unnamed file

4f2200008e2902e14cee c5 q75 864x486 1 scaled

In today’s Core i3-10100 review we’ll take a close look at what this quad-core processor offers at its highly affordable $130 price point. Intel launched its 10th generation Core “Comet Lake” desktop processors at a brisk pace, with the entry-level Core i3 parts following only a couple of weeks behind the flagship Core i9-10900K. It took AMD nine months to launch its “Zen 2” based Ryzen 3 series parts. Intel also had its motherboard partners announce products based on the cost-effective B460 and H410 chipsets because pairing a Core i3 processor with a $150+ Z490 motherboard makes little sense.

Priced at $130, the Core i3-10100 is the entry-level of the 10th generation Core desktop processor family. Keeping in tune with Intel enabling HyperThreading across the board for Comet Lake, this is a 4-core/8-thread processor. Before AMD “Zen” came along, the price of entry for eight threads from Intel was around $300 for the cheapest Core i7 quad-core part. Intense competition from AMD forced Intel to increase core counts generation over generation. The 8th and 9th generation Core i3 were 4-core/4-thread parts, after nine or so years of Core i3 desktop being 2-core/4-thread with 3 or 4 MB of L3 cache and a lack of Turbo Boost. The consumer always benefits from competition.

unnamed file 321

The reviewed Core i3-10100 is based on the “Comet Lake” microarchitecture by Intel, and built on their existing 14 nm++ silicon fabrication process. The per-core performance (IPC) of these chips is identical to “Skylake” from 2016. The Core i3-10100 features four cores and HyperThreading, enabling eight logical processors. These four cores, however, are cushioned by only 6 MB of L3 cache. Historically (7th generation and prior), Intel used 6 MB of cache on 4-core/4-thread Core i5 SKUs, while reserving 8 MB for the 8-thread Core i7 parts. Intel is trying something different with the 10th generation. While the Core i3-10100 has 6 MB, the slightly pricier Core i3-10300 and Core i3-10320 have 8 MB of cache and slightly higher clock speeds.

The i3-10100 ticks at 3.60 GHz and has a maximum Turbo Boost frequency of 4.30 GHz. It also features an integrated graphics solution in the form of the Gen 9.5-based Intel UHD 630, clocked up to 1.10 GHz. Across the competitive landscape, the Core i3-10100 faces the recently launched AMD Ryzen 3 3100 processor at $100 and the Ryzen 3 3300X at $120. Both are 4-core/8-thread parts based on the “Zen 2” microarchitecture with large 16 MB L3 caches, and some value additions, such as PCIe gen 4.0 and an unlocked multiplier. Both chips also earned critical acclaim for providing high performance per dollar with gaming. That’s why we will put a special focus on those two SKUs in our Core i3-10100 review. It’ll be interesting to see whether Intel’s $130 offering has the potential to gravitate the e-sports and entry-level gaming PC crowd away from AMD. It’s also worth exploring if buying a Core i5 processor still makes sense for paper-pushing office desktops that mostly deal with web-browsing, MS Office, etc.

