CPU terminology explained
- Also known as a semiconductor fabrication plant, this is where all of a semiconductor’s electronic components are interconnected onto a single die of silicon.
- Very Large Scale Integration: the process of placing hundreds of thousands (between 100,000 and one million) of electronic components on a single chip. Nearly all modern chips employ VLSI architectures, or ULSI (ultra large scale integration).
- Ultraviolet Light has very short wavelengths and is just beyond the violet end of the visible spectrum. It is used to expose patterns on the layers of the microprocessor in a process much like photography.
- Ultra Large Scale Integration: more than one million transistors on a chip.
- A device used to amplify a signal or open and close a circuit. In a computer, it functions as an electronic switch, or bridge. The transistor contains a semiconductor material that can change its electrical state when pulsed. Invented in 1947 at Bell Labs, transistors have become the key ingredient of all digital circuits, including computers.
- Translation Lookaside Buffer: a small piece of associative memory within a processor which caches part of the translation from virtual addresses to physical addresses. Such translations can often be very large and complex and the data structures that implement them too large to store efficiently on the processor. Instead, a few elements of the translation are stored in the TLB, which the processor can access extremely quickly. If a required translation for a particular virtual address is not present in the TLB the address will be resolved using the more general mechanism. Also referred to as Address Translation Cache.
- A CPU architecture that allows more than one instruction to be executed in one clock cycle. Processors can do this by fetching multiple instructions in one cycle, deciding which instructions are independent of other instructions, and executing them.
- A family of high-performance RISC-based microprocessors from Intel. StrongARM chips are used in handheld devices such as PDAs and palmtops. The StrongARM technology was jointly developed by Digital Equipment Corporation and Advanced RISC Machines (ARM). In 1997, Intel acquired Digital’s chip manufacturing facilities and continues to make the Alpha and StrongARM chips.
- Silicon-On-Insulator: silicon wafer with a thin layer of oxide – into which integrated circuits are built – buried in it. SOI substrates achieve superior isolation between adjacent devices in CMOS devices.
- Symmetric Multiprocessing: a computer architecture that provides fast performance by making multiple CPUs available to complete individual processes simultaneously (multiprocessing). Unlike asymmetrical processing, any idle processor can be assigned any task, and additional CPUs can be added to improve performance and handle increased loads.
- Codename for AMD’s x86-64 design for extending the iA-32 architecture to support 64-bit code and memory addressing.
- Intel uses wafers of pure silicon cut from a silicon ingot to make microprocessors. Silicon, the primary ingredient of beach sand, is a semiconductor of electricity. Semiconductors are materials that can be altered to be either a conductor or an insulator.
- Grown on a wafer during chip fabrication to serve as an insulating layer.
- A large, cylindrical, single crystal made from purified silicon. The cylinder is sliced into thin wafers which are used for making computer chips.
- A solid-state substance with conductive properties that can be altered with electricity. Silicon performs as a semiconductor when chemically combined with other elements. A semiconductor is also halfway between a conductor and an insulator. When charged with electricity or light, semiconductors change their state from non-conductive to conductive or vice versa. The most significant product built from a semiconductor is the transistor.
- Single Edge Connect: a form of processor packaging first used by Intel’s Pentium II CPU. Comprising a hardware module (cartridge) that contains the CPU and an external L2 cache, it plugs into a socket (Slot 1, Slot 2, etc.) on the motherboard which bears greater resemblance to a bus slot than an individual chip socket.
- Reduced Instruction Set Computer: a microprocessor architecture that recognises a relatively limited number of instructions, favouring the speed at which individual instructions execute over the richness of the instruction set. See also CISC.
- An electronic component that resists the flow of current in an electronic circuit.
- The Registers are a mini-storage area for data used by the Arithmetic Logic Unit (ALU) to complete the tasks the Control Unit has requested. The data can come from the data cache, main memory or the control unit and are all stored at special locations within the Registers. This makes retrieval for the ALU quick and efficient.
- The Prefetch Unit decides when to order data and instructions from the Instruction Cache or the computer’s main memory based on commands or the task at hand. When the instructions come in the most important task for the Prefetch Unit is to be sure all the instructions are lined up correctly to send off to the Decode Unit.
- Conductive material used as an interconnect layer on a chip.
- A category of techniques that provide simultaneous, or parallel, processing within a CPU. It refers to overlapping operations by moving data or instructions into a conceptual pipe with all stages of the pipe processing simultaneously. For example, while one instruction is being executed, the computer is decoding the next instruction.
- A material which becomes soluble when exposed to ultraviolet light. Used to help define circuit patterns during chip fabrication where it prevents etching or plating of the area it covers; also called resist.
