Compare two processors: AMD Ryzen 5 5500U vs Intel Core i5-1135G7 @ 2.40GHz [cpubenchmark.net] by PassMark Software

Compare Processors — SightLine Applications

  • The OEM-Video Processor Options document presents an overview of OEM options and comparative technical specifications.
  • The Integration Options document presents a range of OEM, SOM, and Software options for customers to incorporate SightLine functionality.

CRITERIA

Summary

  • Single channel processing (multiple inputs and switching)
  • Multiple video outputs – single-stream H.264 IP video, analog
  • Processing and streaming to 720p25
  • Tiny size for use in the smallest camera systems
  • OEM and SOM integration options
  • Multiple video inputs – dual channel processing
  • Multiple video outputs – dual-stream H.264 IP video, analog, HDMI, HDSDI
  • Processing and streaming to 1080p30
  • Small size (business card footprint)
  • SOM style, board-to-board Interface
  • Multiple video inputs – single or dual channel processing
  • Multiple video outputs – dual-stream H. 264/H.265 IP video, HDMI, HDSDI
  • Most powerful option – processing and streaming to 4Kp30
  • Smaller and lower power than 3000-OEM
  • OEM and SOM integration options

Processor

Texas Instruments DM3730

Texas Instruments DM8148 and Texas Instruments C6657

Qualcomm Snapdragon 820

Multi-Camera

Switching between inputs

Dual Processing with multi-camera display options: picture in picture, 2-up, blending, and switching

Dual Processing optional with multi-camera display options: picture in picture, blending, and switching

Digital Inputs

HDMI, Sony FCB-EH6xxx, Eh41xx, EV7xxx, Hitachi DI-SC120R, Tamron 10x, Airborne 720p GS, Camera Link, FLIR Tau, FLIR Quark, FLIR Boson, DRS Tamarisk

HDSDI, HDMI, Sony FCB-EH6xxx, Eh41xx, EV7xxx, Hitachi DI-SC120R, Tamron 10x, Airborne 720p GS, Camera Link, FLIR Tau, FLIR Quark, FLIR Boson, DRS Tamarisk

3000-OEM cameras + MIPI and USB-3

Analog Inputs (NTSC/PAL)

3 (using dual analog to digital input adapter boards)

2 (using two 3000-AB adaptors, one on OEM and one on MIPI adapter)

Frame Size and Rate Out

SD @ 30fps
720p @ 15-30 fps dependent on SW configuration

720p @60 fps single channel
1080p @30 fps + SD @ 30 fps
2 x 720p @ 30 fps

2x 1080p @30 fps with full SW
4K @30 fps with encoding only
4K @ 15-30 fps other SW functions

Serial Ports Available

3 (@3. 3V)

5 (@3.3V)

4 (@3.3V) ) + 4 with MIPI-Input adapter

Additional IO

I2C (1), GPIO (3+)

I2C (3), GPIO (4+)

I2C, GPIO (3) + 3 with MIPI-Input adapter

Ethernet Interface

10/100 BASE-T Ethernet PHY.  UDP, TCP, and RTSP connectivity, unicast, multicast.
1500-OEM and 3000-OEM with capacitive coupling

Same Ethernet interfaces as 1500 and 3000, but with magnetic coupling

Encoded Video Output

H.264/MPEG4/M-JPEG encoding,
MPEG2 TS/RTP encapsulation

H.264 encoding,
MPEG2 TS/RTP encapsulation

H.264 and H.265 encoding, MPEG2 TS/RTP encapsulation

KLV / Metadata

System metadata can be inserted into KLV IP stream, used in OSD, with JPEG EXIF headers, full pixel snapshots, and KML or NITF files.   KLV metadata is generated in accordance with MISB standards.

HDMI Output
HDSDI Output
Analog Output

No
No
Yes

Yes
Yes (with with HDSDI-output board)
Yes

Yes
Yes (with HDMI-HDSDI-output board)
No

Recording

Micro SD. Class 10 SDHC cards up to 400 GB

Interface for external Micro SD card
Class 10 SDHC cards up to 400 GB

Micro SD. Class 10 SDHC cards up to 400 GB

Voltage In / Power Consumption

4.5 – 6.5 VDC OEM (5 VDC nom)
Some adapter boards = 6.0 V max
3 W (max) 2.5W (typical)

8 – 15 VDC (12 VDC nom)
10 W (typical)

8 – 15 VDC (12 VDC nom)
5 W (startup current 3A per Smart Wireless Computing)

1.04 x 1.48 inches (26.5 x 37.7 mm)
0.27 ounces (7.6 grams)

3.47 x 1.97 inches (88 x 50 mm)
1.4 ounces (39 grams)

2.0 x 1.5 inches (50.5 x 38mm)
0.45 ounces (13 grams)

Environment – Temperature

Temp: Demonstrated with basic delivered heatsink: -40°C to + 55°C
Temp Components: -40°C to + 85°C

Screened: -20°C to + 55°C ambient with delivered passive heatsink
-40°C start-up with heater circuit

Environment – EMI

MIL-STD-461 and CE confirmed as part of customer assembly

SightLine support of Customer Tests is TBD

Environment – Shock Vibe

MIL-STD-810 qualification confirmed as part of customer assembly

SightLine support of Customer Tests is TBD

Fabrication Quality Assurance

Boards are assembled to IPC-A-610 Class2 specifications by a facility certified to ISO 9001 and AS 9100 standards and using ROHS Directive 2011/65/EU compliant materials and processes.

