qué es, cómo es y para qué sirve
Vamos a explicarte qué es exactamente la CPU de un ordenador y para qué sirve. Se trata de un componente tremendamente complejo, no obstante es el cerebro de cada ordenador o dispositivo, pero vamos a intentar explicarlo todo de una forma lo más sencilla posible para que puedas entender los conceptos básicos de su funcionamiento.
Sobre una CPU se podría decir bastante más de lo que vamos a contarte, ya que se puede profundizar mucho en las características de su funcionamiento. Sin embargo, lo que queremos aquí es que si nunca has oído hablar de ella puedas entender sus fundamentos básicos para por lo menos saber exactamente de qué estamos hablando.
Qué es la CPU
CPU son las siglas de Central Processing Unit, lo que traducido significa Unidad Central de Procesamiento. Se trata de uno de los componentes vitales que te vas a encontrar en tu ordenador, tu smartphone o tu tableta o portátil, e incluso en relojes y prácticamente cualquier dispositivo electrónico. Sin él, simplemente no podrían funcionar.
A la CPU se la suele llamar coloquialmente como microprocesador o simplemente procesador, y puedes considerarla como el cerebro de cualquier dispositivo. Se encarga de procesar todas las instrucciones del dispositivo, leyendo las órdenes y requisitos del sistema operativo, así como las instrucciones de cada uno de los componentes y las aplicaciones.
Así pues, igual que el cerebro, la CPU es la que se encarga de que todo funcione correctamente, y de interpretar todo lo que quiere hacer el sistema operativo o los componentes, estableciendo las conexiones y realizando todos los cálculos precisos para que funcione. Cuanto más potente sea el procesador, más rápido podrá hacer las operaciones y más rápido funcionará tu dispositivo en general.
Es también el encargado de leer, interpretar y procesar las instrucciones primero del sistema operativo, y después de los programas o aplicaciones que tienes instalados en el ordenador.
Cuando abres Word o tu navegador, este tiene que cargarse, y para ello necesitar realizar una serie de pasos cuyas instrucciones le manda al procesador. Cada acción que hagas luego dentro, también enviará instrucciones para poder ejecutarse y realizarse.
Tu ordenador o portátil, tendrán una CPU dedicada para realizar todas las operaciones y procesos que requiera el sistema. Mientras, los móviles tienen el denominado SoC o System on Chip, que es un chip que contiene tanto la CPU como otros componentes, como chips gráficos integrados. Con ello, los SoC suelen poder hacer más que las funciones de las CPU estándar.
Cómo es la CPU
La CPU suele ser pequeña, con forma cuadrada, y tiene conectores y pines metálicos. Por lo general, suele estar instalada en la placa base de un ordenador, en un zócalo especial que puede estar en diferentes sitios dependiendo de la placa base. El procesador no va soldado a la placa, por lo que puedes cambiar la CPU de un ordenador por otra si quieres.
La CPU tiene un mínimo de un procesador o núcleo en su interior, un chip para realizar los cálculos y operaciones. Al principio todos tenían sólo uno, pero con el tiempo las CPU han ido aumentando el número de estos núcleos, y hay diferentes maneras de llamarles dependiendo de este número.
Por ejemplo, si alguna vez has escuchado que un procesador es Dual-Core o de doble núcleo, es porque tiene dos de estos chips en su interior. Luego evolucionaron a cuatro, los Quad-Core, y se han ido aumentando con seis, ocho o incluso 12 núcleos o más. Todos ellos se identifican por ese nombre que incluye la palabra Core, pero antes se le pone una referencia al número de núcleos.
Que la CPU tenga varios procesadores o núcleos suele implicar que es capaz de realizar más operaciones a la vez, por lo que su funcionamiento puede ser más rápido y mejor. Sin embargo, esto no siempre quiere decir que cuantos más núcleo haya más rápido será todo, ya que esto es algo que también depende de cómo sean cada uno de los núcleos.
Por ejemplo algunos procesadores de Intel y AMD, los dos principales fabricantes, tienen tecnologías que les permiten ejecutar dos hilos de ejecución en un único núcleo, que son las secuencias de instrucciones pertenecientes al sistema operativo o aplicación. Los hilos de ejecución, o threads en inglés, son las secuencias de instrucciones pertenecientes al sistema operativo o una aplicación.
