CPU Compare in Productivity and 80+ Gaming Benchmarks
How to Compare Processor Benchmarks
When shopping around to upgrade your fleet of business computers, the metric that stands out on labels and in ads is processor — or CPU — performance or speed. You should not, however, try to compare performance based on this one number. Processor performance is dependent on a number of factors and can vary depending on the hardware environment and the task being performed.
Clock Rate
CPU clock rate, sometimes inaccurately called «speed,» is a figure that reports the number of basic calculations, such as adding two numbers together, per second that a single processor core is capable of. The figure is based specifically on the frequency of an oscillator crystal which regulates the processor’s speed and temperature. A processor which runs at a higher temperature can run at a higher clock rate. Clock rates are measured in gigahertz, or billions of wave oscillations, per second. Typical processor cores have clock rates of between 1 GHz and 4 GHz. While clock rate is almost always reported as a spec for a machine, rarely does the clock rate alone mean much for a CPU’s overall performance, much less a computer’s.
Cores
Most contemporary CPUs contain between two and eight «cores.» Multiple cores allow processors to break tasks up and distribute them for higher performance. Multiplying the number of cores times the clock rate can give you a rough estimate of the processor’s maximum performance, but the number would not be totally accurate as some processing has to go into distributing tasks, and some applications make better use than others of multiple cores.
Benchmarking
To truly establish a processor’s performance, the processor has to be put to work on real-world tasks. Good benchmarking tests exercise processors by collecting information from a number of different hardware environments, each running a specific set of tasks. The results can then be compared to the results of other similar processors. Results of these tests, which represent fairly accurate estimates of processor performance, can be found at sites like Tom’s Hardware and CPUBenchmark.
Other Bottlenecks
The performance of a CPU is only one of the factors that determines overall computer performance. Disk drive access, network speed, video card performance, amount of memory and memory access speed can all affect system speed and responsiveness. Some combinations of hardware perform better than others as well. Besides looking at the CPU benchmarks and metrics reported by the manufacturer, it’s a good idea to read reviews of specific systems to find out how they perform in the real world.
You can compare processors in the same or different collection, generation, i9, i7, or i3. You can also compare different type of processors, like Intel® Core, Pentium, and Celeron, or any other type of Intel processors.
The Central Processing Unit (CPU), also known as a processor, is the brain of the computer and is thus the most important component. Unfortunately, comparing two different processors side-by-side can be tough, which can complicate any purchases you might make.
The bad news is that you can’t just rely on clock speed or cores, which are the two most heavily advertised aspects of processors. The good news is that you don’t need to know how a CPU works, although that can prove useful.
The other good news is that there are sites out there that make such comparisons easier. In this article, we’ll tell you exactly what matters and what doesn’t when comparing different processors, and how to compare them the right way.
Clock Speed Isn’t Everything
Clock speed and cores are the most heavily advertised aspect of processors. Clock speed is usually noted in hertz (e.g. 3.14 GHz) while the number of cores is usually advertised as dual-core, quad-core, hexa-core, or octa-core.
For a long time, it was this simple: the higher the clock speed, the faster the processor, and more cores meant better speeds. But processor technology today isn’t dependent as much on the clock speed and cores because CPUs now have several other parts that determine how fast they can perform.
In a nutshell, it comes down to how much computing can be done when all parts of a CPU come together in a single clock cycle. If performing Task X takes two clock cycles on CPU A and one clock cycle on CPU B, then CPU B might be the better processor even if CPU A has a higher clock speed.
Compare clock speeds only when you are trying to decide between two CPUs from the same family and same number of cores. What this means is that if you’re looking at two quad-core Intel Core i5 Skylake processors, then the one with the higher clock speed will be faster.
For any other scenario, the clock speed or cores don’t always indicate performance. If you’re comparing Intel Core i3 vs. Core i5 vs. Core i7 processors or Intel Core i5 vs. Core i7 vs. Core i9 processors, then clock speed and number of cores don’t matter. And if you’re comparing Intel vs. AMD or an AMD A10 vs. AMD A8 vs. AMD FX, then clock speed alone won’t tell you much.
Check Single-Threaded Benchmarks
The dirty little secret in the computer world is that even though you’re buying a processor with four cores, all four of those cores might not actually be used when you’re running applications.
Most software today is still single-threaded, which means the program is running as one process and a process can only run on one core. So even if you have four cores, you won’t be getting the full performance of all four cores for that application.
That’s why you also need to check the single-threaded (or single-core) performance of any processor before buying it. Not all companies explicitly release that information, so you’ll need to rely on third-party data from reliable resources like Passmark benchmark tests.
Passmark’s full list of CPU benchmarks has a single-threaded rating for each CPU.
Cache Benchmarks Is King
The cache is one of the most under-appreciated parts of a CPU. In fact, a cache with poor specs could be slowing down your PC! So always check the cache specs of a processor before you purchase it.
Cache is essentially RAM for your processor, which means that the processor uses the cache to store all of the functions it has recently performed. Whenever those functions are requested again, the processor can draw the data from the cache instead of performing it a second time, thus being faster.
Processors have different levels of cache, starting with L1 and going up to L3 or L4, and you should only compare cache size at the same level. If one CPU has L3 cache of 4 MB and another has L3 cache of 6 MB, the one with 6MB is the better choice (assuming clock speed, core, and single-threaded performance are all comparable).
Integrated Graphics Matter, Too
Intel and AMD have combined the CPU and the graphics card into an APU. New processors can usually handle the graphics requirements of most everyday users without requiring a separate graphics card.
These graphics chipsets also vary in performance depending on the processor. Again, you can’t compare an AMD to an Intel here, and even comparing within the same family can be confusing. For example, Intel has Intel HD, Intel Iris, and Intel Iris Pro graphics, but not every Iris is better than HD.
Meanwhile, AMD’s Athlon and FX series come without graphics chips but cost more than the APU-centric A-Series, so you’ll have to buy a graphics card if you’re getting an Athlon or FX processor.
In short, graphics processing on CPUs is still quite confusing, but you still need to pay attention to it! The best option is to consult third-party benchmarks and look for recommendations.
Futuremark developed the 3DMark graphics test, which is one of the best free Windows benchmark tools out there. You can check the 3DMark Physics Score of any processor and compare it to others in Futuremark’s processor list, which should give you a fair idea of which CPU has better graphics.
The Best Way to Compare CPUs
All of these factors come together to make CPU comparisons a difficult proposition. How do you know which one you should buy? Here are a few tips that may help.
The easiest and best way is to head to CPUAgent. This site compares two processors and gives ratings and explains the differences between the two in terms that any non-techie can also understand.
Benchmarks come from different sources like PassMark, PCMark, CompuBench, GeekBench, SkyDiver, and more. We basically save you the trip of going to many sites.
The score is a safe parameter in making your purchase decision, with the simple idea that whichever processor scored higher is the better one. CPUAgent also compares integrated graphics, telling you which APU has the better graphics performance.
In case you are looking for more details than what CPUAgent provides, I’d recommend the PC Builds Comparison Tool. Here you can browse in-depth benchmarks conducted by one of the best independent hardware review sites and even compare two processors side-by-side.
Other Factors That Affect Benchmarks
When it comes to overall performance, keep in mind that your processor is only as good as the rest of the hardware. If you buy a great processor and only stick in 2 GB of RAM, then it will be bottlenecked in speed.
Compare All CPU Specs, Benchmark Scores
CPU comparison helps you find the differences between CPUs. Using this tool, you can compare upto 3 CPUs side by side and see how they’re different from each other. Compare your selected CPUs now and find the right one for your computing needs.
Just enter the name of the CPUs you want to compare (2-3) and see the differences between them side-by side. You can compare the core count, threads, clock speeds, TDP, performance per watt and 100 other specs. Additionally, our tool also has benchmark score comparisons, so if there’s a score you trust such as Passmark, Geekbench, CineBench, then you can find their comparison scores too.
How to Compare CPUs?
Our CPU Comparison is straightforward to use. Add up to 3 processors of your choice in the search bar to compare with each other. Just start typing the name of the CPUs and our tool will find the perfect model you’re searching for. Do that for 2-3 processors and click the compare button. Our tool would now display the technical comparison of the CPUs you’ve choosen.
- Check their key differences, like how many cores they have and what is their turbo boost frequency, base clock and turbo clock speed. Which one has more CPU threads to handle multitasking better.
- Use the General Info section to compare their performance features like CPU temperature, Cache size, TDP, 64-bit support, semiconductor size, etc.
- Then compare how much memory they have to offer. The more the memory, the better & faster the processor for multitasking.
- You can also compare CPU benchmarks to see which one is leading and then buy the right one directly using the “Buy on Amazon” button.
There is no limit on using this CPU compare tool in a day. So compare your desired options (upto three at a time) till you find the right one.
You can also check our “Similar Comparisons” based on your selection to see which one really fits well with your requirements.
What CPU Specs are Compared?
Spec | Description | Range |
---|---|---|
Cores | Responsible for multitasking |
Higher the better |
Clock Speed/Frequency | Indicates how fast a processor is |
Higher the better |
Threads | Responsible for multi tasking and fast processing |
Higher the better |
Cache Size | Works like computer’s RAM to process data faster |
Higher the better |
TDP | Determines the maximum heat generated by the processor under high workload and how much cooling it needs to perform optimally |
Lower the better |
Integrated GPU/APU | Offer better performance through less battery consumption and heat generation |
NA |
Generation | Represents the technical upgrades and advancement in the processor |
Higher the better |
Motherboard Socket & Chipset Compatibility | Required for optimal performance of processor |
Must be of same generation |
Benchmark Scores | Industry standard for processor’s performance |
Higher the better |
Coolers | Requires to dissipate generated heat to keep the system cool for optimal performance |
Small or mid-sized coolers for low wattage, i. e., 40-70 Watts Large sized coolers for high wattage, i.e. above 70 watts |
Warranty | Covers processors components’ maintenance, repair and replacement |
Higher the better |
Intel vs AMD | Both offers single-core and multi-core processors |
Both are good brands and have a name when it comes to Central Processing Units. However, the right one depends on the needs. |
The performance of any processor depends upon some crucial internal specs, which you should not neglect while buying or comparing. So, to compare CPU performance of two or more processors, consider these specs, factors and features before choosing the one:
Number of Cores
Range: 1-8 Cores
Previously, most CPUs used to have a single core, meaning all the data processing was handled only by one core. However, now CPUs come with multiple cores, from dual-core to eight-core, to split up the tasks between the various cores for faster processing.
