Blender For the Faint Hearted - 02 : Basic modeling and getting around

[Sphynx]

Blending for the faint hearted - Sphynx's guide to Blending

Tutorial 2: Basic modeling and getting around

If you've followed the Introduction to Blending for the faint hearted, then you've got a grasp now of some of the basic concepts behind 3D and Blender in particular. So now, we'll take a look at some of the basics of modeling in Blender, and how we get around in the 3D world.

Primitives

In 3D modeling, we of course have 'meshes' - the interlinking of vertices, edges and faces into shapes. We also have a very specific type of mesh however, called a 'primitive'. The primitives represent the simpler shapes that can be constructed in the 3D environment, and which effectively provide the building materials for our more complex models. In Blender, the primitives are the plane, cube, circle, UV sphere, icosphere, cylinder, tube, cone, grid and.., er.., the Monkey (said in a quite whisper).


Image: The Blender primitives

OK, so the monkey is not a primitve as such and needs some explaining. The monkey is a true mesh, but the decision was decided some time ago to include the monkey mesh in the primitives section. Why? It has been suggested that it was some strange sense of humour that ended up with the wrong type of 'primitive' and that, like many programmers, the Blender-mob are completely out of their heads... Well, as a programmer myself, I can agree that this is probably true but it also has a more serious side.

If you want to experiment with materials, shaders and lighting environments, a simple primitive is all well and good but does not tell you everything - you need a good mesh that can cast shadows on itself, has curves and alcoves and is generally a more complex mesh. The answer of course is the monkey (more accurately called 'Suzanne'). In two keystrokes you can add a complex model to the 3D environment and start to experiment. Many of the Blender tutorials and examples that you will find around the internet will use the monkey for exactly that purpose.

For the rest of this tutorial, we need something to play with. Many of Blenders menus and many of the general ways of using Blender can only be appreciated once we add something to the 3D view and select it for use - though I think we'll avoid using the Monkey for now. Follow the next exercise to get rid of any model in your default Blender scene.

Exercise - Deleting an object

If you've started Blender and have the default scene, we need to empty it. Blender is nice enough to add some basic elements to the scene but we don't really want them for this exercise, or else we're not learning anything. If you've got a cube displayed in the middle of your 3D view (remember that if you are just looking from one plane, it may just appear as a square), select it with your mouse using whichever selection button you decided to use in Tutorial one. If you use the wrong one then you'll know immediately as the 3D cursor will suddenly jump to a different place. Don't bother trying to put it back as you won't be using it very often and it usually just gets in the way, so just leave it where it is.


Image: Deleting an object

Once you've selected the cube it will change colour. The default is a mid-pink colour. Once the colour changes, you know that your cube is now selected. Press [Del] or [X] on your keyboard, or if you are a glutton for punishment, press [Space] and navigate on the context menu that is displayed to [Edit] > [Delete]. Personally, I'd use the keyboard shortcut, and I'll always assume this in the text.

Once you've pressed [Del], Blender obligingly displays a requester directly under the current position of you mouse. If your move you mouse, the menu or requester being displayed will disappear. Confirm that you want to [Erase Selected] and the cube will disappear.

OK, so now we should have a nice, empty 3D scene that we can add a primitve to. First thing to do however, is move to a different layer. You have 20 layers to play with - arranged as two groups of ten, and in two rows. The first layer will be selected at present, and it is this layer that will contain the camera - you've already seen it, its the pyramid-like object with it's base pointing to the centre of the view. This shape is pretty common in 3D applications - the pyramid shape emulates the cowling that is on the front of film-cameras.


Image: Layers

Personally, I prefer to leave this first layer for cameras and lights and do my modeling elsewhere simply because it prevents clutter and objects overlaying each other that I'm not actually working on. Just click the next layer in the row and we'll work on that. You'll notice that everything left in the 3D view will disappear but this time they haven't been deleted - we just left them on the first layer.

