Home / Technology / What's better for 2019 and beyond?

What's better for 2019 and beyond?

Given Computex and E3 2019, now is a good time to look at the tug-of-war between AMD and Intel. Both introduced new products. AMD is scheduled to launch new CPUs and GPUs in July. Intel will not launch the new tenth-generation processors until the end of 2019. Intel's new discrete GPUs will be released in 2020.

The status of AMD

<img class = "size-large wp-image-997597 aligncenter" src = "https://cdn57.androidauthority.net/wp-content/uploads/2019/06/AMD- Ryzen-9-3950X-E3-2019-840×472.jpg "old =" AMD Ryzen 9 3950X E3 2019 [19659004] AMD entered the x86 processor market as a subcontractor for Intel, which allowed AMD to use the x86-based 8086 These chips would help Intel to fulfill orders for IBM's new PCs.

When Intel introduced its 32-bit processor, AMD's contract came to a standstill, forcing the two companies into one AMD reinstated clones until 1

996 when the first internal x86-based processor, the AMD K5, hit the market.

In 2003, AMD introduced the first 64-bit consumer processor, the Athlon 64. In addition, the Athlon was launched 64 FX for enthusiasts and the Opteron chip for servers Thlon 64 X2, the company's first dual-core consumer chip, hit the market in 2005. The first 4-core chip, Phenom, came two years later. The Athlon and Phenom desktop parts seemed to be AMD's main focus.

This changed when Apple launched the mobile boom with its first iPhone and iPad has restarted its mobile efforts with Fusion. This initiative introduced AMD's very first Accelerated Processing Unit (APU), which puts CPU cores and GPU cores on a single chip. It also triggered a chain reaction in which AMD fell behind Intel on the desktop until 2016. AMD's APU efforts essentially dominated the bulldozer years.

For example, between 2011 and 2016, AMD's only desktop efforts were the FX brand chips. The code names Zambezi and Vishera were based on the bulldozer architecture of AMD (Piledriver was a revised bulldozer). In the meantime, Intel has released desktop and laptop chips every year. Given the uncertainty of the desktops, the company also focused on the corporate sector.

Naysayers predicted that tablets and smartphones would destroy the desktop and laptop market. But ultrabooks, 2 in 1 devices and removable devices have apparently saved the PC industry.

In defense of AMD, the naysayers predicted that tablets and smartphones would destroy the desktop and laptop market. But ultrabooks, 2-in-1 devices and removable devices seem to have saved the PC industry, almost killing tablets. Nevertheless, OEMs tend to stick to Intel-based chips in PCs while using ARM-based solutions in handheld mobile devices.

Despite the strong APU focus, AMD had a master plan.

Consoles and Graphics

Custom APUs based on the Graphics Core Next GPU architecture landed on Xbox One, Xbox One X, PlayStation 4, and PlayStation 4 Pro. Developers working on x86-based PCs can now create games that work without porting on the console PCs and . Finally, high-definition PC gaming returned.

This is the disadvantage of AMD's minimal CPU presence during the bulldozer years: it's also a graphics card manufacturer. AMD acquired ATI Technologies in 2006 and started producing graphics cards for desktops. Intel will participate in the discrete GPU race until 2020.

AMD's first generation Graphics Core Next (GCN) architecture was released in the 2012 Radeon HD 7000 "Southern Islands" add-in card family. The Radeon RX Vega series ended the GCN era of AMD with the 7nm Radeon VII graphics card from Gigabyte, Sapphire, XFX and others.

Similar to AMD's mobile focus, the two years between the Radeon RX 300 and Radeon RX Vega focused on mainstream graphics. This allowed Nvidia to dominate the desktop and notebook space with the GeForce GTX 900 and the GTX 10 series. In the meantime, AMD has released budget-friendly RX 400 and 500 cards that bring low-end VR and Full HD graphics to any desktop.

Meet Zen and Vega

In retrospect, AMD noted a four-year gap between FX. Vishera "CPU family and its Ryzen 1000 chips. Two years passed between the final high-end Radeon RX 300 desktop GPU and the budget-friendly Radeon RX 400 family from AMD.