CPU Comparison: Which One is Right for You?

by Christopher McCarthy
May 5, 2015 +Add comment2015-05-05EDT04:00:30.000000-14400
2022-06-08EDT04:46:34.000000-14400

Throughout history, we’ve seen some great PC company rivalries unfurl. We’ve expressed the history behind one of the greatest corporate bouts of all time in my previous blog. Now it’s time to compare the products by both CPU companies, AMD and Intel. When configuring a custom PC, you have a decision that will affect the rest of your computer, from the inside out.

Deciding, configuring, and buying a custom desktop or laptop computer can be a daunting and complex process, on the surface. When selecting what components you desire, probably the most jargon filled component is the processor. When conducting a bit of research, and seeing what the industry has to say about this vital PC component, you’ll find that it’s easy to get lost in the blabber of clock speeds, cores, socket types, chipsets, “Piledriver”, “Steamroller”… etc. Although, once you do make a decision on what CPU to run your PC, a series of other component decisions will come together, saving you a significant amount of time.

Current CPU Landscape

Intel

If performance is a preliminary deciding factor in your next PC, you’ll probably see recommendations from publicists, forums, and social media posted all over internet geared toward making the jump on an Intel processor. If you look at some of our high-end performance based custom towers and gaming PCs, you’ll notice that our high performance workstations and high octane performance computer. The differences between the two can be negligible, and completely dependent on what you’re trying to accomplish with a PC. If your intended applications utilize multiple cores — then AMD can potentially offer equal or better clock speeds for a significantly lower price.

Intel’s Core M series or “Broadwell” CPU family isn’t forecasted to bring any huge changes in single core performance either. It will empower most of the laptops and desktops you’ll see in 2015. The largest gain from this chip upgrade from Intel is the 14 nanometer manufacturing. This cuts power requirements by roughly 30%, allowing tablets and mobile devices to be thinner, and preserve total battery life.

Intel’s recent chip developments could make a notable difference to all new mobile devices launched this year specifically in tablets — also in smartphones. But if you’re looking for a well-rounded desktop solution which delivers great performance, without having to face the price premium of Intel’s high-end chips. Take a look at the Core i5 products as you’ll probably get the most bang for your buck with deciding on an i5 chip.

AMD

As Intel has high hopes for the 2015 fiscal year with the 14nanometer processor, AMD (still at 28nm, with 2o nm cores working in tandem) is taking a different direction from mobile, attempting to restructure and redesign their CPUs to deliver something new to AMD fans in the Kaveri.

The most talked about change by AMD is support for *HSA* or Heterogeneous System Architecture. AMD is attempting to integrate GPU and CPU cores. In the past, the old division of memory – CPUs had one dedicated block, GPU’s had their own, and they were processed in different manners. Kaveri is said to be an update in a new schema from AMD– where both entities share the same random access memory, queuing and executions.

This means that each graphics core and CPU in an AMD Kaveri processor chip can now be running an individual process in its own memory “space”– completely independent from the other currently running processes. This new chip methodology has been deemed “compute cores” by AMD to help separate the terminology from the past. Quite literally, this means that 4 CPU cores and 8 GPU cores can now be valuated at 12 total “compute cores.”

CPU Comparison — Total Product Overview

At the lowest end of the CPU market, where price is critical to your decision, a dual-core system might be the perfect fit. Intel’s Celeron G1840 is a very simple two-core, two-thread CPU, but it’s also a Haswell architecture chip priced at a mere $54. Another considerably attractive budget choice is AMD’s A6 5400K: weaker as a processor, but includes a notably integrated HD 7540D graphics capability considering the $65 price tag.

As we switch gears into the gaming market and the AMD FX-6300 is a impressive product for the price. Six Piledriver cores augment functional multithreaded performance. This chip is majorly overclockable. The end result delivers close to the Core Intel i3 series chip speed for a much lower cost – it’s under $130 right now.

Alternatively, the mid-range AMD Kaveri chip A8-7600 already delivers excellent graphics performance along with minimal power requirements for a slight $122.

Superior single-core performance does mean that standard desktop configurations can, and typically will, be better off with an Intel processor. If you can manage without overclocking then the Core i5 4570 CPU delivers something similar to the power of a Core i7 for a much lower cost – near $235 – although beware: hyper-threading is disabled on these chips.

If that’s not enough power for you, then include the Core i5-4690K $260, while its counterpart Core i7-4790K is a quad cored, eight-thread chip features a standard clock speed of 4GHz (4.4GHz boost). This is great news if you can live with the $440 price tag. The price might be a bit steep here, but you won’t lag behind whatsoever until the next big-time launch from Intel.

Comparison Conclusion

While you’ll have to analyze and consider these CPU options, there’s an easier way to approach the purchasing process. While technical details can assist or leverage the decision: there are fundamental questions to consider: what are you using your computer for — “High performance applications or solely for graphics based gaming?”, “How fast do you desire?”  & “How much can you spend?”

is compared with the previous generation, as well as with Intel Core i9 / Sudo Null IT News I would like to say that the reason for this is the “environmental” situation in the country and the world, but in fact it is the desire to accumulate more practical and test material and, what to do, unlimited procrastination.

Now that the marketing department is writing to me more often than the creditors, and, apparently, will soon begin to come home to inquire about how I feel, I decided that there was nowhere else to put it off. Moreover, enough experience has been accumulated, and AMD has released processors based on the new Zen 3 architecture.

