What do you think?
Rate this book
A Biography of the Pixel
The Great Digital Convergence of all media types into one universal digital medium occurred, with little fanfare, at the recent turn of the millennium. The bit became the universal medium, and the pixel conquered the world. Henceforward, nearly every picture in the world would be composed of pixels. In A Biography of the Pixel, Pixar cofounder Alvy Ray Smith argues that the pixel is the organizing principle of most modern media, and he presents a few simple but profound ideas that unify the dazzling varieties of digital image making. Smith's story of the pixel's development begins with Fourier waves, proceeds through Turing machines, and ends with the first digital movies from Pixar, DreamWorks, and Blue Sky. Today, almost all the pictures we encounter are digital. Smith explains, engagingly and accessibly, how pictures composed of invisible stuff become visible--that is, how digital pixels convert to analog display elements. Taking the special case of digital movies to represent all of Digital Light (his term for pictures constructed of pixels), and drawing on his decades of work in the field, Smith approaches his subject from multiple angles. A Biography of the Pixel is essential reading for anyone who has watched a video on a cell phone, played a videogame, or seen a movie.
558 pages, Kindle Edition
First published January 1, 2022
84 people are currently reading
747 people want to read
Ratings & Reviews
Friends & Following
Create a free account to discover what your friends think of this book!
Community Reviews
Displaying 1 - 30 of 31 reviews
April 2, 2022
An interesting and fascinating look at the element used to hold a digital sample of an image: the pixel. Often misunderstood to mean the picture elements you see on your screen, a pixel is actually an element that is a digital representation (sample) of an analogue image. The book goes through how an analogue signal (sound, images, moving images) is transformed into a digital sample and then used to recreate an analogue output, followed by the history of films and animated films, the creation of computers and the creation of digital images and finally the dream of the author and the others, to create The Movie, a fully computer generated film that would finally appear in the form of Pixar's "Toy Story".
Here's a chapter by chapter look at the book
- Fourier’s Frequencies: The Music of the World: An introduction to Fourier waves is given, along with a history of its creator, Joseph Fourier, including his interactions with Napoleon and other French scientists at the time. The author uses Fourier waves to show how to determine the frequencies that make up sound waves. Then Fourier waves are show to use to describe complex images.
- Kotelnikov’s Samples: Something from Nothing: a history of Vladimir Kotelnikov is given, asserted by the author to have invented the Sampling Theorem, along with interactions with various Soviet people like Stalin. The author shows that when the Sampling Theorem is used with Fourier waves, it can convert analogue values into discrete value samples. Then by using an appropriate 'shaping wave' the digital samples can be used to accurately recreate the analogue values. The author also gives the distinction between a pixel (samples of analogue values) and a picture element (used to display a pixel on a display). He also shows how the misuse of Sampling Theory leads to artefacts in audio and images.
- Turing’s Computations: Eleventy-Eleven Skydillion: this chapter is on Alan Turing and his idea of computing using a device that follows simple rules. This simple device is Turing's Universal computer. Based on Turing's idea, von Neumann would find a way for the universal computer to be created (a computer architecture). The author describes how to program a computer and shows that a computer does it work via the manipulation of symbols, not numbers.
- Dawn of Digital Light: The Quickening: a history of the early British and American computer efforts is covered here. As it turns out, an early use of CRTs (cathode ray tubes) is as a form of computer memory. This would lead to the CRT being used as a way to display images based on data in memory (for example, displaying a graph based on values in memory, rather than showing the values directly). Early forms of computer animation and possibly visual computer games are also shown in this chapter.
- Movies and Animation: Sampling Time: here, the author shows how the Fourier transform and sampling theorem should work for movies: this is based on the speed of movement, details of film projection and on our persistence of vision and perception of motion. A History of film projectors is given, along with the beginnings of film animation. The difference between live and animated film is given and show to be time: live films are shown in real time, while animations can compress or stretch time in films. The art of making animated characters come to live via exaggerations is also shown. Coming back to the Sampling Theorem, the author shows how current films speeds were developed before sampling theorem known, and that current films depend more on the perception of motion by us than accurate sampling to represent movement. This is a reason why sampling artefacts (like wheels spinning backwards) exists in films.
- Shapes of Things to Come: this chapter shows the beginnings of computer graphics, which are representation of lines and surfaces of pictures in computers. The 'spline' line and triangle are shown as the basic elements used to create lines and surfaces that connect discrete points. The difference between CAD and picture based graphics is shown. Next, a history of the first 'epoch' in computer graphics is given, with ideas generated to render smooth lines and surfaces between shapes, to display computer models in 2D and 3D (in perspective) and the ability of computers to interactively modify the computer model in real time. These would set the stage for computer animation.
