Modeling a Sci-Fi Motorcycle in Blender & How to Build Complex Assets

[Music] foreign [Music] hmm hi i'm jonathan lampell and in this video i want to show you how i went about modeling this sci-fi motorcycle for cgcookie.com i started out by getting the dimensions of a real motorcycle off of google and using that as a rough guide then i started blocking out the main shapes because i'm working with mostly flat hard surface shapes you'll often see me outlining the shape inside view and then filling it in with an n-gon and extruding inwards in this case working with n-gons is going to be faster than working with quads because it allows me to select and work with the loops very easily since they don't wrap all the way around the object i like selecting a vertex or edge and using the ctrl and left click shortcut to quickly extrude to the mouse position that way i can super quickly trace pretty much any shape and as usual i'm making heavy use of the mirror modifier so that i only need to adjust one side i'm also not caring about all of the small details yet and not paying too much attention to my topology beyond just making sure that it's easy to work with since at this point i don't know if any of these pieces will actually be in the final model they're just there for making sure that i have the shapes and the proportions correct and if i have to throw them out and start over for better topology later on once those things are set i don't really mind i'm also trying not to spend too much time on any one piece i need to make sure that my proportions are correct and that all the pieces actually fit together before creating any details as you'll see in a bit i'll have to make quite a few large adjustments to how everything is connected and even deviate from the concept art quite a bit in order to make it work if i had created little details bevels and cuts too soon that process would have been much more difficult and since i only have this one view of the concept art to work with and not a side or front or top view then there's a lot of open questions about how thick some of the pieces are and i have to just figure it out and make it up as i go that's just going to be part of the challenge anytime you're taking a 2d idea and trying to flesh it out in 3d especially if it's a technical one coming up here in a sec you'll notice how i made everything too thin along the x-axis and it's causing all the pieces to not fit together quite right and that'll be kind of a recurring theme here where what the thickness appears to be in the concept art doesn't actually work out in three dimensions but that's okay we'll figure it out as we go i just need to rough in some handlebars really quick and then i can move on to the next stage of refining the block out for that i brought in some modeling sheets of a real motorcycle i'd found and compared it so that i knew that i was at least within the realm of possibility when it came to the proportions of all the different parts of course since all these parts are connected if i adjust one part of the bike then that means i have to go through and adjust a bunch of other parts as well in order to make it fit so it's kind of a snowball effect and that's why it's really important to get this work done early when the objects are still relatively simple i then added a plane and rotated it to different angles like 25 or 45 degrees to see if i could get more of the angles on the spike to be parallel with each other so that the design would look cohesive overall it was impossible to get everything to line up in even increments but it did help a bit i did get a little bit impatient at this point and i wanted to see some bevels here but i felt comfortable doing that because i was using the mesh machine add-on so i know i could unbevel it later if i needed to at this point i felt comfortable enough with the overall shapes that i started adding some of the medium-sized details like the part that connects the body of the bike to the axles there's nothing too interesting here to note other than i used the shear tool to tilt the extruded part of the pipe and i'm making sure to rotate everything in object mode so that i can still adjust everything along the local axes even when i rotate it into place on the bike here i'm finishing up the details of the cutout and then i parent it to one piece and move that into place that way if i need to go back to viewing it as flat i can just hit alt r on the parent but once i put it into context it made me realize that the angle of the arm and the angle of the front of the bike didn't quite match up so more adjusting was needed to get it to fit together again better to do all this adjusting now rather than later here i remake the seat by sketching it out in 2d and then moving it along the x-axis and filling it in but i'm going to have to actually redo the seat a few more times here so don't pay too much attention to this but at least i'm getting it blocked out i then adjust the part of the frame that's underneath the seat so that it matches the concept a little bit better i then take another crack at adjusting the shape and thickness of the main engine part because so far it still doesn't quite all fit together especially with how that rear bar is supposed to connect the back wheel and the engine it just doesn't quite go together yet so i decided to leave it for now and then move on to the other parts of the bike and then come back to it a bit later for the wheels i knew i needed six cutouts for the bolts so to make things easy i started with a hexagon and used an array around an empty so that i would only have to make one section i then added a circle with the same array to make the nuts and then took the original object and gave it some thickness to make a boolean to kind of punch out the shape that those circles are sitting in there i just needed to adjust the hubcap and give it a little bit more depth and make it fit the concept art a little bit better where it has that circle inset right in the middle from there i puffed up the tire a little bit because it was looking a little bit sharp and then gave the size of it a sharp crease next i started detailing the blocky booster looking thing that's attached to the back wheel i don't really know exactly what this thing is or what it does so i'm going to have to invent a function as i go or at least model it so that it looks like it's supposed to be doing something