Initially, in the article I wanted to «push foreheads» of two manufacturers — AMD and Intel. But all deadlines were missed. Therefore, instead of explicitly comparing the «red» and «blue», we will test two generations of Ryzen — on Zen 2 and Zen 3. In fairness, we will not forget about Intel either. Moreover, historically we have always offered servers on their processors — both server and desktop. And articles with tests were published exclusively about them. Intel’s monopoly in the line of configurations presented by us ended about a year ago — in December 2019We started offering servers based on AMD Ryzen 7 3700X and AMD Ryzen 9 3900X, then AMD Ryzen 9 3950X in October 2020, and AMD Ryzen 9 5900X in December 2020.

In general, I decided that the processors of both manufacturers on the test bench may not be on time, but definitely not bad.

From the history of my acquaintance with AMD

Before we move on to tests and charts, I would like to digress a little from the main topic. If you are not a fan of nostalgia, feel free to skip this part. In addition to the author’s memories of bygone youth and my experience with computers, including AMD processors, there will be nothing here. nine0005

For me, acquaintance with AMD products began back in 2006, when, as a first-year student, I got a job as an apprentice in a service center and began to “repair computers”. Then in my “personal use” there was a home computer on an Intel Pentium 4 531 processor at 3 GHz, already with Hyper-Threading. On a proprietary Intel motherboard, with SATA1, an 80 GB SATA hard drive from Seagate and an ATI Radeon X300 graphics card. At that time, not the most top-end, but still terribly cool computer among classmates, which allowed you to calmly play on the grid in CS 1. 6. nine0005

While working in the service, for the first time I saw computers and laptops on AMD processors (Athlon, Duron and Turion) — in my subjective opinion, there were about 50/50 of them in the city with machines on Intel. In addition, x86-compatible processors VIA and Cyrix occasionally met. Around the same time, a legend appeared that AMD heats up and even burns out from overheating, which was, in general, not far from the truth. By the way, this fact allowed Intel to successfully pass under the consumer’s radar with all the technical and marketing blunders that were observed during the transition from Pentium 4 to Core 2 Duo. nine0005

Taking advantage of my official position, I quickly assembled a personal computer on an Intel Core 2 Duo E7200 processor on an LGA775 socket and an Asus motherboard — as I saw dozens of Gigabyte motherboards on a 478 socket with a burnout in the south bridge. Everything suited me, except that the processor did not have support for hardware virtualization and the VirtualBox that appeared then worked very slowly. I switched to an AMD Athlon II X2 and, thanks to the forward and backward compatibility of sockets, had a great opportunity to easily upgrade within a couple of years. Intel could not boast of this. nine0005

Later I switched to AMD Phenom II X6 and it was amazing! Gentoo Linux, which was my main OS at the time, took less than a day to compile! None of my colleagues on Intel Core 2 Quad could get such a result. Until a certain moment, I believed that this was the apotheosis of computer thought. So far, the vendor has not offered the new AMD FX-8100 based on the Bulldozer microarchitecture. Eight cores, not some pathetic six! And I updated… The speed of «rebuilding the world» fell, and I found out that the processor can overheat on «boxed» cooling. Later, I upgraded a couple more times — FX-8150, FX-8300 — and settled on the AMD FX-8350 on the Piledriver microarchitecture, and with an update to the motherboard (AM3 + socket). But it still wasn’t the same. Therefore, in my memory, the Phenom II X6 line has remained AMD’s best for many years. nine0005

As time went on, youthful acne faded, Gentoo Linux moved Fedora Linux, and the need for a daily compiler disappeared. I came to terms with the past, started a family, and, as a result, lost the opportunity to constantly upgrade my personal equipment. AMD watched the release of new lines without much enthusiasm, and then completely moved to Apple products, ended his career as a service provider and localhost admin and lost contact with desktops. “History has become a legend, a legend a farce. And then they made up jokes.” nine0005

And in 2018, AMD releases a new generation of processors based on the Zen architecture. I was already all perked up: something new after so many years of stagnation. Even then I worked here, where I work, and as I wrote above, I dealt only with Intel processors. True, hoping that one day we will offer configurations based on AMD.

And now my illusory hope suddenly becomes a reality. Not without my participation, of course.

Remembering the history of the FX-8100, it was decided to start with Ryzens of the third thousand, the tests of which, in my opinion, showed simply gorgeous results. And now a new line of processors based on the Zen 3 architecture has been released, which also did not escape the fate of being tested. nine0005

So, dear reader, discard subjectivism and skepticism. Because, despite all my genuine love for AMD processor products, which, after many years of technical downtime, begins to conquer the market almost from scratch, I am for an objective approach — the test results will tell everything for me.

A few words about the process and nanometers

Many, including us, have repeatedly complained that Intel has long been stuck on its 14 nm. The proof of this was frequent conversations like: “And AMD has long had 7 nm. And soon there will be even less … «. I decided to understand the issue a bit, so here is my brief remark about the process technology and the impact of the declared nanometers on processor performance. nine0003
The number of transistors per square millimeter still remains an important characteristic for comparing technical processes. That is, how many simple logical elements are in the processor chip. The fact is that the technical process does not speak about the size of the transistor itself, but about the size of one of its parts — namely the gate (as far as I can imagine). The rest of the constituent elements of the transistor are not strictly tied to the value of the process technology.

