Loading…

Worked solutions

IB Computer Science HL · Paper 3

Computer Science, Higher Level

Session: November 2016
Marks: 30
Duration: 1 hour
Calculator: Not allowed
Questions: 4

Keep your copy of the paper open alongside. We don't reproduce the questions: each part below is labelled to match the November 2016 Global paper, which your school or the IB can provide.

Practice mode

Hide our working until you've had a go.

Questions

Questions 1–4 · 30 marks

Question 1

4 marks

Surface point coloration factors and geometric primitives in 3D modeling

Modelling and simulation

(a)

2 marks

Identify two primary factors that determine the final rendered color assigned to a surface point in a 3D scene:
1) Surface material properties: The intrinsic base albedo color, texture mapping coordinates, diffuse reflectivity, specular roughness, and transparency coefficients assigned to that surface coordinate.
2) Characteristics of incident illumination: The intensity, spectrum color, direction, and attenuation of light striking the point, along with the orientation angle between the light vector and the surface normal vector.

Mentor tip

State two clear physical factors: the object's surface material properties (diffuse/specular reflectivity) and the properties/angle of incoming illumination.

(b)

2 marks

Define the term geometric primitives in 3D modeling, providing specific examples:
Geometric primitives are predefined, mathematically defined fundamental volumetric shapes and building blocks used by 3D software as starting geometry for complex modeling.
Common examples include polygonal cubes, spheres, cylinders, cones, pyramids, and tori, which artists manipulate, scale, boolean-combine, and deform into intricate objects.

Mentor tip

Define primitives as standard foundational 3D shapes and cite at least two canonical examples such as spheres, cubes, or cylinders.

Question 2

8 marks

Relationship between avars and motion capture alongside inverse kinematics mechanics

Modelling and simulation

(a)

4 marks

Describe the operational relationship between animation variables (avars) and motion capture (Mo-cap):
1) Function of avars: Animation variables (avars) are parameterized mathematical controls, degrees of freedom, or rotational hinges embedded within a 3D digital character's skeletal rig that govern joint angles, facial muscle tension, and limb positioning.
2) Data collection in motion capture: During a Mo-cap recording session, a live performer wears a specialized suit fitted with optical retro-reflective markers tracked in 3D space by calibrated high-speed infrared cameras.
3) Coordinate mapping and retargeting: Tracking software processes the spatial marker trajectories, resolving the actor's anatomical joint rotations and translating them directly into corresponding numerical values for each avar on the virtual character's skeleton.
4) Illustrative example: For instance, when an actor raises their physical forearm, optical sensors calculate an elbow flexion of $7 5^{\circ}$ ; this angular delta directly updates the elbow flexion avar parameter of the 3D model, causing the character's digital mesh to mirror the performer's authentic biological movement frame by frame.

Mentor tip

Define both avars and motion capture, explain the mathematical mapping between them, and provide a clear physical movement example (such as an elbow or shoulder joint rotation).

(b)

4 marks

Explain how inverse kinematics (IK) enables realistic limb positioning when a character places a hand on a wall:
1) Forward kinematics limitation: In forward kinematics, an animator must manually calculate and specify rotation angles for the clavicle, upper shoulder, elbow, and wrist in sequential hierarchy to place the hand against a target coordinate, making precise environmental contact cumbersome and prone to slipping.
2) Inverse kinematics paradigm: Inverse kinematics works in the opposite direction; the animator defines only the desired final 3D spatial coordinate of the end-effector (the character's hand resting on the wall plane).
3) Mathematical joint resolution: The IK solver executes numerical optimization algorithms (such as the Jacobian inverse or cyclic coordinate descent) to calculate the necessary joint angles for the wrist, elbow, and shoulder automatically.
4) Anatomical constraints and realism: The algorithm enforces natural physiological constraints (such as minimum and maximum elbow bend limits and rotational bounds), ensuring the arm bends in an anatomically authentic posture while keeping the palm anchored to the wall surface.

Mentor tip

Contrast FK with IK: with IK, the animator specifies the final end-effector target on the wall, and algorithms automatically calculate intermediate joint angles within anatomical limits.