Exercise - Our first primitive

So now its time to create our first primitive. Drop the cursor somewhere near the centre, but deliberately NOT in the centre. I'm going to show you something later, and we specifically do not want the cursor too perfect for now.


Image: The main context menu

Next, press [Space]. This pulls up quite a sizeable context menu and it is this menu that stores a lot of the tools that you'll use in Blender. Go to the menu called [Add] and you'll see lots of things displayed that we can add to the 3D view. Go to [Mesh] > [Cube] and select it, and a cube should appear centred on your cursor.

Editing Modes

Right, so why does the cube look so different to the one we just deleted? The reason for this is because you've just added a new primitive mesh to the scene and Blender has obligingly put you into a different editing mode. There are two modes that you will use a lot when modeling: Object mode (which is what you saw earlier when you deleted your pink cube), and Edit mode (which is what you are looking at now). There are other modes, but we'll cover them later.


Image: Object mode

Image: Edit mode

There are two ways to move back and forward between these two modes. You can select the mode that you required from the menu on any of the 3D view windows, or you can just press [Tab]. Trust me - use [Tab]. As a side note, in all of the exercises that you do, make sure that you repeat them a few times. There is only one thing more important than knowing how to do something - and that's how to get youself out of it when it goes horribly wrong. With nearly all of the exercises in this tutorial, the shortcuts are toggles. You just press the key once to go into a mode, and then again to leave it.

Exercise - Switching modes

Press [Tab] now. As the cube is the currently selected object, when it drops back into Object Mode, it will already be pink. Press [Tab] again, and it will change back to the black-edge and yellow-vertex version.

Do this a few times and look at how the mesh changes, but also at how the menu bar also changes. You'll see a few more tools appear when you go into Edit mode. We'll look at these in a little while. After you are comfortable with moving back and forth between the two modes, get into Object mode and leave it there.

All primitives have their origins...

You will notice in the centre of the cube a small yellow or pink ball. I'll call it a ball rather than a 'point' as it sort of has a bit of reflection to it to make it more obvious. This ball is the 'origin' of the object. It will be yellow when the object is not selected, and a slightly darker pink when the object is selected. If you are positioning an object, it is this 'origin' to which the co-ordinates refer.


Image: The origin of an object

Note however, that the origin of an object is NOT the centre. It may be the centre purely by circumstance or co-incidence, but it is not defined as the centre. The origin could be placed totally outside the physical bounds of the mesh that you are making - but this will make much more sense later once we start building more complex shapes.

Moving Around

Now that we've got something to look at, lets find out how we move around. There are two main input devices that I use, and these seem to be the most drastically changing of the two so I'll confine myself to these for this tutorial. For probably 95% of the readers, the first will be far more important.

Is that a wheel-mouse, or is it made of plastic?

The first method is by using a combination of mouse and keyboard shortcuts. I'll assume that a wheel mouse is in use. If you have not got a wheel-mouse, then I am afraid that you will have to rely even more on shortcuts.

Exercise - Zooming

Place your mouse over a 3D view and scroll you mouse-wheel back and forth without pressing any buttons. Do it a few times and get comfortable with this zooming facility. The same facility is provided by the numpad [+] and [-] keys. Make sure that you are comfortable with both methods before moving on.

Exercise - Panning

Panning, or moving from side-to-side or up-and-down is performed by using the [ctrl] and [shift] keys in conjunction with the wheel-mouse. Again, do it a few times using alternate combinations of the [ctrl] and [shift] keys. The same facility is provided by using [shift] and [ctrl] in combination with the numpad cursor keys ([2]=down, [4]=left, [6]=right and [8]=up). Make sure that you are comfortable with both methods before moving - chances are that you'll be doing this a lot.