Earlier, in my opinion, before the advent of such technology as FinFET, which began to use three-dimensional rather than planar transistors, it still made sense to compare manufacturers’ technical processes, starting from the size of the transistor. But after the manufacturers went in different directions, developing their own technological solutions. And «nanometers» from a real indicator turned into an abstract one. At the same time, the habit of comparing technical processes has not gone away, although it all came down to comparing “soft with warm”. Now, focusing on the size of the declared technical process is not even so much pointless as not very significant. And it is worth comparing completely different parameters. nine0005

For example, in this article the author gives an interesting table comparing technical processes:

  • Min Metal Pitch: Minimum distance between two metal layers
  • Fin Height: the height of the fins from the Si substrate in the oxide layer
  • Fin Width: Fin Width
  • There is also a distance between the transistors on the substrate. And, it should be noted that it differs among manufacturers such as Samsung, TSMC, Intel and GF, with the same declared process technology. nine0005

    As a result, it turns out that the concept of a technical process at one fine moment became purely marketing and does not speak, as in the old days, about the technical advantage of processors, which have less technology. Thus, the TSMC 7nm FinFET process technology, on which the last two generations of AMD processors are manufactured, cannot be called with firm certainty the best relative to Intel’s 14 nm proprietary process technology. If you want to delve into the topic, then here is one interesting article and an informative video about it. And we are finally moving on to testing. nine0005

    Testing

    In the comments to the previous article, we were reasonably reproached for using different versions of benchmarks, which affects the test results. What can I say. It often happens that the “cartridges” have just been delivered, and they have already run out. In the sense that it is far from always possible to retest for specific processors on current versions of benchmarks. But this time I managed to get a full load of ammunition in my hands, so further you can see not only the results of testing Geekbench of the fourth version, but also of the fifth. nine0005

    And while we are talking about what influences test results, I remember one example from my life. Once a meteorologist was asked in an interview, “Tell me, what affects the weather forecast?”, He answered — it’s easier to say that it doesn’t. So the phase of the moon does not affect the test results, but this is not accurate.

    But seriously, there are more than enough factors that can affect the final test result. And it’s not just the operating system or Linux distribution. The result will also depend on the OS kernel version, system software versions, memory strips, even if they have the same characteristics, chipset, processor power phases and their cooling, BIOS version, benchmark version, especially the phoronix package, whose tests are updated more often than presentations Apple in 2020. Even the drive affects, for example, the passing of the phoronix Apache test. In general, a lot of conditions that are difficult to repeat over time. That’s why we do our best to create the most equal conditions for those processors, the results of which will be included in one article. nine0005

    I would especially like to note the following point. It applies to a lesser extent to Intel processors, which has been doing reflashes of the same architecture for many years, and to a greater extent to AMD processors: after the release of the processors, the first test results are worse than the results of the same tests a year later. I assume that this is due to the addition of support for the corresponding processors and optimizations in the kernel, software and benchmarks. That is why the results of the first tests of AMD processors were not included here: they were good, but the new ones are even better. nine0003
    This time we double-checked everything so that the reader can get the most up-to-date test results.

    So, “what fell into our tenacious paws today?”

    First, these are two AMD processors on the Zen 2 architecture — Ryzen 9 3900X and Ryzen 9 3950X . It would be more honest to compare them with the ninth thousand Intel processors — Core i9-9900K. Almost «the same age» after all. But we decided that another representative from Intel would take part in the comparison, or rather Core i9-10900K , a relatively high-end processor for the LGA1200 socket. On the one hand, we already have comparative tests of the i9-9900K and i9-10900K, on ​​the other hand, both of these processors are built on the same architecture. So everything is fair.

    We also decided to test the latest 5000 AMD — platforms based on them have replenished our line of servers quite recently, so traditional tests are indispensable. We will check three processors on the Zen 3 architecture: Ryzen 7 5800X, Ryzen 95900X, Ryzen 9 5950X .

    Note that we are not just comparing processors of the same manufacturer in the form of refreshes, as we did earlier. This time we have the opportunity, firstly, to compare processors from two different manufacturers, and secondly, processors from the same manufacturer on two different architectures.

    When comparing Intel and AMD, EVERYTHING is different: manufacturer, nuclear architecture, manufacturing process, processor cache, both in terms of volume and execution architecture, quantitative and qualitative solution of core execution, processor frequency, number of computing units. The only thing that these processors have in common is the x86/x86_64 architecture. And technically this is not entirely true. In the case of comparing AMD processors of different generations: these are two different nuclear architectures. Well, perhaps we will now combine them with our tests. nine0005

    It is worth saying a few words about the third and fifth thousand AMD processors. The company claims that with the Zen 3 line, it has managed to make an even greater leap in performance than with the release of previous generations of Ryzen. Thanks to this, new products, according to the manufacturer, should become the fastest solutions on the market, not only in computing tasks, but also in games. AMD says that they have seriously redesigned the crystal architecture, which made it possible to raise the overall performance to 19 without increasing the base frequencies on the same technical process.% relative to Zen 2. Well, we’ll see.

    So, six processors are involved in testing:

    • Intel Core i9-10900K,
    • AMD Ryzen 9 3900X,
    • AMD Ryzen 9 3950X,
    • AMD Ryzen 7 5800X,
    • AMD Ryzen 9 5900X,
    • AMD Ryzen 9 5950X.

    In testing, only single-unit servers (1U) were used. All processors are cooled by liquid cooling (hereinafter referred to as «water cooling», CBO, «water», «dropsy»). nine0005

    AMD processors are water-cooled in the classic version for 1U platforms. The Intel processor is cooled by advanced liquid cooling — a custom solution of its own production, the configuration of which we still do not cover. Otherwise i9-10900K overheat. Well, at least they don’t burn.

