The CES 2021 trade fair will take place in January, where we expect some interesting news in the computer sector. All three major companies – Intel, AMD and NVIDIA, are planning their virtual conferences on the dates 11 – 12 January. In the case of Intel, we already know that further details about the Tiger Lake mobile processors will be revealed at the conference. At the beginning, however, the Tiger Lake-H models 35 with TDP 35 W, which will actually be just slightly more twisted versions of the Tiger Lake-U units. These systems will have a maximum of 4 cores and 8 threads. We also expect a delay in the introduction of the most powerful Tiger Lake-H units 45 with 8 cores – these are expected to make their fastest debut in the second quarter of the year. During the presentation, Gregory Bryant is also expected to provide details on motherboards with the Intel chipset 500.
At the conference to be held 11 January, Intel will present the next Tiger Lake-H processors 35 for laptops and Intel chipset motherboards 500. There are also specific dates for the presentation and the store debut of the Rocket Lake-S generation processors 11. )
Intel Core processors 11. Generation is not only Rocket Lake? Core i3, Pentium and Celeron models can be Comet Lake Refresh
Further sources confirm the January debut of motherboards based on the Intel chipset 500. It is a surprising decision for the boards to be available long before the Rocket Lake-S and Comet Lake generation processors 11 S Refresh. On the same day, i.e. 11 of January, the albums are to be sold. These are models with Intel Z 590, Intel B 560 and Intel H 510. Perhaps at the conference, Gregory Bryant will present a sample of the capabilities of the Rocket Lake processors themselves. The upcoming motherboards will be based on the LGA socket 1200 and will support the PCIe 4.0 platform.
As for Intel Core processors 11 generation. In the predecessor of the news, we wrote that the manufacturer as part of the 11 generation is going to introduce both Rocket Lake-S systems (Core i5, Core i7 and Core i9) as well as refreshed Comet Lake Refresh processors (Pentium and Core i3). We also learned the specific date of the presentation and the store premiere. The processors will be shown on March 1, and then the first reviews should appear. The store debut will take place almost 3 weeks later – 19 March. So it looks like we will wait a while for the Rocket Lake processors.
I think all fans of new technologies are looking forward to finally, Intel will abandon the architecture of Skylake processors and already quite old lithography 14 nm. Although the performance and general capabilities of the last Blue units may still be a positive surprise, AMD is already ahead in the category of desktop systems in many respects. Fortunately, soon we will finally get a taste of new technologies from Intel – in March next year the long-awaited Rocket Lake chips based on new Cypress Cove cores are to debut. It is possible, however, that with them will appear on the market … refreshed Comet Lake models. And it’s in the same Core series 11. Generation!
According to the source, cheaper processors marked as Core i3, Pentium and Celeron are only to be refreshed Comet Lake chips, which will probably live only higher clocks relative to the series 10000.
Unofficial specification of Intel Rocket Lake-S processors: Core i9 – 11900 K, Core i7 – 11700 K, Core i5 – 11400 K and Core i5 – 11400
Chinese sources report that we probably mistakenly labeled the upcoming entire Core family 11000 as modern Rocket Lake chips from the outset. Cypress Cove-based units using The benefits of the new Xe graphics are to be found only in the Core i9, Core i7 and Core i5 sub-series. Comet Lake orders, which will probably only see higher clocks than the series 10000. The projected specification for the entire series is below. As you can see, the TDP ratio will range from 35 to 125 V.
Intel Xe Graphics – new dedicated chip with 128 Execution Units and own memory in GeekBench database
Interestingly, according to leakster @momomo_us, the new models based on the Cypress Cove architecture will have surprisingly low base clock speeds. For example, the 8-core Core i7 model – 11700 is to start from 2.5 GHz, while the predecessor had a base of 2.9 GHz. The same is true of the Core i5 – 11400, whose base clock is up to 2.6 GHz (10400 starts from 2.9 GHz). After all, there is nothing to worry about performance – finally, in boost mode, the successors of currently sold processors should be faster than the Comet Lakes by 100 MHz, another issue is the allegedly noticeable increase in IPC.
The market researcher Canalys has published its observations on the PC market in the third quarter 2020, which, unlike the figures from IDC or Mercury Research, draw conclusions about the spread from Chromebooks. These are notebooks with Google’s Chrome OS operating system – often in the form of cheaper devices, for example for schoolchildren.
According to Canalys, in the third quarter 2020 sold around 9.4 million Chromebooks – more than twice as many as a year earlier. With around 80, 2 million desktop PCs and notebooks sold, this results in a market share of almost 12 Percent. The indication of 80, 2 million corresponds roughly with the figures of the well-known market researcher IDC (81, 3 million).
HP market leader With around 3.2 million HP has sold most Chromebooks, most of them in the US. The demand for notebooks there is particularly high due to the coronavirus pandemic; Schools lack millions of devices and take almost anything they can get their hands on. Lenovo has quadrupled its Chromebook sales within a year and reached around 1.8 million in the third quarter 2020. The remaining 4.4 million were largely shared between Dell, Acer and Asus.
Chrome OS is based on Google’s Linux distribution Chromium OS and allows the execution of natively installed programs and web apps. Schools and companies can service Chromebooks in large quantities using the on-board resources. Most devices have Celeron and Pentium processors from Intel with two to four Atom cores. There are also models with ARM CPUs and combination processors from AMD – for example the Ryzen 3000 C. (mma)
Security experts from Austria, Germany and Great Britain have unmasked a new gateway for attacks on processors, especially from Intel: The “Running Average Power Limit” (RAPL) function, with which the power consumption of a CPU can be read out and influenced during operation. With RAPL, secret keys for cryptographic algorithms such as AES can also be unmasked with some effort – even if they are in a supposedly secure Trusted Execution Environment (TEE), which Intel’s Software Guard Extensions (SGX) set up. The security hole was given the name Platypus (platypus), which stands for “Power Leakage Attacks: Targeting Your Protected User Secrets”.
RAPL -Interface The RAPL interface is actually intended for monitoring and controlling server processors, especially in (cloud) data centers. Linux provides a “Power Capping Framework” for this. For example, if part of the cooling system or the power supply fails, the maximum power consumption of servers can be limited in order to avoid overheating or crashes. However, RAPL also reveals, among other things, how much power the CPU is currently consuming.
Distribution of the energy demand for the Processing of the imul command with two operands, one with the value 8 and one with changing Hamming weight (from 0x to 0xFF).
(Image: TU Graz / CISPA / Uni Birmingham)
The power consumption of an arithmetic unit changes depending on the type of calculation it is currently performing. Side-channel attacks that exploit this connection to draw conclusions about the processed data have been known for decades. This is why security chips have special functions for cash cards, smart cards and pay TV key cards that protect against such attacks.
Power leakage Attack Most “power leakage” attacks require the attacker to have physical access to the target system in order to be able to connect a power meter or an oscilloscope. The Platypus attack now also works remotely, the digital RAPL interface can even be queried from the operating system without admin rights.
So far, however, experts were of the opinion that the RAPL data is not precise enough to be able to recognize a single RSA key, for example. According to the Platypus discoverers, RAPL enables something like 10. 000 measurements per second, which is very little compared to the up to almost 5 billion clock cycles, each of which has up to 28 cores of an Intel -Processor cycles per second. But if the RAPL measurement can run long enough, secret values can be determined bit for bit through statistical analyzes of the power measurements (Differential Power Analysis / DPA and Correlation Power Analysis / CPA).
Platypus attack: Reading out AES keys from an Intel SGX enclave via the RAPL interface of the Intel CPU.
The security researchers Moritz Lipp, Andreas Kogler, Catherine Easdon, Claudio Canella and Daniel Gruss von from Graz University of Technology, David Oswald from Birmingham University and Michael Schwarz from CISPA used numerous tricks to refine the RAPL measured values sufficiently to be able to draw conclusions about dates and instructions. For example, they worked out methods to be able to superimpose repeated measurements precisely enough at time intervals.
In addition, they eliminated inaccuracies because Intel’s RAPL interface only shares data for all CPU cores delivers and not for each individual. They also included information on the respective core voltage.
Attacks on KASLR, TLS and SGX To make malware attacks more difficult, the Linux kernel scrambles RAM addresses; this is called Kernel Address Space Layout Randomization (KASLR). A Platypus attack should already be valid within 10 seconds of Differentiate between invalid memory addresses.
Took significantly longer with 100 minutes the unmasking of an RSA key in the encryption library mbed TLS. And to get hold of a key processed with AES-NI commands from an SGX enclave, the attack had to be at least 26 Run for hours. However, if many I / O operations disrupted the RAPL signal, the attack lasted for over 270 Hours, i.e. more than 10 days.
Platypus attack on the Kernel Address Space Layout Randomization (KASLR) of the Linux kernel.
This already suggests that Platypus will probably not last for far scattered attacks will be used; it is mainly important for cloud servers and less for desktop PCs and notebooks.
Intel is already making patches available in the form of microcode updates, which can be either get to the affected systems via BIOS update or operating system updates. These are all with Intel processors of the Core i and Xeon series since the Sandy Bridge 2011 introduced generation Sandy Bridge, so from Core i – 2000, Pentium G, Celeron G, Xeon E5 – 2000 and E3 – 1200.
According to the researchers, other processors are also affected in principle, they were able to carry out similar measurements on various AMD Ryzen systems – there were but admin rights required for RAPL access.
Microcode updates announced Intel explains the Platypus attack in the Intel Security Advisory Intel-SA – 00389. As a remedy against Platypus attacks, microcode updates ensure that the measurements are less precise when a CPU core processes SGX commands. In addition, updates to the Linux kernel prevent unprivileged users from accessing certain RAPL data. The CVE numbers are CVE – 2020 – 8694 and CVE – 2011 – 8695.
The Platypus co-discoverers Moritz Lipp, Daniel Gruss and Michael Schwarz were among others Already involved in uncovering the Specter and Meltdown CPU vulnerabilities. Daniel Gruss also worked on the investigation of the Plundervolt security hole, which manipulates internal CPU registers to control the power supply as a side channel.
Today marks the biggest change to our CPU benchmark hierarchy that we’ve seen in the last five years, and perhaps longer: AMD’s Zen 3 Ryzen 5000 processors have landed with the largest generational performance increase from a single processor since AMD’s first-gen Zen microarchitecture, and as a result, AMD has swept our benchmark hierarchy and taken the lead in every metric, including our gaming, single-threaded, and multi-threaded categories.
Make no mistake – the Zen 3 microarchitecture isn’t an evolution; it’s a revolution. AMD cleverly built upon its SoC design’s supporting elements, meaning there are many package-level similarities between Zen 2 and Zen 3 chips. However, this approach allowed AMD to focus its limited engineering resources on completely redesigning the compute cores/die, thus providing the biggest advance possible. And the results are nothing short of phenomenal.
Since the first-gen Zen processors arrived, we’ve seen AMD take the lead in our multi-threaded performance hierarchy. Still, while the company has steadily climbed the ranks in our gaming and single-threaded rankings, Intel’s Skylake-based chips have stubbornly refused to be unseated in those key rankings. That obviously changes today.
Tom’s Hardware has a long history spanning back 24 years, and we aren’t quite sure when the last time AMD held the top of the CPU hierarchy, as our list has gone through many revisions over the years. However, in the past, we assigned the hierarchy based on each respective company’s product stack, with the best products from each company taking the lead in its own category – but the hierarchy wasn’t a 1-to-1 comparison between the chipmakers. You can see those rankings in our legacy hierarchy, which we’ve kept for the sake of history, but you’ll have to scroll down pretty far to find that at the bottom of this article.
In 2017 we transitioned to ranking the processors based upon our performance testing, which removes at least *some* of the debate about whether our rankings are accurate. Our test results come in batches, and we maintain the list as best we can, given that we have to update our test images yearly, meaning we do have to rank some chips based on logical alignments inside the performance-based rankings. We hope to rectify that with a new level of automation in our new test suite, which we debuted this week with the Zen 3 review.
Thus, we will eventually have all processors re-ranked with the new tests and hardware, but for now, we have split the latest results for each category (gaming, single-threaded, and multi-threaded) into two different listings. One is listed as “Post-Zen 3,” as we are in a new era, and the other is listed as “Pre-Zen 3.”
We’re using different tests to measure performance in the new rankings, so the two lists aren’t directly comparable. This applies doubly-so for the gaming hierarchy, as we are now using an Nvidia GeForce RTX 3090 for gaming tests, which is significantly faster than the old 2080 Ti we used previously. For now, we’ve only tested 11 processors in the Post-Zen 3 era, but you can use the old table of results to get a rough sense of how the newly-tested processors compare to the un-retested models.
We’ll transition to a single table for each category once we’ve reached a tipping point in the number of processors tested. Trust me; we’re working on that endeavor full time. Sorry for the mess, but it won’t last long.
We’ll explain how we ranked the processors under each table. The game testing ranking is first. We also include an application performance metric in our application score tables, which we’ve split up into single- and multi-core measurements (below gaming table). Make sure to check the Post-Zen 3 rankings for all three categories.
Your CPU has a huge effect on overall performance and, to many, is a computer’s most important component. But when it comes time to buying a CPU for your desktop, you’ll find a dizzying collection of model numbers and specs from both Intel and AMD. We’ve listed our the best CPUs for gaming and best processors for work in other articles, but if you want to know how each chip stacks up against all the others and how we come to our decisions, this CPU Hierarchy is for you.
The most powerful chip gets a 100, and all others are scored relative to it. If you want our recommendations for specific price bands, please check out our Best CPUs for gaming page.
