Model – The actual product model declared by the manufacturer. Includes the name of the brand, series, article, has a certain set of characteristics.
Socket – A socket is the connector into which the CPU is installed. The socket model is the first thing to look for when choosing a motherboard and a cooling system for the processor.
For a gaming computer – Power is important for a gaming machine. In this regard, it is recommended to choose processors from the highest cell – Intel Core i5 and Core i7. For the game version, processors that have the letters S and T in the markings are not desirable. They are more economical, but because of this performance suffers.
Type of delivery – The characteristic indicates the type of delivery of this processor. It can be an OEM processor shipped in lightweight packaging, without a cooling system, a BOX processor comes in a branded box, as well as a cooling system for it.
Cooling system in the set – Often the processors are equipped with a regular cooling system, which was originally designed for the thermal package of a particular CPU, which undoubtedly facilitates the life of computer assemblers. True, this option has a couple of significant drawbacks. First, this is a high level of noise in comparison with individual branded SOs, as well as restrictions on overclocking, because overclocking increases heat generation, and standard SD for such bullying is not adapted!
Type – The processor is the central part of the computer that performs the information conversion programs specified by the program and controls the entire computing process.
Ruler – In the line, you can first determine whether the processor belongs to a certain performance class and price range.
Kernel and architecture:
Cache volume L3 – L3 cache is the least fast, but it can be very large – more than 24 MB. L3 is slower than previous caches, but still much faster than RAM. In multiprocessor systems is in common use and is designed to synchronize data of various L2 ..
Core – The core is the part of the processor responsible for executing one sequence of commands; Accordingly, the presence of several cores allows the CPU to work simultaneously with several tasks, which has a positive effect on performance.
Usually the nuclei are an even number; the tri-core architecture is relatively rare and is the exception, and the single-core chips are almost completely out of use. In desktop processors, 2 cores are usually typical for budget models and low-cost middle-class solutions, 4 for the average level, and 6 or more for the advanced, including processors for servers and workstations. At the same time, we note that the actual capabilities of the CPU depend not only on the number of cores, but also on a number of features and technological tweaks: for example, Hyper-threading technology can significantly improve performance in comparison with similar models.
Architecture – The characteristic indicates the architecture of this processor.
The architecture of the processor is a set of properties and qualities inherent in a whole family of processors (in other words – the internal design, the organization of these processors).
L1 cache (data) – L1 cache volume is from 8 to 384 KB.
The value of the first level cache. A first-level cache memory refers to a high-speed memory block located directly on the processor core. In this block, the information extracted from the main memory is copied. Maintaining the main commands makes it possible to increase the processor’s performance due to the faster processing speed. The volume of the cache memory of the first level is small, it is calculated by kilobytes. The “older” processor families usually have a large amount of L1 cache.
For multi-core processor models, the first-level cache value is specified for a single core.
Maximum number of threads – A characteristic specifies the maximum number of concurrently executing computation flows supported by this processor.
L1 cache (instructions) – The characteristic specifies the amount of cache memory of the first level, this processor.
Cache memory of the first level is a block of high-speed memory, located directly on the core of the processor. It copies data extracted from the main memory. Maintaining the basic commands allows you to increase the performance of the processor due to the higher processing speed (processing from the cache faster than from the RAM). The capacity of the cache memory of the first level is small and is estimated in kilobytes.
The L1 level cache is always divided into a data cache (L1D) and instruction or instruction cache (L1I). This is the so-called Harvard processor architecture. The L1 cache always belongs only to a particular processor core.
Number of cores – The more, the better! The performance of the processor in multi-threaded applications depends not only on the architecture, frequency and size of the cache, but also on the number of cores.
Technological process – In the production of semiconductor elements, photolithography technologies are used. The resolving power of photolithographic equipment determines the name of a specific technological process. The lower the value, the more perfect the process technology is used. Reducing the process technology is necessary for the creation of thinner transistors, which makes it possible to increase the density and complexity of integrated circuits, and thereby create more efficient microchips with lower power consumption.
Cache volume L2 – The characteristic specifies the amount of cache in the second level of the given processor.
The L2 cache is a high-speed memory unit that performs the same functions as the L1 cache, but has a lower speed and a larger capacity. The larger the L2 cache, the better.
Frequency and overclocking:
Multiplier – The multiplication factor, or multiplier, determines the clock speed of the central processor by multiplying the specified number by the frequency of the clock.
Free multiplier – The free processor multiplier allows you to change its clock frequency by standard means of the motherboard and chipset. The presence of a free multiplier is necessary for overclocking the processor.
Maximum frequency in turbo mode (MHz) – As you know, not all applications are well optimized for multi-core processors, so manufacturers have recently started equipping their CPU with an interesting and very useful technology – dynamically increasing the processor frequency depending on the load within a specific heat package.
This means that if the program requires only one or two cores, when your CPU has four of them, then due to the fact that the processor will be loaded only by half, the frequency of working cores increases up to a certain threshold of energy consumption, which is equal to the power consumption at full load all four cores. Due to this, you can significantly improve performance in programs with limited support for multithreading.
Base Processor Frequency (MHz) – The higher the frequency, the higher the CPU performance. This is true only for a certain manufacturer and a specific line (as well as architecture) of the processors.
Memory type – The characteristic specifies the type of RAM the processor is working with.