    All test instances are summed up under a single common denominator: 1U, “water”, the same RAM at a frequency of 2933, the same data drive. Memory with a frequency of 2933 is the maximum for the i9-10900K, while the presented «Ryazan» supports memory at 3200. And as you know, the performance of AMD processors, including in tests, strongly depends on the memory frequency. On our part, this is done in order, so to speak, to equalize the chances.

    AMD processors are tested on the same motherboard with the new chipset.

    And one more point that I would like to draw your attention to.

    Comparison of AMD 9 processors3900X, AMD 9 3950X and AMD 9 5900X, AMD 9 5950X is logical and understandable: the last successors of the previous ones. But the AMD 7 5800X is out of this range. The fact is that our tariff line also includes the AMD 7 3700X, which I would love to compare with the new AMD 7 5800X. And that would also be understandable. Alas, it was not possible to test it at the time of preparation of the article. But since there are 5800X results, why not share them.

    Performance characteristics of platforms

    Intel i9-10900k

    • Motherboard: ASRockRack W480D4U
    • RAM: 32 GB DDR4-2933 MT/s Kingston 2 pieces
    • NVMe SSD: 1TB Intel 665P

    AMD processors

    • Motherboard: ASRockRack X570D4U (bios beta)
    • RAM: 32 GB DDR4-2933 MT/s Kingston 2 pieces
    • NVMe SSD: 1TB Intel 665P

    Software part: OS CentOS Linux 8 x86_64 (8. 3.2011).
    Kernel: 4.18.0-240.1.1.el8_3.x86_64
    Implemented optimizations relative to the standard installation: added kernel launch options elevator=noop selinux=0

    Testing is performed with all patches against Specter, Meltdown and Foreshadow attacks backported to this kernel.

    List of tests performed:

    1) Sysbench
    2) Geekbench 4
    3) Geekbench 5
    4) Phoronix Test Suite

    Detailed description of tests

    Geekbench test

    A suite of tests conducted in single and multi-threaded mode. The result is a performance index for both modes. In this test, we will consider two main indicators:

    • Single-Core Score — single-threaded tests.
    • Multi-Core Score — multi-threaded tests.

    Units of measurement: abstract «parrots». The more «parrots», the better. nine0005

    Sysbench test

    Sysbench is a package of tests (or benchmarks) for evaluating the performance of various computer subsystems: processor, RAM, storage devices. The test is multi-threaded, for all cores. In this test, I measured one indicator: CPU speed events per second — the number of operations performed by the processor per second. The higher the value, the more efficient the system.

    Phoronix Test Suite

    The Phoronix Test Suite is a very rich test suite. Almost all the tests presented here are multi-threaded. The only exceptions are two of them: single-threaded Himeno and LAME MP3 Encoding tests. nine0005

    In these tests, the higher the score, the better.

    1. John the Ripper multi-threaded password guessing test. Let’s take Blowfish cryptalgorithm. Measures the number of operations per second.
    2. The Himeno test is a linear Poisson pressure solver using the Jacobi point method.
    3. 7-Zip Compression — 7-Zip test using p7zip with integrated benchmarking feature.
    4. OpenSSL is a set of tools that implement the SSL (Secure Sockets Layer) and TLS (Transport Layer Security) protocols. Measures RSA 409 performance6-bit OpenSSL.
    5. Apache Benchmark — Measures how many requests per second a given system can handle with 1,000,000 requests, with 100 requests running concurrently.

    And in these, if less, then better — in all tests, the time it takes to pass is measured.

    1. C-Ray tests CPU performance on floating point calculations. This test is multi-threaded (16 threads per core), will shoot 8 rays from each pixel for anti-aliasing, and generate a 1600×1200 image. The test execution time is measured. nine0046
    2. Audio encoding. The LAME MP3 Encoding test is performed in one stream. The test time is measured.
    3. Video encoding. The ffmpeg x264 test is multi-threaded. The test time is measured.

    Test results

    5950X is better than 3950X by 160%.
    5900X is better than 3900X by 166%.

    5950X is better than 3950X — 23. 3%.
    5900X is better than 3900X — 20.7%.

    5950X is better than 3950X — 7.3%.
    5900X is better than 3900X — 8.7%.

    5950X is better than 3950X -27.2%.
    5900X is better than 3900X — 25.8%.

    5950X is better than 3950X — 8.5%.
    5900X is better than 3900X -10.8%.

    5950X loses 3950X — 1.1%.
    5900X is better than 3900X — 0.2% (almost equal).

    5950X is better than 3950X — 1.4%.
    5900X is better than 3900X — 3.6%.

    5950X is better than 3950X — 8.1%.
    5900X is better than 3900X — 10.8%.

    5950X is better than 3950X — 3.0%.
    5900X is better than 3900X — 7.6%.

    5950X is better than 3950X — 16.1%.
    5900X is better than 3900X — 16.5%.

    5950X is better than 3950X — 17.3%.
    5900X is better than 3900X — 20.3%.

    5950X is better than 3950X — 2.3%.
    5900X is better than 3900X — 10.0%.

    5950X is better than 3950X — 21. 7%.
    5900X is better than 3900X — 19,eight%.