otherwise it'll just turn into boring basic shapes that i don't really know what to do with the concept art has the wheel attached to a chain so we know it's not the main part generating the torque so to me it looks like it could be used for stabilization and or extra batteries or something like that i wish i thought of making it look more like a futuristic rocket booster because really all that would be required for that is to put an inset in the back and maybe study it up some more but oh well the design is vague enough that it's up to interpretation on that note as i'm modeling i have zero doubt that people who are way more mechanically minded than me and know everything that there is to know about motorcycles and science fiction technologies are going to point out every flaw and tell me how xyz shape is impossible and why didn't i use a flux capacitor to power the rear hub converter and why don't i know that future motorcycles won't use chains but for some reason we'll use wingnuts everywhere and whatever and i'm thinking about all this while i'm working so i have a ton of reference off to the side and i'm trying to make as informed of decisions as i can just by observation and common sense but i know that i'm probably going to get some things wrong and i'm actually okay with that i know that this would probably be way cooler of a model if i stopped to learn everything about cars and trucks and became a mechanic and then doubled majored in physics before even starting a blunder but that's just not realistic i'll work on this project with what i've got and hopefully learn some stuff along the way that i can carry on to the next project and so on if there are some glaring errors then yeah i'll probably be a bit embarrassed but at the end of the day i'll learn from that and move on the real goal of this project is to give people something that they'd be excited to try and texture for the last exercise of the fundamentals of texturing course something that they'd be really proud of finishing and have fun with while also being challenged to tackle something outside of their comfort zone and if i can do that then mission accomplished part of what makes sci-fi stuff fun is that it requires you to use your imagination and i'm going to take advantage of that here as i completely make up how this thing works while trying to stick to the shape of the concept all that to say of course do use a lot of reference and definitely do your homework and learn more about how the world works whenever you can but for the project that's right in front of you right now just do the best you can with what you've got i would absolutely hate to see someone get discouraged and not even try to start a project because they think they need to be an expert in xyz subject matter before even getting started i wouldn't say that to a doctor or mechanic of course but this is computer graphics blender is a software that is specifically designed for making things that look cool and that's pretty much it games and movies are always full of totally absurd things that of course could never happen in reality but we accept them because they look awesome and are just plausible enough to make our brains not get distracted from what's actually going on in the story so when you're at the limit of your technical knowledge just try to make things that look great and that'll be good enough for most reasonable people who like to use their imaginations obviously this is not a step-by-step tutorial but more of a process overview so that you can see how larger projects come together hopefully it's helpful to see how i keep my object separated and how i like to start with large forms and work my way down to medium details and then to small details all while bouncing around the model and how i'm really not doing a whole lot that's very special here the problem i have with making tutorials about more complex models like this bike is that with the exception of a few good tricks or add-ons here or there it's just the same few tools used over and over again if you know how to move extrude fill bevel knife cut shear and work with orientations like we talked about in the video on modeling at awkward angles then you pretty much know how to model this bike 99 of complex objects are really just combinations of a bunch of relatively simple shapes once you get past the basics it's really the process that matters a lot more than each individual click what really matters here is thoroughly understanding the fundamentals like when to use n-gons and when to use triangles and how to keep things simple non-destructive and how to not over complicate the geometry i'd say only about one third of the skill of 3d modeling is about knowing the technical tools the second is about good observation being able to analyze reference effectively and constantly keep an eye on proportions and i'd say the last third is just plain patience modeling stuff takes forever especially when you're just getting started and i would probably bet money on the fact that a good number of the problems that i ran into while starting out were due to getting impatient starting to detail a piece before the rough model had the correct proportions or not even doing a blockout in the first place or getting bored with one part because it was taking forever or what have you obviously the more that you model the faster you'll get and the less that will be an issue but if you take your time as a beginner and become very careful about your proportions and your design you can make some incredible work with some very basic tools again i'm not discounting the idea of learning about the things you're trying to model if you generally don't do that then now is definitely the best time to start but at the end of the deadline it's always better to have a slightly incorrect but finished result than a perfect one that's not even close to being completed that goes for how you model as well i love modeling so of course i try to have clean topology and to do things in an efficient manner but i bet as you watch this you'll see some topology that doesn't strike you as ideal and you'll see me do something and wonder why i didn't do it some other potentially faster way that's the difference between watching a step-by-step tutorial and seeing how most people actually work when i make a tutorial i model the object several times find the best way i can to get that end result and