BUS SLOTS

Bus slot

A socket, usually on a motherboard into which are plugged into the various removable circuit boards of the PC.  Examples include the display-adapter.  The name refers to the fact that the slot provides access to the PC bus.  A bus slot is also known as an 'expansion slot' because it allows the capabilities of the PC to be expanded by the addition of new components.

bus

The set of wires that connect the various parts of the computer together.   The bus extends from the central processing unit and incorporates the memory and the bus slots into which peripheral components are connected.  The defining feature of a bus is its 'width,' which refers to the number of signalling wires it contains.   In computer engineering terms, a bus consists of 'data,' 'control' and 'address' parts, and generally  it is the width of the data component that is quoted.   When we talk of a '32-bit bus' we mean one which has 32 parallel data lines.   Having a 32-bit rather than a 16-bit bus gives the PC a speed improvement because a greater amount of data can be transferred in at given time.

A number of different bus types are used in PCs and, because the bus type dictates the type of components that can be installed, a small number of PCs are fitted with more than one type.  At this time of  the buses usually are ISA, PCI, VESA and EISA.

Of these, nearly all  machines use PCI, and the others are largely obsolete. Almost all these buses are 32-bit.  The exception is the older ISA which is 16-bit.  Because of its rapid rise in the early 1990's, most PCs made before about 1994 have the ISA bus.

The word 'bus' is sometimes also used to refer to the set of signal wires that connect disk drives.

ISA bus

Industry Standard Architecture (ISA).  An 8 and 16-bit bus.   ISA bus slots were developed by IBM back in the late 1980's .   There are a few manufacturers that still produce these adapter cards today (proprietary cards for scanners, video cards, sound cards, etc.) that require an ISA bus slot in order to be installed in your computer.  The ISA design includes a buffering chip between the CPU bus and the ISA slot.  This chip adds wait states to I/O to allow faster CPU's to accommodate ISA bus speeds.  If you look at your motherboard's slots, the longer ones are the ISAs.  If they are all one size, they are all ISAs.

EISA bus  (See picture of board above)

Extended or Enhanced  Industry Standard Architecture (EISA). A 32-bit bus that is compatible with 8 and 16-bit products.  This type of bus is not used very often in desktop machines. It is used mainly in servers, or computers that host networks.  With such a computer, the demands placed on its components are too big for ISA to handle.  The EISA slot speed is 8 MHz to maintain compatibility with ISA, but, because of its wider bus, EISA has faster data transfer rates.  Another reason for EISA bus speed is the EISA bus design allows add-in boards to process independently of the CPU.  That is, they have bus-mastering capabilities and can access the computer's memory and peripherals on their own which allows components attached to the bus to talk to each other without bothering  the CPU.

MCA bus slots

Micro Channel Architecture.  IBM's 32-bit bus for the PS/2 machines.  Like ISA and EISA, the MCA bus is limited to 8- or 10-MHz speed, but you can't stick ISA cards into it.  The MCA bus supports bus-mastering (direct communication with the computer's memory and peripherals without CPU involvement) and allows multiple processors and devices to share the single data bus, plus it could look at other devices plugged into it and identify them, leading to automatic configuration.    MCA also produced less electrical interference, reducing errors.

VESA bus

Video Electronics Standards Association (VESA). The VESA-Local Bus, or VL-bus is VESA's standard 32-bit bus (can be upgraded to 64-bit).  VESA buses are basically an ISA slot with an extra slot on the end. The whole thing is about 4 inches longer than an ISA slot.  Based on the local bus architecture, the VL-bus is connected straight to the CPU's internal bus (hence the name "local") to peripherals such as video, disk, and network devices. The VL-bus supports bus-mastering and works with ISA and EISA buses.

Local Bus:  A 32-bit internal path that connects the CPU directly to memory, video, and disk controllers.   The local bus allows data transfer to and from the CPU to memory and peripherals at CPU speeds.  Prior to local-bus architecture, the microprocessor speeds outpaced the internal bus speeds, which created a narrow data stream in and out of the CPU and slowed the computer. Local bus architecture significantly improves data transfer rates and greatly reduces the I/O bottleneck because peripheral devices are bound only by the clock speed of the CPU.  Local bus architecture is utilized by PCI, VESA and PCMCIA buses.