- The process of reproducing the chip’s circuitry pattern onto the wafer surface by using ultraviolet light and stencils or masks to transfer the image photomechanically.
- Pin Grid Array: a square chip package, either ceramic or plastic, with a high density of pins (typically 200 pins can fit in 1.5in square). In an SPGA (Staggered PGA), the pins are staggered and do not line up in perfect rows and columns.
- The Intel codename for the original 60/66MHz Pentiums introduced in 1993. Subsequent faster clock-speed chips were referred to as P54 and the MMX version as P55.
- The Intel codename for the Pentium Pro, which is optimised for 32-bit applications. The P6 generation includes the Pentium Pro and Pentium II.
- A user -installable microprocessor from Intel for the 486 microprocessor. Many PCs 486-based PCs were built with an “OverDrive” socket, which allowed a processor upgrade simply by inserting a faster OverDrive chip.
- Intel’s pop-out CPU packaging designed for mobile processors which includes an integrated L2 cache, introduced with Mobile MMX processor launched in early 1998.
- MultiMedia eXtensions: Intel’s upgraded Pentium processor which incorporates additional instructions designed specifically for processing multimedia data more efficiently and a larger 32Mb on-board cache. Codenamed P55C.
- Metals, such as aluminium and copper are used to conduct the electricity throughout the microprocessor. Gold is also used to connect the actual chip to its package.
- This is the big store house of data located within the main computer outside of the microprocessor. At times the Main Memory may send in data or instructions for the Prefetch Unit, which often get stored at an address in the Instruction Cache to be used later.
- Large Scale Integration: refers to the placement of thousands (between 3,000 and 100,000) of electronic components on a single integrated circuit. VLSI (Very Large Scale Integration) is between 100,000 and one million transistors on a chip.
- A collection of transistors and resistors that implement Boolean logic operations on a circuit board. Transistors make up logic gates. Logic gates make up circuits. Circuits make up electronic systems.
- Katmai New Instructions: the 70 new Single Instruction Multiple Data (SIMD) instructions supported by the Pentium III – formerly codenamed Katmai – which came to market in the spring of 1999 designed to optimise the performance of multimedia and graphics applications.
- Brand name for the first product in Intel’s IA-64 family of processors, formerly codenamed Merced.
- Instructions Per Clock: a measure of how many instructions a CPU is capable of executing in a single clock. Since different processor architectures have different IPCs, clock frequency x IPC is a much truer measure of processor performance than clock frequency alone.
- Atoms or molecules that have a net electrical charge. In semiconductor manufacturing, ions are the source of chemical impurities that alter the conductivity of silicon.
- The Instruction Cache is a warehouse of instructions right on the chip, so that the microprocessor doesn’t have to stop and look in the computer’s main memory for instructions. This quick access makes processing fast as instructions are “fetched” to the Prefetch Unit where they are put in the proper order for processing.
- Integrated Circuit: a tiny complex of electronic components and their connections that is produced in or on a small slice of material (such as silicon). Its name results from the integration of previously separate transistors, resistors and capacitors – all on a single chip.
- Next generation 64-bit architecture made up of the 64-bit Instruction Set Architecture (ISA) jointly developed by HP and Intel and an IA-32 compatibility component. IA-64 supports 32-bit and 64-bit environments, and provides compatibility with PA-RISC and IA-32.
- Intel Architecture 32-bit. Intel’s 32-bit architecture, also known as x86. IA-32 chips span the early 1990s Intel 486 series to the seventh-generation Intel Pentium 4 and AMD Athlon chips. See also IA-64.
- Frontside bus: the bus within a microprocessor that connects the CPU with main memory. The so-called dual independent bus (DIB) architecture allows a processor to use both this and the backside bus (which connects the CPU and the Level 2 cache) simultaneously.
- Floating Point Unit: a formal term for the math co-processors (also called numeric data processors, or NDPS) found in many personal computers. FPUs perform certain calculations faster than CPUs because they specialise in floating-point math, whereas CPUs are geared for integer math. Today, most FPUs are integrated with the CPU rather than packaged and sold separately.
- A process using a chemical bath (wet etch) or a plasma (dry etch) that removes unwanted substances from the wafer surface.
- Digital Signal Processor: a microprocessor-like device designed to process analogue to digital (& vice-versa) data streams. DSPs are used for a variety of devices in personal computers, including high-speed modems, multimedia sound boards, and real-time audio/video compression and decompression hardware.
- The introduction of an element that alters the conductivity of a semiconductor. Adding boron to silicon will create a P-type (more positive) material, while adding phosphorus or arsenic to silicon will create N-type (more negative) material.