Question 3

6 marks

Computational efficiency of backward ray-tracing from camera to light sources

Modelling and simulation

6 marks

Explain why ray-tracing algorithms trace rays backward from the camera/eye to light sources rather than forward from light sources to the camera:
1) Principle of physical light propagation (Forward tracing): In the physical universe, light sources emit countless billions of photons propagating outward in spherical wavefronts in all directions. Only an infinitesimal percentage of these emitted rays happen to strike scene objects and bounce at angles that pass through the camera aperture.
2) Computational futility of forward tracing: Simulating forward ray paths on a computer requires calculating intersections and reflections for millions or billions of rays that never intersect the camera lens, wasting massive CPU cycles on unviewed, out-of-frame computations.
3) Backward ray-tracing strategy (Eye-to-source): Backward ray-tracing reverses the optical path by originating rays exclusively from the virtual camera viewpoint (the viewer's eye). The camera projects exactly one primary eye ray (or a small set of sub-pixel samples for anti-aliasing) through each pixel on the image plane into the 3D scene.
4) Targeted intersection testing: Primary rays are tested for intersection only against visible geometry in front of the lens. At the nearest intersection point, secondary shadow rays are cast directly toward light sources to evaluate illumination and shadows, along with reflection and refraction rays where materials demand it.
5) Processing efficiency comparison: Because the number of primary rays traced is bounded directly by the viewport screen resolution (e.g., $1920 \times 1080 \approx 2.07 \times 1 0^{6}$ pixels) rather than the astronomical number of rays emitted by light fixtures, backward tracing minimizes computational overhead by orders of magnitude.
6) Algorithmic conclusion: Eye-to-source tracing guarantees that every ray evaluated contributes directly to the visible pixels of the final rendered frame, maximizing computational efficiency.

N_{backward} = Width \times Height \times SamplesPerPixel

Mentor tip

Explain that tracing from the light source calculates billions of rays that never hit the camera, whereas tracing backward from the camera bounds calculation count directly to visible screen pixels.

Question 4

12 marks

Resource allocation and algorithmic trade-offs in high-end commercial computer animation

Modelling and simulation

12 marks

Discuss how computational resource demands influence algorithmic decisions for Xiao-Ling's team producing a three-minute high-end animated commercial featuring anthropomorphic characters, dynamic mountainous terrain, and photorealism:
Project scope and production context:
- Scope metrics: A three-minute commercial at standard cinematic frame rates (24 frames per second) requires generating $3 \times 60 \times 24 = 4320$ distinct photorealistic frames. As an offline commercial rather than a real-time game, frames can be pre-rendered across a dedicated rendering farm, but hardware budgets and strict client delivery schedules place hard boundaries on computational overhead.
Feature 1: Animated anthropomorphic creatures (Modeling and animation):
- Optical motion capture vs keyframe tweening: Capturing genuine human actors with optical Mo-cap records nuanced facial cadence and natural musculoskeletal dynamics, effectively overcoming the uncanny valley. However, tracking many synchronized cameras and hundreds of avars generates massive datasets requiring extensive cleanup and specialized processing infrastructure.
- Algorithmic trade-off: The team can capture primary body cadence via Mo-cap for the three characters, while deploying keyframe animation with spline tweening for facial expressions and non-human features (tails, ears). This minimizes motion noise and hardware tracking requirements while preserving realistic locomotion.
Feature 2: Dynamic movement through changing mountainous terrain:
- Procedural fractal generation: Deploying recursive midpoint displacement or diamond-square fractal algorithms allows the team to generate vast, rugged mountain ranges from simple geometric bounding rectangles. Varying stochastic perturbation factors easily creates diverse terrain profiles.
- Memory and recursion depth management: Because fractal algorithms exhibit recursive memory growth, increasing subdivision depth exponentially expands polygon counts. Generating excessive microscopic triangles across distant horizons would exhaust available system RAM.
- Level of Detail (LOD) optimization: The team should implement dynamic Level of Detail; distant background mountain peaks are generated at lower subdivision depths with normal-map textures, while high-density polygonal meshes are reserved exclusively for foreground terrain in direct contact with characters, keeping memory consumption within server limits.
Feature 3: Achieving a high degree of photorealism (Lighting and rendering):
- Scanline rendering vs ray-tracing: Scanline rasterization computes frames rapidly line-by-line using GPU z-buffering, but struggles with complex reflections, ambient color bleeding, and realistic atmospheric shadows.
- Recursive ray-tracing and path tracing: Ray-tracing recursively evaluates light reflection, refraction, and shadow rays for each surface intersection, delivering authentic physically-based global illumination, soft area shadows, and subsurface skin scattering critical for convincing anthropomorphic characters.
- Managing ray-tracing computational overhead: While ray-tracing is computationally demanding, the offline nature of pre-rendered commercials makes it entirely feasible. The team can optimize render times by using Bounding Volume Hierarchies (BVH) for accelerated ray-geometry intersection testing and applying adaptive Monte Carlo sampling with modern AI denoising to suppress rendering noise at lower sample counts.
Strategic synthesis and execution plan:
Xiao-Ling's team should execute an optimized hybrid pipeline: drive the three anthropomorphic characters using Mo-cap retargeted onto constrained avar rigs, generate the mountainous background via recursive fractals with adaptive Level of Detail, and render the final 4320 frames using an offline ray-tracing engine accelerated by Bounding Volume Hierarchies, achieving cinematic photorealism within production deadlines and studio computing capacity.

