The CPU is one of the most important components in a computer and should be carefully selected. In this article, we share how to do so and what you should pay attention to when buying a CPU.
What is a CPU?
Before we move on to the advice on choosing the right model, it is worth saying a few words, what exactly is a processor and what does it consist of?
The processor (or CPU – Central Processing Unit) is an integrated circuit responsible for most of the tasks performed by the computer. It takes data from memory, interprets it, and then executes the commands in it.
Modern processors are built of many cores (the more of them, the computer is able to perform more calculations at the same time), they have a cache, they can also be equipped with an integrated graphics chip. The most efficient CPUs are very complex and advanced designs. However, it is worth remembering that the more complicated the structure of the system, the greater the demand for electricity and the price of the CPU itself.
The processor in a computer and laptop – differences
The processors in computers and laptops are similar in construction, but there are still significant differences. Processor for a computer – they are usually much more efficient, and for easy modernization, they are mounted in a special socket – a socket. It is true that the set with the CPU usually already includes cooling, but the user can replace it with a more efficient and quieter design. Laptop processor – it already has some limitations (mainly when it comes to power and cooling). So it usually offers lower performance at the cost of less electricity consumption. Additionally, there is no option of modernization – the system is permanently installed, and the cooling has been designed with a specific laptop in mind.
How to Choose the Best CPU for Your Needs
Choosing the best CPU for your needs can be quite a daunting task. There are so many different models with different specs.
- The first category is CPUs for gamers – these are the ones that need the fastest clock speed to run games smoothly with minimal lag.
- The second category is CPUs for professionals – these are fast computers with great graphics capabilities used by professional designers, developers, and video editors.
- Lastly, processors for workstations – these are high-end computers that offer extra features like multiple cores and more memory to handle demanding tasks like running virtual machines or big data analytics software.
Intel processors
Let’s start with Intel processors. For many years, these units dominated the market, offering top performance in virtually all applications. What do the various names and symbols mean?
Processor series
Let’s start with a series of processors. How are the Core, Pentium, and other families different?
- Intel Core i9 – is a series of the most efficient consumer processors. Initially, it was available only for Intel Core X CPUs intended for HEDT (High-End Desktop) computers and workstations, and therefore designed primarily for professional applications. It was only with the 9th generation of Intel Core (Coffee Lake Refresh) that it made its debut as a purely consumer processor (i9-9900K model). Units in this series are characterized by the largest number of cores and threads. They are intended for the most demanding users who, apart from gaming, also deal with film processing, video streaming, or creating 3D graphics.
- Intel Core i7 – this series was the most efficient desktop processors up to the 8th generation Intel Core (Coffee Lake). Like the Intel Core i9 models, the Core i7 is not only suitable for playing demanding games, but can also be used as the heart of a semi-professional computer – for video processing or graphics rendering.
- Intel Core i5 – these processors are especially liked by players because they offer one of the best price-performance ratios. They offer fewer cores and threads than the more expensive series. They are naturally lower in performance than Intel Core i7 and Intel Core i9, especially in multi-threaded computing.
- Intel Core i3 – these are processors designed for less demanding users. Occasional Sunday gaming and multimedia – these are the tasks in which these units will prove themselves. They have the smallest number of cores and threads from the entire Intel Core family, which has a positive effect on their price.
- Intel Pentium – over the years, such Intel Pentium processors as G, D, N, J, Silver, or Gold have appeared on the market, and even earlier they were simply called Intel Pentium.
- Pentium processors were once Intel’s main series of gaming CPUs. Currently, these are units intended for less demanding customers. Although the example of the Intel Pentium G4560 with HT technology support (more on it below) confirmed that this series of CPUs can become a very popular CPU for the cheapest gaming desktops.
- Intel Celeron – these are simple processors with low performance and equally low price. They can be used to build office computers that are used for simple applications.
- Intel Core M – these are mobile processors with very low energy demand. They are primarily suitable for light and thin notebooks or 2-in-1 devices.
- Intel Atom – these are the least efficient and very energy-saving processors designed for simple laptops and computers, including compact designs like mini PC, wearable devices (smartwatches), smartphones, or portable game consoles.
The processor lines described above are not all that could or still can be found in Intel’s offer. We also distinguish series such as Intel Xeon, Intel Itanium, or Intel Quark. These are specific systems not intended for home users or gamers but designed for server rooms, professional workstations, or wearable devices.
Model numbers
At first, glance when you see the name Intel Core i7-11700K, you will think that someone hit the keyboard with his fist and this is how the name was created. You already know what Intel Core i7 stands for. Let’s move on to the numbers, so 11700K – what’s behind it?
According to the manufacturer’s specifications, the first numbers represent the generation – in this case, the eleventh. The rest is called SKU, i.e. simply an identifier for managing a given item. That’s it, that’s the secret.
Processor markings
Let’s go further with our example – Intel Core i7-11700K. What does the K at the end mean? What do the other letters mean?
- K – means unlocked processor multiplier, and therefore the possibility of overclocking the processor.