teach you how to get from point a to point b as quickly as i can that's hopefully helpful for learning best practices and not wasting your time but that's not how anyone actually works because when you're modeling something new on your own you don't have a specific point b even if you have a concept like this the final shapes that you have to make are still unknown once you know the tools the challenge is about exploration and problem solving i'm not taking a linear path to the end result here because i'm discovering that end result as i go that means i'll have to start over on some parts or even start over entirely and that's okay because i learned what didn't work and in the process thought of more ideas of how it could work every time you make something new it's sort of like hacking your way through a jungle with a machete the first time is really tricky and tedious and time-consuming and you may even have to start over several times before you even head in the right direction but the next time that you have to do it there's already a clear path then the more that you continue doing the same thing the more you'll find shortcuts and fixes and that path will get shorter and more direct every time make enough types of things and eventually you'll have a rough map about how to go about making any shape even if you've never made that exact one before it's not a perfect analogy but hopefully it helps that's why when you see some crazy work online that looks like it's just so much better and you feel like you'll never get to that point it's not likely that the author is just inherently more talented than you or anyone else i mean maybe sometimes but more often than not they've just cleared so many paths by practicing a lot and they can just zip down them super fast and speed means that they can make more details or pay more attention to the big picture in that same amount of time i said it before but it really is true once you learn the tools it's more about the process the workflow the order of operations it's why tutorials and certain add-ons are really helpful tutorials show you how other people have carved their pads so you can spend less time on figuring it out all on your own great add-ons take a bunch of steps and then combine them all into one so that you can do the same things but lightning fast which allows you to make more and more complex work if you don't know why you're automating those steps in the first place though then you can run into trouble which is why i always recommend everyone learn the fundamentals before getting too much into add-ons i want to come back to the topic of add-ons in a minute but we should also probably talk a bit about the project at hand here are a few key things to learn if you want to go about creating models like this first organization i'll be honest i don't name every single object as i go because i don't know if it's going to be in the final model or not anyway and even if it is it'll probably be combined with other objects and the name will be lost anyway so i don't really feel like naming hundreds of objects as i create them only to end up with 50 or so objects at the end also i know i can always batch rename stuff with ctrl f2 if need be but if i know i'm going to need to find it in the outliner or in a long list of objects then i will go ahead and name it to make it easier on myself in the future it's entirely utilitarian and i don't get very ocd about it same thing with collections i'm pretty much just creating everything in one big collection called motorcycle and then making sub collections if i need to be able to hide or select a whole section at once i'll organize things as i go as i need to but mostly i'll do it all at the end when things will be more simplified anyway that's not necessarily the best way to work or optimal for everybody but i find it works pretty well for me of course naming becomes a lot more important if you're working on a team so if you know you're going to be sharing this file with other people and they're going to need to know what's what then that is a good time to actually go through and name things but since i know it's just going to be me on this project then i'm not worrying about it too much when it comes to organizing objects if i need to move or rotate a whole group of objects all at once i'll generally parent everything to one main object and just transform that one it makes it easier to move but it also makes it easier to reset with alt r or alt g if i want to start over but i also turned off relationship lines in the viewport because i think that gets kind of messy if you look really closely you'll also notice that i have a collection called recycle bin and i use it to stash stuff in that i don't really need anymore but i think that i might want later if i'm going to make a big destructive edit i'll first put a copy of the original object in the recycle bin so that i know i can go back to it and make any crazy changes i want with peace of mind you can call it a recycle bin or backups or whatever but i find the idea really helpful of course i'd also recommend saving versions of your file as you go so you can always step back to an earlier point in the project if something goes horribly wrong but this is a handy second layer of protection for objects that i specifically think could be useful later one thing that you'll really want to keep on top of though as you're working is the local orientations of your objects you'll see here that it's really easy to adjust all of these different objects that are all at random angles because i'm making sure to rotate things in object mode so that i can grab or slide vertices along that local axis in edit mode i'm also using the normal orientation sometimes and snapping objects to a custom orientation and if i wasn't doing that this whole project would be kind of a nightmare i made a whole video about managing orientations and some cool tricks you can do with them and i'll link that down below if you're interested at this point in the process i'm trying to figure out exactly how this engine should look it looks great in the concept but when applied in 3d it ends up looking kind of blocky and flat so i'm trying to mix it up a bit to make it more interesting and explore different shapes i'm starting out by copying the design pretty literally but you'll see me move away from that a bit as i go on if you're working for a client or a