ThomasMahler
02-27-2005, 12:27 PM
Hello,
I'm currently writing a tutorial - I thought it could be clever to post a first version here to get some comments and critique on the tutorial, so I can change some stuff if I left something unclear or if it's too hard to follow - It's the first tutorial I've ever written, so I could really need some help from my fellow modelers here! :)
Bear in mind that it's not finished yet, I'll try to finish it up as soon as possible, so here we go:
Tutorial: Nerrick, the Dwarf
Introduction:
Alright. Since some people asked me to write a tutorial of how I work and how I approach a modeling job, I want to give you an insight now of the stages I go through while modeling, why I do what I’m doing and what I think is the best way to get a job done within an acceptable amount of time.
First of all, if you’re completely new to modeling within a traditional modeler, you should really read up on the theory behind Subdivision Surfaces and Edge Loops, here are two links that _really_ help you getting started!
http://www.theminters.com/articles/derived-surfaces/
http://maxrovat.sns.hu/subdiv/
I’d suggest that you read these carefully, if you understood the principles behind Subdivision Surfaces, I really think that you could model everything you could imagine. For organic modeling, there’s nothing better out there! :)
Maybe I have to disappoint you for a second, because this won’t be a step by step tutorial – Sorry, but I just didn’t have enough time to write one! But I’ll make you happy again with a video modeling session, which shows the whole modeling process from a cube to a pretty defined cage in the future – I’m currently thinking about some cool stuff that could really help a beginner to get right into the modeling process!
Bear with me if some topics are too general, but this tutorial actually isn’t written for the bloody beginner, it’s written for people who are interested in other workflow techniques than their own, for people who want to have an insight on boxmodeling and so on – Also, it’s the first tutorial I’ve ever written, so please don’t send me hate mails or something like that, if I left something unclear!
Concepts and “soothsaying”:
That one is really important! I can’t stress enough how vital it is to have a strong idea of what you’re going to be working on _before_ you even start to block out your geometry. Always have a sketchpad with you, so you can take notes at anytime… very important! You’ll find inspirations anywhere, so make sure you don’t waste good ideas or you’ll end up sitting somewhere not knowing what you want to work on…
Also I think it’s extremely important to get a “feel” for the thing you’ll model before you even turn on your workstation. You should absolutely think in general shapes, like you would if you’d be quickly sketching ideas around… What do I mean by that? Well, think about things that way: A house is a box and a cone on top of it. A plane is a cylinder, a cone and some fancier rectangular shapes… Reduce everything you see to its bare minimum (like a sculptor does) and you won’t have the problem of “Where to begin?!” or “What to do next?” – You’ll learn to think ahead. Just give it some time, have fun investigating things and you’ll do just fine, believe me.
Modeling Workflow:
Okay, so let’s assume that you already have a strong idea of what you want to model. Now is the time where you’ll fire up a 3d modeling program and start to block out some geometry. This is also an important concept: I’ll teach you how to do “boxmodeling” – It’s the “digital equivalent” to traditional sculpting and is especially fitting for organic modeling. There are no rules of how to approach an organic model, but in my opinion, it’s the best thing to get the general shapes first and _then_ concentrate on the details and stuff. Doing wrinkles and super-defined details will be the easiest part later on, finding the silhouette, contour and form of your model is a completely different beast.
The thing is: If you blocked out the general form and you’re satisfied with it, then your job is already 50% done. After we got the general forms down, we’ll start adding features like the nose, some eyes, a mouth, we’ll start working out some edge loops, try to block out the first muscles, etc. Actually, I split the modeling process into 3 parts, while I’m working in a traditional modeler:
1) Blocking out the form, trying to find the general form, silhouette and contours of your model. Then, we check all of that stuff again, since this is one of the most important steps! Keep rotating around your model, tweaking vertices until you’re happy with the form. One very important thing: DO NOT RUSH THIS STEP!
