Thilakanathan Studios

Creating a Photorealistic 3D Backyard in Blender - Course Trailer

Improve your 3D modeling/rendering skills by creating a photorealistic backyard scene from scratch. Perfect for beginners.

Enrol now!

Creating a Photorealistic 3D Backyard in Blender - Course Trailer

Improve your 3D modeling/rendering skills by creating a photorealistic backyard scene from scratch. Perfect for beginners.

Enrol now!

Creating a Photorealistic 3D Backyard in Blender - Course Trailer

Improve your 3D modeling/rendering skills by creating a photorealistic backyard scene from scratch. Perfect for beginners.

Enrol now!

Creating a Photorealistic 3D Backyard in Blender - Course Trailer

Improve your 3D modeling/rendering skills by creating a photorealistic backyard scene from scratch. Perfect for beginners.

Enrol now!

Creating a Photorealistic 3D Backyard in Blender - Course Trailer

Improve your 3D modeling/rendering skills by creating a photorealistic backyard scene from scratch. Perfect for beginners.

Enrol now!

10 Things You Should Do To Create Awesome 3D Environments (Part 2)

Have you ever attempted to create a 3D environment and found that your final render falls a little bit…flat?

Most likely, you may have missed an important step in your scene creation process. Here are 10 things you should focus on if you want to give your 3D environment the best chance of looking awesome and get one step closer to achieving photorealism.

We continue from Part 1 to show you ways to bring out the best of your 3D environments.

Use a physically-based renderer

A physically-based renderer is a renderer which simulates lighting exactly as how it is represented in the real world. It produces results far better than raytraced renderers. Unlike raytraced renderers, a physically-based renderer knows calculates how to bounce light rays from objects to other objects and this can help easily simulate photorealistic lighting.

It would probably be a bit complicated to understand exactly how it works. From an artist’s point of view, a physically-based renderer helps produce photorealistic lighting and textures with very little effort compared to the old raytraced renderer. Your 3D environments will look really awesome

The downside to physically-based renderers are that they are slow. Really slow! This is due to all the complex calculations that need to occur on your scene. When I created my short films, the average render time per frame for a film made using a raytraced renderer (Blender Internal renderer) was 5 minutes. The average render time per frame for another film made using a physically-based renderer (Cycles renderer) was about 1 hour. It’s not much of a big deal if you are only rendering still images, but rendering an animation using these renderers can be quite tedious.

What to do: If you use Blender, there is a physically-based renderer already inbuilt. On the top menu, just change the renderer from Blender Render to Cycles Render.

Use HDRI Lights

Lighting a 3D environment can be complicated. There are so much complexities involved with simulating realistic and convincing lighting and one would need to understand some lighting theory. In the past, 3D artists would create multiple light sources to simulate sunlight, sky lights, bounce lighting and things like that. The result would be a 3D enviroment scene with a big mess of lights everywhere.

Nowadays, the most popular trend is using HDRI lights in combination with a physically-based renderer. HDRI images are high-dynamic range images and contain more information such as brightness and intensity data within the one image. We can make complete use of this to simulate realistic lighting. Physically-based renderers can help remove the big mess of lights as it helps realistically simulate bounce lighting. Combine that with HDRI images and our 3D world can be realistically lit up without the help of really ANY extra lights at all. You could get away with using the HDRI to control the lighting for your entire environment.

Make use of HDRI images as a light source in your 3D scene to help push for more awesome realism and visual appeal!

What to do: Grab a HDRI sky image on any HDRI image download site. I prefer to use HDRI-Skies.com. Load up your HDRI image in your 3D software and set it up so that the intensity information contained within the HDRI could help light up your 3D world. In Blender, you could load the HDRI image directly in the Environment Texture node of the Sky. You would also need to capture the same HDRI image as an input for the strength/intensity. There is no point in having a HDRI image if you don’t make use of the extra information available within HDRIs. Do note that HDRI images are generally quite large and may crash your scene while rendering. That sacrifice may be worth it, however, depending on the result you are able to get from it.

Make use of photography techniques

If you have ever taken classes in photography, your skills have not gone to waste. You can fully utilize those skills in your 3D renderings as well. If you haven’t, then taking some time to learn some of the basics such as composition, f-stops, depth of field and so on will only do you a world of good.