Intel Core i3-10100 Market Segment Analysis
Price Cores /
Threads
Base
Clock
Max.
Boost
L3
Cache
TDP Architecture Process Socket
Athlon 3000G $50 2 / 4 3.5 GHz N/A 4 MB 35 W Zen 14 nm AM4
Athlon 200GE $55 2 / 4 3.2 GHz N/A 4 MB 35 W Zen 14 nm AM4
Ryzen 3 1200 $60 4 / 4 3.1 GHz 3.4 GHz 8 MB 65 W Zen 14 nm AM4
Core i3-9100F $75 4 / 4 3.6 GHz 4.2 GHz 6 MB 65 W Coffee Lake 14 nm LGA 1151
Athlon 240GE $80 2 / 4 3.5 GHz N/A 4 MB 35 W Zen 14 nm AM4
Ryzen 3 2200G $85 4 / 4 3.5 GHz 3.7 GHz 4 MB 65 W Zen 14 nm AM4
Core i3-10100 $130 4 / 8 3.6 GHz 4.3 GHz 6 MB 65 W Comet Lake 14 nm LGA 1200
Ryzen 3 3100 $100 4 / 8 3.6 GHz 3.9 GHz 16 MB 65 W Zen 2 7 nm AM4
Pentium G5600 $100 2 / 4 3.9 GHz N/A 4 MB 54 W Coffee Lake 14 nm LGA 1151
Ryzen 5 1400 $105 4 / 8 3.2 GHz 3.4 GHz 8 MB 65 W Zen 14 nm AM4
Ryzen 3 1300X $115 4 / 4 3.4 GHz 3.7 GHz 8 MB 65 W Zen 14 nm AM4
Ryzen 5 1600 $110 6 / 12 3.2 GHz 3.6 GHz 16 MB 65 W Zen 14 nm AM4
Ryzen 3 3300X $120 4 / 8 3.8 GHz 4.3 GHz 16 MB 65 W Zen 2 7 nm AM4
Ryzen 5 2600 $120 6 / 12 3.4 GHz 3.9 GHz 16 MB 65 W Zen 12 nm AM4
Core i3-8300 $140 4 / 4 3.7 GHz N/A 8 MB 65 W Coffee Lake 14 nm LGA 1151
Core i3-10300 $150 4 / 8 3.7 GHz 4.4 GHz 8 MB 65 W Comet Lake 14 nm LGA 1200
Ryzen 5 1500X $140 4 / 8 3.5 GHz 3.7 GHz 16 MB 65 W Zen 14 nm AM4
Ryzen 5 2400G $150 4 / 8 3.6 GHz 3.9 GHz 4 MB 65 W Zen 14 nm AM4
Ryzen 5 1600X $150 6 / 12 3.6 GHz 4.0 GHz 16 MB 95 W Zen 14 nm AM4
Ryzen 5 2600X $150 6 / 12 3.6 GHz 4.2 GHz 16 MB 95 W Zen 12 nm AM4
Core i5-9400F $180 6 / 6 2.9 GHz 4.1 GHz 9 MB 65 W Coffee Lake 14 nm LGA 1151
Core i5-10400F $160 6 / 12 2.9 GHz 4.3 GHz 12 MB 65 W Comet Lake 14 nm LGA 1200
Ryzen 7 1700 $170 8 / 16 3.0 GHz 3.7 GHz 16 MB 65 W Zen 14 nm AM4
Ryzen 7 1700X $170 8 / 16 3.4 GHz 3.8 GHz 16 MB 95 W Zen 14 nm AM4
Core i5-10500 $200 6 / 12 3.1 GHz 4.5 GHz 12 MB 65 W Comet Lake 14 nm LGA 1200
Ryzen 5 3600 $175 6 / 12 3.6 GHz 4.2 GHz 32 MB 65 W Zen 2 7 nm AM4
unnamed file 322

Our Core i3-10100 sample came in a tray-only package. The retail packaging includes a heatsink, which will help keep overall system cost down.

unnamed file 323

 

unnamed file 324

The Core i3-10100 looks like any LGA1xxx processor released by Intel in the past decade. The processor is only compatible with socket LGA1200 motherboards because the position of the round notches has been changed. It will not work with an older motherboard.

unnamed file 325

Luckily, socket LGA1200 retains cooler compatibility with all older LGA115x-series sockets. This means you’re going to be spoiled for choice when picking a cooler to go with this processor.

Architecture

unnamed file 326

 

unnamed file 327

Under the hood of the Core i3-10100 is the 4-core “Comet Lake-S” silicon built on the same 14 nm++ process as the previous two generations. The die area is estimated to be 125 mm². This die looks similar to the 4-core “Kaby Lake” die. Certain steppings could also be carved out of the 6-core “Comet Lake-S” die with two cores and half its L3 cache disabled.

unnamed file 328

The “Comet Lake-S” silicon is laid out similar to the past four generations of Intel mainstream processors, with two rows of CPU cores flanked by the iGPU on one side and the system agent (integrated northbridge) on the other, and a Ringbus Interconnect serving as town square between the various components. The last-level cache is scattered across as slices, adding up to 6 MB of unified L3 cache all cores can access equally.

