Lecture 01: Welcome to Computer graphics. What is the rasterization

Who I am

Ivan Belyavtsev

Computer Graphics Developer technology engineer at Nvidia

Researcher and educator in Computer science

djbelyak.ru

Disclaimer

This course was prepared or accomplished by Ivan Belyavtsev in his personal capacity. The opinions expressed in this course are the author’s own and do not represent official policy or position of Nvidia corporation.

Schedule

1. 10/23-10/24 - 5 classes
2. 11/06-11/07 - 5 classes
3. 11/20-11/21 - 5 classes
4. 12/04-12/05 - 5 classes + Final assessment

Syllabus

1. 10/23-10/24. Basics of rendering
2. 11/06-11/07. Basics of ray tracing
3. 11/20-11/21. Basics of DirectX 12
4. 12/04-12/05. Advanced technics and algorithms

Caution

This course requires a lot of discipline and energy!

Expectations vs reality

How we will work

• During classes
  • Research computer graphics with a meta-science framework
• At home
  • Repeat lectures materials
  • Implement experiments by yourself

What we won’t cover

• 3D modeling as art
• Animation
• Game engines
• Music and sounds
• Game design
• Network

Hardware requirements

• PC with Windows 10
• Discrete GPU (For DirectX labs)

Software requirements

• Git
• Visual Studio Community
• Premake5
• github.com/djbelyak/basic-of-rasterization

Materials

• Moodle page: https://moodle.innopolis.university/course/view.php?id=626
• My webpage: djbelyak.ru
• There is no spoilers of code
• Reference images are located in the repo: github.com/djbelyak/basic-of-rasterization

Communication channels

• Moodle page: https://moodle.innopolis.university/course/view.php?id=626
• Course channel in Telegram: https://t.me/joinchat/AAAAAFRoEPOgWZAG5tWJTw
• Course chat in Telegram: https://t.me/joinchat/AAAAAFCYalbYOsLlHW-lXg

Questions

How we will research computer graphics

Viktor Kanke

Types of sciences

• Social
• Natural
• Axiological [1]

What the value of a science?

• Economics
• Physics
• Computer science

Theories row

Knowledge means the theories understood and accepted by people and nothing else [2]

\[T_1 -> T_2 -> ... -> T_n\]

Concepts

• Variables:

  \(\vec{F}\), \(\vec{a}\), \(m\)

• Laws:

  \(\vec{a} = \frac{\vec{F}}{m}\)

• Principals:

  Conservation of energy [2]

Methods of concepts managing

• Deduction
• Experiment
• Induction (data processing)
• Correction of original concepts [2]

Methods of theories managing

• Problematization - revealing certain obstructions in the old concept
• Innovation - invention of new theory
• Interpretation - old theory through new concept [2]

Interoperation of sciences

Donor-acceptor relationships [1]

An acceptor science could you use theories and methods from a donor science

Aim and principles of applied sciences

• Aim to make something the most effective way
• Efficiency maximization principle [3]

Aim and principles of computer science

• Is computer science an applied science? [1]
• Aim to process, store and transfer information by the most effective way [4]

Aim and principles of computer graphics

• Is computer graphics a computer science?
• Aim to create, process, store and transfer visual information by the most effective way

Aim and principles computer graphics in game development

• Aim to create and process an image by the most effective way
• What’s mean “the most effective way” here?

Let’s define what is an image

Physics and anatomy are donor sciences

What is the light

[5]

Human eye

• Rods - light quantity
• Cones - light quality
  • S-cones - blue
  • M-cones - green
  • L-cones - red

[6]

Color perception

[5]

Pixels in depth

[5]

Pixels in depth (iPad 2 and iPad 3)

[7]

sRGB gamut

[8]

What is a rasteriztion

It’s possible to construct any image using the next operation:

• draw a point
• draw a line
• draw a triangle

“Clearing of a render target” experiment

Deduction phase

• Image is a matrix of pixels
• Pixel is 3 colored light sources
• Does the intensity for each light sources setup the color?
• Does black color consist of 0 intensity pixels?
• How long does an operation take?

“Clearing of a render target” experiment

Experiment

Let’s implement it together

“Clearing of a render target” experiment

Reference

“Clearing of a render target” experiment

What is the new knowledge?

• Brightness for RGB light sources setups the pixel color
• Black color consists of 0 intensity pixels
• Clearing of the render target is not free