All articles

Archive of articles

«    October 2017    »
MonTueWedThuFriSatSun
 1
2345678
9101112131415
16171819202122
23242526272829
3031 
» Video cards

Video cards

What does a video card do for me?

The video card is responsible for delivering all a computer’s hard work and resultant output to a connected display/monitor. When you play 3D video games, the video card is hard at work in the background rendering vivid 3D scenes and effects. In fact, all the text and pictures you see are processed by the video card, including the visualization effects that are provided by media player software. Without the video card, you wouldn’t know what your computer is doing and you most likely won’t be able to enjoy any of the fun that your computer can provide!
The applications that the video card can do are multiplex. In other words, different users will be able to use the video card to do different things. We can categorize the applications into the following types:

General purpose 2D

General-purpose 2D applications include web surfing, word processing, and picture viewing. Such applications are widely used by home and office users and do not have any special requirements for video cards.




Web surfing – A general purpose 2D application

3D gaming

3D games have been the driving force behind the video card industry’s phenomenally fast-paced growth. The main difference between each new generation of video card and their predecessors is typically that newer products offer better performance and are equipped with more advanced features to meet the requirements of existing or future 3D games. 3D games are so dazzlingly attractive these days that many users treat their computers as game consoles.




Bleeding-edge 3D games are a tough proving ground for video cards


Video playback

The latest video card products add not only new features for a better 3D game experience, but also features that cater to advanced video requirements, which, for example, include hardware acceleration for High-Definition video decoding. Without the help of the video card, the CPU would have a very tough time playing High-Definition video and the result could be heavy frame skipping.




HDTV playback is a hot new feature of the latest video cards

Multimedia/TV

PCs are rapidly replacing the TV and VCR/DVD player as the multimedia hubs of our living rooms through the help of multimedia video cards. Their built in TV and HDTV tuners plus video in/out functions mean you can watch TV and even HDTV; record and edit video; as well as play movies from a variety of formats all from the convenience of your personal computer.




Video cards help you to build a rich multimedia center out of your PC

Professional 2D

Professional 2D applications include general purpose business and corporate applications and financial trading applications. Generally, professional 2D applications require multi-display capabilities from the video cards.



Multi-display capability is important in professional 2D applications

Professional 3D

Professional 3D applications include Mechanical Computer-Aided Design (MCAD), Digital Content Creation (DCC), Non-Linear Video Editing (NLE), and Visualization Applications. To run professional 3D application smoothly, video cards need to be specifically optimized on both the hardware design and driver programming fronts.




Professional 3D applications require purpose-built video cards

Summary:
For the average user and gamer, general purpose 2D, 3D gaming, video playback and MultiMedia/TV make up the bulk of the applications used on desktop PC, whereas professional users will typically run professional 2D and 3D applications on a workstation.

What do I need to consider before purchasing a video card?

Video card improvements have been moving along at breakneck speed for more than five years. Nowadays we can expect a new generation of video cards to replace the old within 3-6 months. There are also a great number of models on the market and in the same market segment with similar pricing. It certainly can make buying video cards a daunting task.

This section is designed to help you get a handle on key video card specifications, and should hopefully make finding the right card for you an easier process.

GPU

The GPU (Graphic Processing Unit) is the core of the video card - it determines what the video card can do and how well the card will perform. The development of the GPU has lead directly to the fast-paced growth of the video card. At present, there are two GPU manufacturers that stand out from the rest: NVIDIA and ATI (acquired by AMD). Some other manufacturers, such as S3 and XGI are working hard to keep up with the Big 2 in the desktop PC market, while Matrox is devoted to the workstation market (especially in professional 2D field).

Both ATI and NVIDIA have developed their own lineup of GPU products designed for all kinds of applications covering general purpose 2D, 3D game and video playback. The NVIDIA GeForce series and ATI Radeon series GPU, for example are represented by the NVIDIA GeForce 7900 GTX and ATI Radeon X1900 XTX respectively in this domain. For professional 2D applications NVIDIA has the Quadro NVS series and ATI its FireMV series; while for professional 3D applications there are the NVIDIA Quadro FX and ATI FireGL series products.

All GPU series contain several models to cover each market segment, and we can generally expect higher priced items to provide higher levels of performance.

