Jennifer's Classroom

[pappou]

Wonderful! From your clear explanations i see that things are actually working properly; mostly, i seemed to have been looking for different results and could not see the correct way:

Translation Problem:
Being aware of the problem as 'translation' helped; so, this is an Object Movement command;

Rotation Problem:
Likewise, the 'C' movement works fine, which too is an Object Movement command;

Panning Problem:
But with shift alt left click drag, everything moves together; so this must be a Scene View command;

Scaling Problem:
The smaller V does work now; and you were right: My wheel 'scaler' was NOT scaling; it merely brought grid and cube closer or farther away; Hooray! It does work.

Scene and Object Command Problem:
Your very vivid description is well received; So, am i correct: Are “ scene view commands” those in which the Cube to Grid relationship remains the same?
If so, then “object movement commands” those where the Cube moves in relation to the Grid. That seems like a necessary distinction to keep in mind.

Do you detect more errors in my understanding? This has really been a good lesson in overcoming my predisposed notions: I could not see the correct way due to my prejudiced vision.

Very interesting, Teacher Jennifer; thank you for your patience and determination to work it out.

[pappou]

Jojon:
Thank you for uncovering the hidden menus at the top.
But i cannot get them to function.

When you say:
"Here there are separate snap-to-grid buttons for moving, rotating and scaling - make sure they are all unticked..."
are you speaking of the 'View & Controls' button? The Edit Methods button? Language & Font? Etc.? These do not respond for me. How do i get into them [these top controls]?

[Jojon]

Jojon:
Thank you for uncovering the hidden menus at the top.
But i cannot get them to function.

When you say:
"Here there are separate snap-to-grid buttons for moving, rotating and scaling - make sure they are all unticked..."
are you speaking of the 'View & Controls' button? The Edit Methods button? Language & Font? Etc.? These do not respond for me. How do i get into them [these top controls]?

*Firing up blender to make sure I get it right-ish*

Ah, if you drag that pane divider a bit further down, you'll discover even more buttons, above the row you mention. That bottom row of buttons works like section tabs, allowing you to select which buttons show up above them. So if you click the 'view and controls' button, in the top half of the pane you should get buttons relating to view and control options. In this section, the "snap to grid" trio of toggle buttons is the third column from the left (note the "snap to grid" title above them). They are two-state buttons - click once and they go active, turning dark to indicate this. Click again and they are deactivated.

However; reading Jennifer's and your posts above, it sounds clear that your problem had nothing to do with the grid. I've gotten so used to some of the basic concepts in modelling software, that it did not even strike me, that an uninitiated person would not know, that when you move around in the view, it is the observer that is moving around, rather than the objects, nor hold the distinction between this and the operations that actually manipulate objects.

Sorry for adding confusion. :7

[Murray]

OOOH, this Classroom is magic, if I never create anything the learning of how others have done so is worth it to me. I can see you have worked very hard to explain Blender to us and I hope I can return a little for your efforts, Cheers Jenny , Mur

[Sophia]

Yes thanks!!! I have done the first steps installation Now on to the tutorial... can't wait to start blending

[Jennifer_P]

Welcome Murray, Sophia! Now if either of you have a different mouse configuration than the one I've been describing in these lessons, be sure to let me know! These instructions are kind of mouse specific in that you need two mouse buttons and (infrequently) a mouse wheel. There are different ways to do some of these commands if you don't have a wheel or if you have a different button configuration. And could you also let me know if something doesn't make sense or takes an inordinate amount of time?--I want these lessons to be accessible to busy people and to people who aren't 3D modeling savvy (yet).

And pappou, looks like you've got things sorted out with the screen viewing and object transforming commands and the grid. And don't worry, this will all make complete intuitive sense shortly. I'm not sure about the distortion on the cube in the screenshot--I'll have to take a look at that here. And Jojon, I'm glad to have someone who can see these things from a different point of view. You know how it is--you might spend three hours trying to figure out some computer problem, and then another person will walk up, point to the error and say, "Here's your problem." So please keep dishing out the help, we're going to need it I'm sure.