PCI bus slots 

Peripheral Component Interconnect (PCI). A 32- or 64-bit bus based on local bus architecture. Local bus architecture design gives the PCI bus a direct data path to the CPU.  This unit made the bus independent of the CPU, a drawback on the VL-Bus. This direct connection between the CPU and the local bus slots allows data transfers at the speed of the system clock, which can currently reach speeds of 66 MHz and higher.   The PCI bus supports bus-mastering and allows ten peripherals such as video, LAN and SCSI devices to be integrated directly onto the local bus.  The PCI bus provides a throughput rate of 132Mbps, 16 times faster than ISA bus, and four times faster than the EISA bus.   It is designed for high bandwidth applications such as multimedia. The bus is self-configuring, leading to the plug-n-play concept in which each add-on card contains information about itself that the processor can use to automatically configure the card.  To provide compatibility with slower CPUs, the PCI bus slot also functions with ISA and EISA adapter cards.

PCMCIA 

Personal Computer Memory Card International Association (PCMCIA). This is a special socket in which you can plug removable standardized credit-card sized module for laptop and palm-top computers.  Mostly, PCMCIA cards are used for laptops, but many PC vendors have added PCMCIA sockets to their desktop machines.  These circuit cards can contain extra memory, hard drives, modems, network adapters, sound cards, etc.  The socket uses a 68 pin interface to connect to the motherboard or to the system's expansion bus.The PCMCIA I/O bus uses the Local Bus architecture.

There are three types of PCMCIA cards: Type 1 slots are 3.3mm thick and hold items such as RAM and flash memory. Type 1 slots are most often seen in palmtop machines or other handheld devices. Type 2 is 5mm thick and I/O capable. These are used for I/O devices such as modems and network adapters. Type 3 is 10.5mm thick and used mainly for add-on hard drives. When buying PC Card equipment, you must consider the size of the slot. In most cases, Type 3 can handle Type 2 and Type 1.

AGP

Accelerated Graphics Port (AGP) The newest type of bus created for the high demands of 3D graphical software. Since AGP is a hot topic and there is much more   to learn about it, a bit more on the subject:

Today's software is increasing in graphics intensity. Even "mundane" business software uses icons, charts, animations, etc.  When you add 3D games and educational software to the equation, you can see that there is a crunch in bandwidth for graphical information.   With newer software and games getting more and more graphics intensive, the PCI bus is maxed out.   In fact, the PCI bus, once considered very fast, can now be considered a bottleneck.

Intel knew this.  In response, they designed the Accelerated Graphics Port, or AGP. Intel defines AGP as a "high performance, component level interconnect targeted at 3D graphical display applications and is based on a set of performance extensions or enhancements to PCI."   In short, AGP uses the main PC memory to hold 3D images. In effect, this gives the AGP video card an unlimited amount of video memory. To speed up the data transfer, Intel designed the port as a direct path to the PC's main memory.

AGP sounds groundbreaking, and it is, no doubt, the latest craze in the need for graphical speed.  One reason is, cause it is faster than PCI is , while PCI runs at 33MHz, the AGP bus runs much faster.  A 4X AGP bus runs at 4 times 33MHz, or 133MHz!    Also, a normally clocked PCI bus can achieve a throughput of 132MB/s. Yes, this is fast, but when compared to the throughputs of 3D games, one finds that it is not enough.   AGP, running in 2x mode (2 x 33 = 66MHz), can achieve a throughput of 528MB/s!   AGP pulls this off by constantly transferring data on both the rises and falls of the 66MHz clock cycle.   Also, AGP makes use of sideband transfers and pipelining so it can constantly transfer data without depending on other components in the PC. 

The pipelining ability of the AGP bus is a key point that explains why it provides a performance advantage.  Since AGP pipelines operations ,it can process quicker and more efficiently than a PCI bus can.   AGP uses a special organization process for all pending and processing requests.   In effect, the bus can process one instruction while still receiving the next set of instructions.   This allows much more to be accomplished in a shorter amount of time.