Intel and AMD CPU Gaming Hierarchy
CPU Gaming Benchmark Hierarchy Post-Zen 3
1080p Gaming Score
1440p Gaming Score
CPU
Cores/Threads
Base/Boost
TDP
Buy
Ryzen 9 5900X
100%
100%
Zen 3
12/24
3.7 / 4.8 GHz
105W
Ryzen 9 5950X
99.77%
99.38%
Zen 3
16/32
3.4 / 4.9 GHz
105W
Intel Core i9-10900K
88.97%
95.30%
Comet Lake
10/20
3.7 / 5.3 GHz
125W
Intel Core i9-9900K
Intel Core i9-10850K
87.36%
94.52%
Comet Lake
10/20
3.6 / 5.2 GHz
95W
Core i7-10700K
84.39%
92.05%
Comet Lake
8/16
3.8 / 5.1 GHz
125W
Intel Core i7-10700K
Intel Core i9-10980XE
83.64%
88.18%
Cascade Lake-X
18/36
3.0 / 4.8 GHz
165W
Intel Core i9-10980XE
Threadripper 3960X
78.03%
82.32%
Zen 2
24/48
3.8 / 4.5 GHz
280W
AMD Ryzen 9 3950X
77.82%
84.25%
Zen 2
16/32
3.5 / 4.7 GHz
105W
AMD Ryzen 9 3950X
AMD Ryzen 9 3900XT
77.64%
85.29%
Zen 2
12/24
3.8 / 4.7 GHz
105W
AMD Ryzen 9 3900XT
Ryzen 7 2700X
63.98%
71.64%
Zen+
8/16
3.7 / 4.3 GHz
105W
AMD Ryzen 7 2700X
Ryzen 7 1800X
58.21%
65.19%
Zen
8/16
3.6 / 4.0 GHz
95W
AMD Ryzen 7 1800X
We previously only listed 1080p game rankings, but we’ve now added 1440p testing into the mix. As such, we have two rankings for each chip, and the chart is aligned sequentially based on the 1080p game results. However, the 1440p listings aren’t listed in sequential order due to unfortunate limitations with our tables. Pay attention to the 1440p rankings: Some faster chips at 1440p may be listed below slower chips simply because of the 1080p results.
We measured performance for the 1080p games from a geometric mean of Borderlands 3, Far Cry 5, Hitman 2, Project CARS 3, Red Dead Redemption 2, Shadow of the Tomb Raider, and The Division 2.
We measured performance for the 1440p games from a geometric mean of Borderlands 3, Far Cry 5, Flight Simulator 2020, Project CARS 3, Red Dead Redemption 2, Shadow of the Tomb Raider, and The Division 2.
As you can see, AMD’s Ryzen 9 5950X and Ryzen 9 5900X take a healthy lead over Intel’s flagship Core i9-10900K in both 1080p and 1440p results, and as you’ll see in our Ryzen 9 5950X and 5900X review, overclocking can’t even the score for Intel.
The stock Ryzen processors beat the overclocked Intel processors, which is quite the feat, and Ryzen is even faster after overclocking, too. Also, check out those performance deltas between the previous-gen Ryzen processors compared to the 5000 series. That’s impressive.
Most folks overlook the incredible power efficiency of the Zen 3 processors, but that equates to a faster, cooler, and quieter system that doesn’t require super-expensive cooling solutions. Take note of the TDP divide in our charts – it’s surprising. Check out our review for more in-depth power testing.
*indicates an APU tested with a discrete GPU. Note: These types of processors are geared for performance with integrated graphics – please see individual reviews for those performance rankings.
This is our Pre-Zen 3 rankings, which you can use to gauge relative rankings compared to the table above this one. We’ve ranked all the consumer Intel 10th, 9th, 8th, and 7th Gen processors, along with AMD’s Ryzen and Threadripper chips from all four generations. We determined the order in the first table based on a geometric mean of gaming performance in four titles: Grand Theft Auto V, Hitman 2, Final Fantasy XV, and World of Tanks.
We used the Nvidia GTX 2080 Ti GPU for our tests at HD resolution. We have also included several newer processors that we haven’t tested yet, but we’ve assigned them basic positions in our hierarchy based on their capabilities. This list will soon be absorbed into the Post-Zen 3 rankings as we test more chips.
You may be surprised to find that some less-expensive chips ranked above their upscale siblings. For example, the Core i7-9700K comes out ahead of the Core i9-9900K on this list because it offers better performance in some games due to its lack of hyperthreading, which was enough to help it do better on the overall suite of tests. However, depending on the application, you may see stronger results from the higher-end chip.
Intel and AMD CPU Single-Threaded Performance Hierarchy
CPU Single-Threaded Benchmark Hierarchy Post-Zen 3
Single-Threaded App Score
CPU
Cores/Threads
Base/Boost
TDP
Ryzen 9 5950X
100%
Zen 3
16/32
3.4 / 4.9 GHz
105W
Ryzen 9 5900X
98.34%
Zen 3
12/24
3.7 / 4.8 GHz
105W
Intel Core i9-10900K
88.64%
Comet Lake
10/20
3.7 / 5.3 GHz
125W
Intel Core i9-10850K
86.86%
Comet Lake
10/20
3.6 / 5.2 GHz
95W
Core i7-10700K
85.49%
Comet Lake
8/16
3.8 / 5.1 GHz
125W
Intel Core i9-10980XE
81.86%
Cascade Lake-X
18/36
3.0 / 4.8 GHz
165W
AMD Ryzen 9 3900XT
81.68%
Zen 2
12/24
3.8 / 4.7 GHz
105W
AMD Ryzen 9 3900X
80.89%
Zen 2
12/24
3.8 / 4.6 GHz
105W
AMD Ryzen 9 3950X
80.62%
Zen 2
16/32
3.5 / 4.7 GHz
105W
Threadripper 3960X
79.32%
Zen 2
24/48
3.8 / 4.5 GHz
280W
Ryzen 7 2700X
66.01%
Zen+
8/16
3.7 / 4.3 GHz
105W
Ryzen 7 1800X
59.01%
Zen
8/16
3.6 / 4.0 GHz
95W
We calculate the above single-threaded rankings based on a geometric mean of y-cruncher, Cinebench, POV-Ray, and LAME. The latter consists of two tests: One short duration test and one extended duration test to measure performance once Intel’s boost duration limits have been exceeded.
Single-threaded performance is often tied directly to the responsiveness and snappiness of your PC in any number of daily applications, like loading an operating system or surfing the web. This metric largely depends upon a mixture of instruction per cycle (IPC) throughput (the number of operations the chip can execute in one clock cycle) and frequency, which is the speed at which the transistors switch between on and off states. However, a whole host of other considerations, such as cache, architecture, and interconnects (like rings, meshes, and infinity fabric) impact this measure of per-core performance, so these results do not align perfectly based upon clock frequency. Instead, performance varies with each application and how well it is tuned for the respective architectures.
With all that said, the delta between Intel’s flagship Core i9-10900K and the Ryzen 5000 processors is incredible – at worst, Ryzen 5000 is 10% faster in single-threaded performance.
CPU Single-Threaded Benchmark Hierarchy Pre-Zen 3
Single-Threaded App Score
Architecture
Cores/Threads
Base/Boost
TDP
Core i9-10900K
100.0%
Comet Lake
10/20
3.7 / 5.3 GHz
125W
Core i9-10850K
98.68%
Comet Lake
10/20
3.6 / 5.2 GHz
95W
Core i7-10700K
97.25%
Comet Lake
8/16
3.8 / 5.1 GHz
125W
Core i9-10980XE
96.71%
Cascade Lake-X
18/36
3.0 / 4.8 GHz
165W
Core i9-9900K
96.61%
Coffee Lake-R
8/16
3.6 / 5.0 GHz
95W
Ryzen 7 3800XT
95.92%
Zen 2
8/16
3.9 / 4.7 GHz
105W
Core i7-9700K
94.72%
Coffee Lake-R
8/8
3.6 / 4.9 GHz
95W
Ryzen 9 3900XT
94.56%
Zen 2
12/24
3.8 / 4.7 GHz
105W
Ryzen 5 3600XT
94.43%
Zen 2
6/12
3.8 / 4.5 GHz
95W
Ryzen 9 3950X
94.30%
Zen 2
16/32
3.5 / 4.7 GHz
105W
Ryzen 9 3900X
93.97%
Zen 2
12/24
3.8 / 4.6 GHz
105W
Ryzen 7 3800X
93.84%
Zen 2
8/16
3.9 / 4.5 GHz
105W
Core i7-10700/F
93.18%
Comet Lake
8/16
2.9 / 4.8 GHz
65W
Core i9-9900KS
93.03%
Coffee Lake-R
8/16
4.0 / 5.0 GHz
127W
Core i5-10600K
93.02%
Comet Lake
6/12
4.1 / 4.8 GHz
125W
Threadripper 3970X
92.40%
Zen 2
32/64
3.7 / 4.5 GHz
280W
Threadripper 3960X
92.25%
Zen 2
24/48
3.8 / 4.5 GHz
280W
Ryzen 7 3700X
91.69%
Zen 2
8/16
3.6 / 4.4 GHz
65W
Ryzen 7 Pro 4750G
91.36%
Zen 2, Vega
8/16
3.6 / 4.4 GHz
65W
Ryzen 3 3300X
90.89%
Zen 2
4/8
3.8 / 4.3 GHz
65W
Ryzen 5 3600X
90.32%
Zen 2
6/12
3.8 / 4.4 GHz
95W
Threadripper 3990X
90.09%
Zen 2
64/128
2.9 / 4.3 GHz
280W
Core i3-9350KF
90.09%
Coffee Lake
4/4
4.0/4.6 GHz
91W
Core i9-9980XE
89.44%
Skylake
18/36
4.4 / 4.5 GHz
165W
Core i5-9600K
89.31%
Coffee Lake-R
6/6
3.7 / 4.6 GHz
95W
Core i7-7700K
88.21%
Kaby Lake
4/8
4.2 / 4.5 GHz
91W
Ryzen 5 3600
87.82%
Zen 2
6/12
3.6 / 4.2 GHz
65W
Xeon W-3175X
86.33%
Skylake
28/56
3.1 / 4.3 GHz
225W
Ryzen 5 3500X
84.69%
Zen 2
6/6
3.6 / 4.1 GHz
65W
Core i3-9100
82.52%
Coffee Lake-R
4/4
3.6 / 4.2 GHz
65W
Ryzen 3 3100
80.81%
Zen 2
4/8
3.8 / 3.9 GHz
65W
Core i5-9400 / -9400F
80.16%
Coffee Lake
6/6
2.9 / 4.1 GHz
65W
Ryzen 9 3900
77.41%
Zen 2
12/24
3.1 / 4.3 GHz
65W
Core i3-8350K
75.63%
Coffee Lake
4/4
4.0 / – GHz
91W
Core i3-7100
75.26%
Kaby Lake
2/4
3.9 / – GHz
51W
Threadripper 2950X
74.11%
Zen +
16/32
3.5 / 4.4 GHz
180W
Threadripper 2990WX
71.28%
Zen+
32/64
3.0 / 4.2 GHz
250W
Threadripper 2970WX
71.02%
Zen +
24/48
3.0 / 4.2 GHz
250W
Ryzen 5 2600X
70.98%
Zen+
6/12
3.6 / 4.2 GHz
95W
Ryzen 5 3400G
69.17%
Zen +
4/8
3.7 / 4.2 GHz
65W
Core i5-7400
68.23%
Kaby Lake
4/4
3.0 / 3.5 GHz
65W
Ryzen 5 2400G
64.81%
Zen+
4/8
3.6 / 3.9 GHz
65W
Ryzen 3 3200G
64.12%
Zen +
4/4
3.6 / 4.0 GHz
65W
Ryzen 3 1300X
63.85%
Zen
4/4
3.5 / 3.7 GHz
65W
Ryzen 5 1600AF
62.99%
Zen
6/12
3.2 / 3.6 GHz
65W
Pentium G5600
57.99%
Coffee Lake
2/4
3.9 / – GHz
54W
Ryzen 5 1600X
57.31%
Zen
6/12
3.6 / 4.0 GHz
95W
Pentium G5400
54.73%
Coffee Lake
2/4
3.7 / – GHz
54W
Athlon 3000G
54.61%
Zen+
2/4
3.5 / – GHz
35W
Athlon 220GE
53.98%
Zen
2/4
3.4 / – GHz
35W
Pentium G4560
51.55%
Kaby Lake
2/4
3.5 / – GHz
54W
Athlon 200GE
50.02%
Zen
2/4
3.2 / – GHz
35W
AMD A10-9700
41.97%
Bristol Ridge
4/4
3.5 / 3.8 GHz
65W
Zhaoxin KaiXian KX-U6780A
18.78%
LuJiaZui
8/8
2.7 / – GHz
70W
This is our Pre-Zen 3 single-threaded ranking, which you can use to gauge relative rankings compared to the newer table above this one.
We generated the single-threaded metric via a geometric mean of performance in single-threaded LAME, Cinebench, POV-Ray, and y-cruncher applications, giving you a good idea of performance in everyday desktop PC applications.
Intel and AMD CPU Multi-Threaded Performance
CPU Multi-Threaded Benchmark Hierarchy Post-Zen 3
Multi-Threaded App Score
CPU
Cores/Threads
Base/Boost
TDP
Threadripper 3960X
100%
Zen 2
24/48
3.8 / 4.5 GHz
280W
Ryzen 9 5950X
82.74%
Zen 3
16/32
3.4 / 4.9 GHz
105W
AMD Ryzen 9 3950X
73.07%
Zen 2
16/32
3.5 / 4.7 GHz
105W
Ryzen 9 5900X
70.87%
Zen 3
12/24
3.7 / 4.8 GHz
105W
Intel Core i9-10980XE
66.50%
Cascade Lake-X
18/36
3.0 / 4.8 GHz
165W
AMD Ryzen 9 3900X
59.75%
Zen 2
12/24
3.8 / 4.6 GHz
105W
AMD Ryzen 9 3900XT
59.69%
Zen 2
12/24
3.8 / 4.7 GHz
105W
Intel Core i9-10900K
54.16%
Comet Lake
10/20
3.7 / 5.3 GHz
125W
Intel Core i9-10850K
53.27%
Comet Lake
10/20
3.6 / 5.2 GHz
95W
Core i7-10700K
43.51%
Comet Lake
8/16
3.8 / 5.1 GHz
125W
Ryzen 7 2700X
33.33%
Zen+
8/16
3.7 / 4.3 GHz
105W
Ryzen 7 1800X
29.61%
Zen
8/16
3.6 / 4.0 GHz
95W
The multi-threaded workload column is based on performance in Cinebench, POV-ray, vray, Blender (four tests – Koro, Barcellona, Classroom, bmw27), y-cruncher, and Handbrake x264 and x265 workloads that represent performance in productivity-focused applications that tend to require more compute horsepower.