2) Add features and watch the topology. This actually is a very repetitive job. If you modeled a nose once, chances are, you’ll model it very similar the next time. The same holds true for the eyes, mouth, etc. etc. Everyone seems to have his or her own approach on how to add features… Some people use extremely less polys and build up on that, other people already split as many edges as they need and only pull/push verts afterwards and then, there are people who do all the modeling/transformation stuff ‘on the fly’, it’s up to you, really. And again, we lay down “general forms” first. Some modelers even keep “generic” models and just tweak their shape until it’s fitting for their next project. Not a bad idea, actually, since every human being has the same anatomy. If you keep a male and a female model, for example, which have just the standard form and features, you should be able to tweak them to your liking. If there’s a tight deadline, this could actually be a major timesaver! Drifted off a little from what I wanted to say: You’ll get a routine for this step. Model a nose 10 times and you’ll be able to quickly lay down the exact geometry (in this case, for the nose) that you’ll need on other models as well.
3) Adding details, fine tune the topology, use magnets (soft selections) to shape the model into its final form, etc. This step is a hell lot of fun, since it’s the easiest one and you can still chop a lot of stuff around! Most people actually think that adding the fine details is the hardest part of your work, but it’s not, it’s the most fun one, since you should already have a pretty strong base! You’ll just add edges, tweak verts, do a lot of stuff with magnets (since you’ll probably already have a lot of details in your mesh, so tweaking with Magnets is a real timesaver here!), this step really is not too fancy at all! I could really talk for hours how much I like this step… Often your models don’t have this… ‘Magic’ yet, so you’ll characterize it now and breathe life into your work! At the end of this step, the model should really start to look at you, kinda intriguing, isn’t it? :) It’s really not as bad as so many people think, also, most modelers nowadays support functions like selecting Edge Rings or Loops with the press of a button (and you’ll just connect, slide and/or bevel these edges to get the details into your mesh), so that really makes your job a lot easier. I think the fourth step would be to work on Micropoly Displacements (ZBrush, http://www.pixologic.com), the fifth step could be the whole surfacing process (texturing and shading).
Boxmodeling or Poly-per-Poly (Edge Extrude)?
There’s currently a lot of discussion going on, which technique would be the most appropriate for organic modeling. My opinion on that one: Use whatever you need to get the job done!
There are modelers creating great models through boxmodeling (Bay Raitt, Martin Krol, Ken Brilliant, etc.) and there are modelers creating great models with the Poly-per-Poly technique (Kolby Jukes). I gave you some names, so you can read up on each of them – No matter what technique you’ll use in the end, these guys do a great job working that way, so try to adapt some of their tricks and read or watch their tutorials!
Personally, I prefer Boxmodeling, since you can block out your form a lot quicker and get a better impression of your model a lot sooner than with Poly-per-Poly, where you’ll actually start with pretty detailed areas of the face and have to tweak the form afterwards. It feels a lot more like traditional sculpting too! So, after talking about the general workflow and some basic principles, let’s start working!
1) Okay.
http://members.chello.at/thomas.mahler/The%20Cube.jpg
Say ‘hello!’ to the almighty cube, since Mr. Cube is the base for an infinite amount of models. Doesn’t look like much right now, but I’ll get you going! :)
2) I’m pretty much used to subdivide the model once, just to get some geometry to work with. Silo has a neat “Refine SubD LVL” (‘CTRL + C’) Command (In Wings, just hover over the cube in Body mode and press ‘S’), that is nice to let the modeler add the geometry for you. Magic, isn’t it? :)
Start tweaking verts into their place to block out the basic form. It’ll always be the same procedure: Add geometry, tweak the resulting vertices into their place, add geometry, tweak vertices, add geometry and tweak vertices… Magnet Moves (Soft Selections) can be quite neat here, since you don’t have to move components one-by-one to get the desired shape. This one actually should be quite a fun process, go get some coffee and just start doing it. I could work just on this step for hours, because I love to see how I get something decent out of a cube in just a few steps! And always watch your references while blocking in the geometry, that’s why it is so important to get a feel for your model even before you opened your modeling package – It’ll help you find the form of the object that you’ll model! Here’s a screenshot of what I ended up after 2-3 minutes quick ‘n dirty modeling:
http://members.chello.at/thomas.mahler/2mins.jpg
3) I can’t stress enough just how important it is that you’re satisfied with the blocking process. I’ve modeled pretty fast here, just trying to shape the basic skull, lay out a hair line, a brow line, a nose line and a mouth line, just like you’d do it if you’d be drawing a portrait. Something _very_ important: Don’t worry about details yet! Try to really think in basic shapes! Look at the head, the skull is actually a sphere with a small part on each side cut off and there’s also the nose line/chin area, that is represented by half of a cylinder – You should start thinking _that_ way! One last time: The general shape is important here, don’t even dare to think about details yet, or I’ll eat your mouse!