What to do: In Blender, you can select your camera and go to the camera settings in the Properties window. From there, you can play around with some of the settings that you would tend to see on a DSLR camera. You can even simulate the look of a specific camera such as a Kodak camera or a cinematic camera like those powerful RED cameras, by going to the Scene tab and under Color Management, changing the “Look”.

Go subtle with post-processing

Once your render is complete, you should take advantage of post-processing. Post-processing can help to push your renders to the next level and give that sort of “pop” or that extra “wow” factor to your final render.

Some of the post-processing effects you could take advantage of includes adding glare, vignettes, glows, depth-of-field, motion blur, color grading, lens distortion, overlaying images over the original render and the list just goes on.

But remember to not go overboard with this. The bulk of the work should have been done in the modeling, texturing/shading, lighting, and rendering. The post-processing is just a touch up to give that final polish.

You will kill the appeal and realism of your render if you go overboard with the post-processing effects. In the art world, subtlety is always key!

What to do: Use a photo editing software like Photoshop or GIMP. If you use Blender, you can also use the compositor built into Blender. Just change the Screen Layout on the top menu from Default to Compositing and then select ‘Use Nodes’. Each node can be used to create the effect you desire.

Get feedback (and get them early)

The final tip that will really benefit you and help push your 3D environments to the next level is getting feedback! While you work on your 3D environments, show it to other people and see what they think of your work so far. Try to get their opinions early and try to get them often.

When you create your environments, you are so immersed into your world and you will likely focus on one specific area. This may cause you to overlook other areas of your environment and sometimes getting a fresh pair of eyes can be beneficial in this regard. They can help you catch issues with your render that you would have otherwise overlooked.

Getting feedback is really important as it can also help you improve as a 3D artist overall. However, you will need to learn to take criticism with a grain of salt. Sometimes you will get very encouraging feedback and other times the feedback could just be downright vicious. If you want to improve as an artist, you need to observe, understand and learn from the honest criticism. Getting honest criticism is almost always beneficial to you as an artist. Use the encouraging feedback as motivation and the pointless and vicious comments with no real substance as a waste of space (they only do this to get attention or perhaps a little envious of you).

What to do: Be active in communities related to 3D! Try being active on online forum sites like CGTalk, BlenderArtists, and relevant Facebook Groups to name a few. The people there encourage works in progress and are always more than happy to help! That’s it! Incorporate these tips into your next epic 3D environment and you’re 3D renderer will gift you by giving you beautiful renders.

Go back to Part 1 if you haven’t seen the rest of the tips already.

10 Things You Should Do To Create Awesome 3D Environments

Have you ever attempted to create a 3D environment and found that your final render falls a little bit…flat?

Most likely, you may have missed an important step in your scene creation process. Here are 10 things you should focus on if you want to give your 3D environment the best chance of looking awesome and get one step closer to achieving photorealism.

References

Making use of references in the form of taking images of that thing you want to create in the real world can do a whole lot of good in making your 3D environments look believable.

This may sound unbelievable at first, but even the best 3D modelers in the world rely on references. You may be thinking that hey, Pixar artists surely don’t rely on references, they just come up with their own magic. And that may seem true at first. Their characters, environments and props almost always look way off compared to the real world. As a matter of fact, they do this deliberately. Pixar artists do a ton of research, including taking reference images of almost all the enviroments they create before actually going ahead and creating it. Then they give themselves artistic licence to essentially “stuff” up their models and give it that stylized look that we associate as that “Pixar” look. If they didn’t use reference images, they would have missed vital details in their environments and things would have just looked off.

References help exactly with that. They help you flesh out the details that you would have otherwise missed and they also help you capture the right proportions.

What you can do: Go out and take reference photos! It’s better than sitting at home all day modeling stuff and frustrating yourself wondering why things don’t look epic. Getting that breath of fresh air can do wonders to you and help boost your creative juices. Take out your smart phone and take pictures of everything that you want to include in your environment; trees, houses, mountains, rocks, grass, and so on. If there’s something that you really want to focus on, like a lamp post or something, then take multiple pictures of the real thing and try to include multiple angles and close ups. You could also take pictures of the texture as well so that you could use it later in the texturing and shading process.

If you really can’t be bothered going out and away from your computer, than there are alternatives. You could go onto image reference sites like Pixabay.com, Flickr.com or even Google images. Store a couple images of the things you want to create in a folder somewhere and name that folder something fancy like “references”. This will help you later on when you start creating your 3D environments.