Much of the processor’s uncore components are clumped into the System Agent, which contains the memory controller, PCI-Express gen 3.0 root-complex, DMI interface, and memory PHY. The iGPU solution, though present on the silicon, is permanently disabled by Intel. On the other end of the ringbus is the Gen 9.5 integrated graphics, which has practically been carried over for the past three generations, featuring 24 execution units in the GT2 trim. All SKUs in the desktop 10th gen processor series appear to have the top GT2 trim. Don’t expect to play PUBG at 4K on this; the “UHD” moniker only indicates that the IGP can handle 4K Ultra HD displays, features modern connectivity options, such as DP 1.4 and HDMI 2.0, and can playback 4K video in new formats with 10-bpc color and HDR10/Dolby Vision standards.

unnamed file 329

The core itself is identical in design to “Skylake,” and there are hence no IPC increases to be had. As we explained in the introduction, all of Intel’s efforts to increase gaming, single-threaded, and less-parallelized application performance revolve around increasing clock speeds and deploying as many as three intelligent boosting algorithms to achieve the advertised clock speeds.

The Core i3-10100 has a nameplate base frequency (aka nominal clock) of 3.60 GHz and a maximum Turbo Boost frequency of 4.30 GHz. Unlike the top Core i9-10900K part, it lacks Turbo Boost Max 3.0 or Thermal Velocity Boost, and makes do with classic Turbo Boost 2.0. It still has significantly increased power limits over the Core i3-9100. The TDP of the chip is rated at just 65 W, and so with just four cores to go around, nearly all socket LGA1200 motherboards should be able to optimally run this processor.

Intel introduced a handful of overclocking enhancements with the 10th generation, including the ability to toggle HyperThreading on a per-core basis rather than globally. This could be an interesting option for those gaming and streaming, where a certain number of cores have HTT disabled for the best gaming performance and certain cores have them enabled, with Windows process core affinity settings taking care of the rest.

The company also introduced the ability to overclock the DMI chipset bus. DMI is a PCIe-based interconnect that handles transfers between the processor and the chipset (PCH). The LGA1200 platform uses DMI 3.0 (comparable to PCI-Express 3.0 x4 in terms of bandwidth). Intel has apparently decoupled PCIe clock domains to enable you to overclock the DMI and PEG (that topmost x16 PCIe slot) without destabilizing your PCIe setup for graphics cards. Multiplier-based overclocking, however, isn’t possible on the Core i3-10100.

The Z490, H470, and B460 Platforms

unnamed file 330

Z490 is the top 400-series chipset targeted at gaming desktops and PC enthusiasts, as it enables serious overclocking and multi-GPU support. In terms of I/O capabilities, the chipset is nearly identical to the Z390, with 24 downstream PCIe gen 3.0 lanes, six SATA ports, six USB 3.2 gen 2 ports that can be converted to three USB 3.2 gen 2×2 ports, ten USB 3.2 gen 1 ports, and fourteen USB 2.0 ports. Intel is recommending its i225-V 2.5 Gbps Ethernet chip as the wired networking solution to go with Z490, and the company’s AX201 802.11ax WiFi 6 WLAN solution to go with the chipset’s CNVio interface.

unnamed file 331

 

unnamed file 332

You are more likely to pair locked and entry-level processors such as the i3-10100 with the B460 or H470 chipsets. B460 has motherboards start at around the $90 mark. It comes with 16 downstream PCIe gen 3.0 lanes (compared to just 12 on the previous-generation B360). Compared to Z490, you get fewer PCIe lanes (16 vs. 24) from the chipset, and fewer USB 3.2 ports (eight 5 Gbps ports and no 10 Gbps ports compared to six 10 Gbps and ten 5 Gbps ports on the Z490). You also lose out on CPU overclocking features and multi-GPU capabilities (such as SLI). B460 motherboards also come with memory frequency restrictions set to DDR4-2933. The H470 is an interesting middle ground between the Z490 and B460. You still lose out on multi-GPU and overclocking, but get more platform PCIe lanes (20 vs. 16 on the B460 and 24 on the Z490), as well as four 10 Gbps USB 3.2 ports in addition to what you get from the B460.

For multiplier-locked chips like the i3-10100, you could save a lot of money by opting for cheaper B460 or H410 chipset motherboards.