- KF – Unlocked CPU multiplier that does not have an integrated graphics chip.
- X – is actually a separate family of processors. These are processors for HEDT (High-End Desktop) platforms with a large number of cores and threads for professional use.
- T – means lower power consumption and lower efficiency.
- U – means ultra-low voltage processor.
- Y – indicates an extremely low voltage processor.
- H – means, according to what the manufacturer claims, a very efficient graphics card (integrated circuit).
- HF – means there is no integrated graphics.
- G – This means it has integrated graphics. Applies to processors for mobile devices. The number next to this designation indicates the power of the iGPU.
- F – indicates a processor without an integrated graphics chip.
In addition, you can meet the combination of two markings:
- HK – it’s very efficient graphics and an unlocked multiplier.
- HQ – means high-performance graphics and a quad-core processor (in mobile units).
What socket for the Intel processor?
- Intel LGA 1151 (300 series motherboards) – 8th and 9th generation Intel Core Core i3, Core i5, and Core i7 processors (Coffee Lake Refresh, Coffe Lake, Kaby Lake Refresh, Whiskey Lake)
- Intel LGA 1151 (100 and 200 series motherboards) – Core i3, Core i5, and Core i7 processors from the older 6th or 7th generation Intel Core (Skylake, Kaby Lake)
- Intel LGA 2066 – efficient Intel Core i5, Core i7, and super-efficient Core i9 processors from the 7th generation Intel Core
- Intel LGA 2011-3 – super-efficient Intel Core i7 CPU from the older 5th or 6th generation Intel Core
AMD processors
With the launch of Ryzen processors, AMD chipsets have gained popularity. What do the various names and symbols mean?
Processor series
- AMD Ryzen Threadripper – by far the most powerful AMD processors, which are characterized by a huge number of cores and threads. It is a CPU designed for professional applications such as CAD design and 3D rendering. So they are processors for users who work with a lot of very complex and resource-intensive programs.
- AMD Ryzen 9 – the second most powerful “red” processor after Threadrippers, intended for professional programmers or graphic designers or for very demanding gaming. The Ryzen 9 series processors are also characterized by a large number of cores and threads, although significantly inferior to the top Threadripper models in this respect. Ryzen 9 appeared for the first time in consumer processors in the Ryzen 3000 series based on the Zen 2 architecture.
- AMD Ryzen 7 – Until the arrival of Ryzen 9, they were the most efficient AMD processors. They have fewer cores and threads, which has a positive effect on their price. Like Ryzen 9, they are great for applications such as video processing and graphics rendering, but they will also be a great choice for gaming or streaming.
- AMD Ryzen 5 – they are much more hosted in computers by players and semi-professionals. Models in this series reduce the number of cores and threads compared to the Ryzen 7 but usually offer better value for money.
- AMD Ryzen 3 – these are simple and inexpensive designs, intended for multimedia computers, which are played rather sporadically and with less demanding titles. They are characterized by the lowest number of cores and threads of all AMD Ryzen series.
- AMD Athlon – is the perfect choice for a cheap set for work or playing older or not very demanding games. They have even fewer cores and threads than Ryzen 3 processors, but they are also cheaper than them. Plus, you don’t need a dedicated graphics card as they come with an integrated GPU.
- As with Intel processors, there are a few series from AMD that we have left out. They are intended for specific audiences, servers, and professional workstations (AMD Epyc) or are no longer produced and supported (AMD FX).
Model numbers
The numeric designations are quite similar to those used by Intel. Take the AMD Ryzen 7 3700X processor, for example. AMD Ryzen 7 means the series as it was before.
The first digit, in this case 3, does not necessarily mean Intel’s generation. In the case of AMD, it is not so obvious – an example is the Ryzen 5000 series, which is not the fifth, but the fourth generation of Zen architecture-based processors (Ryzen 4000 are APU Renoir units). For the rest, the rule is simple: the higher the number, the higher the performance.
Processor markings
Of course, there are letters at the end of the name. In our example, the AMD Ryzen 7 3700X is an X. What do these and other symbols mean?
- X – indicates increased performance (clocking) relative to the model without this symbol.
- U – is used to denote low voltage mobile processors.
- G – means that it is an APU (Accelerated Processing Unit), which consists of a processor and an integrated graphics chip.
- PRO – appears next to business processors.
- GE – means energy-saving processor with integrated graphics chip.
- H – full-voltage laptop processor with higher performance than U units.
- HS – defines a high-performance mobile CPU with lower power consumption and clock speed than the H.
- HX – means the most efficient laptop processor with an unlocked multiplier (OC option). It is the HK equivalent of Intel.
What socket for the AMD processor?
- AMD TR4 – high-performance Ryzen Threadripper processors
- AMD AM4 – CPUs from the AMD Bristol Ridge generation and efficient AMD Ryzen models
- AMD AM3 + – older CPU from the AMD FX generation
- AMD FM2 + – older CPU / Athlon chips from the AMD Kaveri and Godavari generation
Processor components
The naming is behind us, so let’s move on to the next thing. What to look for first when you look at the CPU specs?