- An electronic component that acts primarily as a one-way valve.
- The formal term for an area of silicon containing an integrated circuit. A die has many layers, each designed for a specific function, and refers to a semiconductor component or part that has not yet been packaged. The popular term for a die is chip.
- Substance that is a poor conductor of electricity and will sustain the force of an electric field passing through it. Also called an insulator.
Depleted Substrate Transistor
- A new generation of transistor built into an SOI substrate in such a way that its active silicon layer between source and drain is fully depleted so as to create a high conductivity channel.
- The Decode Unit does just that – it decodes or translates complex machine language instructions into a simple format understood by the Arithmetic Logic Unit (ALU) and the Registers. This makes processing more efficient.
- The Data Cache works very closely with the “processing partners”, the ALU, Registers and the Decode Unit. This is where specially labelled data from the Decode Unit are stored for later use by the ALU and where final results are prepared for distribution to different parts of the computer.
- Central Processing Unit: a formal term for the microprocessor chip that powers a personal computer. The Intel Pentium chip is one example of a CPU. The term sometimes also refers to the case that houses this chip. See also FPU.
- The Control Unit is one of the most important parts of the microprocessor because it is in charge of the entire process. Based on instructions from the Decode Unit, it creates control signals that tell the Arithmetic Logic Unit (ALU) and the Registers how to operate, what to operate on, and what to do with the result. The Control Unit makes sure everything happens in the right place at the right time.
- Boosts CPU performance by increasing the internal CPU clock, while maintaining the same I/O speed (for compatibility).
- The number of pulses emitted from a computer’s clock in one second; it determines the rate at which logical or arithmetic gating is performed in a synchronous computer.
- Pronounced “sisk” and standing for Complex Instruction Set Computer, this relates to a microprocessor architecture that favours the richness of the instruction set (typically as many as 200 unique instructions) over the speed with which individual instructions are executed. See also RISC.
- An electronic component that holds a charge.
- The Bus Unit is the place where instructions flow in and and out of the microprocessor from the computer’s main memory.
- Named after the nineteenth-century mathematician George Boole, Boolean logic is a form of algebra in which all values are reduced to either TRUE or FALSE. Boolean logic is especially important for computer science because of its suitably for use with the binary numbering system, in which each bit has a value of either 1 or 0. Another way of looking at it is that each bit has a value of either TRUE or FALSE.
- Ball/Column Grid Array: a specification for pin lay-outs on micro chips, such as those used on a CPU chip.
- A dedicated channel between the CPU and a Level 2 cache. The so-called dual independent bus (DIB) architecture allows a processor to use both this and the frontside bus (which connects the CPU with main memory) simultaneously. See also Frontside Bus.
- Arithmetic and Logic Unit: the smart part of a processor chip that performs commands like adding, subtracting, multiplying and dividing. It also knows how to read logic commands like OR, AND, or NOT. Messages from the Control Unit instruct the ALU what to do and then it takes data from its close companion, the Registers, to perform the task.
- Flip Chip Plastic Grid Array: a micro CPU package, for socketable boards, consisting of a die placed face-down on an organic substrate. The package uses 478 pins, which are 2.03 mm long and .32 mm in diameter.
- Flip Chip Ball Grid Array: a micro CPU package for surface mount boards consisting of a die placed face-down on an organic substrate. Instead of using pins, the package uses small balls, which acts as contacts for the processor. The advantage of using balls instead of pins is that there are no leads that bend. The package uses 479 balls, which are .78 mm in diameter.
- An industry standard high-speed, high-performance, point-to-point connection method for integrated circuits pioneered by AMD. It initially allowed for connection speeds of up to 6.4GBps.
- The brand name given to AMD’s family of “extra performance” processors, dating from the time of the Athlon CPU’s transition from the Thunderbird core to the Palomino core in 2001.
- Extended Memory 64 Technology: an enhancement to Intel’s IA-32 architecture which allows a processor to run newly written 64-bit code and access larger amounts of memory when used with a 64-bit OS and application. These extensions do not run code written for the Intel Itanium processor.
- Enhanced Intel Speed Step technology: an enhanced version of Intel’s Speed Step technology which dynamically scales the speed of a processor between its default clock setting and a minimum speed, based on how much CPU horsepower is needed at that moment, so as to both reduce power consumption and heat.
- An Intel technology – also referred to as virtualisation technology – that allows a single CPU to run more than one operating system at any given time.
- Support for so-called “No Execute” or NX technology is a joint venture hardware/software mechanism to defend against buffer overruns and consequent vulnerability to virus attack. AMD calls this technology “Enhanced Virus Protection”, while Intel refers to this functionality as the “Execute Disable Bit”.