Introduction to GPU
World’s first computer graphics card
Graphics cards are considered as old as personal computer, starting with IBM’s Monochrome Display Adapter in 1981. Nvidia entered the market in 1993 and is considered the first professional display chip vendor.
The invention of GPU
Nvidia invented the world’s first GPU (graphics processing unit) in 1999, the Geforce 256, which was specially used for computer display processing. The display of the computer screen and the processing of display data are all processed by the CPU. This is a major event that changes the computer world and computing. In the same year, Nvidia was listed on the US stock market.
Differences between GPU and CPU
Since the invention of the GPU is for the display of computer screens, the computer screen is very large and all images on the computer screen must be displayed at the same time in a very short time. Therefore, the strength of the GPU is that it can process a large amount of data in parallel at the same time.
In contrast, the tasks assigned to traditional CPUs focus on processing complex but required accurate data.
Because of this fundamental difference:
- The GPU core is not as powerful as the CPU core and has less memory.
- Although the CPU can “quickly switch” between different instruction sets, the GPU only obtains a large number of the same instructions, but requires high-speed processing. Therefore, GPU capabilities play an important role in “parallel computing.”
Types of GPU
GPUs currently on the market can be mainly divided into the following categories:
CPU built-in GPU
Most low-end and mid-range CPUs launched by Intel and AMD have built-in GPU functions, which are sufficient to handle most display functions, video playback, and general low-end or entry-level computer games.
Because the GPU function built into the computer CPU shares the system memory with the computer CPU. Traditionally, the peripheral I/O processor responsible for memory and display functions in computers is located on the north bridge. This CPU-built-in display method is called “Unified Memory Access (UMA)”, which means that the built-in motherboard GPU uses the same main memory as the CPU.
Discrete display chip
In order to cope with professional graphics displays and games on mid-to-high-end personal computers, Nvidia, AMD (that is, ATI acquired by AMD in 2006), and Intel have all launched independent CPUs for these two fields. Another GPU. This type of GPU is characterized by being very expensive but extremely efficient and belongs to a highly profitable market.
Other devices
For example, the screens of smartphones, various mobile devices, and general small game consoles all need display functions. In particular, power saving and chip miniaturization need to be taken into consideration, so this field is the same as computers. There will be display functions built into the CPU, and there will also be GPUs that specifically handle display functions independently.
Top GPU vendors
Personal computer
Nvidia, AMD, and Intel currently have a global market share of 80%, 20%, and less than 1% in independent display chips respectively.
GPUs used in supercomputers and artificial intelligence are all discrete GPUs.
Mobile device
ARM’s Mali
In terms of mobile devices, most of them use ARM’s CPU, so most of them will directly use ARM’s GPU. ARM’s GPU is called Mali.
Qualcomm’s Adreno
In addition, the Qualcomm platform is very commonly used on mobile phones. Most Qualcomm platform customers will directly adopt the Adreno display GPU solution developed by Qualcomm.
Imagination
PowerVR is the software and hardware image processing development arm of Imagination Technologies (UK ticker: IMG), and is also the common name for the graphics chips it designs.
Before Apple developed and produced its own CPU with built-in display capabilities, Apple’s mobile phones used to use Imagination’s display solution. Because Imagination lost Apple, a super-large customer, Imagination encountered financial difficulties and was acquired by a Chinese company.
As the RISK-V platform has become increasingly popular in recent years, Imagination has also joined hands with Andes Technology to certify GPUs with RISC-V CPU IP to enter this new market.
New uses for GPUs
Cryptocurrency mining
Mining cryptocurrencies such as Bitcoin using GPUs is much more efficient than using CPUs due to the higher hashing rates, which is the standard method of mining Bitcoin today. The hash rate and processing speed of modern top-end GPUs are incomparable to the processing power of CPUs, which is why GPUs are commonly used for mining.
GPUs are used in supercomputers
Due to the natural architecture design of GPU, it is suitable for “parallel computing” of large amounts of data. Therefore, Nvidia’s GPU is gradually used in supercomputers that require parallel computing and large amounts of data processing in a short period of time, so that supercomputers no longer rely solely on the CPU for all processing. Such evolution in data processing has significantly improved the speed and performance of supercomputers.
GPUs are used for artificial intelligence
In 2006, Hinton and his student Simon Osindero published an important paper “A Fast Learning Algorithm for Deep Belief Nets” recommends using GPU to increase the speed of training neural networks.
The publication of this paper made many people realize the possibility of using GPUs to break through computing power bottlenecks. As a result, neural network research finally came back to life after being dormant for many years.
Related articles
- “Top vendors and uses of GPU“
- “ASIC market is getting bigger, and related listed companies in the US and Taiwan“
- “How does nVidia make money, Nvidia is changing the gaming rules“
- “The reasons for Nvidia’s monopoly and the challenges it faces“
- “Why nVidia failed to acquire ARM?“
- “Revisiting Nvidia: The Absolute Leader in Artificial Intelligence, Data Center, and Graphics“
- “Will Intel go bankrupt?“
- “How does Intel make money? and the benefits to invest in it“
- “Intel’s current difficult dilemma“
- “How AMD makes money? A rare case of turning defeat into victory“
- “Why is AMD’s performance so jaw-dropping?“
- “Qualcomm diversifies success, no nonger highly dependend on phone“
- “How does the ubiquitous Arm make money?“
- “Arm relisted successfully and it’s ushering in its best days“
- “Data center, a rapidly growing semiconductor field“
- “Top five lucrative artificial lucrative intelligence listed companies“
- “The artificial intelligence bubble in the capital market is forming“
- “Artificial intelligence benefits industries“
- “OpenAI, the Generative Artificial Intelligence rising star and ChatGPT“
- “Major artificial intelligence companies in US stocks market“
Disclaimer
- The content of this site is the author’s personal opinions and is for reference only. I am not responsible for the correctness, opinions, and immediacy of the content and information of the article. Readers must make their own judgments.
- I shall not be liable for any damages or other legal liabilities for the direct or indirect losses caused by the readers’ direct or indirect reliance on and reference to the information on this site, or all the responsibilities arising therefrom, as a result of any investment behavior.