    In general, the results turned out to be predictable — the latest generation of 5,000th AMD is confidently outperforming its predecessors and leaving the relatively fresh Intel Core i9-10900K far behind. At the same time, it is worth noting that the Ryzen 9 3950X from the «third thousand» proved to be very worthy — according to the results of Geekbench tests, it ranks second after new products, and in the John the Ripper multi-threaded test, which measures the number of operations per second, even outperformed Ryzen 95950X.

    The Ryzen 7 5800X also proved to be quite interesting, becoming a leader not only in the results of Geekbench tests in both single-threaded and multi-threaded modes, but also in other tests — for encoding audio data (encode mp3) and for the number of requests per second ( Apache). I would recommend taking a closer look at servers with this processor. Especially, for processing media content or as a web server.

    Well, since we promised a comparison with Intel, I’ll say a few words about them. Judging by the test results, the i9-10900K has a chance to compete for the championship in single-threaded Geekbench tests (most likely thanks to a couple of additional GHz), but only with AMD «third thousand» — the performance of «fifth thousand» is an order of magnitude better. Moreover, even the “third thousand” does the i9-10900K in most tests.

    Since the «party secretariat» hinted to me that loud cheering is not our style, I’ll just calmly express my opinion. In my opinion, for two years now, Intel, if not catching up, is at least on a par with AMD in the desktop and gaming segment. As soon as Intel releases a new generation of processors, AMD immediately “beats this card”. Apparently, the superiority of the «blue» over the «red» is coming to an end. The Reds, in my opinion, are like Phoenix — they burned out when they released the FX series, and were reborn from the ashes with the release of Ryzen. nine0005

    As you can see, my genuine love for AMD is caused not only by romantic feelings, but also by banal cold calculation. If you follow the news, according to PassMark Software, in early 2021, AMD had 50.8% of the world’s desktop processor market. The share of Intel, respectively, fell to 49.2%. This means that the competition of giant manufacturers is reaching a different level, which will keep both companies in good shape. Therefore, I assume that 2021 will be no less dynamic than the outgoing 2020 in terms of breakthrough news in the processor market. Moreover, both companies have something to improve — Intel still has to figure out the 10 nm process technology, and AMD, at least, to solve supply problems so that it doesn’t turn out the way it did in December, when not everyone got what they ordered. nine0005

    In testing, we used servers based on AMD Ryzen and Intel Core processors from 1dedic.ru. Dedicated servers with these processors can be assembled in the configurator and ordered with a 7% discount for the selected payment period — 1, 3, 6 or 12 months using the promo code HABR1DEDIC21 . The discount does not apply to additional services connected to the server. The promo code is valid until February 28, 2021.

    How did MediaTek beat Qualcomm? Parsing

    If I ask you: who is the king of mobile processors? Probably, you will say, of course Qualcomm. And what about the fact that MediaTek has been overtaking Qualcomm in sales of mobile processors for two years in a row. A few years ago they had a reputation for heating and problematic chips, and now this brand is in first place. And the flagships of the future can completely switch to MediaTek. How is this possible? nine0005

    In this material, we will figure out where such sales growth comes from, what cool and unique technologies the company has, and why the flagship MTK is better than the top Snapdragon.

    Where did MediaTek come from?

    The company was founded in 1997 in Taiwan. It first made chips for CD and DVD drives in computers, and then entered the mobile market in 2004 by making chipsets for phones in China.

    The company became truly famous in 2013-14, when cheap smartphones flooded the market. Many of them are based on MediaTek chips based on the ARMv7 architecture. Basically, due to the low price and acceptable performance. It was MediaTek that then made it possible to get 8 real cores in a mid-budget smartphone. For example, the first Xiaomi Redmi Note worked on a similar processor. nine0005

    How did sales increase?

    How did they come to success? First, let’s see how the mobile processor market works.

    The biggest players on it are Apple, Qualcomm, Samsung, MediaTek, Huawei and UNISOC. Recently, Google broke into it with its Tensor. If you look at the percentage distribution of the share of each of them, it turns out that MediaTek is ahead of everyone.

    But what made the company take the first place?

    The key consumer of MediaTek processors is Xiaomi. And as we know, Xiaomi is one of the largest players in the mobile device market. It is the third on the market in general for all smartphones and the second among Android devices. Above it is only Samsung, who use both their Exynos and Snapdragon with MediaTek. nine0005

    Xiaomi puts MTK processors in most budget Redmi smartphones and even in some devices of the older series, which we used to know under the letter combination Mi. Recently, by the way, they got rid of it completely. You can often find Helio or Dimensity chips in them. What is the difference? More on this later.

    It turns out that MediaTek is most often put in state employees and middle peasants. This is shown by market analytics. But why exactly do vendors choose these chips? nine0005

    Cost

    The fact is that MediaTek offers efficient chipsets that are cheaper than competitors. For example, according to insiders, even the flagship MediaTek Dimensity 9000 costs less than Snapdragon 8 gen 1 from Qualcomm. But the chips are on the same level.

    But MediaTek dominates in the lower price segment, where devices cost from 100 to 300 dollars. All this thanks to inexpensive processors with support for the 5G standard, which is now actively spreading. For example, the Dimensity 700 and Dimensity 9 models00. The cheapest smartphone of 2020 with support for new generation networks, Realme V3, was built precisely on the basis of MediaTek Dimensity 720. Or not?

    Let’s take a look at what’s inside each SoC and find out what the differences are.

