questions what's the albert's website okay i'm going to unwrite them that's very old website i haven't updated it for i think uh seven years but the information i was mentioning it's still there um john you're muted again oops um there's a there's someone asking about how houdini is used in a lighting environment um is and regarding solaris can you talk to that at all um i can say that solaris is a very new thing karma is still in beta i know that some studios as far as i remember dreamworks were showcasing their use of solaris from very get go so some studios have a very nicely implemented already some students are slowly looking into that um usually studios are not jumping into new versions of houdini straight away because implementing a new version of houdini takes time sometimes a lot of libraries have changed sometimes you have to redo some kind of a tool set it takes time so it's not so easy hey we're going to update to 18.5 tomorrow and also a new version of software sometimes mean that they're not necessarily as stable so it is a very big risk so i would imagine that most of studios will be using 18.5 in a year or maybe two um i'm not sure about the solaris like overall i i'm i i've yeah i think only dreamworks and maybe rising sand pictures i'm just guessing here i have no idea and um this talks to usd right this is all to do with the implementation of this new uh system it stands for universal scene description and um well you know what i'm not gonna try and explain any more about it please tell us a little more about usd uh well to be honest i'm new to usd as well so uh usd was created by pixar they came up with this kind of universal description of the scene file because we have a universal description of the um of the file formats like uh fbx was one of the attempts uh to to to be shareable alembic was the second attempt but there was nothing that really encapsulates everything and uh usd was was this kind of a answer to it just let's standardize it let's make this universal um so so renderman um as far as i remember came with that idea first and they started to to spread out and people noticed that yes there is a big potential it would be awesome if we could use exactly the same power format in every software you know no conversions needed no loss of information because everybody is using the same standards so that it's slowly getting implemented the same like asus uh it's kind of a standardization of the color across multiple um software it's just to to make a workflow of the of the company whichever the company is a little bit easier right so the idea is that you could build a scene in houdini uh you're not exactly exporting it as a usd you are simply just sort of that information you are writing you're writing a file and then unreal can access that or yeah that thing can access that or unity or anything and then you're getting the same you're getting all the same uh points attributes that you get from a houdini like that is a standard that cascades through so if you were to write point attributes in your usd you will be able to access that in unreal in some way yes that's that's a general idea so usd as a standard has some specifications you have to follow and as soon as you follow the specification it will be universally accessible for any other software that's that's the idea i am time for any more we're coming on to two hours here [Laughter] um i think we answered most of the questions yeah i mean there's there's one other that was sort of maybe piqued my interest a little bit um realism versus cartoon stylization in terms of is houdini better at one or the other i would imagine that working on cartoons in houdini would be fantastic well look at the sunny and how they achieved their effect in spider-verse right sure i mean a lot of that looks like they looked like a lot of comp too well yeah of course there is a lot of work there is a lot of you know uh lighting and it's arnold and there's a lot of compositing on top but there's also a lot of tools in in houdini so so you know i used it for that dragon ball v exactly of course he did it was amazing i wasn't sure if you used maya or houdini for that but i haven't definitely used for the month surrender but the aura for instance for like a dragon ball z kind of aura so you have to flatten it out and i talked to albert like i do always do try to look for ideas like there are so many ways to do it and then suddenly yeah well yeah houdini could definitely be used for like her toony effect the second i did that but like there was no chance i could have done that in any other program because i had to literally dig into vexed and bottled with that so that that's the answer and i think the procedural nature of houdini will just allow for so much more expression in the animation you can try so many things out with your you know your stretches and your squashes and your anticipation and all that kind of stuff like you could really leverage houdini as a massive time saver in the animation process because of the rigs that you build yeah i think i'd like to hand it back to scott who would like to say some final words before we close this out well first of all i just want to say thank you to uh orion and albert and of course john and carlos for taking time today to share this information with us i know there's always more i mean it's like anything uh this can go on and on and on um but i would also just like to mention that you know one of the key things that came out of this relationship between orion and albert is there that they are connected now and uh orion referenced a moment ago working on another project who does he reach out to he reaches out to albert you know albert's one of those guys he's always getting a paycheck for this he does it because he's a great guy he does it because he really cares and i think you'll find that the vast majority of people in the industry do give back you know they do share their secrets they do uh encourage the up and coming artists to be to work hard and to i mean you know give them direction and give them advice i would always encourage all of you to never be afraid to ask those questions as a school think tank i often help people that have really nothing to do with my school um because i'm gonna it's gonna come back one day i'll get it back in one way or another and i'm not really too concerned about that i'm more concerned that that they get the right information and that they're well informed uh i i just it always pains me when i feel that people are going in the wrong direction and and i know that uh you know money is money and time is time and it's a it's a big deal and uh i often say too with orion is a good example this isn't just career changing this is life changing you know you get an opportunity like working at ilm in vancouver i think vancouver is a wonderful place to live i've worked and lived in in europe as well and i had a great time there but these things and what you know can change your life so don't be afraid to take on something because it could open a huge door for you uh and working with people like albert uh and orion and john and uh you know these these are the guys that can really help you get there so i'm just gonna quickly share my screen not not this isn't a big cell i'm not really concerned about that i'm we're just concerned i just want to share this let's see here if i can just get this to go up um let's just share this one i'm not sure if this is showing the right screen if i'm showing my mentor selector am i yes yes you are scott yep so you can see there's a vast number of mentors working with us but this isn't the limit you know i