¿Y esto qué quiere decir? Pues que en los procesadores que tengan esta tecnología, cada núcleo físico puede actuar como si fuera dos núcleos virtuales, ya que cada uno puede trabajar con dos hilos o secuencias de instrucciones diferentes.
Y luego está la frecuencia del procesador, que se marca en gigahercios. La frecuencia, también llamada el reloj, es la que marca el ritmo con el que trabaja el procesador. Un hercio es una acción que se repite una vez por segundo, y así subiendo, hasta el Gigahercio, que son 1.000.000.000 de acciones por segundo.
El ritmo de trabajo de un procesador viene marcado por la capacidad de realizar cuantas más acciones por segundo mejor, ya que cada operación de una aplicación puede requerir varios ciclos de reloj o acciones.
En definitiva, ya no sólo importa el número de procesadores, sino cómo trabaja cada uno. Cuantos más GHz tenga un procesador más rápido será. Esto, simplificándolo al máximo, significa que si tienes un procesador de cuatro núcleos a 1 GHz, este será más lento que un procesador de dos núcleos a 2,5 GHz.
Para qué sirve la CPU
La CPU de tu ordenador es la que ejecuta una secuencia de instrucciones y procesa los datos de las mismas. Estas secuencias de instrucciones son las que realizan los programas que tienes instalados en el ordenador. Vamos, que son los encargados de realizar las operaciones que necesitan los programas o aplicaciones para realizar las tareas que les pides que hagan.
Tú le dices a tu aplicación fotográfica que aplique un cálculo.
Esta aplicación le da al ordenador las instrucciones con los cálculos o acciones que necesita realizar para llevar a cabo la tarea. Las instrucciones se ponen en la RAM, donde son recogidas por la CPU, que es la que las lleva a cabo.
Para ello se siguen varios pasos diferentes. El primero es el de leer los datos e instrucciones para realizar cada una de las tareas de las aplicaciones. Los datos se decodifican y se dividen para que puedan ser entendidos por las diferentes partes de la CPU, las cuales ejecutan las acciones, realizan los cálculos, y escriben los resultados en la memoria principal, ya sea como resultado o para utilizarlos después en otras operaciones.
Pero esta es una simplificación extrema, ya que estas funciones básicas del procesador se realizan por cada instrucción, y cada programa puede requerir de varias instrucciones para realizar una acción. Además, estas no se realizan en orden y una detrás de otra, sino que el procesador puede estar trabajando con muchas funciones a la vez para los diferentes programas activos.
Tienes que entender que todo en tu ordenador está representado por números, por lo que la CPU es la encargada de recopilar y calcular las operaciones de estos números. Con estas operaciones se realiza exactamente todo lo que haces en tu ordenador, desde abrir el menú de inicio de Windows hasta ver un vídeo en YouTube, todo requiere una serie de instrucciones y cálculos para poder realizarse.
En los ordenadores modernos, la CPU no es la que lo realiza todo, ya que suelen interferir otros componentes especializados, como por ejemplo puede ser la tarjeta gráfica para todo lo relacionado con lo que ves en pantalla. También entra en juego la memoria RAM para almacenar datos de rápido acceso, y el resto de componentes que hay dentro de tu ordenador.
En este enjambre de dispositivos y componentes, la CPU es el cerebro o el director de orquesta, pues es quien lo organiza todo para que las tareas se hagan donde toca y de la manera que deben realizarse para obtener los resultados deseados.
La CPU no es más imprescindible que otros componentes básicos de un ordenador, ya que este no funcionará sin fuente de alimentación u otros componentes. Tampoco es el todo del que depende la ejecución de tareas, ya que para todo lo relacionado con la producción de vídeos o la visualización de videojuegos, también son necesarios tener una buena tarjeta gráfica o GPU y una buena memoria RAM.
Pero de la velocidad y potencia de la CPU sí que puede depender lo rápido que vaya un ordenador, y tener una buena o mala CPU afectará al rendimiento de tu equipo.