Tip: Make sure to do this as the ultimate first step before even touching your computer. You will thank yourself (and hopefully me as well) later!

Scale

A Scene from Big Buck Bunny. A BlenderFoundation Open Source Short Film.

There’s nothing more that will make a person look at a render and scream “FAKE” than a 3D environment full of models that are way off in scale. You could have the most detailed looking 3D models in the world and the most complicated texturing/shader/light setups, but people will still call out fake as we humans have this innate ability to spot things that are off in real-world scale. You could probably get away with it if you’re creating a stylized environment, but if you’re going for photorealism…that’s a big no no!

If you have a house model that seems way too big for any human to live in (that is, where you’d need 3 or 4 humans to stand on top of each other just to turn the handle to open the front door), then it would look fake. Even if it’s something really subtle like the length of the arms being a little bit smaller compared to the body or the eyes being a little lower on the face would be picked up fairly easily by the average human.

When all the 3D models in your environment have correct scale, it screams believability and realism. It also helps to give people the sense of scale of your environment. That is, they could tell if your scene is miniature or really epic!

What you can do: Measure everything! Ok, this could be a little tedious. Try to measure the size of as much objects as you can. For the objects that you can’t really measure, like a tree or a mountain, then just google it! Google is the answer for everything. Search for something like “average size of a tree” and you will get an idea of the correct scale to use in your scene.

Make use of the scaling system in your 3D software. If you use Blender, you can change the scale from the default “blend” units to the metric system so that you can speak in cm, m, km just like the normal human.

Details

You’ll want to make sure your environments contain as much detail as needed. The more detailed your models, the more realistic it tends to look.

For example, if you have a plain road with a texture of a road applied, that would look ok, but a bit boring and perhaps a bit unrealistic. If you have a road and used techniques like sculpting or normal/displacement bump maps to add the tiny rocky bumps that you’d find on a road, NOW you have a scene that’s less boring and more realistic.

It doesn’t need to be overly complicated either. It could be something as simple as adding a clock or even a photo frame on an otherwise boring plain wall.

When you look for tiny little details like that and add them to your environment, your 3D renderer will reward you with a pleasing image.

What you can do: When you took reference photos, hopefully you didn’t just take it and let it lie around collecting digital dust. This is your best chance to closely observe the images and pick out any detail you can. It would be best to pick out details that you (or any normal person) wouldn’t usually pick out. Even if the normal person doesn’t pick out those tiny details, subconsciously, they will feel the environment looks more realistic than if it didn’t have those details. Interesting we are, the human species.

Bevel (almost) everything

Here’s a fact: Nothing in the real world is ever 100% sharp. Not even the sharpest knife in the world. Put that knife under the microscope and you will notice, it too has some thickness.

Here’s another fact: Almost everything (by default) in the 3D world is 100% sharp. The edges of the cube, that house you are trying to model. A very little known fact by (especially beginner) 3D modelers is that the object that they deem to be final, ends up being a little too harsh and sharp than it needs to be.

Bevel helps sort that out by smoothing out the edges a bit. Softer edges help to catch the light better and just makes your scene more visually pleasing. It’s very hard to explain why but it just seems to be true. Try to take advantage of bevel whenever you can and make sure the edges almost never looks completely sharp. When I say almost, I mean you could probably get away with it for something like a knife.

What you can do: Once you’re happy with your 3D model, try to smooth out any edge or point that looks sharp where it really shouldn’t be. In Blender, you can enable Bevel by first selecting your object and either going to the Modifiers tab and selecting Bevel or by tabbing into Edit Mode, selecting the edges and pressing Ctrl+B (for edges) or Ctrl+Shift+B (for vertices).

Use real textures where possible

Procedural (left) and Image (right) textures. Which one looks more realistic?

Procedural textures are essentially textures that your 3D software creates for you by using some maths formulas. Image textures are textures that are predominantly taken by a camera of real world stuff.

From my personal experience, I’ve found image textures really helped to push my renders closer to a level of photorealism. To me, that’s what separates a fake procedurally generated render to a realistic one. Procedural textures do have its benefits in terms of helping add variations to my renders. What I mean by that, is it helps me to define parts of the model to use one texture like mud and another texture like grass. It also helps for other things non-texture related like helping me create random waves of an ocean or procedurally decide places to add grass, flowers, trees and so on.