So how many cores do you need in a CPU? It totally depends upon your requirements. Generally, a dual-core processor is suitable to handle day-to-day work tasks, surfing, etc.
You can go for quad-core or six-core processors for extensive gaming or designing requirements like 3D video rendering and other demanding tasks. If you want to run heavy programming or 3D modeling software, first research how many cores they utilize and then you can opt for six or even eight-core processors depending upon the usage.
Clock Speed/Frequency
Range: 0-5.5 GHz
Clock speed is generally measured in Gigahertz (GHz) and represents any processor’s processing or operational speed. For example, a processor with a 2GHz clock speed indicates that it can handle/execute up to 2 billion instructions per clock/second.
So, it simply refers to how fast a processor can process instructions in a single cycle without lagging. Along with core count, clock speed significantly impacts CPU performance. For simple task processing, 2.4 GHz clock speed works fine. However, for gaming purposes, 3.5-4.0 GHz is considered good.
What is Overclocking and Maximum Clock Speed? – Should you go for it?
Every processor has a certain maximum clock speed. But in some processors, there is a feature of Overclocking that allows you to turn up the maximum clocking speed intended by the manufacturer.
Note that only “Unlocked CPUs” come with the ability to overclock and are generally more expensive than locked CPUs (cannot overclock). Processors in Intel designed for overclocking are represented by “K” in the last, for example, Intel Core i7-11700K.
So, if you think your requirements will increase significantly in upcoming years, you can go for the unlocked CPU instead of the locked ones. However, the overclocking process builds up additional heat in your CPU, so you need a better cooling system if you go for it.
Threads
Range: 2-16 Threads
While the core acts as the individual processors in the CPU, the threads in each core decide how many processes that core can handle significantly at a time. You can consider them as virtual components that divide the physical core of a processor into multiple virtual cores.
Modern CPUs generally have more threads than cores, indicating that each core can handle multiple tasks or data processing simultaneously. A single core processor can have a maximum of two threads. So, a dual-core processor comes with quad-core threads while the eight-core processor can have a maximum of 16 threads.
For better processing power, go for the CPU with hyper-threading capabilities. Clock speed, Core and Thread count together decide the processing speed of any processor.
Processor’s Cache
Range: L1 Cache – 0KB – 2 MB, L2 Cache – 256 KB – 8 MB, L3 Cache – 32 MB – 64 MB
A processor’s cache works similarly to a computer’s RAM to store temporary files/data. The bigger the processor cache size, the more cache files it can store and access quickly. Besides the main cache (L1), processors have additional L2 and L3 cache built between the CPU and RAM for alternative high speed. L2 and L3 caches take more time to access than L1.
While L1 cache is 100 times faster than your computer’s RAM, L2 cache is 25 times faster. Processors with large L2 and L3 cache sizes are generally faster and suitable for multitasking, gaming, designing and programming.
CPU temperature & Thermal Design Power (TDP)
Range: 60 Watts- 137 Watts
Every CPU has a certain optimal temperature which is decided by its thermal design power (TDP). TDP measures how much heat a processor gives off to function optically. Knowing TDP is crucial to decide how much cooling and power wattage the processor requires to avoid overheating. The lower the TDP, the less power consumption of the processor.
Especially if overclocking is your requirement, you might have to use a third-party cooling system instead of the stocked one, for which you have to check its TDP and how much power is drawn by CPU components.
Integrated GPUs (Graphic Processing Units) or APU (Accelerated Processing Units)
Range: NA
Most modern CPUs come with integrated GPUs to share the system’s memory for processing graphics and rendering videos. The benefit of integrated GPUs is that it takes less power and generates less heat to save battery.
However, the integrated GPUs can sometimes not serve you excellent graphic performance compared to a discrete CPU and GPU, especially if you are a designer or professional gamer.
In that case, AMD processors come with APUs that work similar to integrated GPUs but offer better graphic performances for gaming laptops.
Generation
Range: 1st Generation – 12th Generation
The latest versions and upgrades of the CPU come in the form of generations. The higher the generation, the more advanced the CPU’s technology. But you don’t always need the latest generation CPU to handle your tasks.
Even if you go for an older generation with more cores and threads, you will get optimal performance compared to the latest generation with low cores and threads.
For normal multitasking and browsing experience, you can go for the i3 or i5 generation with multi-cores, but for gaming, designing or extensive programming work, try to go for the i5, i7 or beyond.
Motherboard Socket & Chipset Compatibility
Range: NA
While choosing any CPU based on your requirements, don’t forget to check its socket compatibility with your motherboard. CPU sockets are available on motherboards where the CPU gets plugged, and not all sockets are compatible with all CPUs.
For example, an Intel-based CPU cannot fit with the AMD CPU socket and vice versa. Similarly, if an Intel Core i7-4770K processor is designed to function with an LGA 1150 socket, you cannot use it with an LGA 1200 socket. So, pick only the right one to avoid compatibility issues.
Compatibility of your chosen CPU with the motherboard chipset and the socket is crucial for optimal performance. As a thumb rule, you must pair the CPU with the motherboard of the same generation to avoid compatibility issues.
Benchmark
Range: Depends on the benchmark scores.
The tool also provides you with CPU benchmark scores by some reputable benchmark apps like Passmark, Geekbench 5, Cinebench R20, etc.
The CPU benchmark compare scores given by these apps are considered the industry standard for measuring the performance of different processors, graphic cards and computers. The higher the score is, the better the performance of that CPU.
Performance
The performance of any processor is determined by core & threads counts, CPU speed, Cache size and frequency. The tool compares all these aspects of various CPUs to help you pick the right one.
Coolers
Range: Small or mid-sized coolers for low wattage, i.e., 40-70 Watts, Large sized coolers for high wattage, i.e. above 70 watts
Most modern CPUs come with in-built stock coolers, which work fine for normal multitasking. However, if you want to run heavy gaming, designing or programming software or want to utilize it for overclocking, the stock cooler might not function well for you.
If looking for third-party coolers for overclocking, don’t forget to catch their compatibility with motherboard sockets.
Price
Range: $60-$37,000
Budget is the crucial aspect to look for while comparing various CPUs. Expensive doesn’t always mean good. Pay for the features you really require instead of just paying for the extra ones you won’t use. With our comparison tool, you can find the required features in a budget-friendly CPU after comparing it with the best-performing ones.
Depending upon your requirements, choose the right CPU. If your work requirement is less, you can go for a low or mid-range CPU (i3 or i5 generation). However, if your requirements are extensive, you need a high-range gaming CPU (i5 or i7 generation).
Purpose & Workload
Knowing your work requirements and load can help you choose the best CPU for you.
Home users: If you have basic work requirements like web browsing, sending emails, excel processing, etc., a basic CPU will work fine for you. In that case, you can go for either AMD Ryzen 3 or Intel Core i3 with a 2GHz Clock speed.
Startups and SMEs: If you are a freelancer or working in a startup with normal data processing needs like presentations, video calls, surfing, etc., the i3 or i5 generation is enough for you, depending upon the software you want to operate. However, if you have high-end usage like 3D modeling, heavy video rendering, etc., you can go for AMD Ryzen 5 or 7 or Intel core i5 or i7 with 2GHz or above clock speed and 64-bit RAM.
Gaming: If gaming is your main requirement, the baseline CPU is either AMD Ryzen 5 or Intel core i5 to run titles like Fortnite, PUBG, etc.
However, if you are a professional gamer who wants to run high-end games with excellent graphics quality, go for an i7 or AMD Ryzen 7 or above with an excellent cooling system.
Warranty
Range: 1-3 years
Not all brands offer the same warranty for the maintenance and repair of CPU components. So, while buying a processor, don’t forget to check the complimentary brand warranty duration to get the best price for a performance CPU.
Intel vs AMD
So, which one should you choose between Intel and AMD? Both AMD and Intel CPUs offer great performances with a very slight difference. In terms of single-core performance, Intel usually offers better performance than AMD single-core processors; however, it comes at a little expensive price.
On the other hand, for gaming purposes, AMD offers excellent performance through its multi-core processors under an affordable range to beat Intel multi-core processors. So both have something great to offer, depending upon your budget & requirements.
Types of CPU
Depending upon your usage type and budget, there are several types of CPU available in the market, like:
1. Mobile CPU
As the name indicates, these are compact CPUs generally designed for laptops and mobile devices. While they offer optimal performance, they are generally slower than the desktop and other counterparts.
2. Server CPU
Server CPUs are designed to run actively 24/7 without any lag for extensive usage, and thus, they are tested under high-temperature and high-stress conditions. They are ideal for large data processing and hosting multiple software and applications at a time but come at a high cost than mobile or desktop CPUs.
3. Desktop CPU
Unlike Server CPUs, Desktop CPUs are designed for all kinds of users to process moderate workloads. Depending upon their components, they can be used for overclocking, gaming, surfing, programming, and multiple other high-end requirements.
Desktop CPUs are generally more affordable than Server CPUs but a little more expensive than Mobile CPUs. As there are various types of desktop CPUs available in the market for different needs, you must select according to your required specifications.
Conclusion
By comparing the industry’s top processors with our comparison tool, you can get the best value CPU for your requirements under the desired price range. All you need to check is the aforementioned crucial specs and features provided to you in a simple tabular format for comparison. Compare as many processors you want to before making the final purchase to pick the right one.