Memory

Working closely with the GPU is memory, which is the second most important factor determining the performance of a video card. Memory is used to store graphics data temporarily and the faster and larger it is, the better the maximum performance of the video card.

Memory clock, type (e.g. DDR, GDDR2, GDDR3 and GDDR4), and the width of the memory interface are all key factors that affect each other.

The type of memory installed correlates directly to a certain memory clock range - for example, the maximum clock of DDR memory is around 700MHz, while GDDR2 memory can support around 800MHz, GDDR3 will easily exceed 1GHz (1000 MHz), and GDDR4 can reach or even exceed 2 GHz. A higher clock usually means better performance given a memory interface of equal width.

The memory interface actually refers to the width of the memory interface, for which currently there are several common standards: 32-bit, 64-bit, 128-bit, and 256-bit for video cards. Some current flagship video cards support much wider memory interfaces. For example, the latest GeForce 8800 GTX and GeForce 8800 GTS are equipped with 384-bit and 320-bit memory interfaces respectively.

The width of the memory interface can be understood in the same way as the width of a length of highway - the wider the highway is, the greater the traffic that can pass through at any one time. In the same sense, the wider the memory interface is, the greater the amount of data that can be transferred between the GPU and memory at a given clock speed.

Interface

Interfaces, also known as the bus connector, act as the bridge between the video card and the motherboard. Currently, video card interfaces include the PCI, AGP and PCI Express standards.

PCI has been used for a very long time but can still be found on a handful of video cards. The AGP interface was the replacement for PCI and has been developed over the years to encompass the AGP 1X, AGP 2X, AGP 4X, and AGP 8X interfaces. AGP 8X is the fastest and is downwards compatible with AGP 4X; while AGP 4X is downwards compatible with AGP 2X, and so on. Currently, AGP is being rapidly replaced by PCI Express as the latter provides higher bandwidth for future requirements.

The PCI Express standard can be broken down into the PCI Express x1, x2, x4, x8, and x16 standards, with PCI Express x16 used mainly for video cards. Therefore, we recommend getting a PCI Express video card if you are contemplating building a new system from the ground up. However, if you are only upgrading your existing video card it becomes important to select the interface based on the interface sported by your motherboard as installation of your new video card becomes impossible without matching interfaces.

I/O ports

The I/O ports are used to link your computer up with your display device. On a video card, you may find a combination of D-Sub, DVI, TV-out, VIVO, or CATV ports.

The D-Sub port is for connections to analog CRT and LCD monitors, while DVI is used for display devices utilizing the DVI connector, such as LCD monitors and projectors. The S-Video, Composite, and Component ports are all classified as TV-out ports. VIVO means Video-In and Video-out, and is multi-purpose port. The CATV port connects to a CATV cable to receive TV signals, and is provided only by video cards equipped with a TV tuner.

Generally, the D-Sub, DVI and TV-out are the most typically-used ports on contemporary video cards. Please see the following guide for more detailed information about I/O ports.

Nvidia Videocards

Most users are looking for two things when they purchase a video card for the desktop PC: 3D gaming capabilities and the ability to run ordinary 2D applications (e.g. editing and viewing pictures, document processing and video playback). Since all contemporary video cards pack sufficient 2D power, video card performance is naturally all about 3D gaming. That performance also determines a video card’s market positioning (high-end, mid-range or low-end).

The Radeon X1600 XT and GeForce 7600 GT names may sound very familiar – but what are they exactly? For those just getting to grips with computer hardware, they represent the GPU model name. The GPU directly determines the performance and features of the video card and should be the first factor to look at when selecting a video card.


There are a ton of GPU models out there it is pretty easy to get confused especially when only a handful are a good fit for your requirements. This guide provides the lowdown on GPUs from NVIDIA – one of the most famous graphics leaders in the world.







All about the GeForce



 

All current NVIDIA’s GPUs for desktop PC application go by the same name: GeForce. This is then followed by a series of numbers and letters that help to distinguish between the different product series, e.g. GeForce 8 series, GeForce 7 series, GeForce 6 series and GeForce FX series.



The model number designations are used to draw clear lines between products with differing performance levels, which helps to give users a clearer picture of the market segments the products are intended for. For example, the GeForce 7900 GPUs are intended for the high-end segment, while the GeForce 7600 GPUs are prepared for the mid-range segment. At the entry-level segment are the GeForce 7300 and GeForce 7100 GPUs; but as you can already see, each of these still belongs to the GeForce 7 series.