Coreinfo — Sysinternals | Microsoft Learn
Edit
Twitter
LinkedIn
Facebook
Email
-
Article -
- 4 minutes to read
-
By Mark Russinovich
Published: September 29, 2022
Download Coreinfo (531 KB)
Introduction
Coreinfo is a command-line utility that shows you the mapping between logical processors and the physical processor, NUMA node, and socket on which they reside, as well as the cache’s assigned to each logical processor.
It uses the Windows’ GetLogicalProcessorInformation function to obtain this information and prints it to the screen, representing a mapping to a logical processor with an asterisk e.g. ‘*’. Coreinfo is useful for gaining insight into the processor and cache topology of your system.
Installation
Extract the archive to a directory and then run Coreinfo by typing from that directory Coreinfo in the console on a 32 bit Windows version or Coreinfo64 for a 64 bit version.
For each resource it shows a map of the OS-visible processors
that correspond to the specified resources, with ‘*’ representing the
applicable processors. For example, on a 4-core system, a line in the
cache output with a map of shared by cores 3 and 4.
Usage: coreinfo [-c][-f][-g][-l][-n][-s][-m][-v]
| Parameter | Description |
|---|---|
| -c | Dump information on cores. |
| -f | Dump core feature information. |
| -g | Dump information on groups. |
| -l | Dump information on caches. |
| -n | Dump information on NUMA nodes. |
| -s | Dump information on sockets. |
| -m | Dump NUMA access cost. |
| -v | Dump only virtualization-related features including support for second level address translation. |
| (requires administrative rights on Intel systems). |
All options except -v are selected by default.
Coreinfo Output:
Coreinfo v3.03 - Dump information on system CPU and memory topology Copyright (C) 2008-2011 Mark Russinovich Sysinternals - www.sysinternals.com Intel(R) Xeon(R) CPU W3520 @ 2.67GHz Intel64 Family 6 Model 26 Stepping 5, GenuineIntel EM64T * Supports 64-bit mode VMX - Supports Intel hardware-assisted virtualization SVM - Supports AMD hardware-assisted virtualization HYPERVISOR * Hypervisor is present HTT * Supports hyper-threading SMX - Supports Intel trusted execution SKINIT - Supports AMD SKINIT EIST * Supports Enhanced Intel Speedstep NX * Supports no-execute page protection PAGE1GB - Supports 1GB large pages PAE * Supports > 32-bit physical addresses PAT * Supports Page Attribute Table PSE * Supports 4-MB pages PSE36 * Supports > 32-bit address 4-MB pages PGE * Supports global bit in page tables SS * Supports bus snooping for cache operations VME * Supports Virtual-8086 mode FPU * Implements i387 FP instructions MMX * Supports MMX instruction set MMXEXT - Implements AMD MMX extensions 3DNOW - Supports 3DNow! instructions 3DNOWEXT - Supports 3DNow! extension instructions SSE * Supports Streaming SIMD Extensions SSE2 * Supports Streaming SIMD Extensions 2 SSE3 * Supports Streaming SIMD Extensions 3 SSSE3 * Supports Supplemental SIMD Extensions 3 SSE4.1 * Supports Streaming SIMD Extensions 4.1 SSE4.2 * Supports Streaming SIMD Extensions 4.2 AES - Supports AES extensions AVX - Supports AVX intruction extensions FMA - Supports FMA extensions using YMM state MSR * Implements RDMSR/WRMSR instructions MTTR * Supports Mmeory Type Range Registers XSAVE - Supports XSAVE/XRSTOR instructions OSXSAVE - Supports XSETBV/XGETBV instructions CMOV * Supports CMOVcc instruction CLFSH * Supports CLFLUSH instruction CX8 * Supports compare and exchange 8-byte instructions CX16 * Supports CMPXCHG16B instruction DCA - Supports prefetch from memory-mapped device F16C - Supports half-precision instruction FXSR * Supports FXSAVE/FXSTOR instructions FFXSR - Supports optimized FXSAVE/FSRSTOR instruction MONITOR - Supports MONITOR and MWAIT instructions MOVBE - Supports MOVBE instruction PCLULDQ - Supports PCLMULDQ instruction POPCNT * Supports POPCNT instruction SEP * Supports fast system call instructions DE * Supports I/O breakpoints including CR4.