director that wants you to create everything as close to the concept as possible then there's less room for creativity and hopefully that just takes out some of the decision making for you but since that's not a constraint i have here i'm trying to find a layout that works a little bit better since i don't have a clear picture in my head of how this sci-fi engine works my design continues to be flat and blocky and not that interesting for quite a while but i can start by altering the design in ways that still adhere to the overall design language for example the part of the engine that sort of looks like an upside down f is really flat and blocky in the concept but i decided to make that part a metal pipe instead to expose how the frame is being reinforced and keep that general shape cut down on the blockiness and provide a framework that i can fit the other pieces around and this is an idea that i got from all of the reference that i had off to the side of my screen this one decision to deviate away from the concept art a little bit is going to cascade and help me make more decisions down the road i didn't necessarily have to do it at this point in the process but i also felt like fusing the pipes together with an edit mode boolean which is something i talked about in a previous video about modeling pipes so if you want a trick for that then check out that video as well to work on a project like this you'll probably also want to be familiar with the bevel and boolean modifiers and how to use auto smooth i'm using hard ops here to make those processes a little bit faster but it's not doing anything that i haven't done manually a million times i'm also occasionally using mesh machine to adjust my edit mode bevels after the fact which is one thing that's just not possible in normal blender like i mentioned before though i really only recommend getting add-ons like this when you have a specific problem that you want solved and you know an add-on will help you do exactly that if you're still starting out i'd really recommend learning the fundamentals first i mean i get excited about add-ons just as much as any computer graphics geek and i definitely don't want to discount the value of finding that cool new tool that does exactly what you've spent so many hours doing manually that can be worth its weight in gold and sometimes those tools just feel like magic and as someone who's working on learning how to create add-ons that's the kind of experience i want people to have but as i'm also a teacher i see beginners get way too wrapped up in the tools and attribute way too much to them for example we posted a video earlier this year about a super complex construction vehicle getting put together piece by piece this insanely detailed model was created by kent leading several cg cookie members who all contributed to the project at the end i helped kent animate the pieces flying together into place and when we posted it online there were a bunch of questions about whether we used animation nodes or what add-ons we used to create the effect and it was kind of funny because we hadn't really used any special tools at all just normal keyframing animation yeah it was tedious as heck to animate every single piece by hand and we could probably come up with a tool to make that process faster in the future but sometimes nothing can be just good old keyframes or good old poly modeling because with a lot of shortcuts often come limitations and that just wasn't working for us we wanted the most amount of flexibility and the most amount of control so we could fine-tune each individual piece if you're just used to automating everything then that level of attention to detail just isn't available so that's my soapbox and i'll get off it now but one other add-on that i will mention that i used here is the add curve extra objects add-on that is bundled with blender already so you just have to take it on and it helps me create those springs in the front and i'll use it again for the back creating springs is otherwise kind of a pain so it's a good one about 30 seconds ago or so and throughout this video you may have seen a rigged guy kind of flash on screen as i'm working a little bit and that's a guy that i rigged specifically to sit on this motorcycle to help me with proportions because i realized that obviously a lot of the design decisions about a motorcycle have to deal with whether or not it's comfortable for a person to sit on so i actually added that to my scene and that's helped me quite a bit throughout this process to make sure that i get the placement of the foot pedals and the handlebars and the seat and all that stuff actually in the right spot otherwise i'm just kind of guessing and eyeballing it and i could be way off without knowing it it's really easy to be off in your proportions while modeling because there's not really a whole lot of context if it's just an object floating in 3d space with only the grid as a reference then it's really hard to tell exactly how tall something should be or how thick or how wide but if you add a human into the scene even if he's just standing off to the side a little bit then it becomes much more intuitive to size things correctly because we all know how big things are in relation to ourselves because that's our own reference point so i'd really recommend adding a person to most of your scenes going forward and see if that helps the size of things if you're modeling a room or a car or pretty much anything that's centered around humans using it i'd really recommend bringing in a human into the scene even if it's just a bass mesh like this you'd probably be surprised if you do this to some of your old works and just import this human character in and let's say you're modeling a room if you put the guy over by the door you'll probably notice that maybe the doorknob isn't quite the right size or it's way too high on the wall and it would be awkward to reach or something like that and if you slide them over by the window you'll notice that oh this window is really small or if you sit next to the chair it's like oh man this seat is really really low but it's really deep and so it'd be super uncomfortable and just things like that and by simply just having him in there and moving him around then your proportions are going to be way better now back to what i'm doing in the video that super thick tube that's in the concept art that goes from