What is the definition of a video card?

The full name of the GPU is graphic processing unit, which is translated as "graphics processor". The concept first introduced by NVIDIA when publishing the GeForce 256 graphics processing chip. The GPU makes the graphics card less reliant on the CPU and does some of the original CPU work, especially when it comes to 3D graphics processing. The core technology used in GPU is hardware t&l (geometric transformation and illumination processing), cubic environment material mapping and vertex mixing, texture compression and bump mapping, dual texture quad pixel 256-bit rendering engine, and hardware t&l technology can be said to be the symbol of GPU. The production of GPU is mainly produced by Nvidia and ATI two companies.

Memory

Video memory is the abbreviation for the display of RAM (similar to the motherboard's memory). The main function is to temporarily store the display chip to deal with the data and processed data. The higher the performance of the graphics core, the more memory you need. The previous video memory is mainly SDR, the capacity is not large. Most of the video cards in the market are GDDR3 memory, and now the latest graphics card has a better performance of GDDR4 or GDDR5 video memory.

Video card Bios

The graphics card BIOS (similar to the motherboard BIOS) is mainly used to store the control program between the display chip and the driver, as well as the type, specification, manufacturer and time of the display card. When you turn on your computer, the information is fed back to the screen by displaying a section of the control program within the BIOS. Early display BIOS is cured in ROM, can not be modified, and most of the display card is a large volume of eprom, that is, the so-called Flash BIOS, can be rewritten or upgraded through a dedicated program.

Graphics Card PCB Board

Is the circuit board of the video card (similar to the motherboard PCB board), it connects the other parts on the video card.　　Features similar to the motherboard. Video card classification integrated graphics integrated graphics card is will display chip, video memory and related circuits are done on the motherboard, and the motherboard into one; Integrated graphics display chip has a separate, but most of the integrated in the main board of the North Bridge chip; some motherboard integrated graphics cards are also installed on the motherboard with a separate video memory, but the capacity is small,　　Integrated graphics display and processing performance is relatively weak, the graphics card can not be hardware upgrades, but through the CMOS adjustment frequency or brush into the new BIOS files to achieve software upgrades to explore the potential of the display chip.　　The advantages of integrated graphics: low power consumption, small heat, the performance of some integrated graphics can be comparable to the entry level of independent graphics, so no extra money to buy graphics cards. Integrated graphics shortcomings: Can not change the new graphics card, to say must be replaced, only and the motherboard, CPU one-time exchange.

Standalone graphics

Independent video card refers to the display chip, memory and its associated circuit on a single circuit board, self-contained and as a separate board exists, it needs to occupy the motherboard expansion slots (ISA, PCI, AGP or PCI-E).

The advantages of the independent video card: The memory is installed separately, generally does not occupy the system RAM, technically also more advanced than integrated graphics card, than integrated graphics can get better display effect and performance, easy to video card hardware upgrades.

The disadvantage of independent video card: The system power consumption is increased, calorific value is also larger, need to spend extra money to buy video card.

PCI interface

The PCI (Peripheral Component Interconnect) interface was introduced by Intel (Intel) in 1991 to define the local bus standards. This standard allows up to 10 expansion cards that comply with PCI standards to be installed on your computer. The first proposed PCI bus work under the 33MHz frequency, the transmission bandwidth reached 133mb/s (33MHZ * 32bit/s), basically meet the development needs of the processor at that time. With the requirement of higher performance, the 64bit PCI bus was proposed in 1993, and then the PCI bus frequency was raised to 66MHz. The PCI interface has the highest rate of 266mb/s,1998 years and is replaced by the AGP interface. However, there is still a new PCI interface of the graphics card, because some server boards do not provide an AGP or PCI-E interface, or the need to build a multiple-screen output, the purchase of PCI graphics card is still the most cost-effective way.

AGP interface

The AGP (accelerate graphical port, Accelerated image processing ports) interface is a technical standard for video interface developed by Intel Corporation and is an interface technology developed to solve the low bandwidth of PCI bus. It directly opens up a faster bus between the CPU and the graphics processor by connecting the graphics card to the system main memory. Its development has undergone AGP1.0 (agp1x/2x), AGP2.0 (agp4x), AGP3.0 (agp8x). The latest agp8x has a theoretical bandwidth of 2.1gbit/seconds. By 2009, it has been largely replaced by the Pci-e interface (most manufacturers have stopped production in 2006).

PCI Express Interface

PCI Express (PCI-E) is a new generation of bus interface, and the use of such interface graphics card products, has been officially published in 2004. As early as the 2001 Spring "Intel Developer Forum", Intel Corporation proposed to replace PCI bus and multiple chip internal connections with a new generation of technology, and called third generation I/O bus technology. Then, at the end of 2001, more than 20 industry-leading companies, including Intel, AMD, DELL and IBM, began to draft new technology specifications, which were formally named PCI Express in 2002.

Common brands

The graphics industry is becoming increasingly competitive. Various brands, the following are some common brand, for reference only:

Sapphire, Asus, Dylan, Li Tai, Sotec, News, Gigabyte, the audience, MSI, Elsa, Fuji Kang, Tepo, Pan is, Ying Tai, Geng Sheng, Jing Yu, Shadow, Ming Szeto, Xiang Sheng, Ying Tong, Qi Xiang, seven Rainbow, Spartacus, double sensitive, fine thunder, Ang Tatsu jcg, Golden Chen Light

Sapphire, Asus is in the independent research and development of the good brand, sapphire only do a card, Asus's A and n cards are core partners, compared to the seven rainbow such a channel brand, with independent research and development of manufacturers in the workmanship and characteristics of a better, and the channel graphics card prices are cheaper (note: seven Rainbow , Shuang-Min, Ying Tong, Ming Szeto and Onda are all by the same manufacturer of the foundry, so the difference is only in the graphics card stickers and packaging, we need to pay attention to buy, each manufacturer has its own brand characteristics, such as ASUS "for the game", seven Rainbow "game graphics experts" are familiar to everyone.