Multi-monitor

Multi-Monitor, also called Multi-Display and Multi-Head, is the use of multiple physical display devices, such as monitors, televisions, and projectors, in order to increase the area available for computer programs running on a single computer system. Dual-Display, Dual-Screen and Dual-Monitor are names commonly used to describe a particular type of multi-monitor display, specifically one with two display devices.

Contents

History

Contemporary windowing systems, such as the X Window System (used by GNU/Linux) and those used by Microsoft Windows and Mac OS, all support simultaneous use of multiple monitors. Multi-monitor support once depended on specialized proprietary video drivers and support from windowing systems and window managers. Support for multiple monitor configurations was added as a standard feature in Microsoft Windows in Windows 98, and has been a standard feature of Apple's Mac OS since the introduction of the first color Macintosh II in 1987.[1]

Configuration and placement

Single PC multi-monitor

A multiple monitor setup increases the net display area of a system and can be an inexpensive way of improving computer usage.[2] Resulting display area after upgrading to a multi-monitor configuration is limited by the size, resolution and number of monitors. The monitors used for multi-monitor can be different types (LCD or CRT) and sizes. The operating system manages the monitors' resolutions independently.

Video output on a computer is generated by a video graphics device, typically on a removable card but which may also be integrated into the motherboard as a discrete device or as part of the chipset logic. The output is interpreted and displayed by a variety of devices. Video outputs are generally connected to a monitor (of either the CRT or LCD type); however, they are increasingly being connected to projection equipment or television sets. As a result of this trend, manufacturers have produced video cards which can connect to several types of display devices using the appropriate interface. A "Dual Head" configuration utilizes a video card that supports two discrete outputs. Users may also utilize two discrete video cards, and sometimes even an integrated motherboard video socket plus a second video card, though often the motherboard disables the integrated video when a discrete video card is used (a limitation that was common on older chipsets featuring integrated AGP graphics and an AGP upgrade slot).

More than two monitors

Prior to mainstream adoption of the PCIe bus, configurations of more than two monitors were either achieved with an AGP card with dual video outputs or by using an AGP graphics adapter as the primary device and a conventional PCI graphics adapter as a secondary device. However, given the bandwidth limitations of the older PCI bus, such setups were not common, and maximum overall graphics performance could be obtained only by using specialty solutions such as the Matrox G450, which features four outputs from one graphics adapter. Now that computers with two or more PCIe interfaces are popular, middle- and high-end computers are no longer limited to two monitors driven by a single main graphics adapter. If a dual PCIe interface is not available or is otherwise occupied, a standard PCI graphics card can still be used to provide additional video outputs, albeit with performance trade-offs. Specialized application environments such as CAD, day trading of corporate stocks, and software development are increasingly using six or more monitors on one production system.

Additional monitors can also be connected to PCs via a USB connection such as DisplayLink.

Multiple PC multi-monitor

Software such as Maxivista for Windows and ScreenRecycler for Mac OS X let one set up multiple PC multi-monitor mode through virtual display drivers and client-side software. Similarly, software such as InputDirector or Synergy allows one to use multiple PCs, each with their own monitor or monitors, and transition from one to the other on screen edges as if they were one machine. This allows each machine to be doing a different task, freeing up resources.

Linux users may use Xdmx, which is an X Window proxy. It is possible to have multiple monitors displaying as a single virtual desktop. Multiple university display wall projects use this capability,[3] such as The LambdaVision display by the University of Illinois at Chicago's Electronic Visualization Laboratory, with 55 LCD monitors which are connected to 32 PCs.[4] This results in a 17600 x 6000 pixel display. As long as the xinerama extension is enabled, GNOME can use the entire desktop.

An additional and different approach to multiple monitor systems involves using the monitors of networked computers to display the output of a central computer. By using the graphic cards of the networked computers, stability and speed are dramatically enhanced. This is often a preferred choice for systems in which adding additional graphics cards is problematic, such as laptops.