Different products in the same GPU series share similar key features, but products for the lower price points may lack some of the advanced features that the higher models pack. For example, while all GeForce 7 series GPUs support SLI multi-GPU technology, DirectX 9.0, and Shader Model 3.0, the GeForce 7100 models do not feature HDR (High Dynamic Range) rendering, which is an impressive lighting enhancement feature supported by such models as the GeForce 7300, GeForce 7600 or GeForce 7900 (with PCIe bus interface).


Although most new GeForce GPUs are native PCI Express designs, the lasting popularity of the AGP video card interface has caused a portion of new GeForce cards to be produced in both PCI Express and AGP versions; the latter using the HSI bridge chip to convert from PCI Express to AGP.







GeForce 8 Series





Released on November 8, 2006, NVIDIA’s GeForce 8 series, the first DirectX 10 GPUs in the world, feature a powerful unified architecture to deliver true-to-life gaming experiences.


GeForce 8800 Models


The GeForce 8800 GPUs, currently including the GeForce 8800 GTX and GeForce 8800 GTS, were immediately available following NVIDIA’s launch announcement. Both models feature a revolutionary unified shader architecture which is comprise of 128 (GTX) or 96 (GTS) parallel stream processors for unmatched gaming performance, and NVIDIA Quantum Effects physics processing technology that ushers in a new generation of visual effects and graphics realism. Simultaneous 16x anti-aliasing and 128-bit High Dynamic Range (HDR) lighting for unparalleded image quality also makes its way into the series.



Note: the listed core and memory clocks are NVIDIA’s stock settings. They may be different depending on the card manufacture.







GeForce 7 Series





NVIDIA’s GeForce 7 series GPUs represent the next most recent additions to the company’s long line of high-performance products. Featuring CineFX 4.0 Shading Architecture to support the latest and future 3D games better, the GeForce 7 series supplanted the GeForce 6 series.


GeForce 7950 Models












A GeForce 7950 GX2 card


In the GeForce 7950 series, the GeForce 7950 GX2 is the only model of the 7900 series that features dual-GPUs on a single card for the ultimate 3D performance. This card offers the largest onboard memory capacity available today (512MB per GPU for a total of 1GB) and support for resolutions up to 2560x1600, With support for SLI, two GeForce 7950 GX2 can be paired up to build a Quad GPU setup. According to NVIDIA, “It’s a beautiful thing.”



Unlike the GeForce 7950 GX2, the newer GeForce 7950 GT applies traditional single GPU board design. Clocked at 550 MHz with 24 pixel shader processors and 512MB of memory running at a blazing 1.4 GHz, this video card delivers exceptionally smooth frame rates at maximum settings.



Note: the listed core and memory clocks are NVIDIA’s stock settings. They may be different depending on the card manufacture.


GeForce 7900 Models


The GeForce 7900 GPUs are flagship models that are SLI-ready for extreme HD gaming and video on the PC. Currently, there are three available models: the GeForce 7900 GTX, GeForce 7900 GT, and GeForce 7900 GS.


The GeForce 7900 GTX offers the fastest core and memory clock settings in this series, followed by the GeForce 7900 GT with slightly lower clock settings. The GeForce 7900 GS comes with fewer pixel/vertex shaders, and the clock settings are similar to those of the GeForce 7900 GT.



Note: the listed core and memory clocks are NVIDIA’s stock settings. They may be different depending on the card manufacture.


GeForce 7800 Models


The GeForce 7800 series were NVIDIA’s flagship series before the arrival of the GeForce 7900 series. To start off, we have the GeForce 7800 GTX 512, which is the top model with full pipeline architecture (24 pixel pipelines and 8 vertex pipelines), the highest clock settings and of course the largest video memory capacity of 512MB.


The GeForce 7800 GTX also has a full pipeline architecture, but the core clock setting is slightly lower than the “512” edition, and the memory capacity is scaled back to 256MB.


Next up is the GeForce 7800 GT, which is a more affordable version with slightly fewer pixel pipelines and vertex pipelines. Following that is the GeForce 7800 GS for AGP application, although it sports a PCI Express GPU. It works using the HSI (High Speed Interconnect) chip to make it the first AGP card in the GeForce 7 series.