Test Setup

  • All applications, games, and processors are tested with the drivers and hardware listed below—no performance results were recycled between test systems.
  • All games and applications are tested using the same version.
  • All games are set to their highest quality setting unless indicated otherwise.
Test System “Comet Lake”
Processor: All Intel 10th Generation processors
Motherboard: ASUS Z490 Maximus XII Extreme
Intel Z490, BIOS 0508
Memory: 2x 8 GB G.SKILL Flare X DDR4
DDR4-3200 14-14-14-34
DDR4-2666 Test at 16-16-16-36
Graphics: EVGA GeForce RTX 2080 Ti FTW3 Ultra
Storage: 1 TB SSD
Cooling: Noctua NH-U14S
Zadak Spark 240 mm AIO
Power Supply: Seasonic SS-860XP
Software: Windows 10 Professional 64-bit
Version 1903 (May 2019 Update)
Drivers: NVIDIA GeForce 430.63 WHQL
AMD Chipset 1.07.07.0725

 

unnamed file 333

 

unnamed file 334

 

unnamed file 335

 

unnamed file 336

 

Test System “Zen 2”
Processor: All AMD Ryzen 3000
Motherboard: ASRock X570 Taichi
AMD X570, BIOS v2.80 AGESA 1.0.0.4B
Memory: 2x 8 GB G.SKILL Flare X DDR4
DDR4-3200 14-14-14-34
All other specs same as above

 

Test System “Coffee Lake”
Processor: All Intel 8th & 9th Generation processors
Motherboard: Core i9-9900KS: ASRock Z390 Phantom Gaming X
All other Coffee Lake: ASUS Z390 Maximus XI Extreme
Intel Z390
Memory: 2x 8 GB G.SKILL Flare X DDR4
DDR4-3200 14-14-14-34
All other specs same as above

 

Test System “Zen”
Processor: All AMD Ryzen 2000, Ryzen 2000G and Ryzen 1000
Motherboard: MSI X470 Gaming M7 AC
AMD X470, BIOS 7B77v19O
Memory: 2x 8 GB G.SKILL Flare X DDR4
DDR4-3200 14-14-14-34
All other specs same as above

Super Pi

SuperPi is one of the most popular benchmarks with overclockers and tweakers. It has been used in world-record competitions since forever. It is a purely single-threaded CPU test that calculates Pi to a large number of digits—32 million for our testing. Released in 1995, it only supports x86 floating-point instructions and thus makes for a good test for single-threaded legacy application performance.

unnamed file 834

 

wPrime

While SuperPi focuses on calculating Pi, wPrime tackles another mathematical problem: finding prime numbers. It uses Newton’s Method for that. One of the design goals for wPrime was to engineer it so that it can make the best use of all cores and threads available on a processor.

unnamed file 837

Rendering — Cinebench

Cinebench is one of the most popular modern CPU benchmarks because it is built around the renderer of Maxon’s Cinema 4D software. Both AMD and Intel have been showing this performance test at various public events, making it almost an industry standard. Using Cinebench R20, we test both single-threaded and multi-threaded performance.

unnamed file 838

 

unnamed file 839

 

Rendering — Blender

Blender is one of the few professional-grade rendering programs out there that is both free and open source. That fact alone helped build a strong community around the software, making it a highly popular benchmark program due to its ease of use as well. For our testing, we’re using the Blender “BMW 27” benchmark scene.

unnamed file 840

 

Rendering — Corona

Corona Renderer is a modern photorealistic renderer that’s available for Autodesk 3ds Max and Cinema 4D. It delivers physically plausible and predictable output due to its realistic lighting algorithm, global illumination, and beautiful materials. Corona does not support GPU rendering, so CPU performance is very important for all its users.

unnamed file 841

 

Rendering — KeyShot

The standalone KeyShot rendering software features fast and efficient workflows that help you get high-quality realistic product shots in the shortest possible time frame. Real-time raytracing, multi-core photon mapping, adaptive material sampling, and a dynamic lighting core provide high-quality images that update instantly even when interactively working on the scene. KeyShot is optimized for usage on CPUs only, which lets them use more complex algorithms than a GPU-based renderer. Unlike our other rendering tests, we record “frames per second” in KeyShot while rendering, so higher numbers are better.

unnamed file 842

Game Development — Unreal Engine 4

Unreal Engine 4 is one of the leading multi-platform game engines in the industry. Not only advanced, it also has lots of features to help you get results quicker than with competing products—time is money. Before a game is shipped, the lengthy process of “light baking” has to be executed. It takes all static geometry and fixed light sources in the scene and precalculates lightmap textures for them, which results in a tremendous performance speed boost of the final game because these calculations no longer have to be performed in real time on the user’s system. For our benchmarks, we generate baked light maps for a relatively simple scene, which usually takes several hours.