But in terms, of the actual texture itself, nothing beats a real-world image texture (unless you want to go for a cartoon stylized look). Just compare these two images. Which one do you think would help make your render photorealistic?

What to do: Use image photos wherever you can. You can get good quality free textures from Textures.com (formerly CGTextures.com) or you could take photos yourself. There are general guidelines on how to take photos which would need to be a separate topic on its own but generally, try to obtain high resolution images that are seamless and don’t contain too much highlights or lighting in them.

  Phew! This is probably too much reading just for one post. Continue onto Part 2 here for the remaining tips!

Here are my latest renderings of a backyard scene made with Blender from scratch. My aim was to shoot for photorealism and found it quite a challenge.

Merry Christmas everyone :) 

I'm giving away the Introduction to Blender course for free as a present to you guys (It won't be free for long though)! Here is the link: http://thilakanathanstudios.teachable.com/p/introduction-to-blender-for-beginners

Enjoy!

In this tutorial, you will learn how to make a 3D object follow another 3D object in Blender!

Perhaps you may have wondered how to make your 3D models stick on, or follow, another 3D object. For example, you may have wondered how to make a hat model stick on to your character model, even if he moves. You may have wondered how to have your character(s) stay in a boat as the boat moves, or stay put inside a car, no matter which way the car drives. You might have wondered how to make a ball stay in the hands of a person no matter where that person’s arm moves.

You could do everything manually. Like with the hat, you could manually animate the hat movements as the character moves. But that is so time killing, boring, repetitive, and just unproductive. There is a better way..

In Blender, we can simplify this process using constraints! In particular, the constraint I use the most often is the “Child Of” constraint. This can help you too!

Here are the steps to pulling this off in Blender:

Open up Blender and have 2 models ready and position them to follow. This could be a hat and 3D character. But to simplify the process, I’m going to use a cube and a sphere. The cube is going to act as our hat, and the sphere is going to act as our character.

Select the object that you want to follow. In my case, I select the cube.

Go to the Object Constraints tab in the Properties Window on the right.

Press Add Object Constraint and choose “Child Of”.

Your object now has a constraint to follow. But we need to specify which object that we want to follow. Select that object next to Target. I chose the sphere.

Your cube might pop out somewhere else. I don’t really know why it does this, but to get it back just press Clear Inverse and then Set Inverse. It should pop right back.

Now select your main object (the object that you want to be followed). In this case, I select the sphere.

Move the object anywhere and observe the following. Congrats! You can now make objects stick and follow!

If you go back to the “Child Of” settings, you’ll see there are extra settings you can play with. For example, un-check the Z location and see what happens. If you move the sphere now, the cube will still follow but will not follow any of the Z-movements.

Under that, you’ll see a slider called “Influence”. 0 influence means the object won’t follow at all, and 1 means it will follow it strictly. If you put it somewhere in the middle like 0.5, it will follow but not really, it will lag behind. Like a dog not really willing to go for its daily walk.

In this tutorial, you will learn how to do camera shakes in Blender!

If you have seen one of the high-impact action scenes from any Hollywood film, you would notice that there’s a lot of camera shakes. Not only action films, but also in one of those natural disaster films, especially those earthquake films, there’s always a lot of violent camera shakes. Camera shakes help to add more impact to your scene. If you have a character who punches another guy in the face, it just seems a bit more violent and aggressive if the camera shakes as he makes impact. Or the lead up as well. If you have a big tough guy fuming in anger and his face turn red in anticipation of a fight scene, having the camera shake can make him seem more angrier. You tend to feel his inner explosive anger more.

While you could manually animate camera shaking frame by frame, there is a better (and quicker) way. Save that keyframing for more important camera movements like zooming in and panning and things like that..

Here is how you can employ camera shakes in your own scenes in Blender:

Open up any scene you want in Blender (or use the default cube if that suits).

Drag out a new window in the 3D viewport and the change the type to Graph Editor.

Select your camera and add a keyframe anywhere (by pressing i or automatic keyframe insertion). You need to have at least 1 keyframe on your camera for this to work.

In the Graph Editor, you should see that your f-curves are visible (those colored lines). Select an f-curve by left-clicking a name on the left. I chose the Z-location. This means the camera will shake up and down only on the Z-axis.

Hover your mouse in the Graph Editor and press N to open up a toolbar on the right.