Note: the listed core and memory clocks are NVIDIA’s stock settings. They may be different depending on the card manufacture.


GeForce 7600 Models


Featuring the GeForce 7 series architecture, the GeForce 7600 models are very serious offerings for the mid-range market and currently consist of two available products: the GeForce 7600 GT and the GeForce 7600 GS.


Both the GeForce 7600 GT and GeForce 7600 GS own the same pixel/vertex configurations, while GeForce 7600 GS runs at lower core and memory clock settings. The two products are SLI-ready.



Note: the listed core and memory clocks are NVIDIA’s stock settings. They may be different depending on the card manufacture.


GeForce 7300 Series


Designed for budget applications, users will be able to commit less than 100 dollars to enjoy the features of the GeForce 7 series. The GeForce 7300 GS is the first member of the 7300 line, with the GeForce 7300 GT, GeForce 7300 LE and GeForce 7300 SE coming afterwards. Of the current three models, the GT version is the most powerful with 8 pixel shader pipelines, while the GS, LE and SE versions offer faster clock settings and TurboCache support.to allow the cards to expand video memory through dynamic access to system memory.



Note: the listed core and memory clocks are NVIDIA’s stock settings. They may be different depending on the card manufacture.


GeForce 7100 Models


Only one model, the GeForce 7100 GS is available in the series so far. Sporting the same specs as the GeForce 7300 LE in terms of clock settings and pipeline configuration, it does not, however, support HDR and is actually based on the NV44 core which is also known as GeForce 6200 with TurboCache.



Note: the listed core and memory clocks are NVIDIA’s stock settings. They may be different depending on the card manufacture.







GeForce 6 Series





The first of NVIDIA’s products capable of supporting DirectX 9.0 Shader Model 3.0, the GeForce 6 series also became a launch pad for NVIDIA’s innovative new technologies including SLI, PureVideo and TurboCache.


GeForce 6800 Models


Born for the high-end market, top model GeForce 6800 Ultra was positioned in the same way as the GeForce 7800 GTX at the time of its own launch. Although the GeForce 6800 series were launched with only three models available, more models have been introduced over time. While certain 6800 series models are actually very new releases, others are being phased-out gradually.



Note: the listed core and memory clocks are NVIDIA’s stock settings. They may be different depending on the card manufacturer.


GeForce 6600 Models


The GeForce 6600 series are NVIDIA’s main combatants for the mid-range market. Currently they are gradually being replaced by the GeForce 7600 series. This family consists of the GeForce 6600 GT, GeForce 6600, and GeForce 6600 LE.


At the top is the GeForce 6600 GT with the best performance thanks to the highest clock settings and pixel pipelines. The GeForce 6600 is essentially the same as the GeForce 6600 GT but with slightly lower clock settings. It is easily overclocked, however, and many manufacturers have produced factory overclocked editions.


The highlight of GeForce 6600 DDR2 edition card is definitely the DDR2 memory which provides higher clock speeds over DDR while keeping prices affordable. Both the core and memory clock speeds for the GeForce 6600 DDR2 are improved for better performance although it uses the same graphics core as the GeForce 6600.


The GeForce 6600 LE is 4 pixel pipelines down on the GeForce 6600, but possesses very exciting overclocking abilities, as it is built on the same high-quality PCB as the GeForce 6600.



Note: the listed core and memory clocks are NVIDIA’s stock settings. They may be different depending on the card manufacturer.


GeForce 6500 and GeForce 6200 Models


The GeForce 6200 and 6500 series are products for the budget-end of the market. They are excellent choices for budget users and provide a variety of attractive features such as DirectX 9.0, Shader Model 3.0 support, HDR, and PureVideo. There are GeForce 6200, GeForce 6500, GeForce 6200 with TurboCache and GeForce 6200 LE in this family.



Note: the listed core and memory clocks are NVIDIA’s stock settings. They may be different depending on the card manufacturer.



ATI Video cards

Most users are looking for two things when they purchase a video card for desktop PC: 3D gaming capability and the ability to run ordinary 2D applications (e.g. editing and viewing pictures, document processing and video playback). Since all contemporary video cards pack sufficient 2D power, video card performance is naturally all about 3D gaming. That performance also determines a video card’s market positioning (high-end, mid-range or low-end).