Related Tools: GPU Comparison | Monitor Comparison | Motherboard Comparison | Phone Comparison |
CPU-Z Benchmark — CPU-Z VALIDATOR
Best CPU performance — 64-bit — September 2022
Intel Core i9-12900KF
Intel Core i9-12900K
Intel Core i7-12700K
Intel Core i7-12700KF
Intel Core i5-12600K
Intel Core i5-12600KF
Intel Core i9-11900K
AMD Ryzen 9 5900X
AMD Ryzen 9 5950X
Intel Core i7-11700K
AMD Ryzen 7 5800X
Intel Core i7-11700KF
Intel Core i5-11600K
AMD Ryzen 5 5600X
Intel Core i5-11600KF
Intel Core i7-11700
Intel Core i7-11700F
AMD Ryzen 7 5700G
AMD Ryzen 9 5900HX
Intel Core i5-11500
Intel Core i9-10900KF
AMD Ryzen 5 5600G
Intel Core i7-11800H
Intel Core i9-10900K
Intel Core i9-10850K
Intel Core i5-11400
Intel Core i7-10700KF
Intel Core i5-11400F
Intel Core i9-10900
Intel Core i5-11400H
AMD Ryzen 7 5800H
Intel Core i9-9900KF
Intel Core i7-9700KF
Intel Core i7-10700K
Intel Core i7-9700K
AMD Ryzen 5 5600H
Intel Core i9-9900K
Intel Core i5-10600KF
Intel Core i5-9600KF
Intel Core i5-10600K
Intel Core i7-10700
Intel Core i7-10700F
Intel Core i7-1165G7
AMD Ryzen 7 3800XT
Intel Core i7-9700F
Intel Core i7-9700
AMD Ryzen 9 3950X
Intel Core i5-9600K
AMD Ryzen 9 3900X
AMD Ryzen 5 3600XT
AMD Ryzen 7 3800X
Intel Core i7-8700K
AMD Ryzen 7 3700X
Intel Core i5-1135G7
AMD Ryzen 5 PRO 4650G
Intel Core i5-8600K
AMD Ryzen 7 5700U
AMD Ryzen 5 3600X
AMD Ryzen 7 4800H
Intel Core i7-7700K
Intel Core i7-10875H
Intel Core i5-7600K
Intel Core i7-8700
AMD Ryzen 5 3600
AMD Ryzen 5 3500X
Intel Core i3-1115G4
AMD Ryzen 5 3500
Intel Core i7-10870H
AMD Ryzen 5 4500U
AMD Ryzen 5 5500U
Intel Core i7-10750H
Intel Core i3-10105F
Intel Core i3-9100F
Intel Core i5-8500
Intel Core i7-6700K
Intel Core i3-10100
AMD Ryzen 5 4600H
Intel Core i5-10400
Intel Core i5-9400
Intel Core i3-10100F
Intel Core i5-6600K
Intel Core i5-9400F
Intel Core i5-10400F
AMD Ryzen 3 3100 4-Core
AMD Ryzen 7 2700X
Intel Core i7-4790K
Intel Core i5-10300H
Intel Core i7-9750H
Intel Core i5-8400
AMD Ryzen 5 2600X
Intel Core i7-7700
Intel Core i5-4690K
AMD Ryzen 5 3400G
Intel Core i7-8750H
AMD Ryzen 3 3200G
Intel Core i5-9300H
AMD Ryzen 5 2600
Intel Core i3-8100
Intel Core i7-10510U
Intel Core i5-7500
Intel Core i5-8300H
Intel Core i5-4670K
AMD Ryzen 5 1600X
Intel Core i7-8565U
AMD Ryzen 3 2200G
Intel Core i7-4770K
Intel Core i5-10210U
Intel Core i5-4690
Intel Core i5-1035G1
AMD Ryzen 5 2400G
Intel Core i7-4790
AMD Ryzen 7 1700X
Intel Core i7-6700
AMD Ryzen 7 2700
Intel Core i7-4770
Intel Core i5-8265U
AMD Ryzen 7 1700
AMD Ryzen 5 1600
Intel Core i3-1005G1
AMD Ryzen 5 3550H with
Intel Core i5-4590
Intel Core i5-7400
Intel Core i5-3570K
Intel Core i7-3770K
Intel Core i5-6500
Intel Xeon E3-1231 v3
Intel Core i7-8550U
Intel Core i5-4570
Intel Core i3-7100
AMD Ryzen 3 1200
Intel Core i5-2500K
Intel Core i5-3570
AMD Ryzen 7 3750H with
Intel Core i7-3770
Intel Core i7-2600K
Intel Core i7-7700HQ
Intel Core i5-8250U
Intel Core i5-7300HQ
Intel Core i5-6400
Intel Core i3-6100
AMD Ryzen 5 1400
Intel Core i5-4460
AMD Ryzen 5 3500U with
Intel Core i5-3470
Intel Xeon E5-2640 v3
Intel Xeon E5-2678 v3
AMD Athlon 3000G
Intel Core i5-4440
Intel Core i3-4170
Intel Xeon E3-1230 V2
Intel Core i5-2500
Intel Pentium G4560
Intel Core i7-2600
Intel Core i3-4160
AMD Ryzen 5 2500U with
Intel Xeon E5-2689
Intel Core i7-7500U
Intel Core i7-6700HQ
Intel Core i5-2400
Intel Xeon E5-2620 v3
Intel Core i3-4130
Intel Xeon E5-2650 v2
Intel Core i5-3330
Intel Core i5-7200U
Intel Core i3-3240
Intel Core i7-6500U
Intel Core i3-3220
Intel Core i5-6300U
Intel Xeon E5450
Intel Core i3-2120
Intel Core i5-3230M
Intel Core i3-2100
Intel Core i5-6200U
Intel Core 2 Duo E8400
Intel Core i5 750
Intel Core i5-5200U
Intel Core 2 Quad Q9550
Intel Core i5-2520M
Intel Core 2 Duo E7500
Intel Core i5 650
Intel Core i5-3210M
Intel Core i7 920
AMD FX -8350
Intel Core i5-2450M
Intel Core 2 Quad Q9400
AMD FX -4300
AMD FX -8320
AMD FX -6300
Intel Core i5-2410M
Intel Core i5-4210U
Intel Core 2 Quad Q6600
AMD FX -8300
Intel Core i3-7020U
Pentium E5200
Intel Core i5-4200U
Intel Core i3-3110M
Intel Core i3-5005U
Intel Core i3-6006U
AMD Phenom II X4 955
Intel Core i3-4005U
Compare CPU: Best CPU for gaming 2022
What is the best processor for gaming?
If you want a cheap processor at a low price, then we track the top deals, sales, Prime offers and biggest discounts on Amazon to find the best budget gaming CPU for the money.
241
Processors found
See more
Network N earns commission from qualifying purchases via Amazon Associates and other programs.
Price
AnyUnder $500Under $300Under $200Under $100
Brand
AnyAMDIntel
CPU Rating
AnyUnder 80%Under 70%Under 60%Under 50%
Cores
Any16 cores8 cores6 cores4 cores2 cores
Filter by game
Min = game will run
Rec = game will run well
- GTA 5
- FIFA 23
- Valorant
- Elden Ring
- Marvel’s Spider-Man Remastered
- Red Dead Redemption 2
- Call of Duty: Warzone
- Fortnite
- Genshin Impact
- Apex Legends
- CSGO
- Microsoft Flight Simulator
- Minecraft
- Stray
- FIFA 22
- Fall Guys
- Tower of Fantasy
- Farthest Frontier
- Cult of the Lamb
- MultiVersus
AMD Ryzen 9 3900XT
Available with Prime Delivery
$476 (Save $23) |
AMD |
100% Rating |
12 cores |
Min | Rec |
---|---|
Intel Core i9-10940X
$819 (Save $131) |
Intel |
100% Rating |
14 cores |
Min | Rec |
---|---|
AMD Ryzen 9 3900X
Available with Prime Delivery
$415 (Save $70) |
AMD |
99% Rating |
12 cores |
Min | Rec |
---|---|
Intel Core i9-10900X
Available with Prime Delivery
$598. 88 (Save $83) |
Intel |
99% Rating |
10 cores |
Min | Rec |
---|---|
Summary
The document contains tables of CPU performance from 80486 to Intel Core i7 and AMD Phenom II,
In analysing results of the different benchmarks, in virtually all cases (except for RAM), CPU speed, including using data in cache, is shown to be proportional to CPU MHz for a particular type of processor. So it has been decided to show the MHz of available CPUs and provide simple tables with one set of numbers for each type of CPU. For results depending on memory speed, results are provided for a range of RAM types. As the number and type of instructions used in the benchmarks is known, performance results can be easily converted to Millions of Instructions Per Second (MIPS) or Millions of Floating Point Operations Per Second (MFLOPS). Comparison numbers represent percentages of MIPS or MFLOPS divided by CPU MHz.
Contents
Introduction Benchmark results used Intel CPU Characteristics
AMD CPU Characteristics
Performance Comparisons
1.CPU Performance Tables — Introduction
In analysing results of the different benchmarks, in virtually all
CPUs can have more than one Integer or Floating Point Pipeline where
Overall processor performance depends on the type/version and speed
The introduction of the Intel i7 processor leads to a major complication as it
Some information on Cyrix CPU Characteristics is also included. Note that these benchmarks reflect performance of one CPU on multi-core processors. 2. CPU Characteristics
Some of the following have been gathered from benchmark results and magazines
A special CPUID assembly code instruction enable details of processor
Intel model codes are readily available via their Processor Spec Finder but
Intel
Back To Contents List
AMD
Athlon MP varieties are essentially the same as XP.