unnamed file 843

 

Software Development — Visual Studio C++

Microsoft Visual C++ is probably the most popular programming language for creating professional Windows applications. It’s part of Microsoft’s Visual Studio development suite, which has a long history and is widely accepted as the gold standard when it comes to IDEs. Compiling software is a relatively lengthy process that turns program code into the final executable, and programmers hate waiting for it to complete. We run a medium-sized application through the C++ compiler and linker, and execute the resource compiler, too. The build is executed in “release” mode with all optimizations turned on and multi-processor compilation enabled.

unnamed file 844

Web Browsing — Google Octane

Google Octane tests web browser performance by running a suite of Javascript-based tests that represent typical use cases in today’s dynamic, interactive web applications.

unnamed file 845

 

Web Browsing — Mozilla Kraken

Mozilla Kraken is similar to Octane in that it measures the execution time of Javascript, but uses a different set of tests based on the SunSpider benchmark. Test cases include audio processing, search algorithms, image filtering, JSON parsing, and cryptography.

unnamed file 846

 

Web Browsing — WebXPRT

WebXPRT 3 is a browser benchmark that measures the performance of typical web applications, like photo enhancement, media management using AI, stock option pricing, encryption, OCR, charting, and productivity. This is in contrast to our other two browser benchmarks which focus more on microbenchmarks, testing specific algorithms.

unnamed file 847

Machine Learning — Tensorflow

Artificial Intelligence is everywhere these days. Machine-learning-based algorithms are taking the grunt work out of many manual tasks that could previously only be performed by humans. In order for Deep Learning AI to solve problems, it has to be trained first through a large set of training data that is evaluated repeatedly to generate a neural network that can later be put to work (also called inference). Google’s Python-based Tensorflow is one of the most popular machine-learning software packages that supports both CPUs and GPUs. Setting up Tensorflow for the GPU is a bit complicated, so lots of algorithm development and training on small data sets still happens on the CPU. Training performance on the CPU can also beat the GPU when problem sizes exceed typical GPU memory capacities.

unnamed file 848

 

Physics Simulation

Engineering widely uses the finite element method (FEM), which is able to simulate the flow of liquid (CFD), heat transfer, and structural stability to verify whether a final product is able to meet design requirements. Solving such a problem breaks the system up into a large number of simple parts called finite elements that are all able to interact with each other. This is a highly complex mathematical task that requires a lot of processing power, which is very difficult to parallelize on GPUs. Our Euler3D benchmark test is fully parallelized to make the most of multiple CPU cores, but it also puts a lot of stress on the memory subsystem.

unnamed file 849

 

Brain Neuron Simulation

In order to better understand how brains work, biology and medical research uses software to simulate neurons and their interactions with each other. Scientists hope that this can ultimately lead to an understanding of how biological intelligence emerges. Just like our physics simulation test, this is a highly complex, memory intensive problem that is best solved on CPUs—GPUs aren’t well suited to these algorithms.

unnamed file 850

Microsoft Office

Microsoft’s Office suite needs no introduction as it’s probably the most widely used PC software on the planet, installed on every office computer no matter the industry. Our tests cover a wide range of editing and creation tasks in Word, PowerPoint, and Excel.

unnamed file 851

 

unnamed file 852

 

unnamed file 853

 

Image Editing — Adobe Photoshop

Adobe Photoshop has become the industry standard for photo and image processing. We run the latest Photoshop CC through a battery of typical editing tasks, like image resize, various blurs, sharpening, color and light adjustments, and image export.

unnamed file 854

 

Video Editing — Adobe Premiere Pro

Adobe Premiere Pro CC is the workhorse of the video production industry to create high-quality content for film, TV, and the web. It can handle pretty much every recorded file format and supports workflows for editing Full HD, 4K, 8K, and virtual reality content. Unfortunately, most of Premiere Pro is single-threaded, and media encoding is highly GPU accelerated, so benchmarking “export” on the CPU makes little sense. For our testing, we’re using the software’s “object tracking” functionality, which automatically scans through a video to follow a specific person or object—this task does indeed use more than a core, but doesn’t fully scale. A lot of memory is consumed and accessed in the process (over 10 GB for our test scene).

unnamed file 855

 