4) After that’s done, clap your hands and eat some chocolate, since the rest will be easy! We’ll start adding features now. Since eyes and ears really depend on each other, we’ll work on them simultaneously. You’ll probably not have enough detail on the front side of the face to block these features out (in the last screenie I already added some geometry for the eyes/nose area), so just connect some edges and push/pull the resulting points (or even better: Slide the edges!) into place.
5) The eyes are very easy to block out, just split one or two edge loops into the mesh, delete the edges from the “eye socket face” and extrude it in two times. Afterwards, tweak the “outer rings” again, until you’ll find the shape that fits your needs. Magnet translations can take away a great deal of your work here! Try to already define some good edge loops, so you don’t have to fix so much topology afterwards. Oh, and: If you’re adding a sphere at this stage and put it into your eye socket, it’ll be a lot easier to tweak the loops around the eye, since you’ll see this wonderful, curvy thingy, that will later turn into an eye, while tweaking!
Here’s what I ended up with:
http://members.chello.at/thomas.mahler/eyeSpy.jpg
6) The nose is a little trickier to deal with: Connect and tweak until you got the base laid out, then extrude two times: once to get some geometry for the tip of the nose and once to get some geometry to play around with for the nostrils. Just tweak the verts on the nose until you’ll find the form you were looking for! Maybe you have to use connect on the nostril area, so you get some more geometry to play around with. Concentrate and only add edges where they could contribute to the shape! If you’re adding geometry, think about your loops and try to keep most of your faces 4-sided. Pretty easy stuff, huh? (And fun too!) :)
http://members.chello.at/thomas.mahler/nosyMosy.jpg
I'm currently writing a tutorial - I thought it could be clever to post a first version here to get some comments and critique on the tutorial, so I can change some stuff if I left something unclear or if it's too hard to follow - It's the first tutorial I've ever written, so I could really need some help from my fellow modelers here! :)
Bear in mind that it's not finished yet, I'll try to finish it up as soon as possible, so here we go:
Tutorial: Nerrick, the Dwarf
Introduction:
Alright. Since some people asked me to write a tutorial of how I work and how I approach a modeling job, I want to give you an insight now of the stages I go through while modeling, why I do what I’m doing and what I think is the best way to get a job done within an acceptable amount of time.
First of all, if you’re completely new to modeling within a traditional modeler, you should really read up on the theory behind Subdivision Surfaces and Edge Loops, here are two links that _really_ help you getting started!
http://www.theminters.com/articles/derived-surfaces/
http://maxrovat.sns.hu/subdiv/
I’d suggest that you read these carefully, if you understood the principles behind Subdivision Surfaces, I really think that you could model everything you could imagine. For organic modeling, there’s nothing better out there! :)
Maybe I have to disappoint you for a second, because this won’t be a step by step tutorial – Sorry, but I just didn’t have enough time to write one! But I’ll make you happy again with a video modeling session, which shows the whole modeling process from a cube to a pretty defined cage in the future – I’m currently thinking about some cool stuff that could really help a beginner to get right into the modeling process!
Bear with me if some topics are too general, but this tutorial actually isn’t written for the bloody beginner, it’s written for people who are interested in other workflow techniques than their own, for people who want to have an insight on boxmodeling and so on – Also, it’s the first tutorial I’ve ever written, so please don’t send me hate mails or something like that, if I left something unclear!