Back To Contents List
Cyrix
Back To Contents List
3.Benchmark Results Used
CPUID (WhatCPU.zip) has tests using 1, 2, 3 and 4 registers, simply
BusSpd2K (.zip) is read only using streamed sequential integer data
RandMem (.zip) uses C code with of eight & xi[xi[i+0]] | xi[xi[i+2]]
BenchNT.zip provides Whetstone, Dhrystone, Linpack and Livermore
SSE3DNow (.zip) uses the same read and read/write functions as
FFTGraf (.zip) calculates Fast Fourier Transforms using single and
BusSpd2K, RandMem and SSE3DNow run tests with increasing memory
4. CPU/L1 Cache Speed %MIPS/MHz — 32 bit integers
Note: assuming CPUs of the same MHz, Pentium 4 is faster than AMD 64 on only one example and AMD CPUs have larger L1 cache. Pentiums could also be faster than P4s, probably due to shorter pipelines. Pentium 4E has even longer pipelines and performance characteristics are different to earlier P4s, some tests producing slower results and others faster. Pentium M/Celeron M and Core 2 Duo CPUs show superior performance to the P4 range on many of these tests, influenced by shorter pipelines. This is the first example with i7 results, mainly showing exceptionally good MIPS/MHz results if we were unaware that the processor is running at a higher Turbo Boost speed. Assuming that all tests are run at the elevated speed, on the basis of these comparisons, the i7 processors are no better than those for Core 2 and much worse in two cases, maybe due to the particular code used in the benchmarks. The first is using BusSpd2k streaming data from L1 cache and ANDing to the same register. Here, performance has degraded to old Pentium standards, where ANDing to two registers is twice as fast. The other case is reading and writing using RandMem, where performance has returned to Pre-Core 2 levels. Results for a newer i7, running at the same GHz, have been added, but this has lower Turbo Boost frequencies. However, it shows some significant improvement in tests involving reading/writing (RW — comparing #3 and #1). The Phenom II performance profile is at least as good as earlier AMD processors. Adding Turbo Boost would make it faster than i7, based on average speed of these tests. 5. CPU/L1 Cache Speed %MFLOPS/MHz — 32/64 bit SP/DP floating point
AMD CPUs show superior performance to P4 of same MHz on all these floating point tests, average 1. 9 times. They also have a larger L1 cache. Pentium M/Celeron M and Core 2 Duo CPU scores are similar to AMDs but L1 cache sizes are smaller. The Phenom II profile is a little better than earlier processors and at least comparable to i7 boosted MFLOPS/MHz. Assuming that the CPU is running at Turbo Boost GHz, i7 performance characteristics are similar to Core 2 but there is some degradation with reading and writing using SSE3DNow, corrected on the newer 930 model.
6. CPU/L1 Cache Speed %MFLOPS/MHz — SSE SP, 3DNow SP, SSE2 DP FP
Again AMD performance is superior to P4s but similar to Pentium M/Celeron M. When introduced, Core 2 scores were by far the best on these SSE/SSE2 tests. Again assuming i7 run at Turbo Boost GHz, speeds reading cached data are comparable to Core 2 but slower when reading and writing. AMD performance on SSE/SSE2 has been doubled with Phenom, in some cases and is, on average, as good as i7 Turbo Boost on the basis of MFLOPS/MHz.
7. CPU/L2 Cache Speed %MIPS/MHz — 32 bit integers
Pentium 4 is faster than AMD CPUs of the same MHz up to Athlon XP on these L2 cache tests. Athlon 64 shows improvements on these speeds. Core 2 Duo is better than both P4 and Athlon 64 on most tests and has a much larger L2 cache. Relative Turbo Boosted i7 to Core 2 results are similar to those using L1 cache, with RandMem reading and writing being much slower. All Phenom speeds are faster than older AMD processors, L2 cache efficiency being much higher, and MIPS/MHz calculations are now better than i7 using Turbo Boost. Phenom and i7 speeds via L3 cache are also shown.
8. CPU/L2 Cache Speed %MFLOPS/MHz — 32/64b SP/DP floating point
Latest AMD CPUs have similar average performance to P4s of the same MHz, the former benefiting from faster floating point and the latter from more efficient L2 cache. Pentium M/Celeron M are relatively better but top marks go to Core 2 Duo. The new i7 processor has similar characteristics as Core 2, assuming that Turbo Boost is used. The more efficient L2 cache improves Phenom relative score with some better than i7. With data in L3 cache, the i7 has the edge.
9. CPU/L2 Cache Speed %MFLOPS/MHz — SSE SP, 3DNow SP, SSE2 DP FP
Performance of these SIMD streaming instructions is superior on the P4 versus AMD CPUs of the same MHz. Athlon 64 shows improvements over earlier AMD CPUs. Again, Core 2 Duo was best at the time. Core i7 provides further relative gains and is still better than the vastly improved Phenom. This relativity is increased further with data in L3 cache. See above comments for slow Core i7 #1/2 results for reading/writing and FFTs. 10. RAM Speed %MIPS/MHz — 32 bit integers
Performance via data in RAM may also depend on CPU MHz besides
Comparing CPUs of the same MHz and the same memory speed, P4s show superiority performance of speed via data in RAM compared with AMD with CPUs prior to Athlon 64. Core 2 Duo performance is outstanding on some of these tests but poor on others, perhaps due to using the nForce 570 chipset. Later results via Intel 965 chipset confirmed this suspicion, giving rise to Core 2 Duo with DDR2 RAM taking the lead on performance at that time. This i7 with DDR3 RAM pushes this processor into the lead with data in RAM. Core i7 #1/2 RAM is faster but see above comments for slow results for reading/writing.
11. RAM Speed %MFLOPS/MHz — 32/64b SP/DP floating point
Examples — see paragraph 10 notes, but FP variations are less
Superior floating point performance with AMD CPUs helped to overcome inferior RAM speeds compared with Pentium 4. Except for those from nForce 570 chipset, Core 2 Duo scores came out best. Core i7 provided further gains in relative speeds and Phenom improvements failed to close the gap in many cases. Core i7 #1/2 RAM is faster but see above comments for slow results for reading/writing and with FFTs. 12. RAM Speed %MFLOPS/MHz — SSE SP, 3DNow SP, SSE2 DP FP
Examples — see 10. notes, but FP variations are less than with
These results could justify that a 2 GHz Athlon 64 is equivalent to
END
Back To Contents List To Start
More Historic Computer Speeds
To Start Produced by Roy Longbottom December 2004 Updated May 2014 |
Intel vs AMD Processor Comparison: Which is Better?
Intel and AMD have been the two primary processor companies for more than 50 years now. Although both use the x86 ISA to design their chips, over the last decade or so, their CPUs have taken completely different paths. In the mid-2000s, with the introduction of the Bulldozer chips, AMD started losing ground against Intel. A combination of low IPC and inefficient design almost drove the company into the ground. This continued for nearly a decade. The tables started turning in 2017 with the arrival of the Zen microarchitecture.
The new Ryzen processors marked a complete re-imagining of AMD’s approach to CPUs, with a focus on IPC, single-threaded performance, and, most notably, a shift to an MCM or modular chiplet design. Intel, meanwhile, continues to do things more or less exactly as they’ve done since the arrival of Sandy Bridge in 2011.
It all Started with Zen
First and second-gen Ryzen played spoiler to Intel’s midrange efforts by offering more cores and more threads than parts like the Core i5-7600K. But a combination of hardware-side issues like latency, and a lack of Ryzen-optimized games meant that Intel still commanded a significant performance lead in gaming workloads.
Things started to improve for AMD in the gaming section with the introduction of the Zen 2 based Ryzen 3000 CPUs, and Intel’s gaming crown was finally snatched with the release of the Zen 3 based Ryzen 5000 CPUs. A drastic improvement to IPC meant that AMD was able to offer more cores, but also match Intel in single-threaded workloads. Buying into Skylake refresh-refresh-refresh-refresh wouldn’t necessarily net you better framerates.
AMD and Intel have (or used to have) fundamentally different processor design philosophies. Here’s an annoying elementary school analogy that might help you understand the difference. Which one’s more fruit: A watermelon or a kilo of apples? One’s a really big fruit. And the other’s, well, a lot of small fruit. You’ll want to keep that in mind as we take a deep dive here in the next section.
Intel
Monolithic Processor Design vs AMD Ryzen Chiplets
Intel follows what’s called a monolithic approach to processor design. What this means, essentially, is that all cores, cache, and I/O resources for a given processor are physically on the same monolithic chip. There are some clear advantages to this approach. The most notable is reduced latency. Since everything is on the same physical substrate, different cores take much less time to communicate, access the cache, and access system memory. Latency is reduced. This leads to optimal performance.
Intel’s monolithic design
If everything else is the same, the monolithic approach will always net you the best performance. There’s a big drawback, though. This is in terms of cost and scaling. We need to take a quick look now at the economics of silicon yields. Strap in: things are going to get a little complicated.
Monolithic CPUs Offer Best Performance but are Expensive and…
When foundries manufacture CPUs, (or any piece of silicon for that matter) they almost never manage 100 percent yields. Yields refer to the proportion of usable parts made. If you’re on a mature process node like Intel’s 14nm+++, your silicon yields will be in excess of 70%. This means you get a lot of usable CPUs. The inverse, though, is that for every 10 CPUs you manufacture, you have to discard at least 2-3 defective units. The discarded unit obviously cost money to make, so that cost has to factor into the final selling price.
At low core counts, a monolithic approach works fine. This in large part explains why Intel’s mainstream consumer CPU line has, until recently, topped out at 4 cores. When you increase core count though, the monolithic approach results in exponentially greater costs. Why is this?
On a monolithic die, every core has to be functional. If you’re fabbing an eight-core chip and 7 out of 8 cores work, you still can’t use it. Remember what we said about yields being in excess of 70 percent? Mathematically, that ten percent defect rate stacks for every additional core on a monolithic die, to the point that with, say a 20-core Xeon, Intel actually has to throw away one or two defective chips for every usable one, since all 20 cores have to be functional. Costs don’t just scale linearly with core count–they scale exponentially because of wastage.
Furthermore, when expanding your 14nm capacity, the newly started factories won’t have the same level of processor yields as existing ones. This has already led to Intel’s processor shortages and the resulting F series CPUs.
- Intel 14nm Processor Shortages Explained: How Ryzen Made it Worse
The consequence of all this is that Intel’s process is price and performance competitive at low core counts, but just not tenable at higher core counts unless they sell at thin margins or at a loss. It’s arguably cheaper for them to manufacture dual-core and quad-core processors than it is for AMD to ship Ryzen 3 SKUs. We’ll get to why that is now.