Scroll down to where you see it says Add Modifier. Click it and choose noise. This is what will shake your camera.

Playback the animation (or press Alt+A) and observe your camera shake. Go to the Camera View to see it live in action!

You can also make modifications to your camera. In the modifier settings, you have a bunch of settings you can change. The Scale changes how quick your camera shakes, higher values makes your camera shake slower. The Strength changes how violent your camera shakes.

Click the Restrict Frame Range. Start allows you to decide at what frame the shaking should start. End allows you to decide when the shaking should stop. So if I choose Start as 50 and End as 200, the camera will shake at frame 50 and continue to shake till it reaches frame 200. In and Out pretty much mean fade in and fade out. For example, if you choose say 5 for In, as soon as the camera is set to start shaking, it wouldn’t start shaking immediately but gradually make its way over 5 frames.

The final option Influence, determines how much influence the shaking will have. That is, 0 influence will have no shaking effect, 1 influence will have full shaking effect.

How To Switch Cameras in a Blender Animation

Sometimes one camera is not enough and you want to work with 2 or more cameras in a scene. This tutorial will teach you how to switch between cameras during an animation in Blender.

If you’ve ever put on the director’s shoes and want to showcase your new animation with awesome cinematography, you may have come across a small problem: how do I cut to the next camera in this animation? And how do I switch back and forth 2 different cameras for this shot of 2 people having a conversation? You’ve probably approached this in one of 2 ways:

Render your animation in 1 camera angle. Once the rendering is done to the point where you want the switch to occur, re-position your camera in another angle and continue to render from that. There will be a lot of stop render, start render, stop render, start render, especially if it is a dialogue shot.

Manually animate the 1 camera. At the point where you want the switch to occur, you animate the same camera to quickly to move to the next angle. This way, you don’t have to do partial rendering like in 1 and you can just render it straight out and edit it later on in a video editor to remove the quick switch frames.

I used to follow the latter method. But I stumbled across an issue. I enabled motion blur in Cycles. This created a big issue actually. At the point where I animate the camera to quickly move to the next angle, there was tons of motion blurring going on. This added a HUGE amount to the rendering times. I mean, it took FOREVER to render that 1 frame where the switch happened. Then I realized there was a better way..

Here are the steps to do camera switching in Blender:

You can use your own Blender scene, but I’m going to start with the default scene. Add in another camera (Shift+A -> Add -> Camera). You should have 2 cameras in your scene. Position each of the cameras to have 2 different angles. One could be a mid-shot of the cube, the other a low-angle close-up shot of the cube.

In the timeline, we’re going to add markers where the switch will happen. Add a marker by hovering your mouse inside the timeline and pressing M.

Go to another frame and add another marker by pressing M again. This is the part of the animation where the switch will happen.

Select the first marker by right-clicking the first marker.

Now select a camera in the 3D view.

We’re going to bind this camera with the selected marker and switch cameras at this point. Hover your mouse in the timeline again and press Ctrl+B to bind.

Now select the next marker and the other camera.

Bind this camera with the selected marker by hovering over the timeline and pressing Ctrl+B.

Playback your animation and you should see your camera switch where you added the markers.

If you have a dialogue shot between 2 characters, you probably just want to switch between the 2 cameras a couple of times in the same scene. Instead of creating more markers and binding them again, you can just duplicate the existing markers as this is much quicker and pain-free. Just select the marker where you want the camera to switch back to (Camera 1 for example), and press Shift+D to duplicate the marker. Move the marker past the 2nd marker (where Camera 2 is binded) and place it somewhere. You should see that your camera switches back again (from Camera 2 to Camera 1 again).

This method of camera switching is the best and most efficient way to switch cameras in Blender. You can animate your camera movements such as zooming in and out. The switch will still occur wherever you place the markers in the timeline. This works well for more than 2 cameras. Just keep making more markers and binding the cameras with Ctrl+B.

How To Create Mountains In Blender (In 5 minutes or less)

In this tutorial, you will learn how to quickly generate mountains in Blender!

If you are working on a scene in Blender, and realise you need to make mountains, you can either use this method from the Noob to Pro Wikibooks or use the method below which will generate it quicker:

Open up Blender and delete everything if you want.

We need to use a plugin that will generate mountains for us. So go to File -> User Preferences -> Addons.