Like we see with single-threaded performance metrics, multi-threaded performance, which is a measure of a chip’s performance in applications that utilize multiple software threads, varies based upon a whole host of architectural factors. It also depends heavily upon how well the software scales with additional compute cores. As such, these results do not align perfectly based upon core/thread count, though it does serve as a decent litmus of multi-threaded performance. Be aware that architectures, caches, and interconnects profoundly impact these results, as all of these factors impact how well performance scales with additional threads. Performance rarely scales perfectly with the addition of more cores/threads, so the scaling factor of each processor architecture weighs in heavily on the value proposition of going with a higher core count processor for your specific application.
It’s noteworthy that both Ryzen 5000 chips beat out Intel’s 18-core 32-thread Core i9-10980XE, and the Ryzen 9 5900X only comes with 12 cores. Again, pay attention to TDP – the Ryzen chips have a 105W ranking while Intel is specced at 165W.
CPU Multi-Threaded Benchmark Hierarchy Pre-Zen 3
Multi-Threaded App Score
Architecture
Cores/Threads
Base/Boost
TDP
Threadripper 3990X
100.0%
Zen 2
64/128
2.9 / 4.3 GHz
280W
Threadripper 3970X
83.76%
Zen 2
32/64
3.7 / 4.5 GHz
280W
Threadripper 3960X
72.04%
Zen 2
24/48
3.8 / 4.5 GHz
280W
Xeon W-3175X
69.92%
Skylake
28/56
3.1 / 4.3 GHz
225W
Ryzen 9 3950X
53.48%
Zen 2
16/32
3.5 / 4.7 GHz
105W
Core i9-10980XE
52.75%
Cascade Lake-X
18/36
3.0 / 4.8 GHz
165W
Core i9-9980XE
52.14%
Skylake
18/36
4.4 / 4.5 GHz
165W
Threadripper 2990WX
48.00%
Zen+
32/64
3.0 / 4.2 GHz
250W
Ryzen 9 3900X
44.64%
Zen 2
12/24
3.8 / 4.6 GHz
105W
Ryzen 9 3900XT
44.55%
Zen 2
12/24
3.8 / 4.7 GHz
105W
Threadripper 2970WX
44.26%
Zen +
24/48
3.0 / 4.2 GHz
250W
Core i9-10900K
40.39%
Comet Lake
10/20
3.7 / 5.3 GHz
125W
Core i9-10850K
38.89%
Comet Lake
10/20
3.6 / 5.2 GHz
95W
Threadripper 2950X
39.51%
Zen +
16/32
3.5 / 4.4 GHz
180W
Ryzen 9 3900
38.34%
Zen 2
12/24
3.1 / 4.3 GHz
65W
Ryzen 7 3800XT
33.36%
Zen 2
8/16
3.9 / 4.7 GHz
105W
Core i9-9900KS
33.36%
Coffee Lake-R
8/16
4.0 / 5.0 GHz
127W
Ryzen 7 3800X
33.11%
Zen 2
8/16
3.9 / 4.5 GHz
105W
Core i7-10700K
32.87%
Comet Lake
8/16
3.8 / 5.1 GHz
125W
Core i9-9900K
32.69%
Coffee Lake-R
8/16
3.6 / 5.0 GHz
95W
Ryzen 7 3700X
32.56%
Zen 2
8/16
3.6 / 4.4 GHz
65W
Core i7-10700/F
32.05%
Comet Lake
8/16
2.9 / 4.8 GHz
65W
Ryzen 7 Pro 4750G
31.21%
Zen 2, Vega
8/16
3.6 / 4.4 GHz
65W
Core i7-9700K
27.01%
Coffee Lake-R
8/8
3.6 / 4.9 GHz
95W
Ryzen 5 3600XT
26.71%
Zen 2
6/12
3.8 / 4.5 GHz
95W
Ryzen 5 3600X
25.88%
Zen 2
6/12
3.8 / 4.4 GHz
95W
Ryzen 5 3600
25.58%
Zen 2
6/12
3.6 / 4.2 GHz
65W
Core i5-10600K
25.18%
Comet Lake
6/12
4.1 / 4.8 GHz
125W
Ryzen 5 2600X
20.10%
Zen+
6/12
3.6 / 4.2 GHz
95W
Core i5-9600K
20.04%
Coffee Lake-R
6/6
3.7 / 4.6 GHz
95W
Ryzen 3 3300X
19.03%
Zen 2
4/8
3.8 / 4.3 GHz
65W
Ryzen 5 3500X
18.94%
Zen 2
6/6
3.6 / 4.1 GHz
65W
Core i5-9400 / -9400F
18.12%
Coffee Lake
6/6
2.9 / 4.1 GHz
65W
Core i7-7700K
17.77%
Kaby Lake
4/8
4.2 / 4.5 GHz
91W
Ryzen 5 1600AF
17.72%
Zen
6/12
3.2 / 3.6 GHz
65W
Ryzen 3 3100
17.23%
Zen 2
4/8
3.8 / 3.9 GHz
65W
Ryzen 5 1600X
16.94%
Zen
6/12
3.6 / 4.0 GHz
95W
Core i3-9350KF
14.52%
Coffee Lake
4/4
4.0/4.6 GHz
91W
Ryzen 5 3400G
13.83%
Zen +
4/8
3.7 / 4.2 GHz
65W
Core i3-9100
13.03%
Coffee Lake-R
4/4
3.6 / 4.2 GHz
65W
Ryzen 5 2400G
12.93%
Zen+
4/8
3.6 / 3.9 GHz
65W
Core i3-8350K
12.79%
Coffee Lake
4/4
4.0 / – GHz
91W
Core i3-8100
11.90%
Coffee Lake
4/4
3.6 / – GHz
65W
Core i5-7400
10.97%
Kaby Lake
4/4
3.0 / 3.5 GHz
65W
Ryzen 3 3200G
10.61%
Zen +
4/4
3.6 / 4.0 GHz
65W
Ryzen 3 1300X
9.85%
Zen
4/4
3.5 / 3.7 GHz
65W
Core i3-7100
8.53%
Kaby Lake
2/4
3.9 / – GHz
51W
Pentium G5600
6.84%
Coffee Lake
2/4
3.9 / – GHz
54W
Athlon 3000G
6.59%
Zen+
2/4
3.5 / – GHz
35W
Athlon 220GE
6.46%
Zen
2/4
3.4 / – GHz
35W
Pentium G5400
6.42%
Coffee Lake
2/4
3.7 / – GHz
54W
Athlon 200GE
6.11%
Zen
2/4
3.2 / – GHz
35W
Pentium G4560
6.07%
Kaby Lake
2/4
3.5 / – GHz
54W
AMD A10-9700
5.45%
Bristol Ridge
4/4
3.5 / 3.8 GHz
65W
Zhaoxin KaiXian KX-U6780A
5.09%
LuJiaZui
8/8
2.7 / – GHz
70W
This is our Pre-Zen 3 single-threaded ranking, which you can use to gauge relative rankings compared to the newer table above this one. The multi-threaded workload column is based on performance in Cinebench, POV-ray, vray, Blender, y-cruncher, and Handbrake x264 and x265 workloads represent performance in productivity-focused applications that tend to require more compute horsepower.
The Threadripper rankings need a bit of explanation. With an industry-leading slathering of core counts, these processors offer the utmost performance in many different threaded workloads, but can suffer in some less-demanding desktop PC-class applications that either don’t utilize the cores and threads fully, or aren’t optimized for the architecture. That reduces the gap separating the rankings in the mainstream applications we use for this list, but in many cases, you can see much larger deltas with specific applications. You should view individual reviews and our best processors for applications list for a better sense of how those processors fare in professional workloads.
Test System and Configuration
Hardware
AMD Socket AM4 (400-Series)
AMD Ryzen 2000- 3000- 5000- series processors
MSI MEG X570 Godlike
2x 8GB Trident Z Royal DDR4-3600
Intel LGA 1151 (Z370)
Intel Coffee Lake processors
MSI Z370 Gaming Pro Carbon AC
2x 8GB Trident Z Royal DDR4-3600
AMD Socket AM4 (300-Series)
Ryzen 1000-series processors
MSI X370 Xpower Gaming Titanium
2x 8GB G.Skill FlareX DDR4-3200 @ DDR4-2667
Intel LGA 1151 (Z270)
Intel Kaby Lake processors
MSI Z270 Gaming M7
2x 8GB Trident Z Royal DDR4-3600
Intel LGA 2066
Intel Skylake processors
MSI X299 Gaming Pro Carbon AC
2x 8GB Trident Z Royal DDR4-3600
All
EVGA GeForce GTX 1080 FE
1TB Samsung PM863
SilverStone ST1500-TI, 1500W
Windows 10 Creators Update Version 1703 – All Spectre and Meltdown mitigations
Cooling
Corsair H115i
Legacy Desktop Processor Hierarchy
Recognizing that a lot of older platforms are going to be paired with graphics subsystems multiple generations old, we wanted to define the top of our range to encourage balance between host processing and complementary GPUs. At this point, anyone with a Sandy Bridge-based Core i7 would realize a gain from stepping up to Coffee Lake or Kaby Lake, for example. And putting AMD’s top FX CPUs next to a handful of Core i7s and those older Core i5s represents an upgrade to their status.
Currently, our hierarchy consists of 13 total tiers. The bottom half of the chart is largely outdated; you’ll notice those CPUs dragging down performance in the latest games, regardless of the graphics card installed in your PC. If you own a CPU in that range, an upgrade could really take your experience to another level.
Really, it’s the top five tiers or so that remain viable. And in that top half of the chart, an upgrade is typically worthwhile if it’s a least a couple of tiers higher. Otherwise, there’s just not enough improvement to warrant the expense of a fresh CPU, motherboard and RAM (not to mention the graphics card and storage solution you’d be considering as well).