    This is what the fresh Snapdragon 8 gen 1 from Qualcomm looks like. It is made up of a bunch of blocks. In addition to the obvious CPU and GPU, there are several other important modules here — ISP, Sensing Hub, 5G Modem and FastConnect. About everything in order. nine0005

    Let’s start with the most interesting — ISP. Why is it useful?

    The Spectra ISP is a signal processor dedicated specifically to photo processing.

    What it does: controls autofocus, adjusts exposure and white balance. And all this in real time, before you take the picture. The ISP also does more important things: it turns the raw information from the sensor into a color image. After that, Spectra applies HDR, corrects the image geometry and removes noise. That is, it finishes the image so that it looks good. In general, a really useful thing. nine0005

    The latest ISP Spectra 680 has a throughput of 3.2 gigapixels/s. This indicator means the amount of data that it can process at a time. Pixels are used as data, as ISP works with images. That is, the Spectra 680 processes 3.2 billion pixels per second. Thanks to this, the chip can process photos up to 200 megapixels and shoot HDR video in 8K. The numbers are impressive.

    What does MediaTek have? Let’s take a look inside the Dimensity 9000.

    He also has an ISP — Imagiq 790. It does everything the same, but differs in capabilities. Its bandwidth is 9 Gp / s, that is, 3 times higher than that of Snapdragon. Therefore, such a chip can process a picture no longer at 200, but at 320 megapixels. But according to the company itself, HDR video can only be processed in 4K. For 8K, only normal 24-frame recording is available.

    In modern smartphones, ISP does not work alone, but in conjunction with machine learning. For AI to work in Snapdragon, both the CPU and the GPU are immediately involved. However, the development of neural networks over the past few years has forced the company to add the Sensing Hub, which we mentioned earlier, to the chips. This is a separate module dedicated specifically to AI. With its help, neurons analyze video, recognize objects, faces during shooting. And all this in order to get the perfect picture. nine0005

    MediaTek has an APU 590 for the same purpose. AI works on this module to process noise in pictures and videos, as well as lossless double zoom. By the way, the same chip is engaged in optimizing system processes in order to increase performance in games.

    Finally, the last two modules in Snapdragon are the 5G modem and FastConnect. Thanks to them, Qualcomm’s flagship SoC supports 5G, Wi-Fi 6E and Bluetooth 5.3.

    In MediaTek, these two modules are called simply: Connectivity and Modem. But their functionality is almost the same. The exception is the maximum download speed through a 5G modem. The latest Dimensity hits 7 Gb/s, while the latest Snapdragon can hit the full 10 Gb/s. Otherwise, there is still the same set of interfaces as Snapdragon, but at a lower price. In addition, modern MediaTek processors have a unique feature that is still not even in the flagship Snapdragon. nine0005

    We are talking about hardware decoding of the AV1 video codec. We had a separate material about him, we advise you to take a look, why is he revolutionary. In short, it is used by all popular streamers and compresses video very efficiently. A smartphone without a hardware decoder will be able to open video in such a codec, but its processor will be heavily loaded, and therefore heat up more and drain the battery.

    Qualcomm is rumored to add codec support only in the next flagship chip, and MediaTek has had this feature for 2 years now. nine0005

    Performance

    Technology is great, but just as important is chip performance. How is MediaTek doing with this?

    Let’s look at the results of the Dimensity 810 processor. This is one of the most popular chips in the entry-level segment with 5G. It competes with Snapdragon 695. This is the only Qualcomm chip with 5G on devices at that price.

    And here Dimensity is 15% behind. But if we compare MTK with the more popular Snapdragon 680 in the segment, then it is already ahead of Qualcomm. nine0005

    Another significant chip for the company is Dimensity 1200. Last year it helped MediaTek increase its share in the middle segment by almost 20%. Let’s take a look at the results of Geekbench 5.

    We will see that smartphones with it give the same performance as the Snapdragon 865+ and 870. If we look for smartphones with these chips, we will see that the prices for both of them are approximately the same — from 300-350 dollars.

    Not a strong gap in other segments. Even in the flagship, where MTK broke only recently. And there, the company’s chips turned out to be somewhat better than Qualcomm, but more on that later. In short, MediaTek really drags. But it’s one thing to give out good performance, and quite another to work stably. Previously, the middle peasants of the MTK were notorious for overheating, but what now? nine0005

    Let’s test the company’s latest mid-range chip. To do this, we have a TECNO Camon 19 Pro smartphone, which has an eight-core Helio G96. How we will test: first, we will run a throttling test in 100 threads and hold it for an hour.

    Lines

    We compared the performance of the mid-range and sub-flagship lines and also tested the gaming series. But the range of the company is not limited to this. What does MediaTek actually have?

    Throughout the history of the company, there have been many lines: MT65, Helio A, Helio P, Helio X, Helio G and Dimensity. nine0005

    Let’s start with the first MT65s. It was the processors of this series that brought the company both popularity and a bad reputation at the same time. They were cheap, with a large number of cores, so they were massively installed in inexpensive smartphones. In normal tasks, they could give acceptable smoothness, but smartphone manufacturers often scored on any optimization. As a result, such devices slowed down, the graphics in them were very weak, and some chip models also got very hot. Because of this, the company could not earn the trust of users for a long time. nine0005

    To get rid of the bad reputation and at the same time mark the transition to the ARMv8 architecture, MediaTek has moved to release Helio lines. Their naming should have become more understandable for buyers.

    The Helio A series was designed for the most budget devices. This line was created in 2018 and lived for only two years — new CPU models have not been released since 2020.