the back to the engine just isn't going to fit in this design so i've kind of thinned it out a bit but also bent it around the wheels to try to get a similar effect then to move on from that and give myself a break from the engine i knew that at this point i was pretty comfortable with how everything connected to the main part of the body so i started creating the booleans and figured out how the cutouts were going to work for the arms and handlebars also when it comes to the arm in the front or whatever it's called i don't know the technical term but when i posted a work in progress shot on twitter a few people mentioned that the motorcycle looks like it couldn't turn and that's a pretty fair criticism i mentioned earlier that not everything has to be exactly realistic or work mechanically like all the way down to the nuts and bolts but a vehicle does kind of have to turn it needs to go forward and needs to turn it's kind of one of the main things so i think that's a pretty fair criticism but in fact this design actually does work so instead of turning like a normal motorcycle where it has just one large bar then as you like turn the handlebars it physically turns the wheel what you can actually do here is use the arms and sway those left and right and then tilt the wheel axle as that's happening and you can get actually a pretty sharp right and left turn here it's definitely not the most mechanically efficient way of doing things but hey it's science fiction and it looks really cool so i'll keep it another great comment from people who actually know how to ride motorcycles unlike myself was that it didn't look like the seat was great for sitting in and actually turning because apparently when you turn you really grip it with your knees and that's not something i'd really thought about when i had my human character posed so i changed that a little bit later on and altered the design to make it a bit more comfortable really grabbing the frame with your knees that kind of feedback is really valuable and it's helpful to get that earlier on in the design process so that you don't have to change a whole bunch of things that you spent a lot of time modeling so again while you don't have to be a mechanic to model mechanical things and you can make it up as you go do listen to good feedback when you get it some feedback is helpful some is unhelpful and it's kind of up to you as an artist to figure out the difference between the two there have been times where i've ignored feedback because it didn't really fit with my overall goals for the project but other times i've ignored feedback out of laziness and paid for it later at this point in the process i'm still trying things out experimenting with different shapes and seeing what works but because i'm pretty confident that the larger scale details are all pretty set and i'm not going to change them going forward then i know at least in what direction i have to head with these smaller details it's always way easier to come up with new designs when it's in a context of something and you already have some constraints laid down otherwise if you just have to think of new things new shapes new little gizmos to add just add a thin air then that just becomes way more tricky also notice how this phase of working with the smaller details also just naturally leads into optimizing the mesh for the things that i've already created so this part of the process isn't necessarily just about adding details everywhere it's also sometimes about taking away detail from areas that don't really need it and might negatively impact performance i'm still trying to keep things as non-destructive as possible as i go but at this point since things are a little bit more finalized if i do need to apply some boolean modifiers or apply some bevels then i'll go ahead and do that but i'll just make sure to make a copy of that mesh before i do that like i mentioned before and that way i'm making sure that i don't accidentally lose any work and i can always go back to the pre-applied object now this chain that i'm modeling here you can see the model itself is very simple and it seemed to work great but the problem that i ran into a little bit later is that when i apply the curve modifier to this and as it takes a sharp turn then the chain itself actually bends like the metal links um it should only you know really bend at the at the point of rotation but the entire object bends as it goes around the curve and that's not really what we want since it's going to be kind of a smaller detail i decided not to worry about it too much and just kind of ignore that fact and so for this project i think that was fine but going forward if we were wanting to do a close-up for this then i'd probably want to look at using a different technique and make it so the mesh itself isn't actually bending around those sharp corners because it does kind of get warped near the top and the bottom and you'll also probably notice that even though i've been using blender for forever and the curve modifier has been there from the beginning it still takes me some time to set it up every once in a while just because getting everything aligned properly and making sure that it's not like shooting off in the wrong direction still takes me a couple minutes uh if i haven't used the modifier for a while once i had that chain finished then i thought i was ready to go ahead and add in those final touches for some of these objects i liked the way the design was heading so just getting those final bevels just right and making sure that the shading on all of these different objects is good that things are marked sharp where they need to be marked sharp and smooth where they need to be smoothed and all the little tiny things that should be adjusted uh were done so that's kind of the next part of the project and it took a long time um just because i'm going through you know every single object over again so i'll skip over some of the video here because it's just super repetitive and really there's nothing new here it's again just making sure that all the bevels are fine and things line up and just the the very last tweaks but it's really important to get these last tweaks done because sometimes the extra little polish that you do in you know the last like 10 of the project might take a good amount of time but that's really what pushes something from pretty good to really good and that's what