During this time, AMD secretly worked on a new, completely new CPU architecture. Called Zen, it would put the company back on a competitive course.

AMD also developed the Vega graphics platform based on the fifth-generation GCN architecture. This design was AMD's high-end successor to the RX 300 family. AMD has also developed Radeon DNA (RDNA): the company's first new GPU architecture since the introduction of GCN in 2012.

According to AMD, Ryzen CPUs can achieve half the price of Intel processors.

According to AMD Ryzen CPUs, the performance of Intel processors can reach half price. This sounds good, but has a big disadvantage: Ryzen desktop chips do not contain integrated graphics. If you're a PC player, this probably does not matter because you need a specific video card. If you do not need a separate GPU, most Intel desktop and mobile CPUs contain integrated graphics. The Ryzen brand of AMD APUs also contain integrated graphics.

The original Ryzen 1000 series is based on AMD's first Zen design with 14nm process technology. The Ryzen 2000 Desktops Series is based on enhanced Zen architecture (also known as Zen +) and 12nm process technology.

On the mobile front, AMD's Ryzen 2000 APUs for laptops and desktops use the original 14nm Zen architecture. The new Ryzen 3000 APUs arriving in July are based on the 12nm Zen Plus design (or Zen +). Compared to desktop chips, AMD's Ryzen APUs are one step behind Zen's Gen1 refresh-Gen2 update model.

Ryzen 3000

 AMD Ryzen 3000 Series

The new Ryzen 3000 desktop CPUs are based on AMD's second-generation zen architecture (Zen 2) and the 7nm + Process technology from TSMC. Given the fact that Intel's 10nm Ice Lake chips will not be released until the end of 2019, this is remarkable. AMD's new batch also includes the upcoming Ryzen 9 3950X, a 16-core chip clocked at $ 749 to 4.7 GHz. The equivalent of Intel is the Core i9-9960X, which costs at least $ 1725. Ouch.

But wait! There is more! The new Ryzen 3000 series from AMD supports PCI Express 4.0, but not current Intel products. PCI Express, short for Peripheral Component Interconnect Express, is a standard for high-speed connections between CPU, graphics card, memory and more. The PCI-SIG approved the PCIe 4.0 specification in October 2017 and allowed data transfers up to 64 GB per second (16 GT / s).

Twenty months later, PCI Express 5.0 is now ready for hardware manufacturers. As with PCIe 4.0, devices that support this standard may not appear until after 20 months. With an x16 configuration (32GT / s), up to 128 GB per second is promised. AMD, Intel, Nvidia and many others have already agreed to adopt this new standard.

The formula for power per watt

The bottom line is that AMD strives to produce a performance per watt at Ryzen, enabling more cores and frequencies for half the cost. For enthusiasts, AMD offers its Ryzen thread-ripper CPUs like the 32-core 2990WX for $ 1,799. Currently, the highest core number in the Intel X-Series CPU family for enthusiasts is 18 in the Core i9-9980XE for $ 1,999.

In the desktop area, AMD is now in an excellent position. The company is not overloading its Ryzen portfolio with an insane number of products. With the Ryzen 2000 series, AMD offers eight desktop processors, four HEDT processors, ten mobile APUs and twelve desktop APUs. In the ninth generation of Intel's Coffee Lake Refresh family alone, the company sells 34 desktop CPUs and nine laptop CPUs. We expect HEDT chips to arrive later in the summer.

However, keep in mind that while AMD chips are cheaper, they consume more power. Look at this comparison between AMD and Intel:

Basic Speed ​​ Top Speed ​​ Power Price
Ryzen 7 3800X 3.9 GHz 4.5 GHz 105 Watts $ 399
Core i9-9900K 3.6GHz 5.0GHz 95 Watts $ 488

With Intel Turbo Boost technology, the Intel chip can deliver the 5.0GHz upper limit with two reach cores. The number of boosts drops to 4.8 GHz with four cores and to 4.7 GHz with eight cores. Meanwhile, AMD's Precision Boost 2 technology increases the speed of any number of cores. The increase is based on an analysis of the current environment, which includes thermal, electrical and headroom usage. AMD calls this the "reliability triangle".