Exercise - Rotating

Looking at one side of an object is fine, but sometimes it's just not enough. You really need to see a quick 3D view of it to make sure it is correct. To do this there is no way to comfortably do it with the mouse so you'll always need to use the keyboard shortcuts. These are numpad 2, 4, 6 and 8 - the numpab cursor keys - but without any keyboard modifiers like [ctrl] or [shift]. You use then by themselves. Do this now in one of the 3D views and watch how the object rotates. Keep doing it till you are familiar with the keyboard system, then try panning around as well. Get used to switching between the [ctrl] and [shift] versions of how these keys operate and you will be much more comfortable later when you come to do some serious modeling.

How observant are you? Have you noticed something when you started to rotate the 3D view? No? Take a look at the background grid - previously, this grid had a fainter, 'sub-grid' behind it, but now its gone. What does this mean?

The meaning is simple - it's a visual clue that you are no longer in one of the defined 'standard' views (i.e. Top, front or left). As soon as you started to rotate the object, you entered 'User' view mode. Even if you rotate back to what is effectively one of these views the gird will not come back - the reason is that you are still in a user-defined view, even if it just so happens to be from a particular angle. To get back to the standard mode, either select the appropriate mode from the View menu of that 3D view window as in Tutorial 1, or just use the 7,1 or 3 numpad keys to do the same thing.

Take one tablet, twice a day...

The other main input device that I use is a Wacom graphic tablet. The only reason that I raise this in the tutorial is that it (and probably track-balls) are the main alternative - and this has a big impact on how you move around. Personally, as soon as I noticed the differences between using the mouse and pen tablet, I grab the pen as soon as I want to do some Blending.

For obvious reasons, you only need to follow this part of the tutorial if you are the owner of a tablet!

Exercise - Zooming with the pen

Place your pen in one of the 3D views but do NOT press the tip to the pad (usually this is pen-button 1). Hold down [ctrl] and move your pen around with pen-button 2 pressed in (usually the lower of the two buttons on the side of the pen). I find that for a right-handed person, moving the pen NE-SW gives a good result.

Exercise - Panning with the pen

Place your pen in one of the 3D views but again, do not press the tip to the pad. Hold down [shift] and move your pen around with pen-button 2 pressed in. The view pans around wherever you move your pen - there are no separate key strokes for up-down or left-right movements.

Exercise - Rotating with the pen

Rotating is even easier - just press pen-button 2 with the pen in a 3D view and move your pen around (yet again, without the tip touching the pad). As with panning, this is all free-movement.

As you can see - while the pen is a more simple device, it provides much more direct control over the 3D view with the use of only two keyboard modifiers.

Perfect placement

I mentioned in Tutorial 1 that it was possible to place the cursor at specific points in the 3D view. Well the same is true of objects and points. To do this, we get direct access to the location, rotation and size information of a selected object by pressing [n] on the keyboard.

Exercise - repositioning the cube

We deliberately placed our cube in a position other than the centre - but now we're going to correct that and re-position the cursor at the same time. You should be in object mode - if not, press [Tab] until you are, then make sure that the cube is selected. If it isn't, select it now (remember, the cube will turn pink).

Press [n] on the keyboard and the 'Transform Properties' dialog will be displayed. This is really what this window is for, but it gives us a nice little side-effect of being able to directly manipulate the positions of an object. Do this now by clicking each of the location buttons in turn (LocX, LocY and LocZ) and enter 0.000 in their fields. The change is immediate, so each time you change a value and press [Return] the cube will move even before you click the next button.

Just for completeness, the two other fields in the Transform Properties dialog are the OB and PAR fields. OB is simply the name of the object - you could click on that field here and totally change its name. The only reason it's called 'Cube' is because Blender thought that this was a pretty logical default name. PAR is the name of the parent object, but don't worry about this. We'll talk about parenting objects later.

Snapping

We have already discussed two methods of placing the cursor at a specific known point. Both of these are accessible by pressing [shift-s] on the keyboard to display the 'Snap' context-menu. As we've raised this, we'll look at all of the options now.


Image: The snap context menu

Selection to Grid
The [Selection to Grid] menu item snaps the origin of your current selection to the nearest grid intersection in all 3 dimensions. If your object is just a little off the centre of the 3D views for example, hitting [Selection to Grid] will most likely make it perfectly centred.