i'm hoping for here so even though this part of the process is often skipped over a bit in tutorials including this one because i don't want to bore you to tears because it's not really a particularly sexy topic and there are no new tools to learn here and it's just the same techniques over and over again it really is the important bit of final polish that makes your model look professional clean and crisp here i'm just trying to work out some geometric designs for the bottom part of this bike because it felt a bit too boring and flat and it really needed just something extra to give it some life it took me a couple tries here to actually find something that i thought looked good but again since i had the rest of the bike around it already pretty much set it made the design decisions way easier because i knew what angles i was working with and needed to complement if i ever felt stuck and wasn't sure what to do then i would just bounce over to working on a different part of the bike and that would help refresh my eyes a little bit and just make sure that when i came back to it that i could look at it with a different perspective while also getting other work done that i still needed to do also keep in mind that i was doing this over the course of several days you know this wasn't my entire work for the day i was doing other stuff lots of email and whatever so this wasn't my only task for you know several straight hours there are lots of breaks in between uh focusing on different projects as well and so when you're looking at a tutorial it's easy to think that like somebody is able to just do this all in one go and that's definitely not the case sometimes you really do need to focus on something for a long period of time and it's definitely helpful to kind of get in that flow state and just kind of cruise once you're you know really familiar with the tools and you know what you need to do and you just are able to put on some music and do it and that's great but oftentimes it's also really helpful to know when you need to take a break and work on something else or just do something that's so different that you can kind of reset your brain and be able to come back and look at this a little bit differently by now i'm really getting into the nitty gritty like the different caps for the wires and the clips and things like that and just the really tiny tiny details but i'm careful to only do this in a few select places i definitely want a large portion of the design to be just blank space or negative space to allow the design to breathe a little bit but in some of these areas especially for really technical looking objects we want some areas of really concentrated small details and that gives the impression that the whole thing is very detailed even if it's just those certain areas also notice with some of these circular details the lower in scale that i'm going the less geometry i'm using and i'm doing that on purpose to try to keep it at least reasonably consistent throughout the whole object now there are some cases where if adding a few more vertices to a circle just makes it look so much better then of course i'm going to do it because i don't have a really set poly count limit for this project so i can take some liberties if i need to but in general i'd like to keep the density of the mesh pretty consistent throughout one of the telltale signs of beginner work is giving some random or small or not that important object just way too many polygons so definitely keep that in mind as you're working and again the smaller in scale you go the fewer vertices you want to use one thing you might notice at this point is that i'm still keeping a lot of my n-gons again just because i find them to be easier to edit and change but in some cases they do cause problems with the bevel modifier or if i'm using a subdivision surface modifier then i definitely have to convert them so in those cases then i will convert them to tries or quads but if it's not really causing any harm then i don't really see any big reason to actually go through and edit them because if it's just a flat surface then it's just triangulated by the renderer anyway and it's fine now this piece here was just a little detail that i wanted to add to the bottom of the tail to tie two different objects together and make the intersection of them just a little bit more natural but as you can see it took me quite a bit of time to actually get this to work because the solidify modifier wasn't working quite as expected and ended up with some warping and i spent way too much time trying to fix it and eventually just applied it and edited things manually but sometimes you'll run into those little pieces that look like they should be really simple and the geometry itself isn't complex but your first approach doesn't work quite as expected and you need to spend a little bit more time than you thought that's just bound to happen at some point in a larger project and of course the more practice you have the fewer times that's going to happen but even so as you're learning and as you're working on your different projects don't stress out about it too much because that happens to everybody i also think that i mentioned this once before already but i'm constantly going in and out of local view as i'm working on these different objects and that makes the project feel a lot less unwieldy and messy because i can just focus on one thing at a time and it really helps me to work a lot faster for this next part i didn't really have any information from the concept art for what was going on inside of the top of the inset on the bike i thought it would be cool to have some sort of mount system sort of like on the sides of a gun and that way it could look like you could put on some attachments on here and you can use your imagination for what that would be but i just thought that would be kind of a cool system and it seemed to work well with the overall design so i kept it in i then used a boolean to add a slot as seen in the concept that i'm sure something else some other attachment would get stuck into i also had to make that same sort of inset around all of the different corners of the sides of the bike and for that i found it a lot easier to separate into two pieces one for this flat section and one for everything else and that way i could make the inset on just that kind of angled area without disturbing the end gone in