However, the AMD chip has a basic speed advantage, while the Intel chip has a higher turbo speed. And while the Ryzen 7 chip is $ 89 cheaper, it also consumes 10 watts of power. Unfortunately, we have no comparative figures, as the Ryzen 3000 desktop parts will not arrive until July.

A new deal with Samsung.

 The Samsung logo.

AMD is re-entering the handheld market thanks to a new contract with Samsung. After the acquisition of ATI Technologies in 2006, the company was briefly present on the non-gaming handheld market. Now it's back in the game to license its GPU technology to Samsung.

Prior to the acquisition, ATI provided two SoCs (system-on) -a-chip (also known as all-in-one processors). Xilleon accelerated video decompression for broadcast networks. Imageon brought integrated graphics to mobile handheld devices that support the rendering of 2D and 3D graphics.

After the acquisition, the chips were renamed by AMD as AMD Imageon and AMD Xilleon. Two years later, AMD decided to focus primarily on x86-based processors and graphics chips. This meant the spin-off of the production facility as GlobalFoundries and the sale of the ATI-related SoC divisions in 2008. Qualcomm purchased the Imageon technology and gave it to Adurino as a brand, while Broadcom acquired the Xilleon technology.

The new agreement with Samsung brings AMD's Radeon graphics core technology for Samsung's Exynos chips for smartphones and tablets. Samsung typically uses its Exynos chips in internationally-sold devices, while it relies on Qualcomm Snapdragon chips in North America.

A new contract with Google

 Google Stadia Logo

Samsung is not the only company looking for AMD's GPU technology. The company announced in March that Google's future game streaming service Stadia will use custom Radeon data center GPUs. Based on AMD's multi-user GPU technology, introduced in 2015, these GPUs include 56 processors (3,584 stream processors) and dedicated HBM2 memory to produce 10.7 teraflops of graphics processing power.

The big misunderstanding during GDC 2019 was that AMD would provide a custom graphic APU like the consoles. That's not the case. AMD clearly states that Google will use its data center-designed GPUs. APUs or CPU cores made by AMD are not mentioned. Presumably, these are manufactured by Intel with 2.7 GHz custom.

Google's data centers may already be filled with Intel-based CPUs. The company is likely to have signed an agreement with AMD to install Radeon data center GPUs (if they have not already). Buying AMD Opteron APU-based systems may not be ideal because multiple virtual machines need to be run and streamed. In addition, AMD's server APUs target small businesses looking for high performance with low power costs.

Nevertheless, Google Stadia is a big win for AMD and Intel. In addition, console games that are already targeting custom APUs do not compromise between features because games run on AMD's GCN architecture. The only really big loser in this scenario is Nvidia.

The State of Intel

Intel really does not need a backstory. The doors opened in 1968 as N M Electronics and changed a month later to Intel – short for Integrated Electronics. The era of the x86-based processor began with the Intel 8086 chip used in IBM's new PC family introduced in 1981. This was followed by the Intel 80286, 80386 and 80486 microprocessors.

Intel began using a tick-tock production model in 2007. "tock" represented a change in the CPU microarchitecture, while the "tick" pushed the revision into a smaller chip layout. For example, Intel used its fourth generation ("Haswell") 22nm microarchitecture in processors introduced in 2013. The fifth generation Intel ("Broadwell") processors arrived the following year based on a 14 nm version of "Haswell."

The Death of Tick Tock

The shift to 14nm process technology has effectively destroyed Intel's tick-to-tail model and introduced a new model called Intel Process Architecture Optimization . Intel developed a new microarchitecture, codenamed Skylake, whose 14nm process node was already in operation. This design served as the basis for the fifth to ninth generation processor families. Intel has officially killed its tick-tock model with the introduction of its seventh-generation Kaby Lake processors.

Kaby Lake relies on the first optimization of 14nm process technology from Intel (called 14nm +) in 2016. Intel has released Kaby Lake for 2017 in the first wave of mobile processors eighth generation using the same process node. This updated design increases energy efficiency and extends the Intel Core i5 CPU family by four cores. The eighth generation started only with Intel's second Skylake optimization (14nm ++), which was referred to as "Coffee Lake".