Create 3D Model From Photos

Creating 3D models is a tedious and complex task that takes time and requires experienced artists. It’s thus the holy grail of 3D modeling to reconstruct a 3D model from just a series of photos. That’s exactly what Photogrammetry does. This method is also used to reconstruct terrain geometry from photos taken by aerial drones.

unnamed file 856

 

Text Recognition OCR — Google Tesseract

Optical Character Recognition, or OCR, is the task of turning text in scanned images or photos into actual characters for archival, further processing, or editing. While most OCR software is single-threaded, Google’s Tesseract engine can operate on multiple pages of a scanned document at once, spreading the load over several processor cores. The software, which is considered one of the most accurate open-source OCR packages available, automatically runs a spellcheck on the initial recognition results, which adds to the complexity of the workload.

unnamed file 857

Virtualization — VMWare Workstation

A virtual machine is a simulated computer inside your computer that’s completely independent of the host PC. This not only improves security, but also enables software written for different operating systems to execute on one physical machine. Virtualization is the foundation for “the cloud” and helps reduce hardware ownership cost by dynamically spreading out virtual machines over multiple computers to make best use of given hardware resources. We’re testing VM performance using VMWare Workstation, with hardware virtualization support enabled for both Intel and AMD processors. Curiously, many AMD Ryzen motherboards ship with the “SVM” setting disabled by default, so we made absolutely certain we had enabled it.

unnamed file 858

 

Database — MySQL

More data is stored and processed today than ever before in human history. The backbone for this revolution are database systems that manage storage and retrieval throughout large data sets. Whenever you interact with a website or other digital service, it’s almost guaranteed that at least one database is involved in returning the results you are looking for. We benchmark the most popular database system, MySQL, in the TPC-C test, which simulates a large number of warehouses and their constantly changing inventory. The number reported is “transactions per second”, so higher is better.

unnamed file 859

 

Java

The Java programming language is designed to be platform independent, highly scalable, and fault tolerant, which is why it’s very popular for enterprise services that work with large amounts of data and many concurrent users. Our test suite consists of a large mix of individual Java benchmarks, some of them single-threaded, some that scale somewhat, and some that fully scale to as many cores as are available.

unnamed file 860

Compression — WinRAR

Data is compressed almost all the time when it moves over the wire to reduce download time and transfer sizes. WinRAR uses a more advanced compression algorithm than classic ZIP, which is why we chose it for this test. It’s also able to scale across multiple processor cores.

unnamed file 861

 

Compression — 7-Zip

Another popular compression software is 7-Zip, which includes a benchmark that measures the integer instruction rate (MIPS) using the ZIP algorithm. It makes good use of multiple threads, when available.

unnamed file 862

 

unnamed file 863

 

Encryption — VeraCrypt

Encryption is the cornerstone for today’s security on the Internet, ensuring your data isn’t seen by everyone as it travels through various routers on the way to its final destination. VeraCrypt, which is based on the disk-encryption software TrueCrypt, enables open-source encryption for your disk drives without any backdoors. The application comes with a multi-threaded performance test that measures the rate at which data can be encrypted. We are using the AES algorithm for our testing.

unnamed file 864

Media Encoding — H.265 / HEVC

Nowadays, all video that’s consumed is compressed using various codecs, whether on TV, physical media, or streamed over the Internet. Our first video-encoding test uses the fairly new H.265 codec, which is also known as HEVC. We compress a full HD video using the latest version of the X265 encoder, with 8-bit color depth, preset “slow,” and a quality setting of crf 20.

unnamed file 865

 

Media Encoding — H.264 / AVC

H.264, also called AVC, is a slightly older compression format, though probably the most widely used compression format these days because it is well supported in even older hardware. We compress the same video as in the H.265 test using the X264 encoding software, with preset “slower” and crf 20.

unnamed file 866

 

Media Encoding — MP3

MP3 revolutionized the music industry like no other technology. Introduced in the 90s, it enabled massive reduction in audio-file sizes without noticeable impact on sound quality. This made music downloads, and ultimately streaming, a feasible method of content delivery over the Internet. For our benchmark, we convert a 2.5 hour-long 44.1 kHz Stereo recording to a variable bitrate MP3 file. MP3 encoding is a single-threaded process.