The additional monitors can be extensions of the desktop or mirrors of the central display. The arrangement of these monitors can be configured within the properties tab in the windows display dialog box, making horizontal, vertical, or other monitor configurations possible. Further, because the additional monitors are powered by networked computers, they can be located wherever the network reaches, both wireless and hardwired.

Display modes

Clone mode

Since before personal computers existed, video signals have been split with simple Y-adapters to provide duplicate signals to multiple monitors for various reasons. When personal computers came to have video outputs, this naturally carried over—sometimes for the purpose of presentation, and sometimes to provide a different representation of the same output (for example color alongside the higher resolution monochrome interpretation of the output of an Apple II). Later systems—particularly portable machines with built-in displays—provided built-in redundant outputs for this. Even later systems, in addition to being capable of the discrete modes described below, are able to mimic this "cloning" or "mirroring" behavior (typically defaulting to it upon power-up/reset).

Multi-mode

In the early days of the IBM-PC and its clones, there were two main types of display adapter: color (CGA, EGA) and monochrome (MGA). Since these devices used fixed areas of the 1 MiB address space of the 8086/8088 CPU, it was not possible to have more than one of the same type in a machine. However, since color adapters used a different space than monochrome, one of each could coexist. This led to systems using a CGA or EGA for color graphics, and an MGA for monochrome text, on side-by-side monitors.

The Commodore 128 was capable of similar behavior using its C64 compatible 40-column output for graphics on one monitor, alongside its 80-column capable output on another.

Span or extended desktop mode

Most —if not all—current multi-head video cards are able to "span" a single frame buffer across two monitors. The result is one large rectangular desktop space, of double either the horizontal or vertical resolution. (For example, when two 1024x768 displays are used, they can have a combined virtual resolution of either 2048x768 or 1024x1536.) Both monitors operate at the same resolution and depth settings, and usually the same refresh rate. This mode requires very minimal operating system support, as the total display area is a simple rectangle.

In "extended" mode, additional desktop area is created by giving each monitor its own frame buffer. The result is a virtual desktop space comprising multiple adjacent, but not necessarily aligned, rectangular areas. In this considerably more flexible configuration, each monitor can be of different resolution, depth, and refresh rate; and the desktop area is not necessarily rectangular. This mode is actually older than spanning—first appearing in 1986, when Radius introduced the vertical Full Page Display for the Macintosh Plus (which was a dramatic increase over the machine's 9" built-in display).

Both of these modes present the display space to the user as a contiguous area, allowing objects to be moved between, or even straddled across displays as if they are one.

Some problems occur when a user wants to define one large spanned view of two 1024x768 displays into one 2048x768 display in Windows Vista and Windows 7 since only extended desktop is supported and spanned resolutions are non-existent in many applications such as games and presentation software. Windows XP supports both spanning and extended desktops through older NVidia and ATI drivers.

Hybrid Span mode

This is a technique that allows using multiple GPUs to create one single unified display.

Commercial systems

Major players in the visual computing technologies currently include ATI Technologies (now a division of AMD), which supplies graphics hardware and supports its function via ATI's Hydravision Multi-Monitor Management Software; NVIDIA, also a hardware supplier, which includes software support under the moniker of nView Multi-Display Technology; and Matrox, a third hardware supplier providing both multi-display add-in boards and a series of external multi-display upgrade units known as DualHead2Go and TripleHead2Go. The technology provided by these companies was once limited to the professional graphics market, but has gradually become more widespread and affordable in the consumer market. The latest version of Microsoft Windows supports up to 64 monitors.