- Shades of Meaning: Moore's law makes an appearance, leading to an explosion of the ability of computers to display better computer graphics and to animate graphics. 3D shapes are now being rendered, as lots of triangles. Ideas in how to hide 3D surfaces, adding a light source and brightness, doing texture mapping and shading appear. Ironically, the first applications of computer graphics to aid in the production of 2D animation would run into problems. Unlike 3D, 2D computer graphics had no concept of objects, making animation of items that hiding and reappear (like an arm moving back and forth and being hidden by a body) difficult for a computer to handle. At this time, initial work on rendering short 3D animations would also be done.
- The Millennium and The Movie: this chapter would become more personal as it involves the author's personal experience. The ambitions to make The Movie by him and others would being here. Moore's Law was creating computers capable of more sophisticated and better ways of creating images and animating them. Besides generating scenery images, the desire to create animated characters that look alive take root: this is Character animation. Interestingly, the ability of computers to add blurring to images would lead to the ability of computers to represent realistic movement in computer animation. But the wait now is for good enough computation, as determined by Moore's Law. In the meantime, there was the creation of various animation and graphics departments that would lead to various groups like Pixar and others. Eventually, this would lead to the production of The Movie: "Toy Story".
- Finale: The Great Digital Convergence - this closing chapter summarizes the earlier chapters of the book. It also looks to the future of computer graphics in Virtual, Augmented and Mixed Reality. Other aspects include the manipulation of feature of actors. A section also looks at the ability of AI to modify images in ways not known then.
Here's a chapter by chapter look at the book
- Fourier’s Frequencies: The Music of the World: An introduction to Fourier waves is given, along with a history of its creator, Joseph Fourier, including his interactions with Napoleon and other French scientists at the time. The author uses Fourier waves to show how to determine the frequencies that make up sound waves. Then Fourier waves are show to use to describe complex images.
- Kotelnikov’s Samples: Something from Nothing: a history of Vladimir Kotelnikov is given, asserted by the author to have invented the Sampling Theorem, along with interactions with various Soviet people like Stalin. The author shows that when the Sampling Theorem is used with Fourier waves, it can convert analogue values into discrete value samples. Then by using an appropriate 'shaping wave' the digital samples can be used to accurately recreate the analogue values. The author also gives the distinction between a pixel (samples of analogue values) and a picture element (used to display a pixel on a display). He also shows how the misuse of Sampling Theory leads to artefacts in audio and images.
- Turing’s Computations: Eleventy-Eleven Skydillion: this chapter is on Alan Turing and his idea of computing using a device that follows simple rules. This simple device is Turing's Universal computer. Based on Turing's idea, von Neumann would find a way for the universal computer to be created (a computer architecture). The author describes how to program a computer and shows that a computer does it work via the manipulation of symbols, not numbers.
- Dawn of Digital Light: The Quickening: a history of the early British and American computer efforts is covered here. As it turns out, an early use of CRTs (cathode ray tubes) is as a form of computer memory. This would lead to the CRT being used as a way to display images based on data in memory (for example, displaying a graph based on values in memory, rather than showing the values directly). Early forms of computer animation and possibly visual computer games are also shown in this chapter.
- Movies and Animation: Sampling Time: here, the author shows how the Fourier transform and sampling theorem should work for movies: this is based on the speed of movement, details of film projection and on our persistence of vision and perception of motion. A History of film projectors is given, along with the beginnings of film animation. The difference between live and animated film is given and show to be time: live films are shown in real time, while animations can compress or stretch time in films. The art of making animated characters come to live via exaggerations is also shown. Coming back to the Sampling Theorem, the author shows how current films speeds were developed before sampling theorem known, and that current films depend more on the perception of motion by us than accurate sampling to represent movement. This is a reason why sampling artefacts (like wheels spinning backwards) exists in films.
- Shapes of Things to Come: this chapter shows the beginnings of computer graphics, which are representation of lines and surfaces of pictures in computers. The 'spline' line and triangle are shown as the basic elements used to create lines and surfaces that connect discrete points. The difference between CAD and picture based graphics is shown. Next, a history of the first 'epoch' in computer graphics is given, with ideas generated to render smooth lines and surfaces between shapes, to display computer models in 2D and 3D (in perspective) and the ability of computers to interactively modify the computer model in real time. These would set the stage for computer animation.
- Shades of Meaning: Moore's law makes an appearance, leading to an explosion of the ability of computers to display better computer graphics and to animate graphics. 3D shapes are now being rendered, as lots of triangles. Ideas in how to hide 3D surfaces, adding a light source and brightness, doing texture mapping and shading appear. Ironically, the first applications of computer graphics to aid in the production of 2D animation would run into problems. Unlike 3D, 2D computer graphics had no concept of objects, making animation of items that hiding and reappear (like an arm moving back and forth and being hidden by a body) difficult for a computer to handle. At this time, initial work on rendering short 3D animations would also be done.