From then on we saw a third optimization (14nm +++) in 2018 with "Whiskey Lake", a cell phone only successor to Kaby Lake Refresh. We also saw the debut of "Amber Lake," the successor to Kaby Lake, available only on cell phones. Meanwhile, Intel teased a new processor based on 10nm process technology called Cannon Lake. The eighth-generation chip, still based on Skylake, appeared but did not become mainstream. What Cannon Lake has achieved was the restart of Intel's Process Architecture Optimization engine.

Caffeine and icy water

 Intel Ice Lake of the 10th generation

This brings us to Intel's latest processors. Introduced in October 2018, the ninth-generation family is a refresh of Coffee Lake on the 14nm ++ process node. Three desktop CPUs arrived in October, followed by six in January during CES 2019 and another twenty-four in April. This rollout number does not even include mobile, server, and HEDT products.

Intel unveiled the 10th generation of the "Ice Lake" family at Computex 2019, based on a new "Sunny Cove" architecture. This is the architecture section of the Intel Process Architecture Optimization Model. The first eleven chips are targeted at "U" (Ultra-Low Power) and "Y" (Extreme Low Power) mobile devices. You can see up to four cores and eight threads, speeds of up to 4.1 GHz and GPU speeds of up to 1.1 GHz.

Unfortunately, we do not know anything about these chips, unless Intel offers only a few minor things. They have a redesigned integrated GPU architecture (Gen11), which promises smooth frame rates in Battlefield V with 1080p. They also support DDR4 memory at 3,200 MHz. Intel 300 series chipsets offer Wi-Fi 6 connectivity and Intel Optane Memory support.

Ice Lake CPUs and chipsets are now expected to be delivered to OEMS for laptops arriving during the 2019 Christmas season.

AMD vs Intel Showdown

That is, the battle between AMD and Intel sets Ryzen's third-generation "Zen 2" chips against Intel's ninth-generation "Coffee Lake" products. As already mentioned, Ryzen 3000 will be delivered in July, so we have no comparison values.

We can compare a second-generation AMD Ryzen chip with a similar ninth. Generation Intel CPU. We've dug into Geekbench to determine the single and multi-core scores:

Ryzen 7 2700X Core i9-9900K
Cores: 8 8
Threads : [19659042] 16 16
Basic speed (GHz): 3.7 3.6
Maximum speed (GHz) 4.3 5.0 [19659091] Power: 105 watts 95 watts
Single-core score: ~ 4994 ~ 6519
Multi-core score: ~ 29453 ~ 37440
Price: $ 279 $ 484

As the results show, the Ryzen 7 2700X does not outperform Intel's Core i9-9900K part even if it has a slightly higher base speed at a lower cost , This is a big argument in the debate between AMD and Intel: The Intel CPU core simply can do better per cycle instructions. In addition, the AMD chip consumes more power and is not even delivered with integrated graphics. In the end, it may be better to spend the extra $ 205 on the Intel chip.

Let's make another comparison for laptops:

Ryzen 7 2700U Core i7-8559U
Cores: 4 4
Topics: 8 8
Basic speed (GHz): 3.3 2.7
Maximum speed (GHz): 3.8 ] 4.5
Power: 25 Watt 19659039] 28 Watt
Single-Core Score: ~ 4011 ~ 5689
Multi-Core Score: ~ 10929 [19659043] ~ 19133

Here we see that AMD's second generation APU consumes three watts less power. Despite the higher basic speed, the chip falls behind the eighth-generation laptop CPU in the single-core Geekbench test. Also in the multi-core test he falls partly due to its lower maximum speed.

According to Intel's business unit "not all cores are built the same and more cores do not always correspond to better overall performance.

According to Intel, performance also depends on memory and architecture optimizations. This was made clear after AMD compared its new 64-core Epyc second-generation CPU "Rome" with Intel's 28-core Xeon Platinum 8280 CPU "Cascade Lake" for servers. AMD demonstrated in a benchmark that its chip runs 2x faster than the Xeon. According to Intel, AMD did not properly configure the test system and did worse on the Xeon chip.