Chips, Chiplets and Dies
AMD adopts a chiplet-based or MCM (Multi-chip Module) approach to processor design. It makes sense to think of each Ryzen CPU as multiple discrete processors stuck together with superglue–Infinity Fabric in AMD parlance. One Ryzen CCX features a 4-core/8-core processor, together with its L3 cache. Two CCXs (single 8-core CCX with Zen 3) are stuck together on a CCD to create a chiplet, the fundamental building block of Zen-based Ryzen and Epyc CPUs. Up to 8 CCDs can be stacked on a single MCM (multi-chip module), allowing for up to 64 cores in consumer Ryzen processors such as the Threadripper 3990X.
There are two big advantages to this approach. For starters, costs scale more or less linearly with core counts. Because AMD’s wastage rate is relative to its being able to create a functional 4-core block at most (a single CCX), they don’t have to throw out massive stocks of defective CPUs. The second advantage comes from their ability to leverage those defective CPUs themselves. Whereas Intel just throws them out, AMD disables functional cores on a per-CCX basis to achieve different core counts.
For example, both the Ryzen 7 5800X and 5600X feature a single CCD with eight cores. The latter has two cores disabled, giving it 6 functional cores instead of eight. Naturally, this allows it to sell six-core parts at more competitive prices than Intel.
There is a big drawback to the chiplet approach: latency. Each chiplet is on a separate physical substrate. Because of the laws of physics, this means that Ryzen CPUs incur a latency penalty for communication over the Infinity Fabric. This was most noticeable with first-gen Ryzen. Infinity Fabric speeds correlated to memory clocks and overclocking your memory, therefore, resulted in noticeably faster CPU performance.
AMD managed to rectify this with the Ryzen 3000 CPUs and then further improve it with the newly launched Ryzen 5000 lineup. The former introduced a large L3 cache buffer, called “game cache”. L3 cache is the intermediary between system memory and low-level CPU core cache (L1 and L2). Typically, consumer processors have a small amount of L3–Intel’s i7 9700K, for instance, has just 12 MB of L3. AMD, however, paired the 3700X with 32 MB of L3 and the 3900X with a whopping 64 MB of L3.
The L3 cache is spread evenly between different cores. The increased amount of cache means that, with a bit of intelligent scheduling, cores can cache more of what they need. The buffer eliminates most of the latency penalty incurred over Infinity Fabric.
The Ryzen 5000 CPUs went a step ahead and eliminated the four core CCXs in favor of eight core complexes, with each core directly connected to every other on the CCX/CCD. This improves the inter-core latency, cache latency, and bandwidth as well as provides each core with twice as much L3 cache, significantly improving gaming performance:
- AMD Ryzen 5000 “Zen 3” Architectural Deep Dive
Chiplet or Monolithic: Which is Better?
The chiplet approach going to see widespread adoption in the coming years, from both AMD as well as Intel, for CPUs as well as GPUs. This is because Moore’s law–which mandated a doubling in processing power, mainly as a result of die shrinks (56nm to 28nm> 28nm to 14nm>14nm to 7nm) every couple of years–has comprehensively slowed down. Intel has been stuck on the 14nm node for more than half a decade and even now, after 6-7 years, the succeeding 10nm node isn’t twice as faster (and denser) than its preceding 14nm process node.
The ironic thing is that Intel has already adopted the chiplet design for most of its future processors. The Ponte Vecchio Data Center GPU, Xe HP GPUs, and Meteor Lake are all going to leverage chiplets, with the only difference being with respect to the nomenclature. Intel calls them tiles:
- Intel Believes its Tiles are a Better Approach to the MCM Design than AMD’s Chiplets
Video card comparison — Technical City
Top 5 video cards
NVIDIA GeForce RTX 3090 Ti
100.00
AMD Radeon RX 6950XT
96.21
NVIDIA GeForce RTX 3080 Ti
92.78
NVIDIA GeForce RTX 3090
90.49
NVIDIA GeForce RTX 3080 12GB
90. 45
To the performance rating of video cards →
Top 5 video cards for the best price-quality ratio (from $ 300)
NVIDIA GeForce GTX 1080 SLI Mobile
100.00
NVIDIA GeForce GTX 980 SLI Mobile
74.02
NVIDIA GeForce GTX 1070 (Mobile)
46.78
NVIDIA GeForce GTX 1650 Mobile
46.19
AMD Radeon RX 6750XT
46.03
To the rating of video cards for the best price-quality ratio (from $300) →
Top 5 video cards for the best price-quality ratio (up to $300)
NVIDIA A10G
100.00
NVIDIA Tesla P40
100.00
NVIDIA GeForce GTX 1080 Mobile
100.00
NVIDIA GeForce GTX 1070 SLI Mobile
95. 62
NVIDIA TITAN Xp
82.25
To the rating of video cards for the best price-quality ratio (up to $300) →
Top 5 video cards in mining
- Bitcoin
- Decred
- Ethereum
- Monero
- Siacoin
- ZCash
NVIDIA GeForce GTX 1080 Ti
1427 Mh/s
NVIDIA GeForce GTX 1080
1045 Mh/s
AMD Radeon R9 FURY X
858 Mh/s
NVIDIA GeForce GTX 1070
832 Mh/s
AMD Radeon Pro Duo
809 Mh/s
To the rating of video cards in Bitcoin mining →
NVIDIA GeForce GTX 1080 Ti
4.60 Gh/s
NVIDIA Tesla P100 SXM2
3.43 Gh/s
NVIDIA GeForce GTX 1080
3. 09 Gh/s
NVIDIA GeForce GTX 1070
2.83 Gh/s
NVIDIA GeForce GTX 980 Ti
2.40 Gh/s
To the rating of video cards in mining Decred →
NVIDIA TITAN V
79 Mh/s
NVIDIA Tesla P100 SXM2
69 Mh/s
AMD Radeon R9 295X2
62 Mh/s
AMD Radeon R9 390 X2
57 Mh/s
AMD Radeon R9 390X2
57 Mh/s
To the rating of video cards in Ethereum mining →
NVIDIA Tesla V100 PCIe
2.02kh/s
AMD Radeon RX Vega 64
1.90 kh/s
NVIDIA Tesla V100 SXM2
1.68kh/s
NVIDIA TITAN V
1.45 kh/s
NVIDIA Tesla P100 SXM2
1. 44kh/s
To the rating of video cards in Monero mining →
NVIDIA TITAN V
3.44 Gh/s
NVIDIA GeForce GTX 1080 Ti
2.96 Gh/s
NVIDIA GeForce GTX 1080
2.28 Gh/s
NVIDIA Tesla P100 SXM2
2.20 Gh/s
NVIDIA GeForce GTX 1070
1.75 Gh/s
To the rating of video cards in Siacoin mining →
NVIDIA TITAN V
884 Sol/s
NVIDIA GeForce GTX 1080 Ti
630 Sol/s
NVIDIA Tesla P100 SXM2
499 Sol/s
NVIDIA GeForce GTX 1080
470 Sol/s
NVIDIA GeForce GTX 980 Ti
461 Sol/s
ZCash Mining Video Card Rating →
Popular Comparisons
Here are some of the most popular video card comparisons of recent times.
GeForce RTX
3060 Ti
or
GeForce RTX
3060
GeForce GTX
1050 Ti
or
GeForce GTX
1650
GeForce RTX
2060
or
GeForce RTX
3050 8GB
GeForce GTX
1660 Super
or
GeForce RTX
3050 8GB
GeForce RTX
3060 Ti
or
GeForce RTX
3070
GeForce RTX
2060 Super
or
GeForce RTX
3060
Popular graphics cards
These graphics cards have been getting the most attention in the last few months.
GeForce GTX
1050 Ti
GeForce RTX
3090 Ti
Radeon RX
Vega 7
GeForce GTX
1650
Top 10 Intel
Top 10 Intel
RU
Help: You can add multiple processors to the comparison (up to 14 processors per table) . If the processors do not fit in the table, a scroll bar will appear. Add Processors 9341
GeForce RTX
3060
Processor
VsRank points
1
AMD Ryzen 9 3900X
11980
2
AMD Ryzen Threadripper 2950X
11936
3
Intel Core i9-9900KS
11908
4
AMD Ryzen Threadripper 1950X
11868
5
AMD Ryzen Threadripper 2920X
11867
6
Intel Core i9-9900K
11858
7
Intel Core i9-10980XE
11815
8
Intel Core i9-7900X
11811
9
AMD Ryzen 7 2700X
11806
10
AMD Ryzen 7 3700X
11794
#
Processor
VsRank points
1
Intel Core i9-9900KS
11908
2
Intel Core i9-9900K
11858
3
Intel Core i9-10980XE
11815
4
Intel Core i9-7900X
11811
5
Intel Core i9-9980XE
11783
6
Intel Core i7-8086K
11728
7
Intel Xeon E-2176G
11707
8
Intel Core i7-8700K
11707
9
Intel Core i7-8700
11550
10
Intel Core i7-4960X
11524
#
Processor
VsRank points
1
AMD Ryzen 9 3900X
11980
2
AMD Ryzen Threadripper 2950X
11936
3
AMD Ryzen Threadripper 1950X
11868
4
AMD Ryzen Threadripper 2920X
11867
5
AMD Ryzen 7 2700X
11806
6
AMD Ryzen 7 3700X
11794
7
AMD Ryzen Threadripper 2970WX
11779
8
AMD Epyc 7F52
11768
9
AMD Epyc 7F52
11768
10
AMD Ryzen 7 1800X
11679
Processor comparison » BNAME.
Multiple processor comparison
Processors / Features
Quick Add Processor
Short Description
Page
Processor description page on our website
Manufacturer
Main processor manufacturers Intel and AMD
Processor family
Processor series
The line or series to which the comparison model belongs.
Processor model
Bus throughput
System bus — throughput is measured in gigatransactions per second.
The number of nuclei
The number of nuclei may not always indicate the high performance of the processor
The number of flows
How many instructions can process the processor for one beat
Base frequencies
The number of operations that the processor can perform per second.
Comparison of performance of processors
Reference: Ratings and benchmarks can be sorted in descending, ascending and list order. If you select the «by list» method, then the rating will be displayed in the order of the processor table.