- The Millennium and The Movie: this chapter would become more personal as it involves the author's personal experience. The ambitions to make The Movie by him and others would being here. Moore's Law was creating computers capable of more sophisticated and better ways of creating images and animating them. Besides generating scenery images, the desire to create animated characters that look alive take root: this is Character animation. Interestingly, the ability of computers to add blurring to images would lead to the ability of computers to represent realistic movement in computer animation. But the wait now is for good enough computation, as determined by Moore's Law. In the meantime, there was the creation of various animation and graphics departments that would lead to various groups like Pixar and others. Eventually, this would lead to the production of The Movie: "Toy Story".
- Finale: The Great Digital Convergence - this closing chapter summarizes the earlier chapters of the book. It also looks to the future of computer graphics in Virtual, Augmented and Mixed Reality. Other aspects include the manipulation of feature of actors. A section also looks at the ability of AI to modify images in ways not known then.
September 30, 2021
DNF. Life's too short for another book about all the men in computing.
October 6, 2021
Not enough of a story structure for a good narrative and not enough technical details for a useful textbook.
October 25, 2022
One of the best books I have ever read. This book is both for movie nerds or computer science nerds.
The books does an extensive account on the technical and artistic developments that ultimately led to the creation of animated movies and the founding of Pixar, but it is much much more.
One negative aspect of the book is that the account does not shed light on the contributions of women in the field, which is a shame and reflects the gender gaps in tech.
However, it does a brilliant job explaining mathematical concepts such a Fourier’s transformations, kotelnikov’s theorem, the impact of Moore’s law, and my favorite mind blowing concept and idea of the book: a pixel is not a square box in a screen, but a discrete sample representation of a point in space, a point can’t be seen, we can’t see a pixel.
Deeply recommended, this is a long, dense, hard read, but and enjoyable one that has made me look at the world of digital light in a whole new way.
The books does an extensive account on the technical and artistic developments that ultimately led to the creation of animated movies and the founding of Pixar, but it is much much more.
One negative aspect of the book is that the account does not shed light on the contributions of women in the field, which is a shame and reflects the gender gaps in tech.
However, it does a brilliant job explaining mathematical concepts such a Fourier’s transformations, kotelnikov’s theorem, the impact of Moore’s law, and my favorite mind blowing concept and idea of the book: a pixel is not a square box in a screen, but a discrete sample representation of a point in space, a point can’t be seen, we can’t see a pixel.
Deeply recommended, this is a long, dense, hard read, but and enjoyable one that has made me look at the world of digital light in a whole new way.
February 3, 2023
I liked this book, but Smith goes into way too much nitpicky details about things. In some ways, that's a good thing, but this is one of those books where it's hard to tell who the audience for it is. He randomly picks a couple of things and goes into a ton of detail about them (e.g. Fourier transforms, digital sampling), but he doesn't go into enough detail about them that you'd be able to put them into practice usefully. For someone who already knows how it all works, there's some interesting historical tidbits about independent discoveries and Stigler's Law of eponymy (no scientific discovery is named after its original discoverer), but it's mostly a sort of elementary retread of well-known ideas. For someone who doesn't already know these things, I think one would be overwhelmed with the details and lose sight of the bigger picture.
Smith spends a ton of time on Fourier transforms and spacial Fourier transforms, and how graphics people think of straight lines as having high frequencies, but I don't feel like I needed that information at all to understand the later parts of the book. There are parts of the book where it seemed useful to understand Nyquist sampling in time, but I kept waiting for the spacial frequency stuff to pay off and it never did.
It seems to me like this would be a great book to have if society had collapsed and you were trying to re-invent digital image processing, since it covers a whole huge history with enough detail to let a determined group of people re-invent such a thing themselves, but that's a pretty weird audience to shoot for.
My impression from the book is that Smith wanted to write a technical history of digital graphics and he's kind of a nerdy guy, so he couldn't help but go into way too much detail, since he thought, "Well, this stuff is integral to how I think about graphics stuff, so how would anyone else understand it without these details?" It's an understandable impulse, but I think a professional writer rather than a retired geek writing up a technical history of his own field would know not to write the book this way.
Still, there's definitely some interesting stuff in here, and I can only aspire to be the sort of dude who spends his retirement writing stuff like this. ☺
3.5 of 5 stars
Smith spends a ton of time on Fourier transforms and spacial Fourier transforms, and how graphics people think of straight lines as having high frequencies, but I don't feel like I needed that information at all to understand the later parts of the book. There are parts of the book where it seemed useful to understand Nyquist sampling in time, but I kept waiting for the spacial frequency stuff to pay off and it never did.