There is also a growing movement in multi-monitor displays as production companies around the world are using this technology to expand their screen setups without the extravagant costs included with multi-screen processors. Now with the commercial systems mentioned above in place, bands and other entertainment based events are able to create technological setups with ease and low cost.[5]

Disadvantages

The primary hardware disadvantage to multi-monitor use is that common resources of the video card are divided between each display's output duties. For example, if a user is showing a 2D widescreen desktop display at 1680x1050 resolution and 32-bit color depth on a second monitor while playing a game on the primary monitor, nearly 7 MB of video memory (VRAM) will be consumed by the second display image, making it unavailable to the game. In some cases, the decreased processing power and VRAM available to each display may lead to unacceptable performance on both devices. When this situation is encountered, the common remedy is to install an additional video adapter and connect the additional display to the new device. Ongoing improvements in graphics technology have minimized this issue in recent years, with many mainstream graphics adapters now supporting 1 GB or more of VRAM and a graphics core purpose-built around two or more video outputs.

Although multi-monitor configurations are increasing in number, single monitor PC users continue to dominate the market. Cost can be one problem for multi-monitor, as there is the cost of the second display and sometimes an additional video adapter as well. Furthermore, growth in 16:9 and 16:10 aspect ratio wide-screen LCD technologies has led to decreased cost for larger, wider displays, and some users may find that multi monitor does not add significant utility over a similarly priced widescreen display.

Full-screen software can pose a problem on a multi-monitor configuration. Many full-screen applications make use of the absolute edge of the display to control view movement, and may not work properly on a multi-monitor PC, for example by failing to track the mouse cursor properly when it continues onto the next display of an extended desktop. "Edge-scrolling" can frequently be found in full-screen image viewers, 3D model editors, and RTS (real time strategy) genre video games. Newer software is more likely to either be multi-monitor aware, or else not depend on the endpoints of the visual area as a fixed reference, albeit this does not solve all of the ergonomic problems a user may encounter.

For example, many full-screen applications, even when tolerant of multi-monitor displays, only cover one display area and relegate the other display to secondary use. This can create edge-scrolling problems when the cursor crosses between the display fields. One notable exception is flight simulator software, which might take over all available displays, e.g. using one display to show a windshield view of the simulated flight and presenting simulated instrumentation and controls on another display.

Problems can also arise if the user clicks outside of the full-screen application's display area—though this is not directly a multi-display problem. Anything that shifts focus (including clicking elsewhere, pressing the Windows/Super key, typing Alt-Tab or Command-Tab, etc.) away from a game, or any full-screen application, can cause grief—even on a single-monitor system. The full-screen application may drop out of full screen mode. A game may continue running, yet control has been taken away, since the newly focused application is now first in line for console input events.

When software is not multi-monitor aware, or the edge-scrolling problem is encountered, the user must adjust his or her computer usage to minimize it, or engage a work-around solution. One of the most common methods of overcoming the edge-scrolling problem is to set up a multi-monitor orientation on a diagonal. The diagonal orientation means there is no adjacent desktop space on any primary edge, generally preventing the mouse cursor from moving beyond the screen edge. As a downside, a diagonal orientation can make moving the mouse from monitor to monitor difficult, as the user must target a very narrow transition region in order to move the cursor between displays. A diagonal orientation also does not usually correspond to the physical arrangement of monitors, reducing the intuitiveness of the crossover point.

Another method is to temporarily disconnect the unused monitor(s). However it is not always desirable to disable all other displays, and on a Windows operating system platform, the OS will generally compensate for missing displays by reorganizing all desktop shortcuts onto the remaining active monitor(s). This can be overcome by using utilities that can store shortcut locations, such as ATT.

There are also some programs that provide full workarounds to the issue. One such utility is CSMMT.[6]

Graphics cards

These display adapters are built with two or more outputs of various types. Typically these will be DVI ports and/or VGA. The older CRT monitors will usually use VGA, while higher-end CRTs may include BNC or DVI. LCDs - depending on the model - usually support either or both. Conventional or high-definition television outputs are also sometimes provided, although these will commonly disable one of the other display outputs when accessed and cannot be used to create a third display device.