Benchmark tests
Comparison of instructions and technologies
Name of technology or instruction | Short description |
---|---|
Turbo Boost | Intel auto overclocking technology. |
BPT (Burst Performance technology) | Performance enhancement technology. |
IDA (Intel Dynamic Acceleration) | Intel Dynamic Acceleration Technology. |
Precision Boost | Self-overclocking technology from AMD. |
Turbo Core | AMD Auto Overclocking Technology . |
XFR (Extended Frequency Range) | Extended frequency range. |
Turbo Boost Max 3.0 | Intel auto overclocking technology. |
Technology or instruction name | Short description |
---|---|
EIST (Enhanced Intel SpeedStep) | Enhanced Intel SpeedStep Power Saving Technology. |
Stop Grant state | Energy saving status. |
Extended Stop Grant State | Energy saving status (Advanced technology). |
Sleep state | Sleep state. |
Deep Sleep state | Deep sleep state. |
Deeper Sleep state | Deeper sleep state. |
Enhanced Deeper Sleep state | Improved state of deeper sleep. |
AutoHalt state | Auto stop status. |
C2 and C4 states (Stop Clock and Deeper Sleep) | Stop and deeper sleep. |
Dynamic Cache sizing | Dynamic cache resizing. |
Extended HALT state | Extended stop state. |
C5 state | Extended deeper sleep. |
Deep Power-down technology | Deep power off. |
HALT state | Stop status. |
Pure Power | Technology to dynamically reduce the frequency and power consumption of the processor. |
Enduro technology | Energy management technologies. |
PowerNow! | PowerNow! idle frequency reduction technology. |
Halt mode | Stop mode. |
Stop Grant mode | Energy saving status. |
Cool’n’Quiet | Coolness and silence. |
Enhanced PowerNow! | PowerNow! advanced idle frequency drooping technology. |
CoolCore Technology | Complements Cool’n’Quiet. Temporarily disable unused processor blocks. |
DPM (Dynamic Power management) | Dynamic Power Management. |
DDPM (Dual Dynamic Power Management) | Dual dynamic power management. |
CoolSpeed Technology | Technology to protect the processor from overheating. |
Thermal Monitoring | Temperature monitoring. |
Idle States | Idle states. |
Technology or instruction name | Short description |
---|---|
MMX (Multimedia Extensions) | Multimedia extensions. |
SSE (Streaming SIMD Extensions) | Processor streaming SIMD extension. |
SSE2 (Streaming SIMD Extensions 2) | Processor 2 Streaming SIMD Extension. |
SSE3 (Streaming SIMD Extensions 3) | Processor Streaming SIMD Extension 3. |
SSSE3 (Supplemental Streaming SIMD Extension 3) | Additional SIMD extensions for streaming 3. |
EM64T (Extended Memory 64-bit Technology) | 64-bit extended memory technology. |
NX (Execute disable bit) | Execution inhibit bit. |
SSE4 (Streaming SIMD Extensions 4) | Streaming SIMD Processor Extension 4. |
AES (Intel Advanced Encryption Standard New Instructions) | Command system extension. |
AVX (Advanced Vector Extensions) | Command system extension. |
AVX 2. 0 (Advanced Vector Extensions 2.0) | Command set extension 2.0. |
BMI1, BMI2 (Bit manipulation instructions sets) | X86 bit control command set. |
F16C (16-bit Floating-Point conversion) | 16-bit floating point conversion. |
FMA3 (Fused Multiply-Add 3) | Multiply Add with Round One (FMA3). |
SSE4.1 (Streaming SIMD Extensions 4.1) | Processor 4.1 streaming SIMD extension. |
SSE4A (Streaming SIMD Extensions 4A) | Processor 4A streaming SIMD extension. |
AVX 2 (Advanced Vector Extensions 2) | Command set extension 2. |
BMI1 (Bit manipulation instructions 1) | BMI1 bit control command set. |
AMD64 | 64-bit microprocessor architecture developed by AMD. |
FMA4 (Fused Multiply-Add 4) | Multiply Add with Round One (FMA4). |
XOP (eXtended Operations) | Advanced operations. |
3DNow! | Optional MMX extension for AMD processors. |
SSE4.2 | Processor 4.2 streaming SIMD extension. |
FMA (Fused Multiply-Add) | Fused multiplication-addition. |
Technology or instruction name | Short description |
---|---|
SHA (Secure Hash Algorithm extensions) | Secure hash algorithm extensions. |
TXT (Trusted Execution Technology) | Trusted execution technology. |
MPX (Intel Memory Protection Extensions) | Memory protection extensions. |
SGX (Software Guard Extensions) | Intel Software Protection Extensions. |
SMAP (Supervisor Mode Access Prevention) | Supervisor mode access prevention. |
SMEP (Supervisor Mode Execution Prevention) | Prevent execution in supervisor mode. |
EVP (Enhanced Virus Protection) | Improved virus protection. |
Secure Key | Secure key technology. |
Identity Protection | Personal protection. |
EDB (Execute Disable Bit) | Execute trip bit. |
Anti-Theft | Anti-theft. |
Secure Boot | Secure Boot. |
Technology or instruction name | Short description |
---|---|
VT-x (Virtualization technology) | Virtualization technology. |
VT-d (Virtualization Technology for Directed I/O) | Directed I/O virtualization technology. |
AMD-V | AMD-V Virtualization Technology. |
EPT | Extended page tables. |
VT-c (Virtualization Technology for Connectivity) | Virtualization technology for connectivity. |
VT-i (Virtualization Technology for Itanium Processors) | Virtualization technology for Itanium processors. |
IOMMU 2.0 (Input/Output Memory Management Unit 2.0) | I/O memory control unit 2.0. |
Technology or instruction name | Short description |
---|---|
SIPP (Single In-line Pin Package) | Single row memory modules. |
Flex Memory Access | Flexible memory access. |
Fast Memory Access | Fast memory access. |
eDRAM (Embedded Dynamic Random Access Memory) | Embedded DRAM. |
Technology or instruction name | Short description |
---|---|
Hyper-Threading | Hyperthreading technology. |
SMM (System Management mode) | System control mode. |
TSX (Transactional Synchronization Extensions) | Transactional synchronization extensions. |
Dynamic FSB frequency switching | Dynamic FSB switching. |
SMT (Simultaneous MultiThreading) | Simultaneous multithreading. |
TBM (Trailing Bit Manipulation) | End bit manipulation. |
vPro | Intel Technology Suite, vPro is a marketing term. |
FDI (Flexible Display Interface) | Flexible display interface. |
DBS (Demand Based Switching) | Switching on request. |
QAT (QuickAssist) | Quick help. |
DualGraphics | Dual graphics. |
PAE (Physical Address Extension) | Physical address extension. |
AMT (Active Management Technology) | Active control technology. |
HD Audio (High Definition Audio) | High definition audio. |
My WiFi | Intel Pro Wireless Wi-Fi Management Technology. |
OOBM (Out-of-band management) | Out-of-band management. |
Smart Response Technology | Smart Response caching technology. |
GPIO (General Purpose Input/Output) | General purpose I/O interface. |
Smart Connect Technology | Smart connection. |
Matrix Storage | Matrix storage. |
RST (Rapid Storage Technology) | Rapid storage technology. |
SST (Speed Shift Technology) | Shift technology. |
PowerTune | Dynamic frequency scaling technology. |
FreeSync | Free sync. |
Integrated video comparison
Feature or technology |
---|
Enduro |
Intel® Quick Sync Video |
Maximum video memory |
Maximum GPU clock |
Number of execution units |
Clear Video HD |
Switchable graphics |
Clear Video |
InTru 3D |
Number of video processor cores |
VCE |
UVD |
Number of shaders |
Interface name |
---|
Maximum number of monitors |
LVDS |
eDP |
DisplayPort |
HDMI |
SDVO |
CRT |
VGA |
DVI |
MIPI-DSI |
Image parameter |
---|
Maximum resolution via HDMI 1. 4 |
Maximum resolution via eDP |
4K support |
Maximum resolution via DisplayPort |
Maximum resolution via VGA |
API name |
---|
DirectX |
OpenGL |
Vulcan |
Other data
Technology or parameter name |
---|
PCI |
USB version |
Number of SATA ports |
Maximum number of SATA 6 Gb/s ports |
Number of USB ports |
Integrated LAN |
UART |
Embedded IDE |
Technology or parameter name |
---|
Operating temperature range |
Voltage P0 Vcore |
Benchmarks
Overall performance rating
The overall rating is calculated by an internal formula, taking into account all indicators, such as — test results in all benchmarks, architecture, socket, release year, temperature, number of cores, threads, frequency, auto-acceleration technologies, and much more.
PassMark CPU Mark
Perhaps the most common benchmark on the Internet. It includes a large set of tests for a comprehensive assessment of PC performance, including the processor. Among which are integer calculations, floating point calculations, extended instruction checking, encryption, compression, game physics calculations, multi-threaded and single-threaded tests. At the same time, it is possible to compare the results obtained with other configurations in a common database. Almost all processors presented on our website have been tested in PassMark.
Cinebench 10 (32 bit) Single-threaded test
This benchmark for video cards and processors is already very outdated today. Uses the method of geometric optics — ray tracing. Released by MAXON, and based on the Cinema 4D 3D editor. Single-Core — uses only one core and one thread for rendering in its test. The main performance testing mode is a photorealistic rendering of a 3D scene, working with light, simulating global illumination, multilevel reflections, spatial lights, and procedural shaders. It is possible to test multiprocessor systems. Works under Mac OS X, Windows systems.
Cinebench 10 (32 bit) Multi-thread test
Multi Core is another test option in Cinebench R10 that already uses multi-thread and multi-core testing mode. It should be noted that the possible number of threads in this version is limited to 16.
Cinebench 11.5 (64-bit) Multi-threaded test
64-bit version of the CINEBENCH 11.5 test — which can load the processor at 100% using all cores and threads. Unlike older versions, 64 threads are already supported here.