It seems to me like this would be a great book to have if society had collapsed and you were trying to re-invent digital image processing, since it covers a whole huge history with enough detail to let a determined group of people re-invent such a thing themselves, but that's a pretty weird audience to shoot for.
My impression from the book is that Smith wanted to write a technical history of digital graphics and he's kind of a nerdy guy, so he couldn't help but go into way too much detail, since he thought, "Well, this stuff is integral to how I think about graphics stuff, so how would anyone else understand it without these details?" It's an understandable impulse, but I think a professional writer rather than a retired geek writing up a technical history of his own field would know not to write the book this way.
Still, there's definitely some interesting stuff in here, and I can only aspire to be the sort of dude who spends his retirement writing stuff like this. ☺
3.5 of 5 stars
March 29, 2023
I have very mixed feelings about the book. On the one side I enjoyed the stories about the three founding concepts and their creators - Fourier, Kotelnikov and Turing. Also the partial history of how the field of computer graphics emerged I found appealing.
What I didn't like so much, that for the explanation of certain concepts the author has used made up terms, which for me personally, were more confusing than helpful. Also occasionally I found the story to jump too much from one person or concept to the next without fully describing its role.
Still in general, I'm happy that I picked up the book and was able to finish it learning plenty of historical background about computing and "picture-oriented" computer graphics!
What I didn't like so much, that for the explanation of certain concepts the author has used made up terms, which for me personally, were more confusing than helpful. Also occasionally I found the story to jump too much from one person or concept to the next without fully describing its role.
Still in general, I'm happy that I picked up the book and was able to finish it learning plenty of historical background about computing and "picture-oriented" computer graphics!
This entire review has been hidden because of spoilers.
June 6, 2023
Este es un gran libro que, en su mayoría sirve como un crash course de historia de la computación.
December 20, 2021
First, a caveat. At least one of those stars is due to pure nostalgia. I was on the edges of this community from the mid-70s to the early 90s. Some schooling with Csuri's OSU group, master's degree in the field, worked with Don Greenburg (who does indeed have the warmest grin in computer graphics) and his Cornell group for a while , even met the author a time or two. So this was a wonderful trip down memory lane, reminding me of all that and of some of friends who have slipped thru time's sieve. So thanks for that, Alvy. And thanks for reminding me how to compute perspective! (how could I have ever forgotten?)
That said, this is a book chock full of information for those interested in how computer generated movies are made. The history of techniques is extensive, sometimes a little deep, often not deep enough, but a good general overview. You won't be able to compute your own movies from this info but you might just find yourself eager to try. Or to do any of the marvelous things Digital Light can do. (besides, we live the Age of Lookin' Stuff Up so you can find technical details online)
It is not, as I was led to believe from a single review, a philosophical view into "Digital Light" and all it can do. The author touches on that here or there, gives some thoughts, but it's not the Gödel, Escher, Bach: An Eternal Golden BraidGödel, Escher, Bach: An Eternal Golden Braid of Digital Light. Sigh.
It also follows a very narrow path thru the history of Computer Graphics. Alvy's own personal journey, populated mostly with the people he worked closely with and ultimately with those who formed Pixar in its early days. There's nothing wrong with that, of course. It's an exciting journey. What it actually emphasizes, to me, is just how broad and vast the history and field of Computer Graphics really is. You won't find much of computer art here. Some women but not many. Radiosity isn't even mentioned. So many researchers and developers and practitioners don't get a single nod. But they probably don't belong in this book. I mention it just so you won't be disappointed reading it if you don't find exactly what you're looking for. Or don't find yourself! (another caveat: I didn't take much of a look at the online annotations which appear to be quite extensive)
But it's well worth the read. Besides the nostalgia I found much here that captured my interest. Especially, of course, the chapters covering what I did not know. I found it to be thoughtful, well written, generous with acknowledging who did what, and an exciting read. No doubt you will too.
Btw, Alvy Ray Smith does not like Steve Jobs.
That said, this is a book chock full of information for those interested in how computer generated movies are made. The history of techniques is extensive, sometimes a little deep, often not deep enough, but a good general overview. You won't be able to compute your own movies from this info but you might just find yourself eager to try. Or to do any of the marvelous things Digital Light can do. (besides, we live the Age of Lookin' Stuff Up so you can find technical details online)
It is not, as I was led to believe from a single review, a philosophical view into "Digital Light" and all it can do. The author touches on that here or there, gives some thoughts, but it's not the Gödel, Escher, Bach: An Eternal Golden BraidGödel, Escher, Bach: An Eternal Golden Braid of Digital Light. Sigh.