The RAM of the computer is of the DRAM type, a volatile random access memory. DRAM is divided into subtypes (different versions of DDR memory), which differ in both the connector and the data transfer rate (with each generation the speed increases).
Maximum supported memory capacity – A characteristic specifies the maximum amount of RAM in the configuration of the computer with which the processor is running.
The amount of RAM depends on how quickly the processor will be able to process the intermediate data when running programs that are stored in RAM.
Number of channels – A memory controller built into the processor usually supports several 64-bit channels.
Maximum RAM frequency – The characteristic specifies the maximum frequency of RAM that this processor can interact with.
The frequency of RAM is one of the main parameters, and the more it is, the higher the performance.
Minimum RAM frequency – The characteristic specifies the minimum frequency of RAM that this processor can interact with.
The frequency of RAM is one of the main parameters, and the more it is, the higher the performance.
Maximum body temperature – The maximum temperature of the processor case at which the processor will remain operational.
Heat Dissipation (TDP) – Any processor emits a lot of heat during operation and in principle can not do without a cooling system. In order for the collector to correctly select the cooling system, a “heat package” specification was introduced reflecting the maximum heat release of the core at maximum load at standard operating modes specified by the manufacturer (core voltage, core frequency).
Maximum frequency of the graphics core – The clock speed of the video core of the central processor is from 300 to 1350 MHz.
The frequency of the core indicates the frequency with which its simplest element, the transistor, is switched (that is, how quickly it changes its state). If the frequency of the video card is 1100 MHz, then the switching speed of the transistor will be 1100 million times per second, respectively.
GPU model – The actual GPU model.
Embedded graphics solutions are weaker than discrete ones, but they are great for everyday tasks and simple 3D applications.
Integrated graphics core – Almost all modern CPUs are equipped with integrated graphics cores. They consume less energy, in contrast to discrete graphics cards, and are much less heated. If you are not interested in computer games, you can do without a discrete video card, thus saving a decent amount!
Bus and controllers:
Bus bandwidth – The data bus bandwidth (measured in bits per second) is equal to the product of the bit width (measured in bits) and the bus frequency (measured in Hz = 1 / s).
System bus – A bus with which the processor is connected to the rest of the system components. The frequency of the system bus, supported by the processor, is actually the clock frequency at which data is exchanged between the processor and the rest of the system.
This parameter is the key for determining the total clock frequency of the CPU (see above): this frequency equals the frequency of the system bus multiplied by the multiplier (see below).
Number of PCI Express lines – The characteristic indicates the number of PCI Express lines that this processor can communicate with. Today, PCI Express has almost every new computer, it is used to connect a video card, both built-in and external. The PCI Express interface is based on a serial point-to-point protocol. That is, PCI Express requires a relatively small number of conductors. But the interface uses much higher clock speeds compared to parallel buses, which gives high bandwidth. In addition, the bandwidth can be easily increased by linking several PCI Express lines together. The most common types of slots are x16, x8, x4, x2 and x1, where the numbers indicate the number of PCI Express lines.
Built-in PCI Express controller – The characteristic indicates whether or not a built-in PCI Express controller is installed in this processor.
The built-in PCI Express controller significantly reduces delays and improves performance with devices connected via the PCI Express interface.
Teams, instructions, technologies:
Technology to increase the frequency of the processor – The characteristic indicates the technology of increasing the frequency, implemented in this processor.
Such technologies increase the processor’s performance at peak loads, automatically accelerating the processor core to a frequency higher than the base one, if the power, current consumption and temperature do not exceed the maximum values.
A set of instructions and commands – Enumeration of instructions and commands supported by this server processor.
A set of instructions and commands is an agreement about the programming tools provided by the architecture of this processor, namely: certain types of data, instructions, register system, addressing methods, memory models, methods for handling interrupts and exceptions, input and output methods.
Energy-saving technology – The characteristic indicates the energy-saving technology implemented in this processor. All modern processors (both Intel and AMD) and motherboards support technologies that allow to reduce power consumption and, consequently, dissipated heat output.
So, in the case of Intel processors, this technology is called Enhanced Intel SpeedStep (EIST), and for the AMD processors – CooPn’Quiet. In order to reduce the heat dissipation of the processor and its power consumption, you need to dynamically change its clock frequency depending on the load.
Enhanced Intel SpeedStep technology determines the use of several possible supply voltages and frequencies (collectively, work points). This allows for a better voltage / frequency ratio and a more efficient mode of operation when the performance is consistent with the workload.
Multithreading – Multithreading is the technology or concept of multi-threaded software designed to improve system performance in specially optimized applications. Roughly speaking, one core is represented as two virtual processors, so even in the operating system’s task manager, one kernel is represented as two. However, this does not mean that the performance will also grow twice – it all depends on the level of software optimization.
Support for 64-bit instruction set – The characteristic indicates the presence of support for a 64-bit instruction set in this processor.
Processors with 64-bit architecture can work with both old 32-bit applications and 64-bit ones, which are becoming more popular lately. Processors with 64-bit addressing support work with more than 4 GB of memory, which is not available in traditional 32-bit CPUs. To take advantage of 64-bit processors, you need to have your operating system adapted to them.
Implementation of 64-bit extensions in AMD processors is called AMD64, in models from Intel – EM64T.
Virtualization Technology – Virtualization Technology Emulates Multiple Computers