Cinebench 11.5 (64-bit) Single-threaded test
Good old full featured Cinebench R11.5 from Maxon. His tests are still relevant. The tests still use the ray tracing method, rendering a complex three-dimensional room with many glass and crystal balls. In this Single-Core variant, tests are performed using one thread and one core. The result of the test is the «frames per second» value.
Cinebench 15 (64-bit) Multi-thread test
Multi-Thread version of Cinebench 15 — will boot your system to the full, showing what it can do. All cores and processor threads are involved when rendering complex 3D objects. Ideal for modern multi-threaded processors from AMD and Intel, as it is capable of using 256 computing threads.
Cinebench 15 (64-bit) Single-threaded test
Cinebench Release 15 is the most up-to-date tester from Finns from Maxon. The system is being tested: both processors and video cards. For processors, the result of the analysis will be the value of PTS points, and for video cards, the number of frames per second FPS. A complex 3D scene is rendered with many detailed objects, light sources and reflections. In the Single Core version, one thread is involved in rendering.
Geekbench 4.0 (64-bit) Multi-threaded test
This is already a 64-bit multi-threaded Geekbench 4 test. It is the wide support for devices and operating systems that makes Geekbench tests the most popular today.
Geekbench 4.0 (64-bit) Single-threaded test
The current single-threaded version of Geekbench 4 for testing desktop PCs and laptops. The benchmark is still running on systems running Mac OS, Windows, Linux, just like its versions. For the first time, Android and iOS mobile devices are also supported in this release. The Single-Core test uses one thread.
Geekbench 3 (32 bit) Multi-threaded test
Multi-threaded version of Geekbench 3 benchmark — will allow you to stress test your processor and show how stable your system is.
Geekbench 3 (32 bit) Single-threaded benchmark
The Geekbench cross-platform benchmark is often used to evaluate a system under Max, although it will run on both Windows and Linux. The main purpose of testing the performance of processors. The 32-bit version of the test uses one thread and one processor core.
Geekbench 2
A very outdated version of the Geekbench 2 benchmark. Our archive contains almost two hundred processor models that have test data in this program. Today there are newer versions of the actual 4v and 5v.
X264 HD 4.0 Pass 1
This is a hands-on test of system performance by transcoding HD video files to H.264, the so-called MPEG 4 x264 codec. The number of frames processed per second is the result of the test. This test is faster than Pass 2 because it encodes at a constant rate. An ideal test for multi-core and multi-threaded processors.
X264 HD 4.0 Pass 2
This is a slightly different, slower test based on video file compression. The same MPEG4 x264 codec is used, but encoding is already done at variable bit rate. As a result, we get a higher quality video file. The resulting value is also measured in frames per second. You need to understand that a real task is being simulated, and the x264 codec is used in many encoders. Therefore, the test results really reflect the effectiveness of the system.
3DMark06 CPU
Based on DirectX 9.0 by the Finnish company Futuremark. Benchmark program for testing the video system, and the central processor. Processors are tested in two ways: the game artificial intelligence calculates the pathfinding, and the second test emulates the game physics engine using PhysX. This test is very often used by overclockers and gamers.
3DMark Fire Strike Physics
Nearly 200 processors on our site have 3DMark FSP test data. This is a math test that calculates physics
WinRAR 4.0
A well-known file archiver. The tests were performed under Windows OS. The speed of compression into the RAR format was checked, for this large volumes of random data were generated. The resulting speed during compression «Kb / s» — this is the result of the test.
TrueCrypt AES
Not exactly a benchmark, but the results of its work can give an estimate of system performance. Unfortunately support for this project ended on May 28, 2014. The program has a built-in ability to encrypt disk partitions on the fly. Our site presents the results of encryption speed in Gb / s using the AES algorithm. The program can run on Windows, Linux and Mac OS X operating systems.
What is a processor comparison?
Processor Comparison is a tool that allows the user to visually analyze the specifications, performance and test results of two or more CPUs. (Our service allows you to compare up to 14 processors)
How to compare processors?
Start by adding processors to the table. To do this, click on the green button «Add processors» — enter a search query (For example, the name of the model. You can also specify the architecture and socket), and you will get a list of available models.
If you click on the «Checkbox» checkbox, then the processor will be included in the list of selected ones. You can select multiple models and then click on the «Add Selected» button. If you click on the name of the processor, a quick addition will occur.
After you have selected 2 or more models, you can start analyzing processors. There are many performance tests available on the page, there is an overall performance rating, benchmarks, synthetic tests, a list of technologies and instructions, as well as technical specifications.
To our comparison tool — we have added the ability to select processors, for this you just need to click on the name of the processor in the table or in the rating, after which the selected model will be highlighted and highlighted. This is necessary so that you do not get confused in the rating of performance and characteristics.
In addition, this allows you to quickly remove the selected elements or perform an inversion (ie, leave only the selected, and delete the rest). There is a function to change the processor (Click on the «Change» button) as well as a quick deselection function.
Popular comparisons
Ryzen Threadripper PRO 3995WX
Ryzen Threadripper 3990X
Ryzen Threadripper 3970X
Ryzen Threadripper PRO 3975WX
Ryzen Threadripper 3960X
Ryzen 9 5950X
Ryzen Threadripper PRO 3955WX
Ryzen 9 5900X
Ryzen 9 3950X
Ryzen 9 5900
Core i9-10980XE
Ryzen Threadripper PRO 3945WX
Ryzen 9 3900XT
Ryzen 9 3900X
Ryzen Threadripper 2990WX
Core i9-9980XE
Core i9-9990XE
Ryzen 9 PRO 3900
Core i9-9960X
Ryzen 9 3900
Ryzen Threadripper 2950X
Core i9-7980XE
Ryzen Threadripper 2970WX
Core i9-10940X
Ryzen 7 5800X
Core i9-9940X
Core i9-7960X
Ryzen Threadripper 1950X
Core i9-7940X
Ryzen 7 5800
Core i9-10920X
Core i9-11900K
Core i9-11900KF
Ryzen Threadripper 2920X
Core i9-9920X
Core i9-10900K
Ryzen 7 3800XT
Core i9-10900KF
Xeon W-1290P
Core i9-7920X
Core i9-10850K
Ryzen 7 3800X
Ryzen Threadripper 1920X
Core i9-10900X
Ryzen 7 5700G
Ryzen 7 3700X
EPYC 7371
Ryzen 7 PRO 3700
Core i9-11900F
Core i9-11900
Xeon W-1290E
Ryzen 5 5600X
Core i9-10910
Xeon W-1290
Core i9-9900X
Core i9-7900X
Core i9-10900
Core i9-10900F
Ryzen 7 PRO 4700G
Ryzen 7 PRO 4750G
Core i7-11700K
Core i9-9820X
Core i7-11700KF
Ryzen 5 5600G
Ryzen 7 4700G
Core i9-9900KS
Core i7-10700K
Xeon W-1370P
Core i7-10700KF
Xeon W-1270P
Core i9-9900KF
Core i9-9900K
Ryzen 5 3600XT
Core i7-11700F
Core i7-11700
Ryzen 7 4700GE
Core i7-9800X
Ryzen 5 3600X
Ryzen 7 PRO 4750GE
Xeon W-1270
Ryzen 5 PRO 4650G
Ryzen 7 2700X
Ryzen 5 4600G
Core i5-11600KF
Core i7-10700
Core i5-11600K
Core i7-7820X
Ryzen 7 5700GE
Core i9-11900T
Core i9-9900
Xeon W-1290T
Ryzen 5 3600
Core i7-10700F
Ryzen Threadripper 1950
Ryzen Threadripper 1900X
Ryzen 7 5700U
Core i9-10900T
Ryzen 5 4600GE
Ryzen 7 1800X
Ryzen 7 PRO 2700X
Online performance comparison of processors and video cards
Main specifications, popular benchmark results, performance rating and reviews
ASUS Dual GeForce RTX 2070 SUPER EVO OC
EVGA GeForce RTX 2070 SUPER BLACK GAMING
Intel Xeon E5-2640 v2
Intel Xeon E7-8891v2
AMD Epyc 7501
Intel Core i9-7940X
Intel Pentium G4600
Intel Atom C3958
AMD Epyc 7401
Intel Core i9-7940X
Gainward GeForce RTX 2080 Founders
ASUS Dual GeForce RTX 2070 SUPER EVO OC
Intel Pentium Gold G6500
Intel Core i9-7960X
Sapphire Pulse Radeon RX 6800
EVGA GeForce GTX 1080 Ti SC2 ELITE GAMING GREEN
ZOTAC GeForce GTX 1070 Mini
EVGA GeForce GTX 1070 GAMING ACX 3. 0
Intel Core i9-7920X
Intel Xeon W-2170B
GIGABYTE Radeon RX 570 Gaming 4G
GALAX GeForce GTX 1070 EX
Intel Core i9-7920X
Intel Xeon W-2125
Colorful iGame GeForce GTX 1080 Ti Neptune W-V
Palit GeForce RTX 2080 SUPER GP
AMD Epyc 7302
AMD Epyc 7713P
Processors (CPUs)
TOP 10 processors
PassMark CPU Mark
AMD Ryzen Threadripper Pro 3995WX
100%
AMD Ryzen Threadripper 3990X
99%
AMD Ryzen Threadripper 3970X
98%
AMD Ryzen Threadripper PRO 3975WX
97%
AMD Ryzen Threadripper 3960X
96%
Best Embedded Processors
iGPU
Intel Core i9-9980XE
100%
AMD Ryzen Threadripper 2970WX
99%
Intel Core i9-12900KF
98%
AMD Ryzen Threadripper Pro 3955WX
97%
AMD Ryzen 9 3950X
96%
Top performance
Monero Hashrate kH/s
AMD Epyc 7763
100%
AMD Epyc 7742
99%
AMD Epyc 7702P
98%
AMD Epyc 7702
97%
AMD Ryzen Threadripper 3990X
96%
The best processors in terms of price-quality ratio
Fire Strike and CompuBench OpenCL benchmarks
Intel Celeron 6305
100%
AMD Ryzen 3 3200G
99%
AMD Ryzen 5 3400G
98%
AMD Ryzen 3 2200G
97%
AMD Ryzen 5 2400G
96%
Popular processors
CPU and iGPU benchmarks
AMD Ryzen 9 5950X
99%
Intel Core i5-11260H
95%
Intel Core i5-11500T
85%
Intel Core i7-1160G7
100%
Video cards (GPU)
TOP 10 video cards
FP32 Performance
ASRock Radeon RX 5700 XT Phantom Gaming D 8G OC
100
MSI Radeon RX 5700XT EvokeOC
99
GIGABYTE GeForce GTX 1080 Xtreme Gaming WATERFORCE WB 8G
98