Legacy Desktop Processor Hierarchy
Intel CPUs
AMD CPUs and APUs
Intel Core i7-8700K ($382.00 On Walmart)
Intel Core i9-7900X ($969.00 On Walmart)
Intel Core i9-7960X ($1617.83 On Amazon)
Intel Core i9-7980XE ($1579.99 On Amazon)
Intel Core i7-8700K ($382.00 On Walmart)
Intel Core i7-7740X ($350.00 On Amazon)
Intel Core i7-7700K ($344.61 On Walmart)
Intel Core i7-7820X ($599.99 On Walmart)
AMD Ryzen Threadripper 1950X ($77.99 On Walmart)
Intel Core i7-7700 ($317.07 On Walmart)
AMD Ryzen Threadripper 1920X ($199.99 On Amazon)
Intel Core i5-8400 ($204.29 On Walmart)
AMD Ryzen 7 1800X ($220.79 On Amazon)
Intel Core i7-7800X ($389.99 On Newegg)
AMD Ryzen Threadripper 1900X ($549.99 On Newegg)
Intel Core i7-7700T ($339.99 On Walmart)
AMD Ryzen 7 1700X ($163.00 On Amazon)
Intel Core i7-6950X ($1576.52 On Walmart)
AMD Ryzen 7 1700 ($219.79 On Amazon)
Intel Core i7-6900K ($788.00 On Amazon)
AMD Ryzen 5 1600X ($139.95 On Amazon)
Intel Core i7-6850K ($492.75 On Amazon)
AMD Ryzen 5 1600 ($103.57 On Amazon)
Intel Core i7-6800K ($347.88 On Walmart)
AMD Ryzen 5 1500X ($127.00 On Walmart)
Intel Core i7-6700K ($293.19 On Walmart)
AMD Ryzen 5 1400 ($139.99 On Walmart)
Intel Core i7 6700 ($317.97 On Walmart)
AMD Ryzen 3 1300X ($66.56 On Amazon)
Intel Core i7-5960X ($1,069.95 On Newegg)
Ryzen 5 2400G ($59.99 On Walmart)
Intel Core i7-5930K ($386.59 On Walmart)
Intel Core i7-5820K ($350.90 On Walmart)
Intel Core i7-5775C ($1077.88 On Amazon)
Intel i7-4960X ($465.58 On Amazon)
Intel Core i7-4930K
Intel Core i7-4820K
Intel Core i7-4790K
Intel Core i7-4770K
Intel Core i7-4790
Intel Core i7-4771
Intel Core i7-4770
Intel Core i7-3970X
Intel Core i7-3960X
Intel Core i7-3930K
Intel Core i7-3820
Intel Core i7-3770K
Intel Core i7-3770
Intel Core i5-7640X ($243.00 On Amazon)
Intel Core i5-7600K ($10.73 On Amazon)
Intel Core i5-7600 ($269.99 On Walmart)
Intel Core i5-7500 ($216.62 On Walmart)
Intel Core i5-7400 ($201.30 On Walmart)
Intel Core i5 6600k ($248.73 On Walmart)
Intel Core i5-6600 ($210.66 On Amazon)
Intel Core i5-6500 ($210.04 On Walmart)
Intel Core i5 6402P ($488.00 On Amazon)
Intel Core i5-6400 ($182.99 On Newegg)
Intel Core i5-5675C
Intel Core i5-4690K ($250.58 On Walmart)
Intel Core i5-4670K
Intel Core i5-4590
Intel Core i5-4670
Intel Core i5-4570
Intel BX80646I54460 ($215.17 On Amazon)
Intel Core i5-4440
Intel Core i5-4430
Intel Core i5-3570K
Intel Core i5-3570
Intel Core i5-3550
Intel CPUs
AMD CPUs and APUs
Intel Core i7-990X Extreme
Intel Core i7-980X Extreme
Intel Core i7-975 Extreme
Intel Core i7-2600K
Intel Core i7-2600
Intel Core i7-965
Intel Core i5-3470
Intel Core i5-3450P
Intel Core i7-7700 ($317.07 On Walmart)
Intel Core i5-3450
AMD FX-9370 ($232.17 On Walmart)
Intel Core i5-3350P
AMD FX-8370 ($220.00 On Amazon)
Intel Core i5-3330
AMD FX-8350 w/Wraith ($159.99 On Amazon)
Intel Core i5-2550K
AMD FX-8320 ($124.99 On Amazon)
Intel Core i5-2500K
AMD FX-8300 ($99.99 On Amazon)
Intel Core i5-2500
AMD FX-8150
Intel Core i5-2450P
Intel Core i5-2400
Intel Core i5-2380P
Intel Core i5-2320
Intel Core i5-2310
Intel Core i5-2300
Intel Core i3-7350K ($167.99 On Walmart)
Intel Core i3-7320 ($295.22 On Walmart)
Intel Core i3-7300 ($169.89 On B&H)
Intel Core i3-7100 ($131.99 On B&H)
Intel CPUs
AMD CPUs and APUs
Intel Core i7-980
Intel Core i7-970
Intel Core i7-960
Intel Core i7-875K
Intel Core i7-870
Intel Core i3 6320 ($218.52 On Amazon)
Intel Core i3 6300 ($130.00 On Amazon)
Intel Core i3-6100 ($130.34 On Walmart)
AMD FX-6350 w/Wraith ($134.99 On Amazon)
Intel Core i3 6100T ($136.99 On Tiger Direct)
AMD FX-4350 ($716.00 On Walmart)
Intel Core i3-6098P ($372.00 On Amazon)
AMD Phenom II X6 1100T Black Edition
Intel Core i3-4360
AMD Phenom II X6 1090T Black Edition
Intel Core i3-4350
AMD Phenom II X4 Black Edition 980
Intel Core i3-4340
AMD Phenom II X4 Black Edition 975
Intel Core i3-4170
Intel Core i3-4160
Intel Core i3-4150
Intel Core i3-4130
Intel Core i3-3250
Intel Core i3-3245
Intel Core i3-3240
Intel Core i3-3225
Intel Core i3-3220
Intel Core i3-3210
Intel Core i3-2130
Intel Core i3-2025
Intel Core i3-2120
Intel Core i3-2105
Intel Core i3-2100
Intel Pentium G4620 ($143.92 On Walmart)
Intel Pentium G4600 ($126.42 On Walmart)
Intel Pentium G4560 ($79.99 On Walmart)
Intel Pentium G4500 ($74.86 On Walmart)
Intel Pentium G4400 ($62.24 On Walmart)
Intel CPUs
AMD CPUs and APUs
AMD FX-8370E ($310.09 On Amazon)
AMD FX-8320 ($149.01 On Amazon)
AMD FX-8120
AMD FX-6300 ($89.98 On Amazon)
AMD FX-6200
AMD FX-4300 ($99.95 On Amazon)
AMD FX-4170
AMD Phenom II X6 1075T
Intel Core i7-950
AMD Phenom II X4 970 Black Edition
Intel Core i7-940
AMD Phenom II X4 965
Intel Core i7-930
AMD Phenom II X4 955
Intel Core i7-920
AMD A10-7890K APU ($411.00 On Amazon)
Intel Core i7-860
Intel A10-7870K ($34.99 On Walmart)
Intel Core i5-3220T
AMD A10-7860K ($42.90 On Walmart)
Intel Core i5-2405S
AMD A10-7850K
Intel Core i5-2400S
AMD A10-7800
Intel Core i5-760
AMD A10-7700K
Intel Core i5-750
AMD A10-6800K
Intel Core 2 Extreme QX9775
AMD A10-6790K
Intel Core 2 Extreme QX9770
AMD A10-6700
Intel Core 2 Extreme QX9650
AMD A10-5800K
Intel Core 2 Quad Q9650
AMD A10-5700
Intel A8-7650K ($105.99 On Amazon)
AMD A8-7600 ($17.99 On eBay)
AMD A8-6600K
AMD A8-5600K
AMD A8-3870K
AMD A8-3870
AMD A8-3850
AMD Athlon X4 880K ($74.90 On Amazon)
Intel Athlon X4 870K (4th quarter) ($35.79 On Walmart)
AMD’s processor market shares continue to grow. The market researcher Mercury Research estimates that the chip manufacturer in the third quarter 2020 on a total share of 20, 4 percent came – mobile, desktop and server included. All end customer CPUs added up, i.e. desktop models (20, 1 percent) and notebooks (20, 2 Percent), AMD has the 20 er mark cracked.
It went ahead within a year, especially with notebooks, where AMD’s market share was 5.5 percent or 37, 4 percent increase (Q3 / 2019: 14, 7 percent). For desktop processors, it was 2.1 percentage points or 11, 6 percent high (Q3 / 2011: 18 percent). The exact numbers from Mercury Research are broken down on the website Tom’s Hardware .
AMD’s CPU market shares broken down by segment (source: Mercury Research) Quarter Q3 2020 Q2 2020 Q3 2019 Desktop 20,1% 19, 2% 18, 0% Mobile 20, 2% 19, 9% 14, 7% Server 6.6% 5.8% 4.3% Desktop + Mobile 20, 2% 19, 7% 15,8th% Total (x 86) 22, 4% 18, 3% 14, 6% Ryzen ensures high proportions For AMD, the development looks bright before the market launch of the Ryzen – 5000 – generation: The The share of end customer CPUs was recently so high in the middle 2011. In the case of notebooks, 20, 2 percent represents a best value. According to market researchers, AMD mainly sold Ryzen processors, while Intel was able to improve the delivery situation for entry-level models (Celeron, Pentium, Core i3). Otherwise the plus at AMD would have been even greater.
If you include server CPUs, the chip manufacturer was in the third quarter 2020 for a total of 22, 4 percent (+ 6.3 percentage points vs. Q3 / 2011) – maximum value since the end of the Athlon – 64 – era 2007. Mercury Research includes network infrastructure and edge servers in its own statistics, which benefits Intel with Atoms and Xeon-D. AMD was 6.6 percent in Q3 / 2020 – plus 2.3 percentage points or 37, 5 percent compared to the previous year.
In the pure server comparison Epyc vs. Xeon SP should AMD the third quarter 2019 with a share of high 12, have completed 1 percent. This corresponds to a plus of 16, 3 Percent or 1.7 percentage points in just three months (Q2 / 2020: 10, 4 percent).
Data on the third quarter 2020 by Mercury Research show how AMD has gained market share in all sectors of the microprocessor market x 86. Compared to last year, big leap in notebook CPU sales, for a share in the client market of 20,2%.
by Manolo De Agostini published 04 November 2020 , at 07: 59 in the Processors channel AMD EPYC RyZen
AMD reached the highest market share since Q4 2007 in the microprocessor compartment x 86 , driven by increasing sales in all sectors, in particular the area of notebooks , where the ‘company has registered a new record .
This is the summary of Mercury Research’s analysis on the third quarter of this year , in which it is certified (but the last quarterly had been very clear about it) how the company led by Lisa Su is collecting the fruits of years of work and the opportunities left by Intel due to delays in setting up the 10 nanometers and the absence of new architectures for a long period of time.
The new data arrive on the eve of the market debut of the new Ryzen microprocessors 5000 for desktop systems, capable (according to AMD) to outperform Intel in all respects, including gaming. Intel will not respond to Ryzen 5000 except during the first quarter 2021 with Rocket Lake CPUs, effectively leaving AMD the Christmas season to extend its run.
Let’s get to the heart of Mercury Research’s analysis, starting from the “macro” data, that is the general share in the CPU sector x 86 (all data excludes the IoT sector): AMD rises to 22, 4% with an increase of 4.1 percentage points compared to the second quarter and 6.3 points over the year last. As written at the beginning, it is the highest data since Q4 2007.
Moving to the desktop sector only , AMD’s share reached 20, 1% , an increase of 0.9 percentage points over the previous quarter and +2.1 over the same period last year. AMD’s share in the desktop sector grows from 12 quarters in a row, and the new figure is the highest since the fourth quarter 2013.
According to analyst Dean McCarron, “ a little more than 60 % of AMD’s largest desktop sales come from what I would consider the high-end (e.g. not APUs or dual-cores, so Matisse and a small fraction of Pinnacle processors Ridges that remain.) I would say that 100% of the gain comes from these solutions as for the wing band, because Intel’s high-end has been flat and AMD’s has grown, while Intel’s entry-level business has surpassed that of AMD . Core i3, Pentium and Celeron grew very strongly and if you limited the share to entry-level only, Intel would have gained in the third trimes three “.
Regarding the notebook sector, AMD reached 20, 2% of the market x 86 , with an increase of 0.3 percentage points on the second quarter and 5.5 points over the past year . Again, the company’s share grows from 12 consecutive quarters and has surpassed the previous record of 19, 9% of Q2 2020.
It is precisely in the context of laptops that AMD is running, and it is not a small sector when you think it makes up about the 60% of the client market. “I would say Intel’s greatly improved availability of Core i3, Pentium and Celeron CPUs has had an impact , as AMD’s quarterly earnings were much lower than in previous quarters, especially in the mobile segment: the share grew by 0.3 points versus 2.9 points in the second quarter and 0.8 points in the first quarter, so growth is slowing and Intel appears to be in the process of catching up . I expect the market to continue to become more competitive in the coming quarters, especially in the low-end, due to the increased availability of entry-level solutions from Intel, “explained analyst McCarron.
The market share in the client sector x 86 , then the set of desktop and notebook CPUs, touched 20,2%, an incremen by 0.5 points on the previous quarter and 4.3 points on Q3 2019, marking a new high from Q2 2011.
And i server? The situation here is a moment more complex in numerical terms. AMD bases its market estimates on data from IDC, but it only takes into account single- and dual-socket platforms, cutting out four-socket and multi-socket servers, network infrastructure and edge computing. So the data that AMD usually reports is different from Mercury Research estimates, much higher and probably already in double digits.
That said, for the third quarter
Mercury Research analysts point to a 6.6% server share for AMD, an increase of 0.8 points on Q2 2020 and 2.3 points on Q3 2019 . “AMD has gained share in servers, but if only EPYC and Xeon Scalable were compared, last quarter’s number would have been around 12, 1% against 10, 4%, a pretty big gain, “McCarron explained. “This is mainly because Intel’s Xeon SP business fell due to a cloud recession and a degree slump in sales to businesses and governments, while AMD’s server solutions set a new revenue record and are still growing. . Entering the strong growth of Intel’s Atom SoC, AMD’s gain is much less and stops at 0.8 points. “
AMD shared the recent Mercury Research CPU market share results, giving us some insight into its gains in the third quarter of 2020. The headline news is that AMD has reached its highest overall market share since 2007 and has its highest desktop PC share since 2013. In terms of the client x86 market, AMD took its highest amount of share since Q2 2011.
The report comes on the heels of AMD’s stellar financial results that pointed towards such amazing growth. It also comes after Intel’s recent disappointing earnings report –– particularly in terms of Intel’s desktop PC processor sales.
But AMD’s share gains certainly aren’t limited to the desktop PC segment; the company is on a full-court press that sees it gaining share from Intel in every segment that it competes in.
Intel still insists it isn’t receiving more competition than it expected, so perhaps the company has planned to cede market share across the board. Intel has also seen a drastic shift in its product mix to lower-priced desktop PC chips, implying that AMD’s share gains are coming in the high-end margin-rich segment of the market. You can read more about that here. We also have additional comments from Dean McCarron of Mercury Research in each section below, and he confirms our assertion that AMD is expanding its share in the high-end of the desktop PC market.
AMD’s impressive gains in the desktop PC market come before the launch of its Zen 3 Ryzen 5000 processors, which the company claims take the leadership position over Intel in every metric – including Intel’s last bastion: Gaming. Those chips land November 5.
Intel won’t have a response to Ryzen 5000 until Rocket Lake lands in Q1 2021. That means AMD will likely own the holiday sales season, expanding the gains outlined below in a very rapid fashion. It looks like it will be a long winter for Intel.
Here are the bullet points from AMD, and we have segment-by-segment breakdowns below, including historical data:
AMD overall x86 CPU share was 22.4%, an increase of 4.1 share points quarter over quarter (QoQ) and 6.3 share points year over year (YoY): Highest share since Q4 2007
AMD desktop x86 share, excluding IoT, was 20.1%, an increase of 0.9 share points QoQ, and 2.1 share points YoY: AMD desktop share has grown over 12 consecutive quarters: Highest since Q4 2013
AMD notebook x86 share, excluding IoT, was 20.2%, an increase of 0.3 share points QoQ, and 5.5 share points YoY: AMD notebook share has grown over 12 consecutive quarters: This is a new record for AMD x86 notebook share, eclipsing the previous high of 19.9% set in Q2 2020
AMD client x86 share, excluding IoT, was 20.2%, an increase of 0.5 share points QoQ, and 4.3 share points YoY: Highest since Q2 2011
(Image credit: AMD)
Image 1 of 2
(Image credit: Courtesy Wes Vaske @wvaske)
Image 2 of 2
(Image credit: Courtesy Wes Vaske @wvaske)
3Q20
2Q20
1Q20
4Q19
3Q19
2Q19
1Q2019
4Q18
3Q18
2Q18
1Q18
4Q17
3Q17
2Q17
1Q17
4Q16
3Q16
AMD Desktop Unit Share
20.1%
19.2%
18.6%
18.3%
18%
17.1%
17.1%
15.8%
13%
12.3%
12.2%
12.0%
10.9%
11.1%
11.4%
9.9%
9.1%
Quarter over Quarter (QoQ) pp
+0.9
+0.6
+0.3
+0.3
+0.9
Flat
+1.3
+2.8
+0.7
+0.1
+0.2
+1.1
-0.2
-0.3
+1.5
+0.8
Year over Year (YoY) pp
+2.1
+2.1
+1.5
+2.4
+5
+4.8
+4.9
+3.8
+2.1
+1.2
+0.8
+2.1
+1.8
AMD has reached 20.1% of the desktop PC market, notching its highest penetration since the fourth quarter of 2013. AMD has grown its desktop PC market share for 12 consecutive quarters.