It also follows a very narrow path thru the history of Computer Graphics. Alvy's own personal journey, populated mostly with the people he worked closely with and ultimately with those who formed Pixar in its early days. There's nothing wrong with that, of course. It's an exciting journey. What it actually emphasizes, to me, is just how broad and vast the history and field of Computer Graphics really is. You won't find much of computer art here. Some women but not many. Radiosity isn't even mentioned. So many researchers and developers and practitioners don't get a single nod. But they probably don't belong in this book. I mention it just so you won't be disappointed reading it if you don't find exactly what you're looking for. Or don't find yourself! (another caveat: I didn't take much of a look at the online annotations which appear to be quite extensive)
But it's well worth the read. Besides the nostalgia I found much here that captured my interest. Especially, of course, the chapters covering what I did not know. I found it to be thoughtful, well written, generous with acknowledging who did what, and an exciting read. No doubt you will too.
Btw, Alvy Ray Smith does not like Steve Jobs.
May 8, 2022
I was very conflicted about my feelings for this book, oscillating back and forth between finding it really annoying / frustrating and really enjoying it. Overall though, the good parts far outweighed the frustrating ones, and I'm quite glad I read it.
With any popular book covering a technical subject, things have to be simplified to appeal to a general audience and be more approachable, and I totally get that, but in this case this process went way too far. Many of the technical sections and explanations, which constitute a significant portion of the book, generally felt simplified to the point of losing a lot of content, and occasionally being just wrong. Technical terms that would have been used throughout the book and become familiar were instead switched for easier to understand terms, despite the fact that this makes it harder to integrate these concepts into larger knowledge and understanding. This is always a tough balance to strike, but here it really felt like they were expecting way to little from the reader, and it was disappointing. Also, the Audiobook has no accompanying PDF of figures, which is a huge oversight for a book so heavily reliant on them. Also, for audiobook listeners, I'd particularly recommend trying it at 1.3x speed or more.
All of that said, this book tells an amazing story about the history of computer graphics from someone who was deeply involved from the very beginning all the way up through founding pixar, and this story is deeply interesting and well written. Despite the oversimplification of many of the technical areas, I still learned a lot about how certain kinds of computer graphics are done, and there was a lot of interesting content about correctly sampling and reconstructing the visual world in space and time. Just being able to see all of this history of the field that underlies so much of our modern lives from someone who was so deeply a part of it for so long made the book well worth it on its own.
With any popular book covering a technical subject, things have to be simplified to appeal to a general audience and be more approachable, and I totally get that, but in this case this process went way too far. Many of the technical sections and explanations, which constitute a significant portion of the book, generally felt simplified to the point of losing a lot of content, and occasionally being just wrong. Technical terms that would have been used throughout the book and become familiar were instead switched for easier to understand terms, despite the fact that this makes it harder to integrate these concepts into larger knowledge and understanding. This is always a tough balance to strike, but here it really felt like they were expecting way to little from the reader, and it was disappointing. Also, the Audiobook has no accompanying PDF of figures, which is a huge oversight for a book so heavily reliant on them. Also, for audiobook listeners, I'd particularly recommend trying it at 1.3x speed or more.
All of that said, this book tells an amazing story about the history of computer graphics from someone who was deeply involved from the very beginning all the way up through founding pixar, and this story is deeply interesting and well written. Despite the oversimplification of many of the technical areas, I still learned a lot about how certain kinds of computer graphics are done, and there was a lot of interesting content about correctly sampling and reconstructing the visual world in space and time. Just being able to see all of this history of the field that underlies so much of our modern lives from someone who was so deeply a part of it for so long made the book well worth it on its own.
December 24, 2024
The younger generation today takes the transition from traditional to digital media for granted. We grew up on computers with GUI, high-resolution films with CGI, and digital art and music. We never fully appreciated the advent of computer graphics that made this all possible today.
Having done a computer graphics project in high school, I was already familiar with many of the technical developments that Smith chronicled. I was impressed, however, that he went back to the real origins of computer graphics -- Fourier analysis, the Sampling Theorem, and early film-making. It was very satisfying to see how math, optics, graphical art, and computer science came together. I also liked the author's definition of pixel and enjoyed seeing it generalize well across eras to different mediums.
My biggest takeaway from the second half of the book is that Moore's Law is just magic. Smith and other computer graphics pioneer saw early prototypes of beautiful images rendered on computers, starting as early as the 1960s. At NYIT, even though the first movie was short, jagged, and didn't work well, it proved that computers could help produce movies. Catmull, Smith, and many others set the ambitious goal that one day, they would produce the first movie fully generated by computers. They believed in Moore's Law and waited decades for hardware to catch up. In the meantime, the graphics community invented techniques that refined the quality of rendering and animation.
