Gah I kept messing up today with the UVs and ended up going through 3 different versions of this because I didn’t realise how warped the texture got at the seam. Anyway this is what I have after todays work. The textures are set up so that all the colours can be easily changed in Maya if need be.
We were watching The Wanderer (2015) with Yuan and she pointed out the use of light and tonality (see screenshots below). She suggested that we should think of our animation in terms of greyscale first as it’s easy to get carried away with colour. I think we could do this by either light intensity, shading darker to lighter with materials or Alec also mentioned how we could render using layers but I’d have to look that up again. Once we have our lights and darks sorted out it might be good to consider how we can make depth with a range of warm to cold colours leading into our scene. Abigail and I were discussing having warm colours on the character and colder on background maybe? I also think because our animation is happening around sunset and dusk that the light and shadow contrast would be quite soft. Yuan pointed out that the use of very soft filtered light might add to the romantic mood that’s happening in our scene.
Something similar to this back lighting might happen in ours at some point as Anam is travelling towards the sunset.
Screenshots are from this animation that Clare found:) The Wanderer (2015)
I really love this screenshot from Old Fangs (2009). We should consider how we can bring out the character’s silhouette with rim lighting but maybe we’ll not go too crazy so as to keep the lighting looking quite soft. Abigail also pointed out in this and in the Andy Kehoe drawings that there are some low detailed trees behind the detailed ones which works really well!:)
Yuan also mentioned today that we need to focus more on our cause and effect in the story so that it just doesn’t look like a character moving around from a to b without purpose. She suggested that we give more time for our first and last shot (and less time to middle) so as to give time to establish the setting and then show the effect. As we discussed before, getting the setting across and giving the story context will be very important for our story to work. Yuan agreed that the textures and detail will be very important for this but we need to make sure that the background never overpowers our main focus which is the character.
This is the cloak that I modeled to go over Abigail’s model of Anam and then converted to nCloth. I had started this by cutting a plane into the shape of the cloak out flat thinking that it would be easy to fold loosely in place for falling over the character. Duplicating the faces of the torso and extending the geometry down turned out to be a much easier solution and I then tried to reroute the mesh so that hopefully it deforms more cloth-like.
Ambient occlusion<3 It looks like I might have to apply it to the output mesh separately to the input mesh.
This is what the nCloth looks like on a plane that hasn’t been shaped:
These are tests with Anam. We might get a better result if we apply one of the cloth presets or I can play with the cloth thickness or other attributes.
This is where I tried to make the collar stiffer by adding ‘point to surface’ nConstraints:
This is one of the 2D trees that I didn’t upload last time. I liked the red outline that happened from having an incandescent red material underneath the other colours. I was trying to shape it more so that it would point towards the character in the scene also.
Today I UV unwrapped Clare’s 3D tree and tried to apply similar textures to those on my 2D trees. I think next time I’ll have to plan ahead a little more and place the seams in a place that the camera won’t see.
Look at the difference between 540 and 1080 quality in these two renders. The 1080 one took about 10 times longer at 26mins 5secs. I’ll have to figure out what Alec was telling me about baking textures if we want to render at this quality but quicker.
These are good videos to recap UV mapping.
Also some issues that came up today:
- How to make polygons render 2-sided:
- I didn’t know for a while why my alpha channel textures weren’t showing in the viewport and it was because Clare had the legacy default viewport turned on instead of viewport 2.
Mascelli, J. (1965) The 5 C’s of Cinematography. U.S.A. : Silman-James Press
This book is quite dense with information. It’s split into camera angles, continuity, cutting, close-ups and composition. I read the first four sections after Christmas and picked it up again to finish the composition chapter. I took notes this time for quicker revision of concepts as I think I’ve lost bits of what were in the first 4 sections. It’s probably worth re-reading anyway.
These are my rough notes for composition:
- motivate audience reaction according to scripts intent. Balanced or unbalanced?
- Space and time dimensions: movement holds the viewers attention, guard against insignificant/undesirable movement. Secondary actions such as dialogue also attract attention.
- Compose the shape of motions
- Language: lines, forms, masses, movements. Imaginary transitional lines are created from following action – these are important too.