One can easily see why the need for a new graphical interface is needed.  While PCI served us well, and still continues to do so, it is bogged down by the demand of full screen 3D graphics.  It works great for 2D business software and most games, but intense 3D challenges the bandwidth limitations.  For true 3D, there is much more information that must be transferred for a single image.

AGP, as stated above, uses the main PC memory to store all 3D information, including
textures and the Z-buffer. This rids us of a prime problem of PCI video.  Textures add
reality to what we see on screen.  The Z-buffer creates an illusion of depth.    Both of these take up loads of memory, and they use the same chunk of memory.    Therefore, manufacturers were forced to choose between textures or the z-buffer.   Often, they had to design software that was weak in both areas in order to deal with the PCI bus.   With AGP, this restriction is gone.

To create lifelike 3D images, the CPU must perform intensive 3D calculations. The graphics controller processes the texture data and bitmaps. In many cases, the controller must read elements from 7 or 8 different textures and average them into a single pixel on the screen. When this calculation is performed, the pixel must be stored in the memory buffer.    Because these textures are so large, they cannot be stored on the video card's buffer.  With AGP, they instead are stored in the main system memory.   Because of this, it is recommended that you have a large amount of system memory in your machine. This should be no problem due to the low prices of RAM.   Intel, no doubt, took this into account when they decided to use your RAM for graphics.

To access the texture data from the main memory, AGP uses a technique called Direct
Memory Execute, or DIME. In short, this connects the memory directly to the AGP/PCI
chipset. This lets the graphics card access the textures in the main memory, which is
limited only by the amount of memory you have in your system.

Like PCI, AGP uses a 32-bit connector.   But, there is a difference. The AGP connector has 64 contacts, just like the old MCA adapter.   AGP uses a 64-bit wide data path.  This extra contact provides new roadways for the pipelining and queuing of data requests.  Another difference is that AGP uses an extra eight sideband address lines that allows the controller to issue simultaneous commands while also accessing all 32 of the main data pathways.   This is called Sideband Addressing, or SBA. All this comes together to give AGP a faster throughput then PCI.    In order to use AGP with the P2, you must have a motherboard with Intel's 440LX or BX chipset.  All such boards offer SDRAM support, an absolute must have for AGP.   If you want to use AGP with a Socket 7 processor, you'll find yourself using chipsets like the Via Apollo VP3 and the ALI Alladin V.

AGP also requires software support, including both the OS and graphics drivers. Windows 95 and NT4 can be modified to support AGP, but Windows 98 has built-in support.  NT5 will have built-in support for AGP.  Windows 95 users can get the Windows 95 OEM Service Release 2.1 or a patch program called USBSUPP.EXE. Your current Win95 PCI device driver will support AGP, but you will need to get DirectX5, which is the only version of DirectX to support DIME.  You must make sure your video drivers include VGARTD.VXD as well.  This is a virtual device driver that turns on the DIME feature. 

Most mainstream graphics card vendors have produced AGP versions of their PCI based video cards.  Among these are ATI, Diamond, Matrox, NVidia, STB, and Number Nine.   These AGP cards are not always all they are cracked up to be.  Each vendor implements a different set of 3D instructions and effects.   Some vendors implement these effects through software, a practice that negatively effects performance.

Often, these AGP cards come with a large amount of video memory. With 4 MB as the minimum, and 8 MB being more standard.  Some offer support up to 16 MB of RAM. This video memory gives a large amount of space for texture storage.

The Bottom Line

At this point in the development of AGP, It's not recommend to be going out of your way to upgrade your system to AGP. The benefits over PCI video are definitely staggering, but the hardware required to support it in the first place can be rather staggering to the typical PC user still using a low-end Pentium PC.

Reviews indicate that many AGP video boards do not perform significantly better than their PCI counterparts. With some boards, AGP makes no difference at all. But, the reviews sometimes do not make mention of the fact that, like MMX, AGP requires software to actually take advantage of it. Without proper software support, AGP will provide little improvement.