Total Frames = 3 minutes \times 60 seconds \times 24 fps = 4320 frames

Mentor tip

Structure the discussion around the three project features (anthropomorphic creatures, mountainous terrain, photorealism), evaluating specific algorithms (Mo-cap/tweening, fractals/LOD, ray-tracing/BVH) against CPU, memory, and time constraints.

Our analysis

Where the marks are

Each part sorted by the topic it mostly tests. Modelling and simulation carries the most marks on this paper.

Modelling and simulation30 marks · 100%
Q1, Q2, Q3, Q4

Mentor insights

Traps to watch for

01
Confusing geometric primitives with finished complex 3D meshes · Q1
Primitives are the basic, foundational mathematical geometric elements (cubes, spheres, cylinders, cones) used as starting building blocks, not detailed finished character models.
02
Reversing forward and inverse kinematics workflows · Q2
In forward kinematics, the animator manually specifies each joint angle down the limb hierarchy; in inverse kinematics, the animator specifies the final end-effector position, and algorithms calculate joint angles automatically.
03
Assuming light rays naturally travel from the eye outward in physics · Q3
In the physical universe, light travels strictly from the light source to the eye; backward ray-tracing is a computational optimization that reverses this direction to evaluate only visible rays.

Solutions and tips are GeorgeTheTutor's own; the questions are not reproduced here. Not affiliated with or endorsed by the International Baccalaureate Organization.

Want someone to go through your paper with you?

Our maths mentors took these exams recently. One-to-one, they'll find where your marks are going and fix it before May.

Find your maths mentor More past papers

Worked solutions

IB Computer Science HL · Paper 3

Computer Science, Higher Level

Session: November 2016
Marks: 30
Duration: 1 hour
Calculator: Not allowed
Questions: 4

Q1 Q2 Q3 Q4

Keep your copy of the paper open alongside. We don't reproduce the questions: each part below is labelled to match the November 2016 Global paper, which your school or the IB can provide.

Practice mode

Hide our working until you've had a go.

Questions

Questions 1–4 · 30 marks

Question 1

4 marks

Surface point coloration factors and geometric primitives in 3D modeling

Modelling and simulation

(a)

2 marks

Identify two primary factors that determine the final rendered color assigned to a surface point in a 3D scene:
1) Surface material properties: The intrinsic base albedo color, texture mapping coordinates, diffuse reflectivity, specular roughness, and transparency coefficients assigned to that surface coordinate.
2) Characteristics of incident illumination: The intensity, spectrum color, direction, and attenuation of light striking the point, along with the orientation angle between the light vector and the surface normal vector.

Mentor tip

State two clear physical factors: the object's surface material properties (diffuse/specular reflectivity) and the properties/angle of incoming illumination.

(b)

2 marks

Define the term geometric primitives in 3D modeling, providing specific examples:
Geometric primitives are predefined, mathematically defined fundamental volumetric shapes and building blocks used by 3D software as starting geometry for complex modeling.
Common examples include polygonal cubes, spheres, cylinders, cones, pyramids, and tori, which artists manipulate, scale, boolean-combine, and deform into intricate objects.

Mentor tip

Define primitives as standard foundational 3D shapes and cite at least two canonical examples such as spheres, cubes, or cylinders.