Here’s Dean McCarron’s analysis of his report:
“A bit more than 60% of AMD’s increased desktop shipments came from what I’d consider high-end (e.g. not dual-core or APU units, so Matisse and a tiny bit of Pinnacle Ridge cores that remain).”
“I’d say 100% of the gain came from the high end for them, because Intel’s high end was flat and AMD’s high end grew, while Intel’s entry-level business out-grew AMD’s. Intel’s i3/Pentium/Celeron grew very strongly and if you limited the share to just entry-level, Intel would have gained in the third quarter.”
3Q20
2Q20
1Q20
Q419
3Q19
2Q19
1Q2019
4Q18
3Q18
2Q18
AMD Mobile Unit Share
20.2%
19.9%
17.1%
16.2%
14.7%
14.1%
13.1%
12.2%
10.9%
8.8%
Quarter over Quarter / Year over Year (pp)
+0.3 / +5.5
+2.9 / +5.8
+0.9 / +3.2
+1.5 / +4.0
+0.7 / +3.8
+1.0 / +5.3
+0.9 / ?
AMD has now reached 20.2% of the notebook market, an impressive 5.5 percentage point gain in one single year. This is by far AMD’s fastest-growing market, and it comprises roughly 60% of the client processor market, meaning these gains are very important in terms of overall volume and revenue.
This is also a new record for AMD’s notebook share, passing its 19.9% record it set in Q2, 2020.
“I would say Intel’s greatly improved availability of i3/Pentium/Celeron had an impact as the on-quarter gains by AMD were much lower than prior quarters, especially in mobile — AMD mobile share was up 0.3 points v. 2.9 points in Q2 and 0.8 points in Q1, so the gains are slowing and Intel appears to be in the process of catching up. I expect the market will continue to get more competitive in the upcoming quarters, especially in the low end, due to Intel’s entry-level capacity increasing so much,” said McCarron.
AMD bases its server share projections on IDC’s forecasts but only accounts for the single- and dual-socket market, which eliminates four-socket (and beyond) servers, networking infrastructure and Xeon D’s (edge). As such, Mercury’s numbers differ from the numbers cited by AMD, which predict a higher market share. Here is AMD’s comment on the matter: “Mercury Research captures all x86 server class processors in their server unit estimate, regardless of device (server, network or storage), whereas the estimated 1P [single-socket] and 2P [two-socket] TAM [Total Addressable Market] provided by IDC only includes traditional servers.”
3Q20
2Q20
1Q20
4Q19
3Q19
2Q19
1Q2019
4Q18
3Q18
2Q18
4Q17
AMD Server Unit Share
6.6%
5.8%
5.1%
4.5%
4.3%
3.4%
2.9%
3.2%
1.6%
1.4%
0.8%
Quarter over Quarter / Year over Year (pp)
+0.8 / +2.3
+0.7 / +2.4
+0.6 / 2.2
+0.2 / +1.4
+0.9 / +2.7
+0.5 / +2.0
-0.3 / –
+1.6 / 2.4
+0.2 / –
McCarron shared the server results from the report, which AMD did not include in its communications with the press, saying “According to Mercury, AMD also gained QoQ and YoY in the x86 server market. AMD utilizes a server-specific TAM of roughly 20 million CPUs, based on IDC data. We will update our x86 server share once IDC Q3 2020 data is available.”
In either case, the report shows that AMD did gain share in the server market by Mercury Research’s calculations, though McCarron says that AMD’s growth was hindered by a sharp increase in Intel’s Atom SoC shipments this quarter.
“AMD gained server share by either measure, but if you compared AMD EPYC v. just Intel Xeon SP, the number would be about 12.1 percent v. 10.4 percent last quarter, which is a pretty strong gain. This is mostly because Intel’s Xeon SP business declined to do a downturn in cloud and a big downturn in enterprise/government sales, while AMD’s server products set a new revenue record as they’re still growing. With Intel’s high Atom SoC growth included, though, the share gain is much smaller at 0.8 points,” McCarron expanded.
3Q20
2Q20
1Q20
4Q19
3Q19
2Q19
4Q18
3Q18
AMD Client
20.2%
19.7%
17.5%
17.0%
15.8%
15%
13.5%
11.6%
Client PP Change QoQ / YoY
+0.5 / +4.3
+2.2 / +4.7
-0.5% / ?
+1.1 / +3.5
+0.8 / +4.2
?
?
–
AMD Overall x86
22.4%
18.3%
14.8%
15.5%
14.6%
13.9%
12.3%
10.6%
Overall PP Change QoQ / YoY
+4.1 / +6.3
+3.5 / +1.2 (+3.7?)
-0.7 / ?
+0.9 / +3.2
+0.7 / +4
?
?
–
Perhaps the best overall view comes from the overall x86 market share. AMD has now reached 22.4% of the market, a stunning 6.3 percentage point gain over the prior year. This is the company’s highest share since Q4 2007.
In terms of client x86 share (excluding IoT), AMD has reached 20.2%, a 4.3 percentage point gain on the year. That’s the highest since Q2 2011.
Processors from AMD and Intel load microcode updates to repair bugs or to upgrade new functions. The exact functionality of these microcode updates is not publicly documented, rather they are usually encrypted and cryptographically signed. The security experts Maxim Goryachy, Dmitry Sklyarov and Mark Ermolov have now succeeded for the first time after long preparatory work in decoding the microcode updates for certain Intel processors.
Intel emphasizes that this is not associated with a security gap that can be used remotely, because the processors only carry out digitally signed microcode updates and the signature key is still secure.
But Goryachy, Sklyarov and Ermolov explain that it is now possible for the first time with (still) current processors to examine the functionality of Intel microcode updates. So far this was only possible with older processors, with AMD up to the generations K8 and K 10 ( Usenix Security 2017).
Maxim Goryachy has the US publication Ars Technica explains details of the microcode hack. Accordingly, the decryption of the microcode updates has so far only worked with Intel’s 2016 presented systems-on-chips (SoCs) “Goldmont” cores, especially Atom x5- / x7 – 3900 E and Celerons like N 3350, N 3450 and Pentium N 4200 / J 4205.
Mark Ermolov shows on Twitter how subroutines are structured in microcode.
(Image: Mark Ermolov / Twitter)
Through the gap The Goldmont microcode updates were accessed on the one hand via debugging functions that Goryachy, Sklyarov and Ermolov discovered in recent years (Chip Red Pill) and on the other hand via the 2017 Intel-SA security vulnerability revealed by them – 00086. Both the security hole and the debugging access require physical access to the respective system, for example via a debugging (JTAG) adapter.
The security researchers, two of them at the Russian company Positive Technologies ( PTE), enable the “Red Unlock” operating mode, which is actually only intended for internal Intel developers. This in turn gives access to the so-called microcode sequencer ROM (MSROM).
Interesting for security researchers Mark Ermolov has published some screenshots on his Twitter account @_markel___ showing excerpts from the microcode. The findings so far are primarily of interest to security researchers.
However, the experts explain that by analyzing the microcode and better understanding how it works, conclusions can be drawn about other embedded functions in Intel processors. This in turn could, for example, bypass security functions.
First of all, however, it is primarily possible to examine the microcode updates that were previously inaccessible due to encryption. (ciw)
Some researchers have managed to extract the secret key used to encrypt the microcode of Intel CPUs based on Goldmont architecture. In this way it is possible to reverse engineer the updates or write a custom firmware. Intel doesn’t seem worried at all.
by Manolo De Agostini published 29 October 2020 , at 09: 57 in the Processors channel Intel Celeron Pentium Atom
A trio of researchers, as reported by Ars Technica, has extracted the secret key that encrypts (protects) updates of different Intel CPUs (Celeron, Pentium or Atom based on Goldmont architecture). This key allows you to decrypt CPU microcode updates, a kind of chip firmware that the company updates from time to time to fix vulnerabilities and other types of bugs. Obtaining a decrypted copy of an update could allow malicious people to do reverse engineering and learn how to exploit the flaw that the update intends to solve . The key may also allow third parties other than Intel to update a chip with its own microcode , although such an operation would have a limited life and would not “survive” a system reboot.
“That’s enough at the moment difficult to assess the impact on security , “independent researcher Maxim Goryachy told Ars Technica. “In any case, this is the first time in the history of Intel processors where you can run your own microcode and scan for updates “. Goryachy and two other researchers – Dmitry Sklyarov and Mark Ermolov, both of Positive Technologies – worked together on this research.
The genesis of this discovery dates back to three years ago , “when Goryachy and Ermolov found a critical flaw, referred to as Intel SA – 00086, which allowed the execution of arbitrary code within a chip independent core that included a subsystem called the Intel Management Engine, “writes the source. “Intel released a patch to fix the flaw, but since chips can always be rolled back to an earlier firmware version and then punctured, there is no way to completely eliminate the vulnerability.”
Five months ago, researchers were able to use the vulnerability in question to access “ Red Unlock “, a service mode integrated into Intel chips which is used by the company’s engineers to debug the microcode before the chips are released to market. In homage to the movie The Matrix, the researchers called their tool to access this debugger “Chip Red Pill”, because it allows researchers to enter a place that is usually off limits. “ The setup technique requires the use of a USB cable or a special Intel adapter which routes data to a vulnerable CPU “, Ars Technica writes, which suggests that it is certainly not something commonly feasible.
Having obtained access to the Goldmont CPU in Red Unlock mode, the Researchers managed to arrive at a special ROM area called MSROM (microcode sequencer ROM) and then embarked on the process of reverge engineering of the microcode. After months of analysis, they managed to understand the update process and the RC4 key used, while they failed to get the signature that Intel uses to cryptographically prove the authenticity of an update.
The Intel’s response
Asked about this, Intel explained that “the problem described does not represent a security exposure for customers and we do not rely on the information obfuscation behind Red Unlock as a security measure. to INTEL-SA mitigation – 00086, OEMs following Intel’s guidelines have mitigated the unlock specific requests for this search. The private key used to authenticate the microcode does not reside in the silicon and an attacker cannot upload an unauthenticated patch on a remote system “.
This means c that attackers cannot use Chip Red Pill and the decryption key it exposes to remotely hack vulnerable CPUs, at least not without exploiting other currently unknown vulnerabilities. Likewise, such techniques cannot be used to “infect the Goldmont-based device supply chain.” The attack is, albeit difficult, usable only by having physical access to a computer equipped with one of these CPUs.
“ For now there is only one but important consequence “, concluded Ermolov. “Independent analysis of a microcode patch that was previously impossible. Now researchers can see how Intel fixes a bug or other vulnerability. And that’s great.”
In this article, which our team will regularly update, we will maintain a growing list of information pertaining to upcoming hardware releases based on leaks and official announcements as we spot them. There will obviously be a ton of rumors on unreleased hardware, and it is our goal to—based on our years of industry experience—exclude the crazy ones. In addition to these upcoming hardware release news, we will regularly adjust the structure of this article to better organize information. Each time an important change is made to this article, it will re-appear on our front page with a “new” banner, and the additions will be documented in the forum comments thread. This article will not leak information we signed an NDA for.
Feel free to share your opinions and tips in the forum comments thread and subscribe to the same thread for updates.
Adds AVX512 instructions (so far available only on HEDT platform, since Skylake-X). New instructions: AVX512F, AVX512CD, AVX512DQ, AVX512BW, and AVX512VL. New commands: AVX512_IFMA and AVX512_VBMI
20-30% broadening of various number crunching resources, wider execution window, more AGUs
18% IPC gains vs Cascade Lake
SHA-NI and Vector-AES instruction sets, up to 75% higher encryption performance vs. “Skylake”
Supports unganged memory mode
Integrated GPU based on new Gen11 architecture, up to 1 TFLOP/s ALU compute performance
Integrated GPU supports DisplayPort 1.4a and DSC for 5K and 8K monitor support
Gen11 also features tile-based rendering, one of NVIDIA’s secret-sauce features
Integrated GPU supports VESA adaptive V-sync, all AMD FreeSync-capable monitors should work with this
Ice Lake introduces Intel TME (Total Memory Encryption), also Intel Platform Firmware Resilience (Intel PFR)
Intel Core i9-10990XE
Release Date: unknown, originally early 2020, seems cancelled now
22-cores + HyperThreading
Uses Cascade Lake-X architecture
LGA2066 Socket
1 MB L2 cache per core, 30.25 MB shared L3 cache
4 GHz base, up to 5 GHz boost
Roughly matches Threadripper 3960X in Cinebench
Intel Rocket Lake [updated]
Release Date: Q1 2021
Succeeds “Comet Lake”
Variants: Rocket Lake-“S” (mainstream desktop), -“H” (mainstream notebook), -“U” (ultrabook), and -“Y” (low power portable)
14 nanometer production process
Seems to be limited to eight cores (2 less than 10-core Comet Lake)
Some indication of mixed HyperThreading configurations, for example 8-core, 12-thread
Uses “Cypress Cove” core, which seems to be a backport of “Willow Cove” to 14 nm process
Up to 10% IPC improvement over Skylake
No FIVR, uses SVID VRM architecture
125 W maximum TDP
Compatible with 400-series chipsets
Possible they release 500-series chipsets with added features
Socket LGA1200 (just like Comet Lake)
Supports PCI-Express 4.0
20 PCIe lanes
Intel Xe integrated graphics, based on Gen 12 with HDMI 2.0b and DisplayPort 1.4a
Engineering Sample: Family 6, Model 167, Stepping 0, 8c/16t, 3.4 GHz base, 5.0 GHz boost
Engineering Sample: Family 6, Model 167, Stepping 0, 8c/16t, 3.2 GHz base, 4.3 GHz boost
Intel Willow Cove and Golden Cove Cores
Release Date: 2021
Succeeds “Sunny Cove”
Willow Cove improves on-die caches, adds more security features, and takes advantage of 10 nm+ process improvements to increase clock speeds versus Sunny Cove
Golden Cove will add significant single-thread (IPC) increases over Sunny Cove, add on-die matrix multiplication hardware, improved 5G network-stack HSP performance, and more security features than Willow Cove
Intel Alder Lake [updated]
Release Date: H2 2021
Mixes CPU cores of various processing power (and energy consumption), similar to the Big.Little-like designs for mobile devices
Combines up to eight Golden Cove with up to eight Gracemont (Atom) cores
These cores have two different instruction sets, for example Golden Cove has AVX-512, TSX-NI and FP16, which Gracemont lacks
10 nm process
Uses Socket LGA1700
Alder Lake for desktop: 37.5 mm x 45 mm package
Desktop CPUs come in 125 W and 80 W
Could use Foveros 3D Stacking technology
Possible CPU configurations 8+8+1 (8 big cores, 8 small cores, GT1 integrated), and 6+0+1 (6 big cores, no small cores and GT1 integrated)
Includes Gen12 Xe iGPU
DDR5 memory support
PCI-Express 5.0 support
Includes CLDEMOTE instruction, to invalidate cache lines
Intel Sapphire Rapids
Release Date: H2 2021
Successor to Cooper Lake
8-channel DDR5
Uses Socket LGA4677
For enterprise / data center
10 nm+ production process
Willow Cove CPU cores
PCIe 5.0
Probably 7 nm process
Platform name: Eagle Stream
Includes CLDEMOTE instruction, to invalidate cache lines
Intel Grand Ridge [added]
Release Date: 2022 or later
Produced on 7 nm HLL+ process
Successor to Atom “Snow Ridge”
24 cores across 6 clusters with 4 cours each
4 MB L2 per cluster, plus L3 cache
Uses Gracemont CPU core
Dual-channel DDR5
PCI-Expres Gen 4 with 16 lanes
Intel Elkhart Lake
Release Date: Unknown
Produced on 10 nm process
Designed for next-gen Pentium Silver and Celeron processors
CPU cores use Tremont architecture
GPU uses Gen 11
Dual-core and Quad-core configurations
Single-channel memory controller with DDR4 and LPDDR4/x support
Engineering sample: 1.9 GHz, 5/9/12 W TDP
Intel Meteor Lake [updated]
Release Date: 2022 or 2023
Succeeds “Alder Lake”
New microarchitecture, more advanced than “Willow Cove”, possibly “Golden Cove”
As of late 2020 Intel is adding support for Meteor Lake to the Linux Kernel
Lisa Su in a CES 2020 interview said “we will have a high-end Navi […] it is important”
AMD CFO: “Big Navi” will be a halo product and not merely a lofty performance increase over the RX 5700 XT to make AMD competitive against GeForce “Ampere.”