Look for (or type in) the addon ‘ANT Landscape’. Enable by checking the box on the right (near the scared-looking human figure).

Close the User Preferences window.

Now add the landscape by Shift+A -> Add -> Landscape. You should now have your own little mountain!

But wait! Don’t do anything with the mountain yet! You can customize that mountain to look more like what you want. On the left, you will see mini-panel called Landscapes (if you don’t see it, press T to toggle the toolbar). Here, you will be presented with a bunch of options you can change.

Landscape settings

There’s a lot of settings here! The main settings are described below:

Subdivision: This subdivides your mountain. The more subdvisions, the more detail you can throw at your mountain.

Mesh size: Changes the size of your mountain.

Type and Basis: Gives your mountain different distinctive looks based on textures.

Random seed: Changing this gives a unique mountain while maintaining all the other settings. Good if you need a few of the same type (or breed) of mountains in one scene but want to make each of them look unique.

Noise: The smaller the value, the more distortions and bumpy noise you’ll have on your mountains.

Height: The maximum height of the mountain.

Offset: Offsets the height of the mountain overall.

Plateau: The height at which the mountain should have a plateau (look flat). Good to create Grand Canyon type of scenes.

Sealevel: The height at which the sea level is. The naming is a little odd in my opinion, it should be called ground level. I would generally add the sea above this.

Introduction to Sculpting in Blender

In this beginner Blender tutorial, you will learn about the basics of sculpting!

Quick links: Introduction | Interface | Properties | Navigation | Modeling | Shading | Texturing | Lighting | Sculpting | Animation | Particles | Physics | Rendering | Video Editing | Compositing

Have you ever wondered how these detailed looking renders were made? All those wrinkles, pores and veins? Well, I guess normal and displacement maps can be good at that kind of stuff but for more higher-end awesomeness, you can achieve this look with sculpting! Sculpting can give even better looking detailed renders compared to modeling by hand or relying on bump maps. With the latest developments in Blender, such as dynamic topology, you now have the ability to model a complete detailed character via sculpting by starting with just a cube. Let’s take a look at how we can sculpt in Blender.

Sculpting in Blender

1) First things first, open up Blender. We’ll use the cube so delete the lamp and camera if you want.

2) Of course, you can’t do much sculpting on a cube since it only has 8 vertices. If you want to do high detail sculpting, you’ll need much more than 8 vertices (think 100s to 1000s). So, all you have to do is just add a Multires modifier on the cube. We can’t use the subsurf modifier when sculpting for some reason. The multires does pretty much the same thing anyway.

multires modifier

  3) Bump up the multires to about 4 or so. Won’t give us 100s of vertices but is good enough. The higher the value, the more you can detail your cube (should now look like a sphere) but obviously the higher you push this value, the more you’re going to strain your computer’s processors. I have crashed my computer oh-so many times doing this.

Subdivide to around 3 or 4.

  4) Next, with the cube selected, go into Sculpt mode.

Select Sculpt Mode.

  5) You should see a bunch of sculpting-related goodness in the Toolbar on the left. Here you can choose different paint brushes and stuff. There are all sorts of cool brushes you can use. For example, the Draw brush allows you to draw on your mesh (cool for drawing veins), the Smooth brush will smooth your mesh, and so on. Pretend as if your mesh is one of those clays you used to play with in school, and the brushes as one of the tools you can use to shape your clay.

The different brushes available in Sculpt Mode.

  6) There are a bunch of options you can play with in the brush settings. You will likely see Add/Subtract buttons on a lot of the brushes. The Add button bumps up your mesh as your draw and the Subtract button carves into your mesh as you draw (cool for drawing wrinkles).

Sculpt away!

Dynamic sculpting in Blender

Blender has a pretty awesome feature called DynaTopo which means Dynamic Topology. This means that when you use the brushes on your mesh, the mesh will dynamically create vertices for you as you draw. You don’t need to rely on Multires. The good thing about this is that you save a lot of memory, since a dynamic sculpt will only focus on adding a lot of vertices around wherever the detail is required whereas a Multires will add a lot of vertices all over which is just unnecessary and CPU-hogging. The bad thing about dynatopo, however, is that you won’t end up having a nice topology structure once you’ve sculpted. Vertices will be placed randomly here and there and it will just be a mess. Certainly a nightmare to animate with! But there is a way to fix this using a technique called Re-topology, but that is a more advanced topic.