Cores and threads
It is on these two things that the performance of the system depends to a large extent. Let’s start with the cores. Theoretically, the more the better, although it is not that simple, as a lot depends on their performance, which is mainly based on architecture. Besides, not everyone will need a large number of cores, but more on that later.
Threads, also called logical cores, are slightly less important. By default, the number of threads equals the number of cores. However, there are two technologies that double the number of logical cores. In the case of Intel processors, we are talking about Hyper-Threading (HT). In AMD it is Simultaneous Multithreading (SMT). The additional threads thus created are, however, less efficient than the physical cores. As a result, e.g. a 4-core and the 8-thread processor may be less efficient than a 6-core and 6-thread chip, but this depends on what it is used for.
Clock speed
When reviewing the specifications, pay attention to what clocks are listed – base or turbo mode. The nominal clock speed of the processors is much lower than that achieved in turbo mode.
Keep in mind, however, that post-acceleration clocking is most often maintained in demanding tasks, such as games, often peaking at peak values for a short while. The values in turbo mode depend not only on the type of load but also on the active cores – the fewer of them, the processor can achieve higher clock frequencies.
Try to choose a model with the highest base clock possible, but remember that some processors can be overclocked.
Cache
What is a cache? In the simplest translation, it is a mini-disk in which the processor stores data potentially needed in the near future. It supports operating memory and allows for faster access to specific information than RAM.
There are basically three memory levels – L1, L2, and L3, the latter being the farthest memory, the access time being the longest, and the most capacious. There is also a rare fourth level – L4, which has been given to e.g. CPUs from the Intel Broadwell or Intel Haswell series. Currently, processors usually have from several to several dozen MB of cache memory, and in some models, these values are counted in the hundreds.
Processor manufacturers have or still use cache technologies in their CPUs. Intel Smart Cache is worth mentioning. It divides the memory between the cores of a multi-threaded processor and each of them gets as much as it needs at a given moment. This means that a single core can use the full Level 2 and Level 3 cache if the other cores remain inactive.
AMD also uses its own solutions. For example, the Ryzen 3000 series processors introduced the so-called Game Cache (which defines total L2 and L3 memory), and in the Ryzen 5000 series 3D V-Cache, which increases the number of L3 caches and improves performance in games and applications.
Integrated graphics
If you’re building a low-cost or multimedia desktop, consider an integrated graphics chip. Almost all Intel processors have an integrated Intel UHD Graphics chip, but for Ryzen processors, look for units marked with G (E).
What should the CPU temperature be?
You can see the maximum values for a given processor on the official websites of Intel (ark.intel.com) and AMD (amd.com). You have to go to the product sheet by entering the product name in the search engine (eg Ryzen 5600X). Then find the maximum temperature allowed in the specification (Max Temps for AMD and T-junction and T-case for Intel).
Maximum CPU temperatures may vary by series and model. For example, the critical value for the AMD Ryzen 5000 is as high as 95 degrees, while the safe temperatures for older processors such as AMD FX-8300 or first-generation Intel Core (e.g. i5-750, i7-870) are clearly lower (up to 61 degrees and 72.7 degrees Celsius).
When idle, a well-cooled processor operates at temperatures of 30-45 degrees Celsius. Under heavy load, it reaches significantly higher values - most often in the range of 60-80 degrees Celsius – and they also depend on the CPU model.
Operation in such conditions does not adversely affect the operation of this component. They may differ depending on the scenario. For example, in benchmarks, the temperature will be higher than in demanding games. On the other hand, in older, light-weight titles, it may slightly exceed the values typical for the IDLE mode (idle).
How to check CPU temperature?
You can check the CPU temperature in UEFI and BIOS (look for Hardware Monitor or PC Health Status), but only when idle. Unless you have a unique case equipped with a display that will inform you about it without having to go into the computer’s firmware.
Otherwise, free applications work very well for the ongoing monitoring of temperatures in the Windows operating system. They are also provided by processor manufacturers. I am talking about, for example, AMD Ryzen Master and Intel Extreme Tuning Utility, which are also used to overclock the CPU (so-called overclocking).
What are the best free CPU temperature monitoring programs?
Here are the most popular free programs that you can use to check the CPU temperature:
- HWMonitor
- Core Temp
- HWiNFO
- Open Hardware Monitor
- Speccy
Many of these programs not only monitor the temperature of the processor and each core separately, and record the minimum, average and maximum values that this component achieves during operation. One of the most popular programs today is CPUID HWMonitor or HWiNFO, which also allow you to read the most important information about disks, motherboard, or graphics card. In the case of components such as the CPU and GPU, you can even check the degree of load or timing with which the components are currently working.
Conclusion
When choosing a CPU, it is important to first identify what kind of computing tasks you will be spending the most time on. For example, if you do a lot of gaming or other complex tasks, then a high-end Intel processor may be for you. If you spend most of your time on office work and don’t do any heavy-duty gaming or other tasks that need raw power, then an Intel Core i5 would likely suffice.