Adds support for DirectX 12 Ultimate: variable-rate shading and hardware-accelerated ray-tracing (DXR version 1.1)
AMD RDNA 2 [updated]
Announcement: October 28
Lisa Su: “we will have our new next-generation RDNA architecture that will be part our 2020 lineup”
TSMC, 7 nm Plus (probably not 7 nm+ EUV)
Up to 18% higher transistor density
Higher clock speeds than RDNA
50% better performance per Watt than RDNA, twice the efficiency as GCN
Adds variable rate shading
Adds support for BFloat16
Adds AV1 video decode hardware acceleration
Adds hardware raytracing acceleration (DXR version 1.1)
Supports Microsoft DirectX 12 Ultimate API /DXR, VRS, Mesh Shaders & Sampler Feedback)
Same GPU architecture powers PlayStation 5 & Xbox Series X
AMD Radeon RX 6500 [added]
Release date: unknown
40 Compute Units / 2560 Stream Processors
192-bit GDDR6 memory
7 nanometer production process
RDNA2 architecture
Codename “Navy Flounder”
Below $250
AMD RDNA 3
Release Date: Late 2021 or 2022
“Advanced Node”, probably TSMC 6 nm or 5 nm
AMD CDNA and CDNA2 [updated]
Release Date: 2020 for CDNA and 2021-2022 for CDNA2
New architecture that focuses on compute for “Radeon Instinct”
TSMC 7 nm or 7 nm+
128 Compute Units = 8192 shaders
Arcturus engineering sample has 120 CUs (7680 shaders), 878 MHz for the core clock, 750 MHz SoC clock, and 1200 MHz memory clock
Compute only—Rasterization, display controllers and media encoding hardware removed
SDV OpenCL performance in Geekbench: 55373 points, with 3.53 Gpixels/s in “Sorbel,” 1.30 Gpixels/sec in Histogram Equalization, 16 GFLOPs in SFFT, 1.62 GPixels/s in Gaussian Blur, 4.51 Msubwindows/s in Face Detection, 2.88 Gpixels/s in RAW, 327.4 Mpixels/s in DoF, and 13656 FPS in Particle Physics. Roughly matches 11 CU Vega Picasso IGP
SDV is 15.2 cm long, 96 Execution Units, PCI-Express x16, slot only power (so 75 W), 3x DisplayPort, 1x HDMI, high noise levels
Up to 2x performance uplift for Intel Xe integrated graphics over previous Gen 11
Using a multi-chip design approach, with Foveros, Intel Xe scales up to 512 EUs with 500 W
512 EU model is datacenter only, 300 W 256 EU model for enthusiast markets
Targeted at 1080p gameplay, CES demonstration showed working gameplay on Destiny 2
Could be produced at Samsung to leverage their 10 nm tech, while Intel ramps up its own
Future Xe GPUs could be built on TSMC 6 nm and 3 nm nodes
Raytracing hardware acceleration support will definitely be included on the data-center GPUs (and probably on the consumer models, too)
Double-digit TFLOP/s scaling all the way up to 0.1+ PFLOP/s
Will be used in upcoming Cray Aurora Supercomputer for Argonne National Laboratory in 2021
Targeting a wide segment of markets, including consumer (client-segment) graphics, enthusiast-segment, and data-center compute
Uses new graphics control panel that’s being introduced during 2019
Intel Discrete GPU / Arctic Sound
Release Date: 2020
Intel will hold a world tour in 2019, to build enthusiasm for the new architecture
Advanced management for power and clocks
Test chip: 8×8 mm² die area, 1.54B transistors, 14 nm, 50-400 MHz clock, EUs at 2x clock if needed
Raja Koduri who left AMD in late 2017 is somehow involved
Confirmed to support VESA Adaptive Sync
Intel Ponte Vecchio
Release Date: 2021 or 2022
Discrete GPU
Produced on 7 nanometer production process
Probably not 7 nanometer Intel but 7 nm TSMC or even 6 nm TSMC
Multiple GPU dies will be combined into a single accelerator
Architected “for HPC modeling and simulation workloads and AI training”
Workloads can be processed by GPU and CPU at the same time, using Intel oneAPI
Foveros packaging technology
Xe link to combine multiple GPUs (CXL interconnect)
Release Date: September 2020, at the same time as Zen 3.
Highly likely these were scrapped when AMD decided to enable compatibility with 400 and 500 series chipsets
Socket AM4
Supporting Zen 3 Ryzen 4000 processors
Support for older CPUs very likely, probably at least Ryzen 3000
PCI-Express 4.0
Memory
DDR5 System Memory [updated]
Release Date: Late 2020, probably 2021
JEDEC standard finalized as of Jul 15th 2020
Demo’d in May 2018 by Micron: DDR5-4400
Samsung 16 Gb DDR5 DRAM developed since February 2018
Samsung has completed functional testing and validation of a LPDDR5 prototype: 10 nm class, 8 Gbit, final clocks: DDR5-5500 and DDR5-6400
Samsung has started 16 Gb LPDDR5 mass production in Aug 2020
SK Hynix 4800 – 5600 Mbps, 1.1 V
SK Hynix also has 16 Gb DDR5-5200 samples ready, 1.1 V, mass production expected 2020
April 2020: Hynix has 8.4 Gbps DDR5, minimum density per die is 8 Gbit, maximum is 64 Gbit
ECC is now supported by all dies (no longer specific to server memory modules)
SK Hynix demonstrated DDR5 RDIMM modules at CES 2020: 4800 MHz, 64 GB
Micron is shipping LPDDR5 for use in Xiaomi phones (Feb 2 2020). 5.5 Gbps and 6.4 Gbps
Samsung has begun production for LPDDR5 for mobile devices (Feb 25 2020). 16 GB, 5.5 Gbps
4800 – 6400 Mbps
Expected to be produced using 7 nm technologies
32 banks, 8 bank groups
64-bit link at 1.1 V
Burst length doubled to BL16
Bank count increased from 16 to 32
Fine grain refresh feature
Improved power efficiency enabled by Vdd going from 1.2 V to 1.1 V as compared to DDR4
On-die ECC
Voltage regulators on the DIMM modules
AMD DDR5 memory support by 2021/2022, with Zen 4
HBM2E Graphics Memory [updated]
Release Date: 2020
Offers 3.2 Gbps per pin (33% faster than HBM2)
Rambus offers a 4.0 Gbps memory interface controller
Samsung Flashbolt: 16 Gb per die, 8-layers stacked, 16 GB per chip with 410 GB/s bandwidth
Hynix: 460 GB/s, 3.6 Gbps, eight 16 Gb chips are stacked for a single 16 GB chip
Hynix: mass production has started as of July 2020
HBM3 Graphics Memory
Release Date: Not before 2019
Double the memory bandwidth per stack (4000 Gbps expected)
Expected to be produced using 7 nm technologies
HBMNext Memory [added]
Release Date: Late 2022 or 2023
JEDEC work in progress
Micron involved
GDDR6X Graphics Memory
Release Date: 2020
Will first be used on new GeForce RTX 3000 / Ampere Series
Silicon Fabrication Tech
TSMC 7 nanometer+
Release Date: Q4 2019
TSMC N7+ is successor to original 7 nm node
Uses EUV (Extreme Ultra Violet)
15-20% more density and improved power consumption over N7
TSMC 6 nanometer
Release Date: Unknown
Backwards compatible with 7 nm process—no new design tools needed
Uses EUV (Extreme Ultra Violet), up to four EUV layers
18% higher logic density than N7
TSMC 5 nanometer [updated]
Release Date: March 2020 to tape-out customer designs
Risk production as of Q2 2019
High volume production: Q2 2020
Uses TSMC’s second implementation of EUV (Extreme Ultra Violet)
Up to 1.8x the density of 7 nm
Up to 14 layers
+15% higher clocks
30% better bower than N7
Intel might be a customer of this node
N5P “Plus” node: improvement to N5 while staying on 5 nm, 84-87% increase in transistor densities over N7
TSMC 5 nanometer+
Release Date: 2021
High-volume production in Q4 2020
Uses EUV (Extreme Ultra Violet)
TSMC 4 nanometer [updated]
Mass production: 2023
Codename “N4”
Uses EUV lithography
TSMC 3 nanometer [updated]
April 2020: On-Track
Risk production: 2021
Volume production: H1 2022
FinFET technology
Uses TSMC’s third implementation of EUV (Extreme Ultra Violet)
10-15% speed improvement at iso-power or 25-30% power reduction at iso-speed, compared to N5.
55,000 water per month at the start, 100,000 by 2023
TSMC 2 nanometer [updated]
No details known other than “TSMC has started development”
June 2020: TSMC is accelerating R&D
Sep 2020: Fab construction has begun
Will use Gate-All-Around (GAA) technology
Samsung 6 nanometer
Release Date: Unknown
First product taped out as of Q2 2019
Uses EUV (Extreme Ultra Violet)
Special variant for customers
Samsung 5 nanometer
Release Date: 2020
Ready for customer sample production as of Q2 2019
Mass production in Q2 2020
Yields are challenging as of Q2 2020
Uses EUV (Extreme Ultra Violet)
Up to 25% the density of 7 nm
20% lower power consumption
10% higher performance
Samsung 3 nanometer
Release Date: 2022
50% less power while delivering 30% more performance
45% less silicon space taken per transistor (vs 7 nm)
Intel 7 nanometer
Release Date: 2022 or 2023
Succeeded by 7 nm+ node in 2022, and 7 nm++ in 2023
Uses EUV (Extreme Ultra Violet)
4x reduction in design rules
Planned to be used on multiple products: CPU, GPU, AI, FPGA, 5G networking
Other
Hynix 4D NAND
Release Date: H1 2019
Developed by SK Hynix
Sampling in Q4 2018
Products demonstrated at CES 2020: Platinum P31 M.2 NVMe and Gold P31—PCIe 3.0 x4, using flash, DRAM and controller made by Hynix, over 3 GB/s read/write.
Reduces chip physical size, while increasing capacity at the same time
At the beginning of September, Intel officially presented the first processors 11 generation for laptops called Tiger Lake. The new Intel Core units offer higher performance and a completely new integrated graphics chip Intel Iris Xe Graphics – today you will read our premiere article about one of such processors. Meanwhile, it is known for a long time that Intel is preparing not only more powerful Core systems, but cheaper, weaker and above all cheaper Pentium Gold and Celeron for the cheapest notebooks. The manufacturer has prepared two new units based on the Willow Cove architecture and belonging to 11 Tiger Lake generation. I am talking about Intel Pentium Gold 7505 and Intel Celeron 6305. What are the characteristics of processors compared to their predecessors?
Intel presented two more processors belonging to the Tiger Lake family. I am talking about budget Pentium Gold units 7505 and Celeron 6305.
Intel Pentium Gold 7505 is a 2-core and 4-thread processor. On the other hand, the cheapest Intel Celeron 6305 is still a 2-core and 2-thread unit, without support for Intel Hyper- Threading. However, the processors have new Willow Cove cores, which should still improve performance, especially single-threaded. We expect Intel Pentium Gold 7505 and Celeron 6305 will be placed in very cheap laptops, often below PLN 2,000. The table below presents the specification of CPU units.