[Jennifer_P]

And Monkeyboy, no problemo--go with whatever works for you, of course!

pappou--that cube distortion you got when merging your vertices is supposed to happen. (At least, it's what happened when I merged tow vertices.) So you are good.

And yeah, I know this lesson looks really long, but it shouldn't take more than 20 minutes (unless you have perfectionistic tendencies); mostly it's just stuff to read, or else really detailed instructions about where to find some new buttons. So don't be intimidated, it's not as long as it looks!

Lesson 4
Today is going to be a big day for us! Today we're learn how to create shapes other than the cube, plus we'll briefly touch on a new sort of transformation and we'll even (gasp!) add some color!
Whooooeeeee!!
Okay, we'll start off by creating some new objects. As always, first start off with a fresh cube (File, New, Erase All). Got it? Well, using the translation transformation (the straight line drag), move the cube away from the center of the grid.
A stern admonition: for the purposes of this lesson, do not change the cube's orientation relative to the grid by rotating the cube ( at least until I say so). Scaling and dragging are okay, but not rotation (for now). I'll explain why in the next lesson; it would make this lesson too long if I got into it now.
Placing Objects
To start off, look in the center of the grid where the cube used to be. Recognize that candy cane striped ring? That's the 3D cursor, and it determines where an object is created at. To place it, just left click somewhere and it will move there. Now, find the "File" menu option, and then hop over one menu option to the right of "File" and you'll see the "Add" menu option. Open the Add menu up, and you will see a submenu by the name of "Mesh." (Note that this "Mesh" menu is not the mesh menu from the teal bar!) So, in the Add menu, open the Mesh menu up....Oooh. These are some of the 3D shapes that you are given by Blender to work with. You've got planes, cubes, circles, UVspheres, icoSpheres, cylinders, cones, grids, monkeys, and tori. These shapes are called primitives, since they are the primitive shapes out of which you build more complex shapes. Go ahead and make a whole bunch of primitives! (Just select "OK" in the menu that will pop up when you choose some of them.) Now let's see what we can do with our primitives. Go ahead and start off with a fresh cube (File, New, Erase All, you know the drill). Like before, drag the cube out of the way. And remember, don't rotate the cube.
Since we're Uru players, the first primitive that we'll work with in our first real model will naturally be the cone. So, in the Mesh submenu go ahead and choose "Cone." Immediately you are presented with another menu full of options, i.e. "Vertices," "Radius," "Depth," and "Cap End." These options determine how your cone will end up looking when it is created. For now, the options that we particularly care about are "Radius" and "Depth." The radius is how wide the cone's base will be. The depth is how tall the cone will be. Let's see what the default settings look like, and then we can change our cone from there. So, just click "OK" and your cone will appear at the 3D cursor's location. Move around and check your new primitive out. Okay, so it's not tall enough (or perhaps not thin enough at the base). We'll fix that soon! But first let's take a look at this cone a bit and see what we can see.
Circles and Vertices; Adjusting a Cone
Let's go into Edit Mode and take a look. As you can see, the vertices which define the shape of the cone are visible. (You might also notice that all the vertices are yellow and selected--leave them that way.) Now take a look at the circle which makes up the base of the cone. As you can see, it is not really a circle; it is just a thirty-two sided polygon with each point occupied by a vertex. Hmm...Is 32 vertices really enough to make a smooth circle? That depends...If the circle is distant or small, then 32 vertices is enough, but if the circle is close up or huge, then its polygonal nature becomes extremely obvious, in which case you'll need to add more vertices. How do we do this? Well, let's delete the cone we have now and make a new one. Remember how to delete? You go to that teal colored bar from Lesson 1 (better known as the 3D view header bar--memorize that) and select the menu option "Mesh." In the Mesh menu, select "Delete," and then in the menu that pops up, choose "All." Bye bye cone! Well, let's make another one. This time, when you get to the menu where you specify the qualities of the cone, type in 100 where it says "Vertices" by left clicking on the numbers "1.00" or "2.00" so you can type in numbers directly. By typing in values directly you can get up to 3 decimal points worth of accuracy! You can bet the circle making up the bottom of the cone will be a lot smoother now. While you're at it, also adjust the radius and the depth until you think they will produce a good Uru cone. Click "OK" when you're ready to try your new cone. If the cone looks right, great! If not, delete the cone and continue with trial and error until you have a cone that you're satisfied with. (Don't you wish there were an easier way to adjust your cone's shape? There is--we'll cover it soon.)