All of their efforts culminated in the release of Toy Story in 1995, which propelled Pixar to the biggest IPO of the year, surpassing Netscape! The convergence was just so beautiful, but it also got me to think: what is the Moore's Law of our age? Is it scaling laws? What will be possible in software in 5-10 years, given the advances in hardware? On the other hand, even though Moore's Law is slowing down, computer engineers are coming up with new ways to improve performance, mostly by parallelization and specialization. The same thing recently happened in LLMs. When scaling pre-training has plateaued, we find ways to scale inference. Scientists and engineers always find ways to beat the limitations imposed upon them. In computer graphics, they beat the constraints of physical media and freed the human imagination. I'm excited for what we come up with in the future.
Having done a computer graphics project in high school, I was already familiar with many of the technical developments that Smith chronicled. I was impressed, however, that he went back to the real origins of computer graphics -- Fourier analysis, the Sampling Theorem, and early film-making. It was very satisfying to see how math, optics, graphical art, and computer science came together. I also liked the author's definition of pixel and enjoyed seeing it generalize well across eras to different mediums.
My biggest takeaway from the second half of the book is that Moore's Law is just magic. Smith and other computer graphics pioneer saw early prototypes of beautiful images rendered on computers, starting as early as the 1960s. At NYIT, even though the first movie was short, jagged, and didn't work well, it proved that computers could help produce movies. Catmull, Smith, and many others set the ambitious goal that one day, they would produce the first movie fully generated by computers. They believed in Moore's Law and waited decades for hardware to catch up. In the meantime, the graphics community invented techniques that refined the quality of rendering and animation.
All of their efforts culminated in the release of Toy Story in 1995, which propelled Pixar to the biggest IPO of the year, surpassing Netscape! The convergence was just so beautiful, but it also got me to think: what is the Moore's Law of our age? Is it scaling laws? What will be possible in software in 5-10 years, given the advances in hardware? On the other hand, even though Moore's Law is slowing down, computer engineers are coming up with new ways to improve performance, mostly by parallelization and specialization. The same thing recently happened in LLMs. When scaling pre-training has plateaued, we find ways to scale inference. Scientists and engineers always find ways to beat the limitations imposed upon them. In computer graphics, they beat the constraints of physical media and freed the human imagination. I'm excited for what we come up with in the future.
August 20, 2025
"История пикселя" - на самом деле три книги в одной. Сначала - попытка объяснить ряды Фурье и теорему Котельникова без формул. Затем - история компьютерной техники с небольшим уклоном в генерацию изображений. Примерно до шестидесятых. История останавливается на первых играх и ASCII-картинках, и рассказ возвращается на 200 лет назад. Идёт история кино с самого начала, а после того, как книга переваливает за половину, начинается уже история именно компьютерной графики. Элви Рэй Смит разбирается в вопросе, потому что он - один из основателей Pixar. Но традиционное для современного научпопа желание избежать формул любой ценой делает некоторые объяснения длинными, путаными и неповоротливыми. Можно было хотя бы простую школьную математику оставить. Книга большая. Однако лишнего там нет. Ни одной буквы. Более того, многого там не хватает. Нужен ещё один том - с математикой и с объяснениями, которые автор оставил у себя на сайте, но в книгу не включил. Если верить книге, центральная догма компьютерной графики (воображаемый мир внутри компьютера описывается с помощью трёхмерной евклидовой геометрии и ньютоновской физики; за ним наблюдает виртуальная камера, которая создаёт проекцию своей точки обзора при помощи линейной перспективы для отображения на экране) сформировалась уже к началу шестидесятых, когда для того, чтобы поиграть на компе, приходилось использовать суперсекретную технику за неописуемые миллионы долларов. История цифрового кино начинается с пятой (с конца) главы. Там уже не так интересно, просто подборка фактов из серии кто, что и когда сделал. И автор честно признаётся, что он не знает почти ничего о конкурентах, вроде DreamWorks. История заканчивается выходом первой "Истории игрушек". Последняя глава - краткое содержание книги и авторский взгляд на процесс изобретательства (концепция "идея-хаос-тиран": для того, чтобы идея пробилась, нужен лидер, желающий получить выгоду путём использования этой идеи в условиях неопределённости). Книга цельная. Все байки и факты плотно вписаны в общее повествование. Для описания связей между изобретениями и научными работами созданы схемы. Центральная часть - вообще шедевр.
September 25, 2023
I thought the start of the book was ingenious, tying a modern-day technology to important but lesser-known discoveries in 18th century France and early 20th century Russia.
The histories of the computer and film were also interesting, and mostly new to me.