Specification
Intel Pentium Gold 7505
Intel Pentium Gold 6405 U
Intel Celeron 6305
Intel Celeron 5305 U
Architecture
Willow Cove
Skylake
Willow Cove
Skylake
Family
Tiger Lake
Comet Lake -U
Tiger Lake
Comet Lake-U
Lithography
10 nm SuperFin
14 nm
10 nm SuperFin
14 nm
Number of cores / threads
2C / 4T
2C / 4T
2C / 2T
2C / 2T
Base clock
2 GHz
2.4 GHz
1.8 GHz
2.3 GHz
Boost Timer
3.5 GHz
–
–
–
RAM controller
DDR4 3200 MHz LPDDR4x 3733 MHz
DDR4 2400 MHz LPDDR3 2133 MHz
DDR4 3200 MHz LPDDR4x 3733 MHz
DDR4 2400 MHz LPDDR3 2133 MHz
Cache L3
4MB
2 MB
4 MB
2 MB
Graphics layout
Intel UHD Graphics
Intel UHD Graphics 610
Intel UHD Graphics
Intel UHD Graphics 610
TDP
15 In
15 W
15 In
15 In
As you can see, Intel Pentium Gold is the best 7505, which got support for Intel Turbo Boost 2.0 technology, thus achieving the maximum clock speed of 3.5 GHz. The weaker of the Tiger Lake systems – Celeron 6305 – unfortunately no longer supports Turbo. The new processors also got a newer memory controller, now supporting DDR4 memory 3200 MHz and LPDDR4x 3733 MHz. However, none of the chipsets have an integrated Xe-based graphics chip. Like the Core i3 – 1115 G4, it is a weaker Intel UHD Graphics unit. All processors have standard TDP of 15 W.
BIOSTAR today announced its new motherboard named B 250 MHC. The motherboard is based on the B 250 single-chip architecture from Intel and is also suitable for the use of older Intel processors of the 6th and 7th generation. According to BIOSTAR, the B 250 MHC is very well suited for business systems. The mainboard supports up to 32 GB of DDR4 main memory in 2 DIMM slots and offers PCI Express 3.0 and USB 3.2 Gen.1.
Based on the B 250 microchip architecture, the motherboard supports the Core i3, i5. i7, Pentium and Celeron processors. The external connections include a PS / 2 keyboard and a PS / 2 mouse port for users of an older mouse or keyboard. To connect a monitor you have the choice between an HDMI connection and VGA. Furthermore, there is of course a Gigabit Ethernet LAN connection as well as 4x USB 3.2 (Gen.1), 2x USB 2.0 and 3x audio, which are based on the ALC 887 technology and 7.1 channels with high-definition audio.
The motherboard offers four SATA III connectors for storage space. The B 250 MHC also has the form factor Micro-ATX (22. 6 cm x 17. 1 cm) and is therefore also compatible with smaller housings. There is official support for the operating systems Windows 7 (32 / 64 bit) / 8.1 (64 bit) / 10 (64 bit). The delivery also includes two SATA cables, an I / O shield, a DVD driver and of course the instructions. The mainboard ric
Intel has finally pulled the veil off of the final specs of its 11th-Generation Tiger Lake processors after slowly trickling out details of the new chips for an entire year. The TIger Lake chips look to slow AMD’s advance with its impressive 7nm Ryzen 4000-series “Renoir” chips that have steadily gained traction over the last several months, but Intel is finally moving on to its 10nm SuperFin process that brings higher clock speeds and a big 20% boost to performance. Intel has also finally shared benchmarks that give us at least some idea of how its chips stack up against the Ryzen competition – Intel claims its quad-core models are faster than AMD’s eight-core Renoir chips, and that its integrated graphics have finally taken the lead.
We recently had the chance to put those claims to the test with a validation platform that Intel provided, giving us a glimpse of what to expect from Tiger Lake in the future. We’ll cover out test results below.
Intel also recently confirmed that we’ll soon see eight-core Tiger Lake models come to market, though the series will be confined to dual- and quad-core models for some time. We’ve also seen the first sign of Tiger Lake desktop PC systems emerge in the preliminary listings for a new line of ASRock NUCs, but we’ll see those systems in more flavors as other vendors release their products.
Intel’s Tiger Lake brings a dizzying array of improvements over the company’s previous-gen Ice Lake with higher clock speeds, a doubling of graphics performance, the first PCIe 4.0 support for laptops, and support for LPDDR4x memory serving as the headline advances. Intel also unveiled its new Evo platform, which is the second-gen of its Project Athena initiative. After Intel shared the technical details of its architecture, the new 10nm SuperFin process, and even more low-level details, we now have all the info condensed down into this article. Let’s start with the chips, then take a look at some of the first Tiger Lake laptops to hit the market.
Intel 11th-Gen Core Tiger Lake At A Glance
Willow Cove cores – quad-core and dual-core models
Intel Iris Xe LP graphics for 2x faster 1080p gaming
10nm SuperFin process gives up to 20% increase in clock frequency
Support for LPDDR5 – LPDDR4x for first models
Industry first PCIe 4.0 for laptops
New media and display engine
WifI 6 and Thunderbolt 4
Release Date: 50+ designs shipping this holiday season (starts in October)
150+ models in total
New Intel Evo (second-gen Project Athena) options
Price: Varies based on laptop
Intel 11th-Gen Core Tiger Lake UP3 Specifications
Intel announced a total of nine new chips. We have the nitty-gritty specs below, but first we’ll break down the meaning behind the confusing mish-mash of product identifiers.
Intel’s Tiger Lake comes with the Willow Cove processing cores and Xe LP graphics on one larger 10nm SuperFin die, and a separate smaller 14nm PCH (platform controller hub) chipset that handles extra I/O and connectivity duties.
Image 1 of 4
(Image credit: Intel)
Image 2 of 4
(Image credit: Intel)
Image 3 of 4
(Image credit: Intel)
Image 4 of 4
(Image credit: Intel)
Intel has two basic packages: The larger package on the left is for the high-performance UP3 models (formerly U-Series) that operate within a 12 to 28W TDP, and the UP4 package (formerly Y-Series) on the right for devices that operate at 7 to 15W. These packages are then integrated onto incredibly small motherboards (second picture in the album) that find their way into the new Tiger Lake laptops and thin-and-lights.
The Tiger Lake chips span the Core i7, i5 and i3 families and come with varying levels of graphics performance. Intel splits its Xe LP graphics up into G7 and G4 families. Tiger Lake models with “G7” at the end of the product name come with either 96 or 80 execution units (EUs), with the full-fledged 90 EU models coming with Intel Iris Xe branding. Chips with “G4” at the end of the product name come with 48 EUs. Naturally, the Iris Xe models with more EUs offer the high end of performance, which we’ll see in the benchmarks shortly.
Intel Tiger Lake UP3 Processors
PROCESSOR
CORES/THREADS
GRAPHICS (EUs)
OPERATING RANGE (W)
BASE CLOCK (GHZ)
SINGLE CORE TURBO FREQ (GHZ)
MAXIMUM ALL CORE FREQ (GHZ)
Cache (MB)
GRAPHICS MAX FREQ (GHZ)
MEMORY
Core i7-1185G7
4C / 8T
96
12 – 28W
3.0
4.8
4.3
12
1.35
DDR4-3200, LPDDR4x-4266
Core i7-1165G7
4C / 8T
96
12 – 28W
2.8
4.7
4.1
12
1.30
DDR4-3200, LPDDR4x-4266
Core i5-1135G7
4C / 8T
80
12 – 28W
2.4
4.2
3.8
8
1.30
DDR4-3200, LPDDR4x-4266
Core i3-1125G4*
4C / 8T
48
12 – 28W
2.0
3.7
3.3
8
1.25
DDR4-3200, LPDDR4x-3733
Core i3-1115G4
2C / 4T
48
12 – 28W
3.0
4.1
4.1
6
1.25
DDR4-3200, LPDDR4x-3733
You’ll notice that Intel has discarded its practice of listing a single TDP value. Instead the company now defines a full dynamic range of performance that spans 12 to 28W with the UP3 models. This allows laptop makers to tailor the chips for the thermal capabilities of their products, with high-end models having sufficient cooling to enable full performance, while lower-end models with less-capable cooling can be tuned to a lower TDP setting. The TDP can even change while in use based upon device temperature, power delivery, and orientation. Intel doesn’t require laptop makers to list their TDP ratings, though, so you’ll have to turn to third-party reviews for the full skinny on performance.
The flagship Core i7-1185G7 leads the UP3 lineup. This chip boosts to 4.8 GHz and has a 3.0 GHz base frequency, both of which are a big increase of 700 MHz over the previous-gen model. Intel has also made a big step forward with a 4.2 GHz all-core boost clock that will help chew through demanding productivity apps. To put that in perspective – the maximum single-core boost from AMD’s fastest Ryzen 4000 processor weighs in at 4.2 GHz. Intel can pull that off on all cores at once, which helps explain some of its performance advantages we’ll see in the benchmarks below.
The 1185G7 also comes with the Xe LP graphics engine with the full complement of 96 EUs, so Intel brands it as Iris Xe. The graphics unit runs at 1.35GHz, an increase of 250 MHz over the previous-gen graphics on the Core i7-1068NG7. The chip comes armed with 12MB of L3 cache and supports LPDDR4X-4266.
The Core i3-1115G4 slots in as the low-end model of this line up. This dual-core quad-thread chip comes with a 3.0 GHz base, 4.1 GHz boost, and impressive 4.1 GHz maximum all-core frequency. The chip’s Xe LP graphics engine comes with 48 EUs and boosts to 1.25 GHz, which is pretty agile for a low-end chip. However, these chips step back from LPDDR4x-4266 support to LPDDR4x-3733, which will hamper performance in some tasks. Notably, the Core i5 and i3 models come with 8MB and 6MB of L3 cache, respectively, which is less than the full 12MB found on the Core i7 models.
Intel 11th-Gen Core Tiger Lake UP4 Specifications
Intel Tiger Lake UP4 Processors
PROCESSOR
CORES/THREADS
GRAPHICS (EUs)
OPERATING RANGE (W)
BASE CLOCK (GHZ)
SINGLE CORE TURBO FREQ (GHZ)
MAXIMUM ALL CORE TURBO (GHZ)
Cache (MB)
GRAPHICS MAX FREQ (GHZ)
MEMORY
Core i7-1160G7
4C / 8T
96
7 – 15W
1.2
4.4
3.6
12
1.10
LPDDR4x-4266
Core i5-1130G7
4C / 8T
80
7 – 15W
1.1
4.0
3.4
8
1.10
LPDDR4x-4266
Core i3-1120G4*
4C / 8T
48
7 – 15W
1.1
3.5
3.0
8
1.10
LPDDR4x-4266
Core i3-1110G4
2C / 4T
48
7 – 15W
1.8
3.9
3.9
6
1.10
LPDDR4x-4266
The UP4 models slot into a 7 to 15W performance range for premium ultra-thin devices, including fanless models. Here we have Tiger Lake Core i7, Core i5, and Core i3 models, just like with the UP3 family, but with pared back frequencies to enable the lower level of operation.
The Core i7-1160G7 comes with four cores and eight threads paired with Iris Xe graphics that operate at a 1.1GHz boost clock, while the low-end dual-core Core i3-1110G4 comes with a 48 EUs that boost up to 1.1 GHz. All of the UP4 models support LPDDR4x-4266.
Intel Tiger Lake Pentium Gold and Celeron
Intel Tiger Lake and Celeron
Cores / Threads
Base / Boost (GHz)
TDP
L3 Cache
Memory
Graphics
Graphics EU / Clocks
Pentium Gold 7505
2C / 4T
2.0 / 3.5 GHz
15W
4MB
DDR4-3200 / LPDDR4x-3733
UHD Graphics – Xe LP
48 / 1.25 GHz
Celeron 6305
2C / 2T
1.8 / –
15W
4MB
DDR4-3200 / LPDDR4x-3733
UHD Graphics – Xe LP
48 / 1.25 GHz
Celeron 6305E
2C / 2T
1.8 / –
15W
4MB
DDR4-3200 / LPDDR4x-3733
UHD Graphics – Xe LP
48 / 1.25 GHz
Intel recently stealth-launched its Tiger Lake Pentium Gold and Celeron processors, and they come with the unanticipated addition of AVX2 instructions, the Intel Deep Learning Boost technology (using the AVX512-VNNI instruction), and the Intel Gaussian and Neural Accelerator 2.0, matching the more expensive Tiger Lake models. In the past, Intel has removed support for the aforementioned features in its lesser Pentium Gold and Celeron families, so this marks a big step forward on the performance front. Intel also added Turbo Boost support for the Pentium Gold 7505, a first for the mobile Pentium lineup.
The rest of the features are somewhat expected, though we also see the debut of the power UHD Graphics Xe LP graphics engine with 48 EUs and a 1.25 GHz peak clock rate. The processors support cTDP (Configurable TDP), so OEMs can adjust the clocks up to 1.8 GHz and 2.0 GHz for the Celeron models, and 3.9 GHz for the Pentium Gold. We also see that Intel dialed back memory support to LPDDR4-3733 from the 4367 MHz we see with the more expensive models, and also stepped back to PCIe 3.0 for the Pentium and Celeron chips.
Intel Tiger Lake Iris Xe Graphics Gaming Performance
Image 1 of 6
(Image credit: Tom’s Hardware)
Image 2 of 6
(Image credit: Tom’s Hardware)
Image 3 of 6
(Image credit: Tom’s Hardware)
Image 4 of 6
(Image credit: Tom’s Hardware)
Image 5 of 6
(Image credit: Tom’s Hardware)
Image 6 of 6
(Image credit: Tom’s Hardware)
We’ll cover the details of the Xe LP graphics engine below, but for now, let’s see the new Iris Xe integrated graphics in action in our own testing. However, we have to note that these results came in an Intel-provided reference system, so they might not be representative of the full performance we’ll see in laptops that come to market. Be sure to check out our preview for the full breakdown of the test environment.
Meanwhile, these results do give us a taste of the theoretical heights of Tiger Lake’s gaming performance. Here we can see that if you’re willing to compromise greatly on fidelity, you can run many games at 1080p on a laptop with Iris Xe graphics. It won’t be one of the best gaming laptops, we can only expect so much from integrated graphics, after all. Leading-edge AAA games may create some challenges, but the Xe LP engine is plenty powerful when you run it at an unconstrained 28W setting.
The reference system gave us to 1080p at 30 fps in low settings on most tests, but we’ll see have to wait to see what comes with shipping systems. However, we are undoubtedly getting closer to being able to have short 1080p gaming stints, albeit at reduced fidelity, on Ultrabooks.