How to Flatten a Cube; the 3D Arrow/Color/Coordinate System
Now it's time to add the square base. For this, we need a cube, which we just happen to have. The plan is to flatten this cube and then position it underneath the cone. Time to go into Object Mode and select the cube by right clicking on it ('cause you can't select the cube by right clicking in Edit Mode). You didn't rotate the cube, right? If you did, just start over with a fresh Blender scene and quickly run through the lesson again up to this point. If you didn't, good, we'll move on to the next step. To start off with, in the 3D View header bar, choose the menu option "Object", and in the menu that pops up, choose the submenu "Transform". Within the Transform submenu, you will see several options, including Grab/Move, Rotate, and Scale. Beneath each of these options is a submenu such as "Grab/Move on axis" or "Scale on axis". Open up the "Scale on axis" submenu and observe its contents, particularly noting the X Global, Y Global, and Z Global commands. What do these command names mean?
To answer that question, note the three arrows shooting out of the cube (the red, green, and blue ones). Those arrows point in the X, Y, and Z directions. Red points at X, Green points at Y, and Blue points at Z. RGB = XYZ. It's my theory that Aloys was staring at these arrows when he thought up Ahra Pahts. And if you ever forget how the colors match with the arrows, just look at the bottom left hand corner of the dark grey window in which the cone and cube are located--you will see three colored little lines labeled X, Y, and Z. These coordinates are known as Global Coordinates, because they apply to everything in your Blender scene, all objects, all everything--thus they are considered "global."
Now, try the "X Global" command, the "Y Global" command, and the "Z Global" command. Did you see the difference between them? Each command scales the cube only in a specific direction: X, Y, or Z. Use one of these commands to flatten the cube to make an appropriate base for the cone. It's okay to use the rotation command now too, so go ahead and use the rotation and translation mouse commands to move the base underneath the cone. Once the base has been placed on the cone to your satisfaction, we are ready to add COLOR!
Objects in TECHNICOLOR
Okay, select the cone (you can color the cube later). Now do you see the bar right below the 3D View header bar? The one with the word "Panels" on it? Well, on that bar look for a little "shaded sphere" looking button; if you hover your mouse over it the popup should read "Shading (F5)" . Click the shaded sphere (Shading) button, and you'll get a bunch of shading-related buttons directly to the right of it: a lightbulb, a red sphere, a leopard skin pattern, a radioactivity symbol, and a globe. Click the red sphere button, which reads "Material buttons" if you hover your mouse over it. In the window below the red sphere, you will see two subwindows, one called "Preview" and the other called "Links and Pipeline." Look in the Links and Pipeline subwindow and you will see a button called "Add new" (as in "add a new material to my boring grey object"); click it. Now you have all sorts of subwindows: "Preview," "Links and Pipeline," "Material," "Ramps," "Shaders," "Mirror Trans," "SSS," and "Texture." In the "Material" subwindow, you'll see a longer-than-it-is-wide light grey rectangle sitting stacked on top of two matching white rectangles which have the same shape as the grey rectangle. Left click this grey rectangle, and in the window that pops up, pick a color for your cone (such as orange). Now select the rectangular base of the cone and again click the shaded sphere button (Shading button). This time, things look different--there is no "Add New" button visible. Why? Because the cube already has a material--Blender automatically gives the "Startup cube" a boring gray material without any intervention on your part. But we don't want this material, so let's change it. In the "Links and Pipeline" window, you will see a box called "Ma:Material." Click the little arrows just to the left of it, and in the little menu that pops up, find the material you used on the base of the cone and select it. There, now the base should be the same color as the cone. Ah....your cone is complete! /clap Very good! I hope you are feeling proud of yourself. And do you know what the best thing is? One day here soon, you're going to be so familiar with Blender's interface that you'll be able to make an orange cone like this in less than two minutes, with minimal mental and modeling effort. Just think what you'll be able to make then!