Navi vs Xe in 2020

 AMD Radeon RX 5700 XT

Another problem AMD is facing is the imminent entry of Intel into the market for add-in graphics cards. Former chief architect of AMD Radeon, Raja Koduri, joined Intel in late 2017 to serve as chief architect and senior vice president of a new core and visual computing group. His first task: to expand a discrete graphics card until 2020. In addition, Radeon cores have been integrated into Intel-based modules with Kaby Lake processor cores and HBM2 video memory.

Intel's new discrete GPUs are based on the scalable "Xe" architecture, providing solutions for data centers, enthusiastic desktops and notebooks, parallel computing and real-time hardware-level ray tracing competing with the recently launched RTX 20 Series. Nvidia's GPU Family The Nvidia GTX 10 Series supports ray tracing via GPGPU acceleration or software only.

This is big news, especially for a processor company that is back in the discrete GPU market A consumer-grade desktop is a big step anyway, promising photorealistic rendering without terrible latency, and is the new addition to gaming powered by Nvidia's RTX 20 family for desktops and laptops.

AMD CEO Lisa Su spoke about hardware- and software-based raytracing in January, but did not mention Raytracing June in their E3 2019 keynote She completed the new "Navi" cards that arrived on July 7, 2019:

Radeon RX 5700 XT 50th Anniversary Edition Radeon RX 5700 XT Radeon RX 5700
Computing Units: [BasicSpeed: 1680 MHz 1.605 MHz 1.465 MHz
Game Speed: 1.830 MHz 1.755 MHz 1.625 40 36
Max. Speed: 1,980 MHz 1,905 MHz 1,725 ​​MHz
Memory (GDDR6) 8 GB 8 GB 8 GB
Performance: [196590421014TFLOPS 9 , 75 TFLOPS 7.95 TFLOPS
Competitive Product: GeForce RTX 2070 GeForce RTX 2070 GeForce RTX 2060
Price: [49659904219659044] $ 379

Since Intel in the discrete GPU area, AMD and Nvidia will not be the only competitors fighting for their dollars. For Intel, it may be difficult to penetrate this market, as AMD and Nvidia have a huge, dedicated customer base. Hardware-level ray tracing seems to be an ace in the gap and a great alternative to Nvidia's RTX 20 series. Unfortunately, the new Radeon RX 5700 series from AMD offers no hardware support.

Who will win this war?

AMD vs. Intel in desktops

In the fight between AMD and Intel for the desktop fights Intel should continue to dominate for the foreseeable future. However, AMD poses a significant threat.

AMD's high core numbers and low prices are an attractive selling point. They require more power and contain no integrated graphics. Customers can purchase an AMD 12-core CPU for $ 499, while Intel currently does not sell a mainstream 12-core chip. In the second half of 2019, the third generation Ryzen 3 chips will be launched by AMD. You may even see new Threadripper HEDT parts.

In the meantime, Intel may introduce new X-Series HEDT processors to keep up with new thread rippers. Considering the fact that Ice Lake CPUs will not arrive until the end of 2019, Computex may be the last we hear from Intel in consumer processors. Until then, the project Athena should cause even before the debut of Ice Lake cause a stir: the ultrabook successor based on Intel's 10-nm CPUs.

AMD versus Intel on laptops

In the feud between AMD and Intel laptops, Intel is likely to stay ahead of the pack

The APUs from AMD were in cheap hands by the time Ryzens arrived in 2017 laptops. Intel still outperforms AMD in laptops, but you'll find great solutions such as Acer Predator Helios 500 and Aspire 3. On the Asus front, the ROG Zephyrus and the VivoBook also offer Ryzen APUs.

Unfortunately, despite the many cores in the desktop area, AMD still does not offer an 8-core mobile chip. Instead, Intel is currently taking the lead with its eight-core i9-9980HK and i9-9880H laptop CPUs. You'll probably find them in gaming laptops paired with a discrete Nvidia GeForce graphics chip. Hell, Intel also sells six-core laptop processors.

AMD vs. Intel – Beyond the PC

Despite the battle with Intel in three major markets, AMD will continue to play the dominant role in the console arena. AMD's strong relationship with console and PC partners brings AMD to the forefront of hardware and development. With PlayStation 5 and Project Scarlet using AMD components, this dominance is unlikely to change for another five years. Nvidia now has Nintendo.

Source link