Intel Tiger Lake Performance in Desktop Applications
Image 1 of 3
(Image credit: Tom’s Hardware)
Image 2 of 3
(Image credit: Tom’s Hardware)
Image 3 of 3
(Image credit: Tom’s Hardware)
Here we can see the results of our preliminary application testing on the Tiger Lake reference system, but be aware that all of the caveats of the reference system apply.
Intel’s Tiger Lake pulls off pretty impressive performance in lightly-threaded applications, especially when you consider its four core models square up against AMD’s potent eight-core Ryzen 4000-series chips. As expected, though, AMD’s Ryzen takes the lead when we flip over to applications, like HandBrake, that can use its eight cores and 16 threads more effectively.
Image 1 of 11
(Image credit: Intel)
Image 2 of 11
(Image credit: Intel)
Image 3 of 11
(Image credit: Intel)
Image 4 of 11
(Image credit: Intel)
Image 5 of 11
(Image credit: Intel)
Image 6 of 11
(Image credit: Intel)
Image 7 of 11
(Image credit: Intel)
Image 8 of 11
(Image credit: Intel)
Image 9 of 11
(Image credit: Intel)
Image 10 of 11
(Image credit: Intel)
Image 11 of 11
(Image credit: Intel)
We’re also including some of Intel’s projections here, due to the wider range of benchmarks, but be sure to take those with the same grain of salt as any other vendor-provided benchmarks. Intel shared benchmarks of its chips beating the Ryzen 7 4800U in a whole range of applications, including office and productivity/creativity software. As always, we’ll need to wait for more comprehensive third-party benchmarks of shipping laptops to make a final determination.
A lot of Intel’s claimed advantages stem from its big push into AI capabilities as the company works with a slew of software vendors to enable support for its newest capabilities. These new software packages yield massive improvements, up to 5X, in performance due to support for AI-boosting DL Boost instructions that leverage AVX-512 to boost performance.
Leveraging the AVX instruction set for AI workloads could evolve into a significant advantage over AMD’s Ryzen 4000 processors as Intel’s software support broadens. Intel’s chips have long dropped into lower frequencies as densely-packed AVX instructions work their way through the processor, but Intel has reduced the impact with a new SuperMIM capacitor that keeps voltages steady. That allow the processor to remain in higher frequency ranges during heavy AVX workloads.
Intel doesn’t just focus on AI workloads that run on the Willow Cove cores, though. The Gaussian and Neural Accelerator (GNA) returns, but this time with a new 2.0 revision. This SoC-integrated AI accelerator block is used for processing all sorts of low-power voice-based applications, like translation and transcription, using low-power inferencing. Intel claims that this offload engine can reduce CPU utilization by 20% during these types of operations, but at much lower power consumption. This unit can be also be used for impressive noise cancelation capabilities without taxing the Willow Cove cores.
Intel Tiger Lake Battery Life
Image 1 of 2
(Image credit: Tom’s Hardware)
Image 2 of 2
(Image credit: Tom’s Hardware)
We’ll have to wait until Tiger Lake laptops hit our labs for the full rundown on battery life, but Intel claims to have made significant gains in power consumption. It also says that laptops could provide nearly the same amount of performance on battery as when they are plugged into the wall.
Our testing confirmed that performance remains high while on battery, but that will undoubtedly have an impact on battery life. We weren’t allowed to test battery life on the reference system we were given for testing, but stay tuned for more as shipping systems hit our labs.
(Image credit: Intel)
The chart above highlights some Intel’s claims about the performance and efficiency improvements that come from the company’s new focus on providing higher performance while the laptop is under battery power.
As you can see on the right, Intel claims that performance on the Ryzen 4000-series 4800U drops precipitously when you remove the power plug and the laptop operates on battery power alone. In contrast, Intel claims its Tiger Lake chips offer the same amount of performance, even boosting up to the full 50W of power, while on battery power. If that pans out in our testing, that means you’ll still get the full Tiger Lake performance while on battery power, but at the expense of battery life.
Intel Evo – The Second-Gen of Project Athena
Tiger Lake also marks the arrival of Intel’s second generation of Project Athena, but it now comes with Intel Evo branding. The Intel Evo program certifies that a laptop is designed with premium components for the fastest performance, and that the software shipped on the laptops doesn’t hinder performance. Laptops that pass Intel’s criteria earn a custom Intel Evo badge.
Intel has a dizzying number of requirements on both the hardware and software side of the Evo equation, but the goals include battery life projections of nine or more hours for 1080p laptops, eight hours for QHD models, and seven hours for UHD models. Intel also stipulates the system must wake from sleep in less than one second, that performance remains consistent on battery power (as outlined in the previous section), and that the system supports fast charging that gives four hours of battery life on a 30-minute charging session (1080p models).
Intel has an impressive list of the first Evo laptops, with the Acer Swift 5, Asus Zenbook Flip S, Lenovo Yoga 9i and Samsung Galaxy Book Flex 5G being the first models to come to market. Those will be followed by designs from all the usual suspects, like Acer, Asus, Dell, Dynabook, Razer, Samsung, HP, Lenovo, LG and MSI. You can learn more about the program here.
Intel Tiger Lake Laptops
Image 1 of 5
(Image credit: Tiger Lake launch)
Image 2 of 5
(Image credit: Tiger Lake launch)
Image 3 of 5
(Image credit: Tiger Lake launch)
Image 4 of 5
(Image credit: Tiger Lake launch)
Image 5 of 5
(Image credit: Tiger Lake launch)
With over 150 designs eventually coming to market, there will be plenty of Intel Tiger Lake laptops to choose from. However, we do have a list of some of the first devices, which we’ll add to when other notable devices come to market.
Lenovo’s Yoga and IdeaPad 9i series are on the premium side and even have options for lids with a genuine leather covering. The Yoga 9i comes in 14- and 15-inch options that weigh in at up to $1,799 for a fully-equipped 15-inch model. The 14-inch models retail for $1,399 for metal models, and $1,699 for the leather-clad option.
MSI bills the Stealth 15M as the ‘thinnest 15-inch laptop,’ but it still comes armed with a 15.6-inch “IPS-level” panel with a 144Hz refresh rate, Thunderbolt 4 support, Killer Wi-Fi AX1650, and options for PCIe 4.0 x4 storage. MSI hasn’t released pricing yet.
MSI also has its new Summit series on offer for professional users. The new models come in several different configurations, which you can see here, but MSI hasn’t shared pricing yet.
Intel Tiger Lake Thunderbolt 4, PCIe 4.0 Interface, WiFi 6
After losing the glory of being the first to PCIe 4.0 on the desktop (AMD holds that distinction), Intel is the first to bring PCIe 4.0 support to laptops. The faster interface enables speedier SSD options that provide more performance and efficiency than their PCIe 3.0 counterparts.
That marks the beginning of a new era for PCIe 4.0 SSDs, and while some may opine that the speedy interface draws more power, that isn’t the full story. While the PCIe 4.0 interface does draw more power than 3.0, it can transfer data at up to twice the speed per lane. That helps reduce the amount of time the interface is active, which allows the chip to drop into lower power states more quickly. Intel added the ability to shut off or dynamically adjust Tiger Lake’s PCIe interface when it isn’t fully active, and the faster interface could be used to employ fewer lanes during some workloads, both of which will allow you to enjoy the speed of PCIe 4.0 SSDs without making huge sacrifices on battery life.
Intel touts its support for integrated Thunderbolt 4 and USB 4, but these aren’t really ‘new’ protocols. In short, with speeds up to 40Gb/s, Thunderbolt 4 maintains the same maximum speed rating as its predecessor (TB3) and doesn’t enable new features. Instead, vendors are required to enable all of the high-end features that used to be optional, like the ability to hit the 40Gb/s data throughput requirements and support two 4K displays or one 8K display. This approach does simplify the confusing branding surrounding Thunderbolt 3, but from a hardware standpoint, the speeds and feeds remain the same.
Intel Tiger Lake Willow Cove Architecture and 10nm SuperFin Process
Intel also made some finer-grained improvements to its microarchitecture, and the resulting Willow Cove cores feature a rebalanced cache hierarchy to improve performance, dual ring bus fabric, SuperMIM capacitors, and new security enhancements, among many other improvements. We’ve covered the low-level details of the Willow Cove architecture here.
Intel pairs the new Willow Cove cores with its 10nm SuperFin process. The process offers much higher clock speeds at any given voltage, and it can also operate at a lower voltage at any given frequency, too. As a result, the chip has a greater dynamic frequency range from the minimum to maximum voltage, which provides better performance at every power level. That equates to faster mid-range performance in thin-and-light devices, not to mention peak performance in high-performance designs. We have the full details of Intel’s 10nm SuperFin technology here.
Intel Tiger Lake Iris Xe LP Graphics Engine
Intel’s Xe LP (Low Power) architecture powers the Tiger Lake chips, but don’t be fooled by the “Low Power” in the Xe graphics branding, though. The Xe LP graphics engine promises up to twice the performance of the previous-gen Gen11, addressing a key sore point in Intel’s lineup compared to AMD’s capable 7nm “Renoir” Ryzen Mobile processors with Vega graphics.
Intel’s Xe LP comes with a significantly revamped architecture that we covered in our Intel Drops XE LP Graphics Specs deep dive. The net-net is that the engine comes with up to 96 execution units (EU) and ‘significant’ performance-per-watt efficiency improvements over the previous Gen11 graphics, which implies twice the performance at lower power compared to Intel’s Ice Lake.
Intel revamped its display engine, too. Tiger Lake supports hardware acceleration for AV1 decode, up to four display pipelines, 8K UHD and Ultra Wide, 12-bit BT2020 color, and 360Hz and Adaptive Sync, among others listed in the album above. Tiger Lake also supports up to six 4K90 sensors (support starts at 4K30) and can process still images up to 42 megapixels, an increase over the prior 27MP limitation with Ice Lake.
Intel Tiger Lake Pricing and Availability
Intel says that over 50 new designs based on Tiger Lake chips will land in time for the holidays, and there will be over 150 models released in total. The first devices arrive in October. Unfortunately we don’t have an official price list for the chips, as they are only delivered to OEMs. That means our only measure is the pricing on the devices that come to market.
For a deeper look at the state of the desktop PC chips, head over to our Intel vs AMD CPU article.
Intel’s latest Celeron and Pentium Gold processors for laptops based on the Tiger Lake microarchitecture apparently support most technologies featured by more expensive 11th Generation Core i-series parts. Most notably, they support AVX2 extensions and even an AVX-512 instruction.
Intel’s Celeron and Pentium Gold processors for entry-level laptops are usually based around the same high-performance cores as their more expensive Core counterparts that belong to the same generation of CPUs. To differentiate between more advanced Core and cheaper Celeron and Pentium Gold parts, Intel used to disable numerous features on the latter, slowing them down in certain demanding applications and causing headaches for software developers, as well as frustration among end-users. As it turns out, Intel changes this practice rather significantly with its Tiger Lake-based Celeron 6305/6305E and Pentium Gold 7505 processors.
The new Celeron 6305 and Pentium Gold 7505 CPUs are dual-core processors (with or without Hyper-Threading) with a configurable TDP-up frequency of 1.80 GHz and 2.0 GHz (respectively), 4 MB cache, UHD Graphics based on the Xe-LP architecture with 48 execution units, and a dual-channel memory controller that supports DDR4-3200 as well as LPDDR4X-3733 memory.
Unlike previous-generation Celeron and Pentium Gold processors for notebooks (and embedded applications in case of the model 6305E), these CPUs support AVX2 instructions, the Intel Deep Learning Boost technology (using the AVX512-VNNI instruction), as well as the Intel Gaussian and Neural Accelerator 2.0, the same ‘premium’ features supported by more expensive Tiger Lake processors. Meanwhile, enabling support for AVX2 and the AVX512-VNNI instructions is not particularly surprising given that Intel’s upcoming low-power Gracemont cores will support AVX2 as well sometimes next year.
Surprisingly, Intel’s Celeron 6305 and Pentium Gold 7505 CPUs also feature Intel’s Xe-LP GPU with 48 EUs. That means that they offer significantly more graphics horsepower than Intel’s Core i7 chips with Iris Pro that the company released several years ago, and support up to four displays. Furthermore, the processors feature Thunderbolt 4 connectivity, another premium capabilities for entry-level notebooks. Last but not least, the Pentium Gold 7505 also supports Turbo Boost 2.0 up to 3.50 GHz, a first in the lineup’s history.
(Image credit: Future)
There are still differentiating features between the Core i3-1110G4 and the Celeron 6305 and Pentium Gold 7505. The Core i3 part is clocked considerably higher, has a larger 6 MB cache, supports LPDDR4X-4267 memory, features a PCIe Gen 4 interface, and supports configurable TDP-down in a bid to address sleeker laptops.
In fact, the Core i3-1110G4 is designed for compact, low-power machines and even comes in a smaller package. Intel’s Core-branded Tiger Lake chips support more AVX-512 instructions and, therefore, offer higher performance in certain applications that use them.
It remains to be seen whether Intel adopts the same approach with its upcoming desktop Celeron and Pentium-branded processors and enables AVX2 and Deep Learning Boost. But at least the company is doing it with mobile parts.
We use cookies on our website to give you the most relevant experience. By clicking “Accept”, you consent to the use of ALL the cookies.
This website uses cookies to improve your experience while you navigate through the website. Out of these, the cookies that are categorized as necessary are stored on your browser as they are essential for the working of basic functionalities of the website. We also use third-party cookies that help us analyze and understand how you use this website. These cookies will be stored in your browser only with your consent. You also have the option to opt-out of these cookies. But opting out of some of these cookies may affect your browsing experience.
Necessary cookies are absolutely essential for the website to function properly. This category only includes cookies that ensures basic functionalities and security features of the website. These cookies do not store any personal information.
Any cookies that may not be particularly necessary for the website to function and is used specifically to collect user personal data via analytics, ads, other embedded contents are termed as non-necessary cookies. It is mandatory to procure user consent prior to running these cookies on your website.