The RADEON X1600 XT and GeForce 7600 GT names may sound very familiar – but what are they exactly? For those just getting to grips with computer hardware, they represent the GPU model name. The GPU directly determines the performance and features of the video card and should be the first factor to look at when selecting a video card.


There are a ton of GPU models out there it is pretty easy to get confused especially when only a handful are a good fit for your requirements. This guide provides the lowdown on GPUs from ATI– one of the most prolific graphics leaders in the world.




^ Back to top




ATI Video Card Product Line Preview



Radeon



The Radeon designation is used for all ATI GPUs for desktop PC application. All GPU models under the Radeon family are fully capable of supporting general purpose 2D, 3D gaming, and video playback applications.



Every several months ATI releases new Radeon GPU products with higher performance and more advanced features. The Radeon 1K family is a fine example, consisting of the high-end Radeon X1900/1800 series, mid-range Radeon X1600 series, and low-end Radeon X1300 series. These will be used to usher out the “older” X850/X800 series, X700 series, and X300 series cards respectively.


CrossFire



CrossFire is ATI’s multi-GPU technology. A CrossFire Edition Radeon card can be paired with a standard Radeon card from the same series to form a CrossFire array that can boost 3D performance by as much as 100% and enable special operating modes/effects for better image quality. For example, we can use the Radeon X1900 CrossFire Edition together with a Radeon X1900 XTX or Radeon X1900 XT in CrossFire mode. Please note, that some Radeon GPUs (e.g. the Radeon X1950 Pro and Radeon X1650 XT) with native CrossFire technology provide a multi-GPU experience without the need for a CrossFire Edition card, and a CrossFire supporting motherboard is also required to build a CrossFire array.


All-In-Wonder



All-in-Wonder series cards are all based on Radeon series GPUs and offer the same capabilities as their Radeon counterparts but with added TV tuner and video codecs to enable them to receive and display TV and capture/edit video.


PCI Express and AGP


Although most new Radeon GPUs are native PCI Express designs, the lasting popularity of the AGP video card interface has caused a portion of new Radeon cards to be produced in both PCI Express and AGP versions; the latter using the RIALTO bridge chip to convert from PCI Express to AGP.



^ Back to top




Radeon X1000 Family



Radeon X1950 Series


ATI’s latest flagship graphics card family,, the Radeon X1950 series was released on Aug 23, 2006. As the successor to the award-winning Radeon X1900 series, the Radeon X1950 series sports up to 48 pixel shader processors and supports Shader Model 3.0, Avivo as well as CrossFire technology. GDDR4 memory, with an effective data rate of 2 GHz is a Radeon first and delivers blazing performance.



Note: the listed core and memory clocks are ATI’s stock settings. They may be different depending on the card manufacturer.


Radeon X1900 Series



The crowning jewels in the ATI Radeon GPU product line, the X1900 video cards sport a phenomenal 48 or 36 pixel shader processors and 8 vertex shader processors each. The Ultra-threaded engine, high speed memory support and 90nm process technology give the Radeon X1900 series some serious power to compete for the ultimate performance crown. Furthermore, Shader Model 3.0 and Avivo are also key features of the Radeon X1900 series to improve image quality, 3D gaming performance and video playback.


The Radeon X1900 XTX is the most powerful model with the highest clock settings and pixel/vertex shader configurations, closely followed by the Radeon X1900 XT. The Radeon X1900 CrossFire Edition is the same as Radeon X1900 XT in terms of clock speeds and should be used when building a CrossFire array. The Radeon X1900 GT runs at lower core and memory clocks with fewer pixel shader processors but is offered at a lower price.



Note: the listed core and memory clocks are ATI’s stock settings. They may be different depending on the card manufacturer.


Radeon X1800 Series



ATI’s Radeon X1800 series are GPUs for the enthusiast segment. They are extremely potent and are designed to take on NVIDIA’s GeForce 7800 family. Each GPU supports Shader Model 3.0 and Avivo technology to bring both extreme performance and quality gaming and video playback (especially for HD content, e.g. H.264 format)/display.