Question 2

8 marks

Relationship between avars and motion capture alongside inverse kinematics mechanics

Modelling and simulation

(a)

4 marks

Describe the operational relationship between animation variables (avars) and motion capture (Mo-cap):
1) Function of avars: Animation variables (avars) are parameterized mathematical controls, degrees of freedom, or rotational hinges embedded within a 3D digital character's skeletal rig that govern joint angles, facial muscle tension, and limb positioning.
2) Data collection in motion capture: During a Mo-cap recording session, a live performer wears a specialized suit fitted with optical retro-reflective markers tracked in 3D space by calibrated high-speed infrared cameras.
3) Coordinate mapping and retargeting: Tracking software processes the spatial marker trajectories, resolving the actor's anatomical joint rotations and translating them directly into corresponding numerical values for each avar on the virtual character's skeleton.
4) Illustrative example: For instance, when an actor raises their physical forearm, optical sensors calculate an elbow flexion of $7 5^{\circ}$ ; this angular delta directly updates the elbow flexion avar parameter of the 3D model, causing the character's digital mesh to mirror the performer's authentic biological movement frame by frame.

Mentor tip

Define both avars and motion capture, explain the mathematical mapping between them, and provide a clear physical movement example (such as an elbow or shoulder joint rotation).

(b)

4 marks

Explain how inverse kinematics (IK) enables realistic limb positioning when a character places a hand on a wall:
1) Forward kinematics limitation: In forward kinematics, an animator must manually calculate and specify rotation angles for the clavicle, upper shoulder, elbow, and wrist in sequential hierarchy to place the hand against a target coordinate, making precise environmental contact cumbersome and prone to slipping.
2) Inverse kinematics paradigm: Inverse kinematics works in the opposite direction; the animator defines only the desired final 3D spatial coordinate of the end-effector (the character's hand resting on the wall plane).
3) Mathematical joint resolution: The IK solver executes numerical optimization algorithms (such as the Jacobian inverse or cyclic coordinate descent) to calculate the necessary joint angles for the wrist, elbow, and shoulder automatically.
4) Anatomical constraints and realism: The algorithm enforces natural physiological constraints (such as minimum and maximum elbow bend limits and rotational bounds), ensuring the arm bends in an anatomically authentic posture while keeping the palm anchored to the wall surface.

Mentor tip

Contrast FK with IK: with IK, the animator specifies the final end-effector target on the wall, and algorithms automatically calculate intermediate joint angles within anatomical limits.

Question 3

6 marks

Computational efficiency of backward ray-tracing from camera to light sources

Modelling and simulation

6 marks

Explain why ray-tracing algorithms trace rays backward from the camera/eye to light sources rather than forward from light sources to the camera:
1) Principle of physical light propagation (Forward tracing): In the physical universe, light sources emit countless billions of photons propagating outward in spherical wavefronts in all directions. Only an infinitesimal percentage of these emitted rays happen to strike scene objects and bounce at angles that pass through the camera aperture.
2) Computational futility of forward tracing: Simulating forward ray paths on a computer requires calculating intersections and reflections for millions or billions of rays that never intersect the camera lens, wasting massive CPU cycles on unviewed, out-of-frame computations.
3) Backward ray-tracing strategy (Eye-to-source): Backward ray-tracing reverses the optical path by originating rays exclusively from the virtual camera viewpoint (the viewer's eye). The camera projects exactly one primary eye ray (or a small set of sub-pixel samples for anti-aliasing) through each pixel on the image plane into the 3D scene.
4) Targeted intersection testing: Primary rays are tested for intersection only against visible geometry in front of the lens. At the nearest intersection point, secondary shadow rays are cast directly toward light sources to evaluate illumination and shadows, along with reflection and refraction rays where materials demand it.
5) Processing efficiency comparison: Because the number of primary rays traced is bounded directly by the viewport screen resolution (e.g., $1920 \times 1080 \approx 2.07 \times 1 0^{6}$ pixels) rather than the astronomical number of rays emitted by light fixtures, backward tracing minimizes computational overhead by orders of magnitude.
6) Algorithmic conclusion: Eye-to-source tracing guarantees that every ray evaluated contributes directly to the visible pixels of the final rendered frame, maximizing computational efficiency.

N_{backward} = Width \times Height \times SamplesPerPixel

Mentor tip

Explain that tracing from the light source calculates billions of rays that never hit the camera, whereas tracing backward from the camera bounds calculation count directly to visible screen pixels.