Concepts and “soothsaying”:
That one is really important! I can’t stress enough how vital it is to have a strong idea of what you’re going to be working on _before_ you even start to block out your geometry. Always have a sketchpad with you, so you can take notes at anytime… very important! You’ll find inspirations anywhere, so make sure you don’t waste good ideas or you’ll end up sitting somewhere not knowing what you want to work on…
Also I think it’s extremely important to get a “feel” for the thing you’ll model before you even turn on your workstation. You should absolutely think in general shapes, like you would if you’d be quickly sketching ideas around… What do I mean by that? Well, think about things that way: A house is a box and a cone on top of it. A plane is a cylinder, a cone and some fancier rectangular shapes… Reduce everything you see to its bare minimum (like a sculptor does) and you won’t have the problem of “Where to begin?!” or “What to do next?” – You’ll learn to think ahead. Just give it some time, have fun investigating things and you’ll do just fine, believe me.
Modeling Workflow:
Okay, so let’s assume that you already have a strong idea of what you want to model. Now is the time where you’ll fire up a 3d modeling program and start to block out some geometry. This is also an important concept: I’ll teach you how to do “boxmodeling” – It’s the “digital equivalent” to traditional sculpting and is especially fitting for organic modeling. There are no rules of how to approach an organic model, but in my opinion, it’s the best thing to get the general shapes first and _then_ concentrate on the details and stuff. Doing wrinkles and super-defined details will be the easiest part later on, finding the silhouette, contour and form of your model is a completely different beast.
The thing is: If you blocked out the general form and you’re satisfied with it, then your job is already 50% done. After we got the general forms down, we’ll start adding features like the nose, some eyes, a mouth, we’ll start working out some edge loops, try to block out the first muscles, etc. Actually, I split the modeling process into 3 parts, while I’m working in a traditional modeler:
1) Blocking out the form, trying to find the general form, silhouette and contours of your model. Then, we check all of that stuff again, since this is one of the most important steps! Keep rotating around your model, tweaking vertices until you’re happy with the form. One very important thing: DO NOT RUSH THIS STEP!
2) Add features and watch the topology. This actually is a very repetitive job. If you modeled a nose once, chances are, you’ll model it very similar the next time. The same holds true for the eyes, mouth, etc. etc. Everyone seems to have his or her own approach on how to add features… Some people use extremely less polys and build up on that, other people already split as many edges as they need and only pull/push verts afterwards and then, there are people who do all the modeling/transformation stuff ‘on the fly’, it’s up to you, really. And again, we lay down “general forms” first. Some modelers even keep “generic” models and just tweak their shape until it’s fitting for their next project. Not a bad idea, actually, since every human being has the same anatomy. If you keep a male and a female model, for example, which have just the standard form and features, you should be able to tweak them to your liking. If there’s a tight deadline, this could actually be a major timesaver! Drifted off a little from what I wanted to say: You’ll get a routine for this step. Model a nose 10 times and you’ll be able to quickly lay down the exact geometry (in this case, for the nose) that you’ll need on other models as well.
3) Adding details, fine tune the topology, use magnets (soft selections) to shape the model into its final form, etc. This step is a hell lot of fun, since it’s the easiest one and you can still chop a lot of stuff around! Most people actually think that adding the fine details is the hardest part of your work, but it’s not, it’s the most fun one, since you should already have a pretty strong base! You’ll just add edges, tweak verts, do a lot of stuff with magnets (since you’ll probably already have a lot of details in your mesh, so tweaking with Magnets is a real timesaver here!), this step really is not too fancy at all! I could really talk for hours how much I like this step… Often your models don’t have this… ‘Magic’ yet, so you’ll characterize it now and breathe life into your work! At the end of this step, the model should really start to look at you, kinda intriguing, isn’t it? :) It’s really not as bad as so many people think, also, most modelers nowadays support functions like selecting Edge Rings or Loops with the press of a button (and you’ll just connect, slide and/or bevel these edges to get the details into your mesh), so that really makes your job a lot easier. I think the fourth step would be to work on Micropoly Displacements (ZBrush, http://www.pixologic.com), the fifth step could be the whole surfacing process (texturing and shading).
Boxmodeling or Poly-per-Poly (Edge Extrude)?
There’s currently a lot of discussion going on, which technique would be the most appropriate for organic modeling. My opinion on that one: Use whatever you need to get the job done!