The Radeon X1800 series covers four models at present: the Radeon X1800 XT, the Radeon X1800 CrossFire Edition, the Radeon X1800 XL and the Radeon X1800 GTO. The top Radeon X1800 XT and the CrossFire Edition models run at very high frequency to boost performance, therefore necessitating a dual-slot heatsink for cooling. The Radeon X1800 XL runs at lower clocks compared to the Radeon X1800 XT, but sports the same pixel/vertex shader configurations and features as Radeon X1800 XT otherwise, The Radeon X1800 GTO features the same clock configurations as the Radeon X1800 XL, but with fewer pixel shader processors.



Note: the listed core and memory clocks are ATI’s stock settings. They may be different depending on the card manufacturer.



Radeon X1650 Series



The Radeon X1650 series replaces the Radeon X1600 series in the mid-range market. Unlike the RV530/RV535-based Radeon X1600 series, the Radeon X1650 XT (the top model in the series) utilizes the RV560 core that is built on the latest 80nm process technology and features 24 pixel shader processors and 8 vertex shaders, (twice that of the Radeon X1600 XT. The Radeon X1650 Pro features the same shader configuration and clock settings as the Radeon X1600 XT.



Note: the listed core and memory clocks are ATI’s stock settings. They may be different depending on the card manufacturer.


Radeon X1600 series



Like the Radeon X1800 series, the Radeon X1600 series are similarly new to the scene. They support Shader Model 3.0 and Avivo, but as they are targeted at the mid-range market, the products feature fewer pixel and vertex shader processors (see table below). At present, the Radeon X1600 series involves the Radeon X1600 XT and Radeon X1600 Pro editions. Running at higher frequency settings, the Radeon X1600 XT is the top model in this series, but looks identical to its sibling the Radeon X1600 Pro.



Note: the listed core and memory clocks are ATI’s stock settings. They may be different depending on the card manufacturer.


Radeon X1300 series



For the entry-level segment there are the Radeon X1300 series. The Radeon X1300 series also support ShaderModel 3.0 and Avivo.



Note: the listed core and memory clocks are ATI’s stock settings. They may be different depending on the card manufacturer.




^ Back to top




Other Radeon Products In Focus



Radeon X850 series


The Radeon X850 series was the top dog in the Radeon X series but has been replaced by the more powerful Radeon X1800/X1900 series. Products in the series max out at 16 pixel pipelines and have 2 less vertex pipelines versus the Radeon X1800 series.



Note: the listed core and memory clocks are ATI’s stock settings. They may be different depending on the card manufacturers.


Radeon X800 series



ATI’s Radeon X800 series is an older product line also designed for the high-end segment. ATI has released a few new models for the lower and middle price points of the high-end segment.


Like the X850 CrossFire Edition, the X800 CrossFire Edition is the master card for the X800 CrossFire array, and its specifications closely match those of the Radeon X800 XL (table below). The Radeon X800 GTO is the frequency-enhanced version of the Radeon X800, while the Radeon X800 GT features 8 pixel pipelines and maintains a 256-bit memory interface. This model strengthens ATI’s position in the mid-range market.



Note: the listed core and memory clocks are ATI’s stock settings. They may be different depending on the card manufacturers.


Radeon X700 series



Based on the architecture of the Radeon X800 series, the Radeon X700 series were targeted at the mid-range market. Currently the Radeon X1600 series has taken the place of the Radeon X700 series.



Note: the listed core and memory clocks are ATI’s stock settings. They may be different depending on the card manufacturers.


Radeon X300 series



These were ATI’s earliest native PCI Express GPUs produced using the 110nm fab process. The Radeon X300 series, especially the Radeon X300 SE and Radeon X300 SE HyperMemory (HyperMemory technology accesses a portion of system memory for use as video memory. This means that a reduction in onboard memory and cost savings.) models are still good choices for base entry-level applications.



Note: the listed core and memory clocks are ATI’s stock settings. They may be different depending on the card manufacturers.


Radeon X550 series



The Radeon X550 is actually a factory overclocked version of the Radeon X300, as the internal architecture is faithful to the Radeon X300.There is another Radeon X550 model featuring HyperMemory technology.



Note: the listed core and memory clocks are ATI’s stock settings. They may be different depending on the card manufacturers.





Provided by NewEgg
Up