the middle and that just helped with topology in general and keeping things clean the main challenge here was to make sure that these weren't causing any pinching or issues with the bevel modifier and one thing that helped was just giving them different bevel weights than the rest of the bike and those smaller bevels helped everything to tie together nicely after that i went back to the engine and found that this area seemed a little bit blank and so i wanted to put a detail here to fill it in just a little bit i know this area is going to be covered by a bar so it's going to be obscured from side view but you'll be able to see it from other areas and if there's just nothing there then it just didn't look quite right i also didn't want this detail to be completely flat and so i'm tapering it in a bit and giving it a nice curve to kind of transition from the round detail off to the right to the rest of the bike i found that my initial poly modeling didn't really work out super well here because it didn't taper quite right and didn't give us the very curved smooth look on the side that i was after so i found that using a boolean was a much better solution for that and allowed me to get both the really sharp corners as well as that perfectly round side but even after figuring out how to get the shape i still didn't like the way that it looked on the bike so i put it in the recycle bin and tried something else the next shape i tried didn't work either and i realized i'm gonna have to go back to the drawing board a little more literally so i used the grease pencil to sketch out a shape that i thought would work better and this allowed me to iterate on shapes a lot more quickly as you can see the sketch itself is not particularly pretty but that doesn't really matter all i needed was just to know where i should put the lines once i had that then i was off the races i wasn't really sure about adding this much geometry for a relatively smaller detail but i liked the way it looked and figured that worst case scenario i can always just turn down the subdiv modifier and then we'll be good one thing that had been bugging me in the back of my mind this whole time while looking at this back wheel was how unsupported it looked it looked like just going over one big bump would make it just fall off and fly away so i put in an extra bar for support this meant of course that i had to tweak some of the other aspects of the design where it was connecting but they were all relatively minor changes and it didn't end up being a big deal i also tried a couple other things such as adding this cover over part of the back of the wheel because that's something that was in the concept art but after playing around with it in 3d it just didn't seem to fit quite right and it made it look like it had just zero suspension and would be really uncomfortable going over any big bump and i'm not a motorcycle rider so i could be thinking about this wrong but it just no matter how i shaped it didn't quite feel right so in the end i got rid of it then it was off to the handlebars this was mostly just cleaning up geometry and separating things out and using as few vertices as possible as well as making sure that everything was beveled nicely once that was done i took another crack at adding more details to the engine even once i finished the piece that i'm working on here and finished the project i still think that everything looks a little bit flat here and it just doesn't have the the look of believability that i'd like when it comes to complex mechanical systems all kind of coming together so if i do a project like this again in the future i'm going to try to make it a lot more like a three-dimensional tangle of components and look at a lot more reference even though i already had looked at a ton by this point it still kind of lacks the chunky really intricate mechanical detail type effect that i was inspired by initially and i think that's just because it's going to take some practice to think about these details in three dimensions because if you look at the design it's pretty clear that i'm still thinking about it mostly from side view everything seems pretty tightly laid out and fits together really nicely at least in the y and the z direction there's not really a whole lot of layering going on in the x direction and while there is some it's mostly like just flat layers stacked on top of each other and not a complex like interlocking weave of different structures so that's something that i'll try to get better at in the future but in the meantime since i don't really have time to redesign this whole engine for this project i'll just try to make it look as good as i can with the pieces that i already have and by the way the magic selection that i'm using here to make these tubes a little bit more interesting is selecting two edge loops and then using ctrl shift plus to repeat that selection pattern it's a great trick but i didn't actually like the result so i went back to my recycle bin version of the tubes which didn't have the geometry applied and were still curves that could edit a little bit more easily i also brought the wires back and fit those in behind them another attempted design that didn't quite work was adding covers to the handlebars it seemed like maybe it would make it fit with the body a little bit more and add some cool style but i just didn't quite like it pretty far off from the concept art too but that kind of stuff is definitely worth trying when it pops into your head you'll see i changed my layout a bit here too and i have my outliner over on the left and when i'm working with a ton of objects this often helps next i tried a few different ideas for where he could place his feet the main reason for changing this is that with the initial design he wasn't able to actually hug his knees into the side of the motorcycle or at least very comfortably and that was part of the feedback that i got from twitter so by moving the foot pedals back here now he's able to actually do that and squeeze pretty tightly and since the that part of the body is angled outwards a little bit then it should be pretty comfortable next i use the extra curves add-on again to generate a spring to make the backs look a little bit more like shocks i don't know if this would 100 work in reality but i thought it added a lot to the design because that space kind of needed a little bit more detail anyway