To enable dynamic topology in Blender, just go to the toolbar and under Options, you should see a check box with something like “Enable Dynamic Topology” depending on what Blender version you use. Click to enable and you can dynamically start creating a full detailed character starting from a cube! Oh yeah, make sure you remove the Multires modifier, it is not needed anymore, we have DynaTopo now.

  That’s all! If you find yourself developing a passion for sculpting, I highly encourage you to check out some of the videos on YouTube where artists show timelapses of their sculpts. I find it to be quite enjoyable and you can learn some new techniques and skills just from watching those masters in action!  

Beginner's Introduction to Animation in Blender

In this tutorial, we will learn how to get started with animation in Blender!

Quick links: Introduction | Interface | Properties | Navigation | Modeling | Shading | Texturing | Lighting | Sculpting | Animation | Particles | Physics | Rendering | Video Editing | Compositing

Animation in Blender is very quick and easy. If you want to make a box move from point A to another point B in 3 seconds, all you have to do is move the box to point A on frame 0 and then set a ‘key’. 3 seconds down the timeline, move the box to point B and then set another ‘key’. When you play back the animation, you will see your box move from point A to point B in 3 seconds. Simple as that! You’ve made you’re first animation. You can do a lot more than that, though! You can animate how it rotates, as well as how big or small it scales. All you have to do is set keys anytime you want something to happen. A key (or keyframe) pretty much just sets the properties of the object at that specific time. So say, I set a keyframe on the 5th second of a cube that is rotated 30 degrees, that is located 5 units above the Z-axis and scaled 2 times the original. That means, when I playback the animation, the cube will always be rotated 30 degrees, and be located 5 units above the Z-axis and be scaled twice the original size when the playback has passed the 5th second. By setting a key, we have ‘locked’ our object with those properties in place. Here is a tutorial that covers keyframing better in case I explained it horribly! I guess it’s nice to know that Blender can do all of this, but exactly HOW do you do all this?!?

My character animation..

  There are 2 different ways you can animate in Blender:

Pressing the i key

To set a key (or keyframe) in Blender, you simply press the i key on your keyboard. You may be presented with a number of options. Depending on whether you want to animate the location only, rotation only, scale only, or a combination of the others, then you select the option that corresponds. Suppose you will regularly need to animate location, rotation and scale all at once. You would then select ‘LocRotScale’. This would set keys for the location, rotation and component properties of the object. If you’re a serial animator and you hate having to select an option every time you press i, you can set the Active Keying Set (found in the bottom of the Timeline next to the red circled button) to LocRotScale. Just press that key looking image inside the textbox and you’ll find those options. Now every time you want to animate your object, you simply press i without having a dropdown option to annoy you!

Automatic keyframe insertion

Another way to animate in Blender (and a personal favourite of mine!) is to use the automatic keyframe insertion! This is that red-circled button I mentioned earlier found in the Timeline. Once clicked, it will assume you are animating with the option LocRotScale. If you want to change this to something else, change the Active Keying Set (the box next to it) to another option that you prefer. To set a keyframe, you simply work with your object directly. Say you move an object to point A, a keyframe should already be set (you should see a yellow line indicating a key in the Timeline). If you go to another point in the timeline, and move somewhere else, another keyframe would automatically be inserted. Blender detects any change in the objects properties and sets a key if a property has changed.

A bouncing ball animation in Blender.

  Give it a go for yourself!

Start with a UV sphere.

At frame 1, set a keyframe (using i or automatic keyframe insertion).

Then go to frame 10.

Move the sphere somewhere else and add another keyframe.

Playback your animation (by pressing the play button or Alt+a).

That’s all! Of course, there are more advanced types of animations you can do in Blender such as armature animation (that is, animating characters and so on) but that is best left for another day. Same principles still apply though, you use either the i key or automatic keyframe insertion to animate your characters. Give it a go, animate something cool. Download or make a quick car 3D model and animate it driving around a race track. With a bit more practice, you will get used to Blender’s animation system.

Beginner's Introduction to Particles in Blender

In this tutorial, we will learn how to work with particles in Blender to create cool looking effects!

Quick links: Introduction | Interface | Properties | Navigation | Modeling | Shading | Texturing | Lighting | Sculpting | Animation | Particles | Physics | Rendering | Video Editing | Compositing

If you’ve watched some 3D animated flicks, you may have wondered how they create fireworks, rain, glowing orbs, dust particles or even hair/fur. The secret all lies in particles! In Blender, we use the in-built particle system to re-create these effects. So let’s take a look at how to use this system!