DE DTES64 - Can write history of 64-bit branch addresses DS - Implements memory-resident debug buffer DS-CPL - Supports Debug Store feature with CPL PCID - Supports PCIDs and settable CR4.PCIDE PDCM - Supports Performance Capabilities MSR RDTSCP * Supports RDTSCP instruction TSC * Supports RDTSC instruction TSC-DEADLINE - Local APIC supports one-shot deadline timer xTPR * Supports disabling task priority messages ACPI * Implements MSR for power management TM * Implements thermal monitor circuitry TM2 * Implements Thermal Monitor 2 control APIC * Implements software-accessible local APIC x2APIC - Supports x2APIC CNXT-ID - L1 data cache mode adaptive or BIOS MCE * Supports Machine Check, INT18 and CR4.MCE MCA * Implements Machine Check Architecture PBE * Supports use of FERR#/PBE# pin PSN - Implements 96-bit processor serial number Logical to Physical Processor Map: *--- Physical Processor 0 -*-- Physical Processor 1 --*- Physical Processor 2 ---* Physical Processor 3 Logical Processor to Socket Map: **** Socket 0 Logical Processor to NUMA Node Map: **** NUMA Node 0 Logical Processor to Cache Map: *--- Data Cache 0, Level 1, 32 KB, Assoc 8, LineSize 64 *--- Instruction Cache 0, Level 1, 32 KB, Assoc 4, LineSize 64 *--- Unified Cache 0, Level 2, 256 KB, Assoc 8, LineSize 64 -*-- Data Cache 1, Level 1, 32 KB, Assoc 8, LineSize 64 -*-- Instruction Cache 1, Level 1, 32 KB, Assoc 4, LineSize 64 -*-- Unified Cache 1, Level 2, 256 KB, Assoc 8, LineSize 64 --*- Data Cache 2, Level 1, 32 KB, Assoc 8, LineSize 64 --*- Instruction Cache 2, Level 1, 32 KB, Assoc 4, LineSize 64 --*- Unified Cache 2, Level 2, 256 KB, Assoc 8, LineSize 64 ---* Data Cache 3, Level 1, 32 KB, Assoc 8, LineSize 64 ---* Instruction Cache 3, Level 1, 32 KB, Assoc 4, LineSize 64 ---* Unified Cache 3, Level 2, 256 KB, Assoc 8, LineSize 64 **** Unified Cache 4, Level 3, 8 MB, Assoc 16, LineSize 64 Logical Processor to Group Map: **** Group 0
Download Coreinfo (531 KB)
CPU and GPU: how to choose the perfect pair
- Computer store
- Blog
- Articles on the topic: Computers (PC)
Article author: Sergey Koval
(koval@andpro.
ru)
Published: 17 September 2021
Choosing the perfect combination of processor and video card is a rather difficult and crucial moment, both when building a PC from scratch and when upgrading the system. Today we’ll talk about why you shouldn’t pair a mediocre processor with a powerful video card, and what criteria to use to select pairs.
Why not combine weak and strong components?
It is not easy to find information on the correspondence of video cards to processors on the network. Therefore, most purchases are made on a whim, on the advice of friends, or based on the opinion of a consultant in a store. It is not uncommon for one of the tandem devices to lag significantly behind in performance.
There are two options here:
- Processor power with a weak video card is simply idle. And investing in additional cores turns out to be a waste of money.
- A weak processor does not allow to reveal the full potential of the video card.
In addition, the load on the processor in such an assembly becomes excessive and can lead to overheating and throttling.
Thus, the main feature of every computer system: performance is limited by the weakest device in the assembly. And it doesn’t matter what kind of element it is — a video card, processor, RAM or power supply.
When it comes to apps, an unbalanced build can noticeably impact gameplay, cause stuttering, and result in FPS drops. But in production assemblies, depending on the type of activity, the requirements for the combination of graphics and CPU may differ. In this area, a combination of a multi-core top-end processor with a budget video card is allowed, as well as vice versa: a professional video card, for example, the Quadro line and a budget processor with four cores.