There are modelers creating great models through boxmodeling (Bay Raitt, Martin Krol, Ken Brilliant, etc.) and there are modelers creating great models with the Poly-per-Poly technique (Kolby Jukes). I gave you some names, so you can read up on each of them – No matter what technique you’ll use in the end, these guys do a great job working that way, so try to adapt some of their tricks and read or watch their tutorials!
Personally, I prefer Boxmodeling, since you can block out your form a lot quicker and get a better impression of your model a lot sooner than with Poly-per-Poly, where you’ll actually start with pretty detailed areas of the face and have to tweak the form afterwards. It feels a lot more like traditional sculpting too! So, after talking about the general workflow and some basic principles, let’s start working!
1) Okay.
http://members.chello.at/thomas.mahler/The%20Cube.jpg
Say ‘hello!’ to the almighty cube, since Mr. Cube is the base for an infinite amount of models. Doesn’t look like much right now, but I’ll get you going! :)
2) I’m pretty much used to subdivide the model once, just to get some geometry to work with. Silo has a neat “Refine SubD LVL” (‘CTRL + C’) Command (In Wings, just hover over the cube in Body mode and press ‘S’), that is nice to let the modeler add the geometry for you. Magic, isn’t it? :)
Start tweaking verts into their place to block out the basic form. It’ll always be the same procedure: Add geometry, tweak the resulting vertices into their place, add geometry, tweak vertices, add geometry and tweak vertices… Magnet Moves (Soft Selections) can be quite neat here, since you don’t have to move components one-by-one to get the desired shape. This one actually should be quite a fun process, go get some coffee and just start doing it. I could work just on this step for hours, because I love to see how I get something decent out of a cube in just a few steps! And always watch your references while blocking in the geometry, that’s why it is so important to get a feel for your model even before you opened your modeling package – It’ll help you find the form of the object that you’ll model! Here’s a screenshot of what I ended up after 2-3 minutes quick ‘n dirty modeling:
http://members.chello.at/thomas.mahler/2mins.jpg
3) I can’t stress enough just how important it is that you’re satisfied with the blocking process. I’ve modeled pretty fast here, just trying to shape the basic skull, lay out a hair line, a brow line, a nose line and a mouth line, just like you’d do it if you’d be drawing a portrait. Something _very_ important: Don’t worry about details yet! Try to really think in basic shapes! Look at the head, the skull is actually a sphere with a small part on each side cut off and there’s also the nose line/chin area, that is represented by half of a cylinder – You should start thinking _that_ way! One last time: The general shape is important here, don’t even dare to think about details yet, or I’ll eat your mouse!
4) After that’s done, clap your hands and eat some chocolate, since the rest will be easy! We’ll start adding features now. Since eyes and ears really depend on each other, we’ll work on them simultaneously. You’ll probably not have enough detail on the front side of the face to block these features out (in the last screenie I already added some geometry for the eyes/nose area), so just connect some edges and push/pull the resulting points (or even better: Slide the edges!) into place.
5) The eyes are very easy to block out, just split one or two edge loops into the mesh, delete the edges from the “eye socket face” and extrude it in two times. Afterwards, tweak the “outer rings” again, until you’ll find the shape that fits your needs. Magnet translations can take away a great deal of your work here! Try to already define some good edge loops, so you don’t have to fix so much topology afterwards. Oh, and: If you’re adding a sphere at this stage and put it into your eye socket, it’ll be a lot easier to tweak the loops around the eye, since you’ll see this wonderful, curvy thingy, that will later turn into an eye, while tweaking!
Here’s what I ended up with:
http://members.chello.at/thomas.mahler/eyeSpy.jpg
6) The nose is a little trickier to deal with: Connect and tweak until you got the base laid out, then extrude two times: once to get some geometry for the tip of the nose and once to get some geometry to play around with for the nostrils. Just tweak the verts on the nose until you’ll find the form you were looking for! Maybe you have to use connect on the nostril area, so you get some more geometry to play around with. Concentrate and only add edges where they could contribute to the shape! If you’re adding geometry, think about your loops and try to keep most of your faces 4-sided. Pretty easy stuff, huh? (And fun too!) :)
http://members.chello.at/thomas.mahler/nosyMosy.jpg