and i just thought it looked cool okay now on to the wheel i actually gave this a couple tries before and none of them really worked out because i wanted to give this a really cool tire tread but looking at references of motorcycle tires online a lot of the treads were inset where a lot of car tires are outset and having inset treads makes this a lot harder because i can't just take a smooth tire and then array some details around it and smash it together and call it a day the tricky part here is that when i go to inset things the normals get all messed up and we can of course use the normal transfer modifier to fix that but the topology for that has to be just right what ended up working here was drawing out the patterns with edges manually and then shrink wrapping them onto the surface then once i had that arrayed around then i could go ahead and apply the modifiers and connect all of the geometry together this was incredibly tedious and several times while i was doing this i thought i should have just textured these it would have been just fine as a normal app but because i'm a masochist and sometimes fall a little bit too heavily into the sunk cost fallacy i decided to finish it the way that i started i'm jumping ahead quite a bit in the video here just because this was so tedious and took a very long time and i don't want to subject you to that it was mostly just a topological puzzle what made it difficult was trying to align the front part and the back part of the vertices for this asymmetric design so that it actually all flowed together it also had to be reasonable quad-based topology because i knew i wanted to add a subdivision surface modifier to it afterwards now because the design was asymmetric that means i couldn't use the mirror modifier to mirror one side to the other of course and so i had to actually flip it along the x-axis which meant it inverted the normals so i had to apply the whole thing and so now i just have a bunch of geometry that i'm working with and in the end it worked out but it was definitely a little bit cumbersome i'm actually really happy with how it turned out and i'm proud of the result and it was a good challenge but would i recommend doing it this way probably not once i was finished with the tire itself i used a lattice modifier to squish it a little bit on the bottom that helps give the whole thing a little bit more weight and makes it look like it's sitting on the ground and not just floating in space i probably went a little bit overboard with the effect but i'll fix that later also the actual tire that i'm working on is not the front or the back tire but a third one that's centered in the middle of the world origin this means that instead of making completely new tires for the front or the back because they need to be slightly different sizes but i want them to be the same geometry so i don't have to do the same work twice or if i need to make changes to one it will also make changes to the other so i created a collection instance for the actual front and back tires and then if i need to edit them then i just need to edit this one single tire in the middle of the scene and yes you did see that correctly i managed to crash blender somehow while i was working but this was also an experimental daily build so i wasn't too mad about it for the very last step i'm doing some final cleanup and applying the boolean modifiers and getting all of this geometry ready for uv unwrapping because of course we can't uv unwrap with booleans and stuff so i'm doing that last final work and combining different objects together and getting rid of as many vertices as possible and just kind of cleaning everything up simplifying the geometry getting some of these n-gons turned into quads and all of that stuff and that way i'll just have an easier time with unwrapping one of the last minute details that i decided to add was some actual gears to go into this chain here my first idea was to use the extra objects add-on which has a gear primitive that works really well and so i was able to create some nice gears with that but the problem was that i couldn't get exactly one spoke per link in the chain and it just didn't line up quite right so i just made it manually myself and that seemed to work a bit better my approach was to start with the spoke and then use an array modifier around an empty to rotate it evenly around to make sure that it fit exactly correct that worked great but it looked a little bit weird just hanging out there by itself so i made a couple other gears to connect it with the rest of the bike and that way it just made it look a little bit more complicated i guess but also just more interesting this part is mostly going to be hidden by that plate on the side so we're barely gonna see that but i figure if we were to use this in a game or something and we were seeing this from behind and we actually saw the chain moving and the gears moving that would just be such an extra little cherry on top detail that people would really love so i definitely can't afford to add these kinds of details everywhere but just where i think it would make the most difference if people noticed them and i finished making the sides of the pipes look welded together and with that i called it finished i hope that watching the process of how i went about building this was educational and entertaining and hopefully encouraging as well it's a bit of a different format and one that i'm still experimenting with while i love follow along tutorials and those definitely won't stop i think there should also be a place for longer format more broad picture kind of things so let me know what you think in the comments below and of course if you want to download this model and mess around with it you can see the fully unwrapped version in the fundamentals of texturing course on cg cookie there's also a full motorcycle modeling course that you can check out if you want by chris kuhn thanks for watching subscribe if you'd like to see more and i'll see you soon in the next one you

Channel: CG Cookie

Views: 229,443

Keywords: Blender tutorial, learn blender, CG Cookie, blender motorcycle, how to make a 3d motorcycle, 3d model motorbike, 3d model motorcycle, hard surface, sci-fi, 3d modeling, blender modeling, Modeling a Sci-Fi Motorcycle in Blender, Importance of a Good Workflow

Id: EnEWjuigTUU

Length: 44min 27sec (2667 seconds)

Published: Tue Jun 22 2021