What exactly are the CPU and GPU responsible for?
Now we are talking about custom gaming equipment. This means that the functions of the CPU are the calculation of the parameters of the game world, the processing of the geometric parameters of static and dynamic objects, their movement, physics and deformation.
The more objects present in the frame, the higher the load on the processor. If the game uses artificial intelligence technologies, then support for this option also falls on the processor.
At the same time, the degree of loading of the central chip does not depend on the quality of graphics, detailing of objects, light and reflection parameters, textures and surface properties. Also, the image resolution selected by the user does not affect the processor load. This means that if the processor shows maximum load, and the FPS does not rise above the uncomfortable 30-40 frames, then setting the image resolution and image quality will not help the computer, except perhaps by changing the scene rendering distance.
Also, when starting the game, it should be taken into account that the central chip is loaded at any time not only by the game process, but also by many parallel ones, including background tasks and minimized applications. Therefore, you can slightly improve performance by first disabling some applications.
Be sure to close your browser and messengers before playing. It’s also best not to use the media player.
The tasks of video cards include drawing textures, creating shadows, rays, highlights, and reflections. In other words, visual features of game objects. Also, the video card smoothes the corners and imposes various effects on objects. Therefore, reducing the quality of game settings greatly facilitates the work of the video card.
How to identify the weak link?
To do this, you need any program that can monitor the operating parameters of computer nodes and display the results over the playing field. The most popular is the free MSI Afterburner application.
You can understand which device slows down the system by the load level, which is shown as a percentage. Just compare the indicators in the CPU and GPU lines and it will immediately become clear to you which node is loaded to the maximum and acts as a limiting factor.
How to choose a balanced pair?
For many years, when assembling a computer, users relied on a simple scheme: in one assembly, a video card should cost 2 times more than a processor.
However, starting from mid-2020, this principle has lost its relevance, because due to the lack of video cards, they have skyrocketed in value several times.
Therefore, the most accessible way will be to divide devices into categories:
For processors of the budget segment, you should select younger models in the lines of graphics adapters. For example, in terms of performance, the GeForce GT 730 video card and the Ryzen 3 1300X microchip correspond to each other. As a result, we get a good office assembly without gaming potential.
But we do not recommend taking video cards younger. For example, the performance level of the GeForce GT 710 is so low that the graphics cores built into the processor perform much better. So the best option would be to buy a processor with integrated graphics than a processor without a graphics core and a GT 710 video card. However, such a weak card also has its roles. It is able to replace a failed discrete video card for the duration of the repair or allow you to connect monitors if there is no video card in the system unit at all.
Average models in the lines will allow you to play undemanding games or more complex applications, it is on the minimum settings. Examples include the Radeon RX 550 and Core i7-7700K graphics cards.
Combining a Core i5-10400F or its rival Ryzen 5 3600 with a GeForce GTX 1660 Ti graphics card will make for a good gaming build.
But to play at maximum settings, flagship models are needed. For example, top octa-core processors from Intel demonstrate maximum performance with Radeon RX 6700 XT, GeForce RTX 3070 and GeForce RTX 3070 Ti graphics cards.
You can also use the compatibility tables, which are widely distributed on the network. When choosing a balanced pair of GPU and CPU, be sure to check the information from several sources. Pairs can differ significantly and there is no guarantee that you have come across reliable data. However, in most cases, the choice is limited to the availability of goods in the store, now this statement is especially true for video cards.