[Jojon]

You know how it is--you might spend three hours trying to figure out some computer problem, and then another person will walk up, point to the error and say, "Here's your problem." :roll

Oh, don't i know it!

If you'll excuse me then, I'll try explaining Pappou's distortion problem a little closer, so that he get the option to grasp it right away, should he want to... Pappou: if all this gets too much to take in at once, just ignore the remainder of this post.

This is a basic matter of geometry, which should be right up your alley and software and hardware limitations. First we must understand that the particular 3D paradigm we work with here, can really *only handle flat surfaces*. That is; say you have a sheet of paper laying on a table. As it is, it is perfectly flat - in effect two dimensional. Now lift one corner of the sheet, leaving the other three touching the table. Ok, so now you have a complex curved shape. Since our goal here is to create a representation of the object, using straight lines only, and only one per side, to boot, we have a problem. If the sheet of paper had been cardboard instead, we wouldn't have this problem, since we would have lifted the entire sheet - in effect rotated it in 3D space, but despite our straight line limitation, we *can* move the corners (vertices) individually. So what does Blender do, when it encounters a situation like this and is required to render an on-screen representation? -It folds the sheet's rectangle diagonally, into two triangles. Triangles are always handled as if they are made from a rigid grade of paper by Blender, so you never get a bent shape (no croissants from a single triangle here , resulting in a flat surface every time. Obviously there are two ways to divide a rectangle into two triangles; top left to bottom right or top right to bottom left. I believe what decides whether you get a convex or a concave duo of triangles here, is decided by which of the corners of the rectangle is considered corner number 1 (probably, in turn, decided by creation date).
So you get two triangles drawn on screen, in place of your rectangle, both shaded as if they were facing the same direction, although they are not (the facing direction is what is termed "normal"), but there is no edge drawn between them -- it is still considered a four-sided polygon (or "quad").
If you want this to look correctly in the future, you will have to divide the rectangle into triangles for real.

So, why does it look right when viewed from certain angles? This is because the view we are working in here is a parallel projection - again something I expect to be familiar to your architect mind, where things are drawn the same size, regardless of how far away they are from you. Some would liken this to an infinite viewing distance, but I feel that complicates things too much. This is what make things line up so perfectly in the view.

You can swap between parallel and perspective projection using the "5" key on the numeric keypad (assuming numlock is on).

A little by the by: Another very useful hotkey (these things will speed up your work flow considerably, once Jen begins the transition to them from menu access), is the TAB key, which toggles a view between "Object" and "Edit" mode.

[Jennifer_P]

Good explanation, Jojon. And I'd love to see a screenshot of everybody's cones, when you finish the lesson!

[pappou]

Friday 21Mar08

Jojon, i see that you're right, concerning the JC Distorted figure on Page 4. Those tortured faces of the cube on the sides (after merge) are complex curves, not compound ones as i said. Thank you for your explanation of the mysteries involved. I think you are confirming my guess that triangulation was what was needed, and what was missing, from the figure.

But i see at the start of Jen's new Lesson 4 that quads are also used in the mesh – for instance, in the torus and the monkey:

Blender 02.jpg (26.91 KiB) Viewed 4093 times

What is exciting to me in these (and i love solid geometry) is the resulting interpenetration of figures. We used to spend hours in architecture school working out the correct shapes of merging surfaces. They don't do that anymore, but just because one's Blender program can do it does not mean that one's eye takes in the revealed complexities. Casting shadows is very much the same thing.