I got lost a little when the book got into the specifics of computer graphics, like splines and textures. The author made an admirable effort to explain these things without using math or algorithms, but I couldn't grasp some ideas based on the small diagrams in the book.
The listing of various people who contributed to the field in the 60s and 70s was too detailed for my tastes, including a complicated flow chart. That section of the book would have benefited from more anecdotes about Pixar and the creation of Toy Story, but the author left the company early in the development of that movie.
Overall, it's a quirky but informative book that would appeal more to people more interested in technology history than the current state of the art.
The histories of the computer and film were also interesting, and mostly new to me.
I got lost a little when the book got into the specifics of computer graphics, like splines and textures. The author made an admirable effort to explain these things without using math or algorithms, but I couldn't grasp some ideas based on the small diagrams in the book.
The listing of various people who contributed to the field in the 60s and 70s was too detailed for my tastes, including a complicated flow chart. That section of the book would have benefited from more anecdotes about Pixar and the creation of Toy Story, but the author left the company early in the development of that movie.
Overall, it's a quirky but informative book that would appeal more to people more interested in technology history than the current state of the art.
January 19, 2022
If you're over 25 or so years of age you've lived through a major historical event and probably not noticed. Alvy Ray Smith identifies and explains the Great Digital Convergence that happened around the turn of the century, and explains both the nature and importance of the pixel. I was fortunate to have worked with Alvy and many of the pioneers he covers in this book, and went on to work for decades at Pixar -- and yet I didn't know what a pixel actually was until reading this book.
I'm not named in the book, nor did I expect to be, but it was strange reading about a part of my life in a history book.
Alvy does a terrific job of untangling the history of early computers and computer graphics, and explains things really clearly. If you have even a passing interest you'll want to dive into this book.
I'm not named in the book, nor did I expect to be, but it was strange reading about a part of my life in a history book.
Alvy does a terrific job of untangling the history of early computers and computer graphics, and explains things really clearly. If you have even a passing interest you'll want to dive into this book.
February 23, 2022
For me, this is two books bound under one cover. Book one is a history and survey of the pixel in what I call computing for visual output - computer graphics and signal/image processing. Having had a foot in both camps, I enjoy Alvy’ approach in telling that story. As such, this is an excellent book for those interested in the background of those phenomena which have become pervasive. The second book is a memoir from someone who witnessed many of the places where computer graphics germinated. Here he visits old friends - and some frenemies - in a familiar and chatty manner. For me, this book could only be improved by it being offered as an audio book read by Alvy, as I can hear his voice on every page.
May 30, 2022
This is a wonderfully thorough account of how pixels and, more specifically, digital movies came to be. Ray Smith does an admirable job intertwining the biographies of technological pioneers with the “genesis stories” of the technological innovations themselves. The book is quite long but even so, he keeps it focused on computer displays for movies rather than splintering off into dozens of other related fields.
This should become part of the syllabus for computer graphics and digital media programs. It is a much more engaging read than the average textbook, while also providing just enough background to warrant supplemental learning. I, for one, wish I’d had this book as an undergrad in the early 2000s.
This should become part of the syllabus for computer graphics and digital media programs. It is a much more engaging read than the average textbook, while also providing just enough background to warrant supplemental learning. I, for one, wish I’d had this book as an undergrad in the early 2000s.
July 11, 2022
Excellent history of the pixel
I’m not sure what I was expecting, but this was much better than whatever I thought it was. I loved the detailed history of computers and computer graphics — I studied cg with Jim Blinn and Jim Kajiya (both featured in the book) at Caltech, and was planning to go into it as a career until practical necessities intervened. Still, there was a lot I didn’t know, and it was wonderfully nostalgic to read about the things I did know. I’m encouraged by Smith’s comments at the end about AI, since that’s what I’m doing now, and I can only hope it lives up to its promise.
I’m not sure what I was expecting, but this was much better than whatever I thought it was. I loved the detailed history of computers and computer graphics — I studied cg with Jim Blinn and Jim Kajiya (both featured in the book) at Caltech, and was planning to go into it as a career until practical necessities intervened. Still, there was a lot I didn’t know, and it was wonderfully nostalgic to read about the things I did know. I’m encouraged by Smith’s comments at the end about AI, since that’s what I’m doing now, and I can only hope it lives up to its promise.
December 23, 2022
An odd and sprawling book – part memoir, part explanation of digital computing, and part history of computer animation. It could have benefited from a tighter focus, but on the other hand it has an idiosyncratic and personal charm to it. It’s clearly a labor of love and is refreshingly different from the standard nonfiction template. Plus, the history is fascinating and is rarely told, and the technical explanations for the non-specialist reader, though rather long, are good.