Selection to Cursor
The [Selection to cursor] menu item snaps the origin of your current selection to the cursor, wherever the cursor is currently located.

Cursor to Grid
In a similar way to [Selection to Grid], [Cursor to Grid] snaps the cursor to the nearest grid intersection. Also in the same way, its the easiest way to get the cursor perfectly centred again. Just place is somewhere near the centre (sort of 'as close by eye') as you can get, then select the [Cursor to Grid] menu item.

Cursor to Selection
This is one of the most useful methods of positioning the cursor. On the one hand, it will take the cursor and snap it to the origin of your currently displayed object or vertex. That however, is the least of its uses. We also have an object called an 'empty' (it is called a NULL, a NOTHING and lots of other things in other 3D applications). An empty is just like a placeholder - it has no substance and can never be rendered, but you can position it using the [n] key in exactly the same way as any other object. On the other hand therefore, you can snap the cursor to a perfectly placed 'empty'.

You can also select a single vertex in edit mode and use the same operation. The cursor will get snapped to the location of the vertex.

Why would you want to do this? Well, just consider building something from accurate plans. Unless you have the ability to create and manipulate objects around perfectly placed positions, you would have a hard time getting your model accurate.

Selection to Centre
Selection to centre is the fastest way to centre whatever object you have selected. Hit this menu item and the origin of the object will immediately jump to the centre of the 3D view. In actual fact, this would have been even faster that what we did in the last exercise to re-position our cube. It would only however, have centred the cube and wouldn't have helped if we wanted to place it somewhere else.

Exercise - Perfect positioning the cursor

We have decided that we want the cursor to be perfectly placed at position LocX=2.500, LocY=3.720 and LocZ=4.234 - a painstaking task to do manually. Instead we will use a combination of positioning an empty, and snaping the cursor.


Image: Transform properties dialog

In a 3D view, hit [Space] then navigate to [Add] > [Empty]. You should see a three-arrow object appear indicating the three dimensions of x, y and z. You'll notice that it is pink rather that showing points to edit - pretty logical. If it's an empty, we can't place it into edit mode. Try hitting [Tab] - nothing will happen. Instead, hit [n] and if the Transform Dialog is not already displayed from earlier it will appear now.

Modify the three co-ordinates to those shown above, and the empty will move to its new location in the screen. If you can no longer see the empty in your 3D view, well you know how to zoom now, so zoom out a little!

Finally, keep the empty selected (so that it is still pink) and hit [shift-s]. Select [Cursor to selection], and the cursor will jump to the new, perfectly placed location.

This was a painstaking process, but it helps in that you learned something new. In reality, you would move you cursor to a specific location by using the view menu and entering the cursor coordinates directly.

A final note on perfect positioning

Perfect positioning is not just relegated to objects however - we can use the same process to very accurately position individual vertices as well.

Exercise - Perfect positioning a vertex

If you are not already in Object Mode, switch to it using [Tab]. Select the cube so that it goes pink, then switch into Edit Mode again using [Tab]. You should now see your cube displayed as a series of black lines with coloured vertices at each corner. Unlike Object Mode, these vertices are selected when they are yellow and unselected when they are pink. Select any one of these points now - which one is irrelevant.


Image: Positioning a vertex

If the Transform Properties dialog is not currently displayed, display it now by pressing [n]. You will see that unlike an object, only three fields are visible: Vertex X, Y and Z. These fields display the position of your selected vertex in relation to the origin of the object. Your vertex will probably be at 1.000 in all three dimensions, and it will be at the same place regardless of where in 3D space the object itself has been positioned.

Change one of the vertex locations to 2.000, and see how this is reflected in the 3D view. Play around and try changing the three values again until you are comfortable with how it works. We won't be doing this so often as there are easier ways of doing this with the mouse, its just not so accurate. When you are happy, change all three back to 1.000 to that the cube is square again.