unnamed file 867

Game Tests: 720p

On popular demand from comments over the past several CPU reviews, we are including game tests at 720p (1280×720 pixels) resolution. All games from our CPU test suite are put through 720p using a RTX 2080 Ti graphics card and Ultra settings. This low resolution serves to highlight theoretical CPU performance because games are extremely CPU-limited at this resolution. Of course, nobody buys a PC with an RTX 2080 Ti to game at 720p, but the results are of academic value because a CPU that can’t do 144 frames per second at 720p will never reach that mark at higher resolutions. So, these numbers could interest high-refresh-rate gaming PC builders with fast 120 Hz and 144 Hz monitors. Our 720p tests hence serve as synthetic tests in that they are not real world (720p isn’t a real-world PC-gaming resolution anymore) even though the game tests themselves are not synthetic (they’re real games, not 3D benchmarks).

unnamed file 868

 

Individual Benchmark Scores

unnamed file 869

 

unnamed file 870

 

unnamed file 871

 

unnamed file 872

 

unnamed file 873

 

unnamed file 874

 

unnamed file 875

 

unnamed file 876

 

unnamed file 877

 

unnamed file 878

Game Tests 1080p

unnamed file 879

 

Individual Benchmark Scores

unnamed file 880

 

unnamed file 881

 

unnamed file 882

 

unnamed file 883

 

unnamed file 884

 

unnamed file 885

 

unnamed file 886

 

unnamed file 887

 

unnamed file 888

 

unnamed file 889

Game Tests 1440p

unnamed file 890

 

Individual Benchmark Scores

unnamed file 891

 

unnamed file 892

 

unnamed file 893

 

unnamed file 894

 

unnamed file 895

 

unnamed file 896

 

unnamed file 897

 

unnamed file 898

 

unnamed file 899

 

unnamed file 900

Game Tests 4K

unnamed file 901

 

Individual Benchmark Scores

unnamed file 902

 

unnamed file 903

 

unnamed file 904

 

unnamed file 905

 

unnamed file 906

 

unnamed file 907

 

unnamed file 908

 

unnamed file 909

 

unnamed file 910

 

unnamed file 911

Power Consumption

unnamed file 912

 

unnamed file 913

 

unnamed file 914

 

unnamed file 915

 

unnamed file 916

 

Energy Efficiency

In this section, we measure the total amount of energy consumed for a SuperPi run (single-threaded) and Cinebench (multi-threaded). Since a faster processor will complete a given workload quicker, the total amount of energy used might end up less than on a low-powered processor, which might draw less power, but will take longer to finish the test.

unnamed file 917

 

unnamed file 918

Temperatures

We used a Noctua NH-U14S to measure CPU temperature while running Blender. We picked an actual application as that better reflects real life than a stress-testing application like Prime95.

unnamed file 919

Clock Frequencies

The following chart shows how well the processor is able to sustain its clock frequency and what boost clock speeds are achieved at various thread counts. This test uses a custom-coded application that mimics real-life performance (not a stress test like Prime95). Modern processors change their clocking behavior depending on the type of load, which is why we provide three plots with classic floating point math, SSE SIMD code, and the modern AVX vector instructions. Each of the three test runs calculates the same result using the same algorithm, just with a different CPU instruction set.

unnamed file 920

 

Overclocking

unnamed file 339

 

unnamed file 340

As mentioned before, the Core i3-10100 has its multiplier locked because it lacks the “K” suffix, which means you can’t just set any desired CPU frequency.

Just like on previous Intel CPUs it is possible to increase the BCLK frequency above its default of 100 MHz on most motherboards. This results in a higher total CPU frequency, as BCLK x Multiplier = CPU Frequency.

With Comet Lake, and Z490 specifically, Intel marketing made some noise about the new ability to independently adjust BCLK from PCIe clock. In reality, this doesn’t do anything for non-K processors because all locked CPUs will measure the BCLK they are running at to simply refuse booting if BCLK is 103 MHz or higher.

That’s why our maximum overclock is with the BCLK at 102.9 MHz. The measurement isn’t 100% precise, so with “102.99” set in the BIOS, you’ll often end up with a hang at POST even though the frequency is technically below 103 MHz.

We also increased the power limits for our “Max Turbo” and “Max Turbo + 103 BCLK” runs. By default, the limits are PL1 = 65 W and PL2 = 90 W; we increased that to the maximum for these two tests, which yields no extra performance as the factory power limits are high enough for the processor.

作者 frank

发表评论