GIGABYTE GeForce GTX 1080 Xtreme Gaming WATERFORCE 8G
97
GIGABYTE GeForce GTX 1080 Xtreme Gaming Premium Pack 8G
96
Best video card
3DMark Benchmark
Colorful GeForce RTX 2070 SUPER 8G-V
100
EVGA GeForce RTX 2070 SUPER
99
MANLI GeForce RTX 2070 Super Gallardo (M3445)
98
MSI GeForce RTX 2070 SUPER ARMOR
97
MSI GeForce RTX 2070 SUPER GAMING
96
Top performance
Battlefield 5
PowerColor Radeon RX 6800 Red Devil Limited Edition
100
ASRock Radeon RX 6800 Phantom Gaming D 16G OC
99
ASUS ROG Strix Radeon RX 6800OC
9eight
ASUS TUF Gaming Radeon RX 6800OC
97
GIGABYTE AORUS Radeon RX 6800 MASTER 16G
96
The best graphics cards in terms of price-quality ratio
Fire Strike and CompuBench OpenCL benchmarks
Palit GeForce GTX 1050 StormX 3GB
99%
GIGABYTE Radeon RX 580 Gaming 4G
97%
ASUS TUF Gaming GeForce RTX 3070OC
93%
Gigabyte GeForce GTX 1650 GAMING OC 4G
92%
MSI GeForce RTX 2060 VENTUS 6G OC
90%
Popular graphics cards
CPU and iGPU tests
MSI GeForce GTX 1060 3G OC
96%
MSI GeForce GTX 1050 GAMING X 2G
96%
ASUS ROG Strix GeForce GTX 1050 OC Edition 2GB
94%
GIGABYTE AORUS GeForce GTX 1060 6G 9Gbps
93%
INNO3D GeForce GTX 1050 Ti Twin X2 V2
92%
Popular comparisons
- Processors (CPUs)
- Video cards (GPU)
Raspberry Pi 4B (Broadcom BCM2711)
Raspberry Pi 3 B+ (Broadcom BCM2837B0)
Intel Core i5-12600K
AMD Ryzen 5 5600G
Intel Core i7-8700B
AMD Ryzen 5 1500X
AMD Ryzen 9 5950X
AMD Ryzen 7 2700X
AMD Ryzen 9 5950X
AMD Ryzen 9 3900
Intel Atom E3805
AMD Ryzen Threadripper 3960X
ASUS TUF Gaming GeForce RTX 3060 V2 OC Edition LHR
ZOTAC GAMING GeForce RTX 3060 Twin Edge OC
KFA2 GeForce GTX 1650 Prodigy
ZOTAC GeForce GTX 1050 Ti OC Edition
ASUS Dual GeForce RTX 3060 V2 LHR
Palit GeForce RTX 3060 Dual
MANLI GeForce GTX 1660 Super (M1432+N537-00)
ASUS ROG Strix GeForce GTX 1660 Super Advanced
Gainward GeForce RTX 3080 Phantom+ GS LHR
GIGABYTE AORUS GeForce RTX 3080 Xtreme 10G
ASUS Phoenix Radeon RX 550 4G EVO
NVIDIA GeForce RTX 2060 Founders Edition
Views
- All
- Processors (CPUs)
- Video cards (GPU)
Qualcomm Snapdragon 8+ Gen1
Watch
EVGA GeForce GTX 1060 FTW GAMING 3GB
Watch
AMD Athlon Gold Pro 3150G
Watch
Palit GeForce GTX 1070
Watch
Qualcomm Snapdragon 8+ Gen1
Watch
AMD Athlon Gold Pro 3150G
Watch
AMD Athlon II X3 440
Watch
AMD Epyc 7252
Watch
Palit GeForce GTX 1070
Watch
EVGA GeForce GTX 1060 FTW GAMING 3GB
Watch
EVGA GeForce RTX 2070 SUPER FTW3 ULTRA+
Watch
EVGA GeForce RTX 2070 SUPER XC ULTRA+
View
About project
Our service uses a conditional system for evaluating CPU and GPU performance. The performance data for ARM processors was taken from various sources, mainly based on the results of such tests, for example: PassMark, Antutu, GFXBench.
We do not claim absolute accuracy. It is impossible to rank and measure performance with absolute accuracy of ARM processors, for the simple reason that each of them has advantages in some ways, and lags behind other ARM processors in some ways.
How to compare processor performance
main
Tips
21.03.2016 — 0
When buying a new computer, the same question always arises — how to compare processors, how to understand which one is better. It is known that the power of the processor greatly affects the performance of the system as a whole. But what determines the performance of the processor itself?
Let’s figure it out.
Clock speed — is it so important?
Processor clock speed is measured in hertz, usually gigahertz (GHz). The CPU clock speed is a measure of how many cycles per second it can execute. For example, a 1.8 GHz processor can run 1,800,000,000 clock cycles per second. The clock cycle can be called, the time during which the CPU performs a micro-operation. For example, the operation of reading data from RAM may consist of several micro-ops.
At first glance, it seems logical that the more clock cycles per second it can perform, the more productive it is. Yes and no.
For a reliable comparison criterion, clock speeds can be used when considering similar processors from the same series. Let’s say you’re comparing two Intel Core i5-4460 and Intel Core i5-4670K processors, which differ only in their clock speed. The first operates at a frequency of 3.2 GHz, and the second at a frequency of 3.4 GHz. In this case, the 3.4 GHz processor will run 30% faster when they are both running at their maximum speed. But you can’t compare the frequency of a processor from the Haswell Core i5 family with an AMD, ARM or even i7 processor.
Modern processors are becoming much more efficient. They can do more things per clock. For example, the 3.6GHz Intel Pentium 4 was released in 2006, while the Intel Haswell i7 is clocked at 3.9GHz. Does this mean that CPU performance has only improved a tiny bit in the last 10 years? Of course not.
Cache
The processor cache is a form of memory dedicated to the processor, which works similarly to RAM.
The cache is built directly into the processor die. From there, the process accesses reusable data directly from its own memory rather than re-requesting it from system memory, saving significant time.
Cache memory comes in various levels, L1, L2, L3 and even more in new processors. L1 represents a small amount of very fast memory, and each successive level is larger and slower. The processor uses each level according to the importance of the data, the most important data will be stored in the L1 cache.
Haswell processors have 64KB L1 cache, L2 — L3, 256KB up to 20MB and L4 cache up to 128MB.
It’s hard to say how much cache you need, but 3-6MB is common in modern laptops and desktops. More cache means better performance.
Instruction set
This is the “firmware” of the processor itself, which tells you how to perform actions. You have no control over the instruction set. It is built into the processor and cannot be changed or updated. But along with the architecture of the processor, this affects the performance of processors. The processor architecture determines how many cycles it takes to execute a given instruction.
In other words, some instruction sets are more efficient than others, allowing the processor to do more useful work at a given speed. But you should not climb into such jungle and take this parameter into account when buying a computer for your home.
A few words about the number of cores
Each «core» of the processor is actually a separate data processing unit. Additional cores allow the computer to do several things at the same time. You will return that I had a single-core processor, but it also did a bunch of things at the same time — the browser was running, music was playing, and a course paper was being made.
In fact, this is just an illusion of simultaneity, in fact, the processor switches each time between tasks, performing part of the work from each process, it does it so quickly that we do not notice.
The more processors or cores a computer has, the more things it can do at once. Imagine that we have 100 tasks and 4 engineers (cores), each one does 25 tasks, and of course, all 100 tasks will be done faster than one. But there are such complex voluminous tasks that cannot be divided, and it looks like one engineer is working, and three are resting. What follows from this is that not all software works more efficiently on multi-core processors than on one.
Total — the more cores, the better
In what parrots performance is measured
MIPS units (million operations per second) and FLOPS are often used
MIPS as well as the clock frequency is meaningless to use for processors of different architectures. It is sometimes also referred to as the Meaningless Indicator of Processor Speed. Since the number of operations cannot serve as an indicator of “useful” performance. Exaggeratedly, one processor will multiply two numbers in one operation, and the other in 10, but the operation of the second processor takes two times less time. Obviously, the first one will be faster.
A more universal measure of FLOPS is the number of floating point operations per second. Floating point is the most useful representation of numbers in processors. By measuring FLOPs, they force all processors to do the same thing. Of course, such a test is more universal, and is suitable for measuring performance.
Processor performance value
Few would argue that the overall performance of a computer is largely dependent on the processor. But it would be erroneous to assume that he completely defines it. It is possible to choose a configuration where the full potential of the processor can be ruined
What affects performance:
- Mainboard internal bus frequency
- RAM clock and timings
- Hard disk rotation speed
Benchmarks
What is a benchmark?
The benchmark includes a series of software tests that replicate the kinds of tasks performed in real-life computer use. For example, the CPU in a laptop will be tested to see how fast it can compress or encrypt data. The hard drive will be checked for read/write speed, etc.
How important are benchmark scores?
Benchmark results are used for comparative analysis. They are especially good at showing how performance improves from one generation to the next and can help you evaluate the value for money of a product.
These are best used when you have specific requirements, be it gaming, video editing or anything else resource intensive.
But for solving everyday computing problems — web surfing, social. networking and text editing — the difference in performance is barely noticeable. Because the benchmark shows peak performance values.
What benchmarks exist
- Test with floating point operations. Results are often displayed in milliseconds, so lower numbers indicate better performance. That is, how quickly one operation will be done. Based on this time, you can approximately calculate the number of FLOPS.