Finally, try this. Select two vertices of the cube - again it does not matter which. Having a hard time? Only one vertex stays selected - as soon as you select another one, the first one is de-selected. To select two, select one then press and hold down [shift]. You can then select the second vertex.


Image: Median values

Now look at the information in the Transform Properties dialog. Does it make sense? None of the information actually relates to either point selected. Well, it does but not as you would think. Take a look at the labels as well - they are now labelled Median X, Y and Z. Effectively, it is displaying the perfect centre of the vertices selected (the average of each of the X, Y and Z co-ordinates of each point).

Understanding what you are looking at

The last topic we'll cover in this tutorial is making a little more sense of what we are looking at on screen. A cube is not so bad - we only have 12 lines and usually only two or three cross over each other, even when we are rotating the model in 3D space. Once we get more edges and polygons on screen, it will start to get more confusing.


Image: A cube in 3/4 view

If its not already selected, select the cube and go into Edit Mode using [Tab]. Use your new found knowledge of rotating and zooming to position the cube in a three-quarters style view. Now make sure that you are in the mesh [Editing] menu (you can press F9 to jump straight to this - take a look at the icon that is now selected on the toolbar. Become familiar with this as you'll be using this menu a lot) where you'll see a section called [Mesh Tools 1]. It is a series of buttons in here all prefixed [Draw...] that we want to explore.

I'm not going to explain each one now - just experiment between now and the next tutorial. Toggle each of these buttons on and off and see what effect they have on the cube. Rotate the cube in 3D space with some of these turned on. Remember that you are not modifying the cube - just changing what you are seeing in the modeller. These buttons are to help you model, not change the colour or texture of the mesh.

Finally, hit [z] on your keyboard while your mouse is in the 3D view that you are using and see what effect this has to what you are seeing. The [z] key toggles back and forward between [Wireframe] and [Solid] views and can really help visualise what you are trying to model.

Saving your work

We're running out of space in this tutorial, but there is one more thing that we need to do. We need to save our cube. The 'empty' is not so important, so you can select it and delete it if you want to - you already know how to do that.

Next, go to the [File] menu and select [Save As] (or just press F2). One of the windows temporarily changes into a file-dialog. See what I meant about any window can do any job? This is only temporary however, and as soon as we are finished, it will switch back again.


Image: The file dialog

Change directory using the file dialog to wherever you want to store your blend files, and save it as something like 'Tutorial2.blend' or something similar that you'll recongise. Now that we have something on screen, we'll be using this file in the next few tutorials.

What's in tutorial 3?

In tutorial 3 we'll reload our cube and start to play with it. We'll start to explore the basics of manipulating the mesh, changing its shape and appearance as well as using some of the mesh tools and attaching some materials. We may also delve into the basics of SubDivided meshes to see what we can do with even such a basic mesh as a cube.

Now that you know the basics of using the Blender interface, from here on the pace will pick up a little.

Copyright (c) 2006, Craig Robinson ("Sphynx")
Permission has been granted for inclusion only on www.scifi-meshes.com and www.industrial-meshes.com

[Alareth]

I've got a Wacom Intuos3 and I can rotate as you described but panning and zooming don't seem to be working.

[Sphynx]

It may be a difference in Tablets and drivers - I'm using a Wacom Graphire3 and it all seems fine. I'll take a look at my wife's intuos and get back to you.

[Mauiman]

How do I add another object to the scene that was created as a project?

[Sphynx]

If mean another primitive, just use the [Add] option from the normal menu again. If you mean that resides in another file, use [Append] and drill down into the other blend file to find the object that you require.

If this is not directly about this tutorial however, it would be far better if you opened a query in teh Q&A section - you've got far more chance of other Blender users answering this question even before I see it.

[Spliteagle]

What happened to Blender For the Faint Hearted - 01 ????

[Sphynx]

It was lost during the re-working that Ace has done. Looks like a lot of threads older than a certain age have been dropped.

I'm looking for it in my archive, and will try to replace it.

[Spliteagle]

Thanks