would say that you can go off menu and you can uh in i i think in some cases like in orion's case yeah well name came up and i was like albert yeah now so you know we do mentorships at the campus but obviously with the situation in the world right now we're doing a lot of mentorships online um and i'm not just saying think tank's the only school that you can go to there's lots of good schools around the world but i would say some are better than others so do your homework you know really do your homework ask people ask people that are successful how did you get here who did you talk to what you do uh those kinds of uh questions will get you good answers and lead you in a good place so never be afraid to ask those questions you can ask me you can ask albert you can ask orion or john everybody is going to give you roughly the same message but i have to warn you it's a lot of work you know and you have to be ready for that this industry can be a bear but i got to say i worked in it for about 10 years and i had a great time you know i really had a great time i think it's a real privilege to work in this industry uh the projects that i can't even imagine the projects that albert's had to work on over the years i mean really so many cool things and he must just have this huge shopping list of amazing experiences and orion's just getting started you know he's the he's the albert of tomorrow i think it's a really cool thing so i'm really excited for him he's getting in on the on the top floor i have to say i'm a little jealous yeah yeah it's uh good for you you totally earned it all right so congratulations that would have happened if i didn't came into your office one day at the very very beginning it said listen it's god i'm loved i don't know in the right direction i don't think any of this would have happened if i didn't do that but was too scared to say something like sometimes that's something i want to say to a lot of people just don't even be scared of asking for help i i just i did it and it helped but people were there to help me and think that so yeah it really is true it's an amazingly supportive community of people who are you know learning houdini and 3d in general um it's surprising as long as you know how to write ask the right questions and demonstrate that you've done your best to find the answer yourself once you once you're able once you're at that level where you can ask the right questions in the right way people want to help yeah that's true and you know if you have any questions you want to get to us you can always go on our website and send us the question they do funnel back to me in a lot of cases depending on the question you're asking um but and i i won't always tell you that we're the right school for you in in even in orion's case i i almost i outsteered him to one of my competitors initially but he said no he said no i've my friends at think tank i i understand the way you guys do things the mentorship is a better fit for me and you decided to come despite me sort of saying maybe we're not the right one for you you know we're willing to do that we're willing to say maybe this isn't the right thing for you but in his case it did turn out really well uh i think the relationship with albert was key obviously um but we have lots of talented hard-working really sweet people working with us um albert's currently mentoring two people for us and uh you know i think that it's it's exciting it's really exciting so i love that so anyway that's all from me i don't know if you guys want to add a couple of final words um somebody asked about dli approved what does that mean to be dli approved are we think tank is deal yeah but that's just the uh the name yeah the designated learning i see this is some kind of certification for yeah we are we're sort of certified in almost everything there's a couple things we're not but it is because we're a private school i hate to say it but the private schools tend to be slightly more progressive than the public schools and that's just the that's just the the ergonomics of a public school they're they're bigger slower moving etc whereas i can go and hire albert in a heartbeat you know it takes a phone call and i can and i can hire or an email and it's really it's really easy to do so we do do that a lot and and lastly i think it is possible to hire albert directly you have a website the vfx mentor is that is that right and people can yeah people can sidestep use god directly oh albert you didn't tell scott about that girl sorry to blow you off so um i in addition i i was doing this kind of a schooling for anyone if if it's somebody professional working already in the industry uh so not necessarily have a time for school then um yeah i'm very happy to help that's that's the general idea yeah definitely if you're starting from scratch i would suggest you going for the school because you should have first a a basic knowledge and and know how it works and then with the think tank center you have all the mentors with a different departments and you can learn um overall how it works and then on top of it ting tang center gives you a mentorship what i'm offering um uh separately it's it's something that's um you know if you need a quick tip for or if you're maybe working somewhere or anything um so you know it it's it's it's it depends where you stand i would say yeah if if you for example you know can it cannot go to school for some reason uh then definitely you you can contact me uh but otherwise yeah i would suggest you that school will give you you know even much more because uh scott is going to give you place a computer and other tutors you have an access to to a lot of the stuff yeah we do train a generalist foundation first uh in in every area and then we want you to specialize and work with somebody like albert yeah but and you know what i'm not offended if if you want to go work with albert directly okay there's enough albert to go around but yeah i'm not going to give you a computer and i'm not going to teach you you know um other other aspects it's let's say um you know more more direct and you know what who knows one day orion might be on my mentor list oh i think it's very likely i'm still telling orion to actually start creating his own tutorial maybe you should do a tutorial about re-entry right right i mean you did say that there isn't one out there so yeah yeah so all right well i think that's it from us today everyone uh carlos is posting the link so you can watch the recording um and there are links for the school if you're interested in that and of course uh albert's posted his link so if you want to go to visit his site i'm sure you can get a hold of him there um thank you again for your time everyone appreciate it on a on a nice saturday you're sitting indoors staring at the screen again so we are turning we are turning thank you to everyone involved that basically yeah took me under their wing that's not taken for granted and also the right fit to make that happen uh because things just like felt like the place they puzzled for me and i think for a lot of people they just went to think tank and that happened to them um so i just want to say thank you and yeah that's that's a privilege for me and all the authorities stand up in the great studios like you well thank you ryan that's a win-win yeah so i don't know how to turn this thing off anymore just quit smash through smash the escape button smash the like button bye [Music] [Music] [Music] uh

Channel: Think Tank Training Centre

Views: 2,044

Rating: 5 out of 5

Keywords:

Id: LrXoLbLcJaQ

Length: 116min 38sec (6998 seconds)

Published: Thu Jan 21 2021