Particles in real life

Using Blender’s particle system

1) Delete everything and leave the cube. Select the cube.

2) Select the Particles tab in the Properties window.

Create new particle system

  3) Press the ‘New’ button to create a new particle system for the cube. Now playback your animation (Alt+A) and you should see that your box is now emitting particles. Your cube should sort of act like the nozzle of a shower head.

Showering particles

  4) This is made possible by the bunch of default settings that have been automatically set up for you when you created a new particle system. Play around with these settings to make the particles do weird stuff.

5) A few things. The Emission bit of the settings are used to control how many particles you want, the start time for when the particles for when the particles should appear and the end time for when they should go away. The Physics bit controls how the particles move and their sizes and other physical related stuff. The Cache is used to store your final particle animation either in a separate file or within the blend file. This is good to do after you are happy with the final animation of the particles since when you want to render or preview later on, it would not have to re-calculate the animation.

Particle settings

  6) If you don’t want gravity to affect it’s movement, you can switch off the gravity completely. Either go to the scene tab and change the Gravity Z-setting from -9.81 to 0. This will affect the gravity of your whole scene. So if any object relies on gravity, they too will be affected. Another (better) way is to turn off the gravity for just the emitting cube. In the Particle tabs, you will see a section called Field Weights. Here, slide the gravity to 0.

Set gravity to 0

  By playing around with these settings, you will be able to simulate fireworks or rain or even floating dust particles.

Let’s create hair

The particle system is not only used to simulate fireworks and stuff, you can also make hair and fur using it. Here’s how to do that:

1) Change the type from ‘Emitter’ to ‘Hair’ in the Particle settings, you would notice your particle system now becomes your hair system.

Change type to ‘Hair’

  2) Play with the settings for a bit. Similar to the emitter, changing the emission number will change how many strands of hair your mesh will have. That is, bumping up the emission number to 2000 will give your cube 2000 hair strands.

3) Let’s comb that hair! At the moment, it looks like the cube has suffered an electric shock or something. So we should comb that hair back into place. In the 3D view, switch from Object Mode to Particle Edit.

Change to Particle Edit mode

  4) Just like sculpting, you will see a bunch of brush options in the toolbar on the left (press T to open if you don’t see it). Select the Comb brush.

Select ‘Comb’

  5) You are now the barber! Go to your mesh and left-click drag to start combing. Go crazy and comb that unruly hair to any shape you want.

Stroke along the edges of the hair to comb

6) You might find that your mesh has WAY too much hair everywhere and it probably needs trimming. Well, you are the barber here and you have full control over how you want your hair to look. Select the Cut brush and cut away at the hairs you don’t want. If you want to trim hair and not cut it off completely, make sure you cut along the outer edges.

Stroke along edges to cut hair

  7) Just a final thing that might help beginners as this part might confuse. This applies to those using older versions of Blender. If you want to give your hair a shader, go to the Particle Settings, under Render choose Material: 2. Now in the material settings, create a new material under the already existing material ‘Material’. This new material will be the hair’s material. Why? When you selected Material: 2, you actually mean the second material of the mesh. So if you had like 5 materials for your mesh in total, and you selected ‘Material: 4’ in the particle settings, the hair will render with the 4th material in the slot. A little confusing but that is currently how Blender works. In the future, this might change and this post might become redundant, but for now it is just the way it is.

If you are using a new version of Blender (mine is around Blender 2.76), you will still need to create a new material under the existing ‘Material’. Then in the Particle settings, under Render, change the material from the dropdown directly! Easy as that!

Creating a shader for hair

  I have to mention that the hair system is not ONLY used to make hair and fur. You can replace those hair strands with 3D models (which is done by going to the Particle Settings -> Render -> Change to Object -> Select your object). This means you aren’t just limited to hair but you can create forests of trees, cities, crowds of people, flock of birds, you name it. The particle system saves a lot of memory so rendering 1000s of trees will not break your PC compared to you duplicating a tree 1000s of times and rendering that. The image below was created by me where I used tree models to represent hair strands!

My dodgy attempt at creating an aerial island forest

  And that’s all! Hopefully, you’ve learned about working with particles in Blender!