- All posts
- KVM equipment (equipment)
- Powerline adapters
- security
- Wireless adapters
- Power supplies
- Video cards (videocard)
- Video surveillance (CCTV)
- HDDs and Solid State SSDs
- Disk shelves (JBOD)
- Sound cards
- Instruments
- Uninterruptible power supplies (UPS, UPS)
- Cables and patch cords
- Switches
- Computer peripherals (computer peripherals)
- Computers (PC)
- Controllers (RAID, HBA, Expander)
- PC Cases
- PC Motherboards
- Multifunction devices (MFPs)
- Memory modules for PCs, laptops and servers
- Monitors
- Monoblocks (All-in-one PC)
- Desktop storage systems (NAS)
- Notebooks (notebook, laptop)
- General help
- Cooling
- Tablets
- Plotters
- Printers
- Software
- Software for corporate consumer
- Projectors
- Processors for PCs and servers
- Workstations (workstation)
- Power Distribution Unit (PDU)
- Consumables for office equipment
- Wi-Fi extenders (repeaters, repeaters)
- Routers (routers)
- Servers and server hardware
- Network cards
- Scanners
- Telecommunication cabinets and racks
- Telephony (phone)
- Thin Clients
- Transceivers
- Smart watch
Installing the processor on the motherboard: 3 easy steps
Installing the processor without the help of a wizard is not so difficult.
The main thing is to act carefully and slowly, and then this will save not only money, but also time that could be spent on a visit from a computer specialist. How and what to do correctly, you can find out below. 9Removing the old processor from the motherboard This is easy to do, but you have to be very careful.
In order to get to the processor, you must first remove other parts: the fan and heatsink. Most often they are removed together, but it happens that you have to extract them separately. Does the replacement algorithm consist of such steps?
- Unplug the power cord of the entire system unit from the socket.
- Place the unit on its side and open the cover.
- Gently pull the wires that go to the fan and processor out of the sockets (that is, remove the power).
- Remove cooler. If it is mounted on rivets, they just need to be turned, and if the bolts are visible, then they need to be unscrewed with a screwdriver. Then you need to loosen it slightly to stir up the thermal paste that might have stuck and remove it.
- Pull out the processor retention lever and carefully remove it. It may be that, in addition to the small lever, there is also a metal mounting plate, which also needs to be carefully pry off and removed.
No force is required during all these processes, otherwise parts may be damaged. Everything should move easily and without problems.
Related: Top 5 CPUs from Intel for NVidia GeForce GTX 1060
How to install CPU
Installing a new processor on the motherboard is a fairly simple procedure, but requires attention and care. First you need to purchase a processor with the right socket, that is, a special connector that must be fastened to the motherboard. Further, the question of how to properly install the processor on a new or old motherboard should not arise. The keys are visible on the CPU — small notches or legs that must fall into the socket, so putting it upside down simply will not work. So you need to carefully look at the details in order to understand exactly how the selected processor is installed on a specific motherboard, and do a few things:
- Move the release lever to release the socket.
- After docking the keys, gently and effortlessly put the processor in place.
- Close lever.
If the processor does not become, do not put pressure on it! It is necessary to pull it out and put it again so that all the notches match.
Interested to know: How to choose a motherboard for your computer?
Applying Thermal Grease
Once the processor is in place, apply thermal paste to the die. Often it is already applied or comes with a kit, but many advise to play it safe and remove the old layer and cover the crystal with a new one.
To do this, a small drop of paste is squeezed into the middle of the processor and spread evenly so that a thin layer is obtained. You don’t need to use a lot of it — this is not a jam sandwich: too thick a layer will have a bad effect on the computer.
You may also be interested: How to assemble a gaming computer, what components to buy: 3 options to choose from
How to install a cooler / radiator on a processor , therefore, it is advisable to consider how the cooling system is installed on motherboards of different parameters for each brand.
Intel
Intel coolers usually have a very simple mounting system — four snap-on legs. So, for example, installing an Intel Celeron G4920 processor on a motherboard in the LGA1151 socket will be like this:
- Find the connectors on the motherboard under the legs, insert them and close the latches. It is better to snap them diagonally to secure them evenly.
- Connect the wires to the power supply. They also have keys, so inserting them incorrectly will not work.
- If the wire is too long, then it is better to fix it somehow so that it does not accidentally get into the fan.
You can compare: Comparison of 2 budget gaming video cards — GTX 1050 vs GTX 1050 TI
AMD
AMD’s mounting system is also simple, only the latches are made in the form of loops or levers. In fact, you need to follow all the same steps as in the case of Intel, but with minor changes. For example, to fix the cooler when installing an AMD Athlon 200GE processor, you need:
- Find the cooler connector on the motherboard;
- Insert it;
- Put on the hinges/pull the levers;
- Connect the part to the power supply.