    For more expensive devices, there was the Helio P series, where the letter P stands for Performance. This line came out in 2015. These mid-rangers became a turning point in the history of MediaTek, since it was with the release of this line that the company’s bad reputation began to disappear. Like the Helio A line, P lived until 2020. Although you can now find smartphones with such processors, both lines have supplanted the other two, more successful ones. nine0005

    Helio X (eXtreme Performance), launched in 2014, should have been the flagship. The problem is that they were top only among the company’s models. In fact, they turned out to be of average performance and could compete with the Snapdragon 600 series. The Helio X processors were not bad for their price, but it was wrong to position them as the top.

    The real reason for MediaTek’s success is the G gaming line for the mid-range segment. What is her secret? The fact is that earlier MediaTek processors were far behind the graphics of competitors. Now, Helio G chips are competing on equal terms with the Snapdragon 700 series. nine0005

    TECNO Camon 19 Pro

    For example, the TECNO Camon 19 Pro smartphone runs on a processor from this line. Inside the device is a MediaTek Helio G96. This is an octa-core processor, made according to the 12 nm process technology. A combination of two productive Cortex A76 cores and 6 energy efficient A55 cores gives good responsiveness in everyday tasks. At the same time, the smartphone remains cold and does not spend all resources in an instant.

    The chip is complemented by 8 GB of RAM, which means that you can keep a lot of applications in the background. If this is not enough, you can always expand the RAM by another 5 GB using Memory Fusion technology. In order to accommodate as many programs as possible on the smartphone, Camon 19Pro has 128 GB of permanent memory.

    And since this is still a processor of the G line, that is, Gaming, then for games there is Mali G57 MC2 graphics. The same GPU can render the interface at 120 FPS.

    Unleash the full potential of graphics with a large IPS screen with a 120Hz refresh rate for smooth visuals. It has a Full HD + resolution with a diagonal of 6.8″. On top, you can stick a protective film from the kit. Small but nice bonus. Thanks to thin bezels and a small cutout for the camera, the display takes up almost the entire space of the front panel, which makes the smartphone look compact. nine0005

    Speaking of the cutout. There is a front camera with a 32 megapixel sensor. The core modules are even cooler. There are three of them.

    For portraits, there is a 50-megapixel sensor with 2x optical zoom. And for beautiful background blurring, there is a separate auxiliary sensor. The built-in ISP in Helio 96 can show this effect in real time.

    Finally, the main camera is a 64 MP sensor with f1.65 aperture. Its feature in the form of RGBW technology allows you to shoot bright and high-quality pictures even at night. Paired with it are noise reduction systems built into the SoC. And for smooth videos and shooting at night, optical stabilization is provided. nine0005

    This interesting set is powered by a 5000 mAh battery. To quickly fill such a volume, the smartphone provides fast charging at 33 watts. Thanks to this, 64% is gained from scratch in half an hour. All this stuff is packaged in a slim and stylish case, the design of which received the iF DESIGN AWARD.

    Let’s get back to MediaTek processors. Having turned off the supposedly flagship Helio X in 2017, MediaTek returned to this segment after 3 years. But already with successful Dimensity. And this is the second reason for the current success of the company. Dimensity chips really belong to the upper segment, and are not called that for marketing purposes. nine0005

    Now the range has expanded to medium solutions. But this only improved the position of the company, because the lower models with 5G modems captured the mid-budget market.

    But why are the flagship Dimensity so successful? Why Dimensity was better than the flagship Snapdragon?

    This is how many points the flagship Qualcomm Snapdragon 8 gen 1 gives out in the GeekBench 5 benchmark.

    MediaTek is on par in the single-threaded test and simply breaks in the multi-threaded test. Impressive? nine0005

    And this despite the fact that MediaTek is cheaper, but this is not the most important advantage.

    Remember how reviewers and smartphone owners complained en masse about the high temperature of the Snapdragon 888? So, the same problem also affected the more recent 8 gen 1, albeit to a lesser extent.

    Why did it happen? Indeed, in its flagships Snapdragon 888 and 8 gen 1, Qualcomm switched to a seemingly cooler 5 nm process technology. And as we know, the smaller it is, the more energy efficient the chip is. So right?

    At first, there were suggestions that the Adreno 660 GPU was to blame. However, Snapdragon 8 gen 1 already has a different video chip, but the overheating problem remained.

    It turned out that the seven-nanometer Snapdragon 865 (2019-2020) and 870 are manufactured at the TSMC factory, but the Snapdragon 888 and 8 gen 1 are already made by Samsung. And although the companies did not make official statements, the choice of Samsung as a manufacturer was most likely to blame. First, if we compare two identical chips manufactured at 5 nm by Samsung and 7 nm by TSMC, there will be almost no difference. But the process is different. nine0005

    Secondly, the next flagship, 8+ gen 1, Qualcomm decided to release already at TSMC and for a reason … According to the manufacturer, its power consumption will decrease by 30%. And most likely, this is the effect of switching to another manufacturer.

    Apparently, Samsung has encountered production problems on 5 and 4 nm technical processes.

    As for the Dimensity 9000, it has been manufactured at TSMC’s 4nm process from the very beginning. By the way, Apple Axx Bionic processors are also produced there. And there are no temperature problems with it. Dimension 9000 manages to be not only faster than its competitors, but also colder and more stable. Finally, Dimensity 9000 supports LPDDR5X memory, while Snapdragon 8 Gen 1 only supports LPDDR5.