The different types of lines and their possible effect:
- Lines can lie flat on surface or recede into picture. Diagonals are dynamic. Line sharpness is connected to speed and forcefulness.
- Form: abstract forms can be created from the arrangement of physical objects: triangles wide, based triangles, inverted triangles, circles, crosses, radiating lines, L-shaped composition.
- Mass: the pictorial weight of an object. Light mass on dark bg or vice versa. Large versus small mass.
- Movements: right to left movements are harder to follow, moving against the grain, should be used for actions which are difficult such as moving towards the villain. vertical movment: upward or downward? Diagonal, opposing forces, stress, the use of force. Curved; fear. Pendulum; pacing, monotony. Cascading. Radiating; growth. Interrupted or movement which changes direction. Towards or away from viewer.
- Balance: unbalance upsets the viewer. A large static object can be counter balanced by a small moving object.The two sides of the screen are like a see-saw. A moving object possesses more weight than a stationary object. The upper part of a picture is heavier than the bottom. The left side of frame can support more weight. An isolated object has more weight than crowded or stacked ones. Warm colours carry more weight than cold ones. Light values are heavier than shadowed objects.
- Formal and informal balance: symmetry versus asymmetry. Gravity inflluences balance; an objects center of gravity.
- Unity; perfect integration of elements.
- One center of interest. A group of people or objects can also be one center of interest.
- Positioning center of interest. Avoid vertical or horizontal lines cutting the image in half. Diversify: horizon placement, position of center of interest, player and camera movement.
- Attracting or switching center of interest: Position, movement, action and sound. Lighting, tonal value and colours. Selective focusing.
- Eye scan: within the frame and also from one frame to another. Usually smooth and orderly unless you want to add shock and abruptness to a subject.
- Image placement. lead room. head room is determined by image balance. Avoid cutting joints with the edge of the frame.
- Image size: give a clue (objects of known size) to scale. Framing influences how you perceive the size of a subject.
- Integrate composition and camera angles: continually revise player/background relationship.
- Perspective: linear; convergence of parallel lines. Aerial perspective; gradual lightening and softening of distant objects.
- How to increase perspective effects: camera angles which reveal the greatest number of planes/facets; angle plus angle. Choose angle and lens focal length that gives best linear convergence. Partially overlap players/props so as to convey spatial relation. Overlapping on movement introduces motion parallax. Move toward or away from screen to indicate spatial depth. Light a scene so as to get contrasting planes. Light an interior scene with a ‘hotter’ background for something similar to aerial perspective.
- Backgrounds: actions in foreground should be tied in with background. The background should be a constant subtle reminder of setting.Use lighting, tone and colour to separate the character from the background.
- Frames: Compose an internal frame at an angle for depth. The frame should some tonal contrast with the subject it”s framing. Keep large foreground frames in sharp focus. Avoid wildly moving frames that detract from subject such as leaves in a high wind. Frames relevant/connected tto story may aid story telling and convey setting.
- Dynamic composition: move from a quiet scene to one which contains sudden movement close to the camera. Use this for dramatic situations.
- Suspenseful composition: Show a few frames where the action is hidden from the audience.
- Catalog pictures: grouping and arranging shapes.
- Compositional variety: variety in composition, camera angle, image size.
- Always think in terms of creating depth.
- Simplicity: be economical in terms of line, form, mass and movement. Have one center of interest and employ one unified style. Get rid of anything in frame that’s not needed for the storytelling. “If a vast number of compositional elements must be photographed then they should be harmoniously grouped.”
- “remember that the viewer mst be affected both pictorially and psychologically.
These are some trees that I’ve been experimenting with over the past few days by making alpha channel shapes in photoshop, plugging them into materials and layering the materials in layered shaders. It’s interesting to see how different coloured lighting and camera angles can affect the scene also.
I was occasionally getting this problem where if two planes with transparency maps plugged into their transparency channels were overlapping, the plane nearest the camera would start to show opaque edges again like this:
I tried going through the different options in the material attributes, the group shader node, the lights and the global render settings but no luck. I posted about it on CGSociety and someone suggested that I use mia_material_x and use the cut out opacity attribute instead. It doesn’t seem to work with the .tga files that I’ve created but I’ll try again.