Question 4

12 marks

Resource allocation and algorithmic trade-offs in high-end commercial computer animation

Modelling and simulation

12 marks

Discuss how computational resource demands influence algorithmic decisions for Xiao-Ling's team producing a three-minute high-end animated commercial featuring anthropomorphic characters, dynamic mountainous terrain, and photorealism:
Project scope and production context:
- Scope metrics: A three-minute commercial at standard cinematic frame rates (24 frames per second) requires generating $3 \times 60 \times 24 = 4320$ distinct photorealistic frames. As an offline commercial rather than a real-time game, frames can be pre-rendered across a dedicated rendering farm, but hardware budgets and strict client delivery schedules place hard boundaries on computational overhead.
Feature 1: Animated anthropomorphic creatures (Modeling and animation):
- Optical motion capture vs keyframe tweening: Capturing genuine human actors with optical Mo-cap records nuanced facial cadence and natural musculoskeletal dynamics, effectively overcoming the uncanny valley. However, tracking many synchronized cameras and hundreds of avars generates massive datasets requiring extensive cleanup and specialized processing infrastructure.
- Algorithmic trade-off: The team can capture primary body cadence via Mo-cap for the three characters, while deploying keyframe animation with spline tweening for facial expressions and non-human features (tails, ears). This minimizes motion noise and hardware tracking requirements while preserving realistic locomotion.
Feature 2: Dynamic movement through changing mountainous terrain:
- Procedural fractal generation: Deploying recursive midpoint displacement or diamond-square fractal algorithms allows the team to generate vast, rugged mountain ranges from simple geometric bounding rectangles. Varying stochastic perturbation factors easily creates diverse terrain profiles.
- Memory and recursion depth management: Because fractal algorithms exhibit recursive memory growth, increasing subdivision depth exponentially expands polygon counts. Generating excessive microscopic triangles across distant horizons would exhaust available system RAM.
- Level of Detail (LOD) optimization: The team should implement dynamic Level of Detail; distant background mountain peaks are generated at lower subdivision depths with normal-map textures, while high-density polygonal meshes are reserved exclusively for foreground terrain in direct contact with characters, keeping memory consumption within server limits.
Feature 3: Achieving a high degree of photorealism (Lighting and rendering):
- Scanline rendering vs ray-tracing: Scanline rasterization computes frames rapidly line-by-line using GPU z-buffering, but struggles with complex reflections, ambient color bleeding, and realistic atmospheric shadows.
- Recursive ray-tracing and path tracing: Ray-tracing recursively evaluates light reflection, refraction, and shadow rays for each surface intersection, delivering authentic physically-based global illumination, soft area shadows, and subsurface skin scattering critical for convincing anthropomorphic characters.
- Managing ray-tracing computational overhead: While ray-tracing is computationally demanding, the offline nature of pre-rendered commercials makes it entirely feasible. The team can optimize render times by using Bounding Volume Hierarchies (BVH) for accelerated ray-geometry intersection testing and applying adaptive Monte Carlo sampling with modern AI denoising to suppress rendering noise at lower sample counts.
Strategic synthesis and execution plan:
Xiao-Ling's team should execute an optimized hybrid pipeline: drive the three anthropomorphic characters using Mo-cap retargeted onto constrained avar rigs, generate the mountainous background via recursive fractals with adaptive Level of Detail, and render the final 4320 frames using an offline ray-tracing engine accelerated by Bounding Volume Hierarchies, achieving cinematic photorealism within production deadlines and studio computing capacity.

Total Frames = 3 minutes \times 60 seconds \times 24 fps = 4320 frames

Mentor tip

Our analysis

Where the marks are

Each part sorted by the topic it mostly tests. Modelling and simulation carries the most marks on this paper.

Modelling and simulation30 marks · 100%
Q1, Q2, Q3, Q4

Mentor insights

Traps to watch for

01
Confusing geometric primitives with finished complex 3D meshes · Q1
Primitives are the basic, foundational mathematical geometric elements (cubes, spheres, cylinders, cones) used as starting building blocks, not detailed finished character models.
02
Reversing forward and inverse kinematics workflows · Q2
In forward kinematics, the animator manually specifies each joint angle down the limb hierarchy; in inverse kinematics, the animator specifies the final end-effector position, and algorithms calculate joint angles automatically.
03
Assuming light rays naturally travel from the eye outward in physics · Q3
In the physical universe, light travels strictly from the light source to the eye; backward ray-tracing is a computational optimization that reverses this direction to evaluate only visible rays.

Solutions and tips are GeorgeTheTutor's own; the questions are not reproduced here. Not affiliated with or endorsed by the International Baccalaureate Organization.

Want someone to go through your paper with you?

Our maths mentors took these exams recently. One-to-one, they'll find where your marks are going and fix it before May.

Find your maths mentor More past papers