Multi-display setups in the workplace

In many professions, including graphic design, architecture, communications, accounting, engineering and video editing, the idea of two or more monitors being driven from one machine is not a new one. While in the past, it has meant multiple graphics adapters and specialized software, it was common for engineers to have at least two, if not more, displays to enhance productivity.[7]

Multi-display setups are also very common in investment banks, particularly in market making, where they allow the simultaneous display of several screens of prices as reference data, allowing the trader to keep an eye on the market. Setups of 6 displays (2×3: 2 rows of 3) are common on interest rate trading desks, which involve many numbers, while 8 displays (2×4: 2 rows of 4) are not uncommon. Financial multi-display setups may also incorporate Bloomberg Terminals, or these may be adjunct to the main display.

Now that multi-monitor setups are more budget-friendly, it is not uncommon to see a wide range of business professionals using multiple monitors to increase visual area. This advantage helps promote the concept of a paperless office by increasing the quantity of simultaneous media that can be viewed digitally, although the advantage of viewing two documents simultaneously is also feasible on many larger widescreen monitors.

Programming

Programming multiple monitors is challenging because each screen will have its own graphics buffer. One possible scenario for programming is to present to OpenGL or DirectX a continuous, virtual frame buffer in which the OS or graphics driver writes out to each individual buffer. With some graphics cards, its possible to enable a mode called "horizontal span" which accomplishes this. The OpenGL/DirectX programmer then renders to a very large frame buffer for output. In practice, and with recent cards, this mode is being phased out because it does not make very good use of GPU parallelism, and does not support arbitrary arrangements of monitors (they must all be horizontal). A more recent technique uses the wglShareLists feature of OpenGL to share data across multiple GPUs, and then render to each individual monitor's frame buffer.[8]

Other uses

Older uses

From the mid 1980s through 1990s, a popular configuration for software developers was to employ a general-purpose VGA, EGA, or CGA display for managing the program under development and an independent monochrome Hercules or MDA card driving a second monitor for displaying debugging information. Many DOS debugging applications supported this configuation. It was possible to operate two display cards in this fashion, even with operating systems such as MS-DOS which did not support multi-monitor natively, because the Hercules and MDA cards used a different hardware memory address than conventional graphics cards and could operate concurrently without creating hardware conflicts. Modern hardware is not affected by the limitations of earlier systems like these when running modern operating systems, because the hardware and software are both designed such that the operating system can abstract the various hardware devices from each other and then manage them appropriately. The first Macintosh computer to support multiple monitors was the Macintosh II. The Macintosh SE/30, which had one slot in it, also supported a second monitor which could be color even though the main monitor was monochrome.

Interactive television sometimes coordinates the use of a television screen and a computer display.

Arcade machines

A number of arcade games were released in the 1980s and 1990s which used a multi-monitor configuration. The earliest of these is the game TX-1, a driving game by Atari from 1983, which used a cockpit cabinet with 3 19" CRT monitors side-by-side to give a wide viewing area. It's successor, TX-1 V8 released in 1984, also used the same 3 monitor configuration. Tatsumi released Buggy Boy in 1985, in both cockpit 3 monitor and upright single monitor cabinets. Darius from 1986 used an upright cabinet which was around half as wide again than a standard arcade cabinet of the time. It used 3 14" FST (flatter squarer tube) monitors but in this case, 2 monitors were mounted on the left and right with the screens pointing upwards and one was mounted in a box at the centre rear of the cabinet, and using a two way mirror a seamless wide image was obtained. A similar effect was seen in Darius II, also known as Sagaia, (which was also released in twin larger monitor format) and The Ninja Warriors. As arcade technology entered the 1990s larger cabinets were being built which in turn also housed larger monitors - such as the 3 28" screen version of Namco's Ridge Racer from 1993. Although large screen technology such as CRT rear projection was beginning to be used more often, multi-monitor games were still occasionally released, such as Sega's F355 Challenge from 1999 which again used 3 28" monitors for the sit-down cockpit version.

Multi-monitor support

Application support

Games support

The following games support multiple screens inherently:

Other games such as Guild Wars and World of Warcraft can also be made to run in multi-monitor set-ups, with the addition of third party software to enable the ability, or by running the game in windowed mode.[17] A larger list of games that support dual/multi screen modes is available at WSGF.[18]

See also

References