October 25, 2021
I was expecting a more narrative style book when I picked it up and found that it went through a whole history of computation and the build up of the movies. The last two chapters get into the more saucy details of Pixar, Steve Jobs, etc. Otherwise it reads more like a technical history book. It helped fill a lot of details in the history of computation and computer graphics for me.
Read
December 20, 2023Did not finish - I couldn't get through this book. It was pretty dry and the author's attempts to humanize the critical theories and engineering underlying the pixel with historical biography fell flat. Jumping back and forth between history and science felt more jarring than deftly woven and like filler. Ironically, I actually the parts where he explained the technical theory more.
April 7, 2023
So dense with so much interesting information and the history of computing from the point of view of a pixel. I enjoyed it, learnt a lot, but I feel there are better books out there. If you're in the field of graphics, is a brilliant read. I'm not.
July 2, 2025
Im not a huge computer nerd. I dont understand most of the lingo either. But the only part that stuck out to me was the shape of a pixel; I thought that was interesting. I do like that I now know how he co-founded Pixar. There's some drama with Disney in there, but that's okay.
October 28, 2025
A very, may too thorough history of visual digital arts. I like some of the tedious details, but I'm guessing many others would prefer not to go so deep. Some of it seemed like it might have been a bit self-serving.
Read
December 25, 2021Of interest to anyone curious about visualization
December 31, 2021
Some interesting background on how 'digital light's actually works, along with a lot of history on various computer guys who made it happen step by step.
July 18, 2022
Inspiring book that took me way too long to read because every other sentence made me wonder.
January 16, 2023
I enjoyed this history, perhaps because of my interest in computer graphics and animation in general.
April 14, 2024
way too pedantic.
December 8, 2024
Interesting read with many new insights. A bit lengthy.
August 2, 2025
Fun read! I like how it takes you through the history of graphics, even behind the math that made them possible although that made it a but slower at times, however it's a very insightful read!
January 2, 2023
This very long book would appeal to a very narrow set of specialists (DSP, AI, Imaging, Audio, and Video engineers and scientists). I happen to be one of them. The book explores the origins, evolution and applications of the pixel, the basic unit of digital representation of an analogue image. The book combines historical anecdotes, scientific explanations, personal reflections and artistic insights to tell the story of how the pixel emerged from the interplay of mathematics, physics, engineering, computing, cinema and animation. The book is not only a tribute to the pixel, but also to the visionaries, pioneers and innovators who contributed to its development and use.
The book begins with an introduction to the Fourier Transform, a mathematical tool that allows any complex waveform, such as sound or light, to be decomposed into a sum of simpler waves of different frequencies. The author explains how the Fourier transform enables the analysis and synthesis of signals, and how it relates to the pixel as a sample of an image wave. I also learned a lot I did not know about the life of Joseph Fourier. Next up was the sampling theorem, the conditions under which a continuous signal can be perfectly reconstructed from a discrete set of samples, such as pixels. I discovered that it was Vladimir Kotelnikov who created the sampling theorem and not Nyquist. The author then follows up with computation, the foundations of computer architecture and the computer architects themselves. These early computers led to experiments with cathode ray tubes and oscilloscopes, and on to the development of computer graphics and animation.
I really liked the concept of tying together Fourier Transforms, the Kotelnikov sampling theorem, Turing machines, computer history, and Moore's law to show how they combined together to lead to pixels and digital imagining/processing. Some of the history and job shuffling of the more recent graphics pioneers and programmers may have gone a bit detailed/long however and those later chapters had me wishing the author would get to point more quickly.
The book begins with an introduction to the Fourier Transform, a mathematical tool that allows any complex waveform, such as sound or light, to be decomposed into a sum of simpler waves of different frequencies. The author explains how the Fourier transform enables the analysis and synthesis of signals, and how it relates to the pixel as a sample of an image wave. I also learned a lot I did not know about the life of Joseph Fourier. Next up was the sampling theorem, the conditions under which a continuous signal can be perfectly reconstructed from a discrete set of samples, such as pixels. I discovered that it was Vladimir Kotelnikov who created the sampling theorem and not Nyquist. The author then follows up with computation, the foundations of computer architecture and the computer architects themselves. These early computers led to experiments with cathode ray tubes and oscilloscopes, and on to the development of computer graphics and animation.
I really liked the concept of tying together Fourier Transforms, the Kotelnikov sampling theorem, Turing machines, computer history, and Moore's law to show how they combined together to lead to pixels and digital imagining/processing. Some of the history and job shuffling of the more recent graphics pioneers and programmers may have gone a bit detailed/long however and those later chapters had me wishing the author would get to point more quickly.
Displaying 1 - 30 of 31 reviews