(Bloomberg) -- Intel Corp. fell after saying a new version of its Xeon server chip line will go into production in 2022, rather than by the end of this year as promised, the latest in a series of delays that have cost the company technology leadership of the chip industry.Rival Advanced Micro Devices Inc. rose after Intel said Tuesday that a chip design, code-named Sapphire Rapids, will begin production in the first quarter of 2022 with the “ramp” beginning the following quarter. The world’s largest chipmaker previously said the new version of its most lucrative product line would start to be manufactured this year.Chief Executive Officer Pat Gelsinger has pledged to return Intel to a dominant position in computer processors and reassert the company’s once unassailable lead in semiconductor manufacturing. Multiyear delays in the introduction of new technology by Intel have ceded leadership to rivals. Gelsinger, who promised the new chip would arrive this year as recently as April, last week replaced the head of Intel’s server chip unit in an executive shake-up.
“Given the breadth of enhancements in Sapphire Rapids, we are incorporating additional validation time prior to the production release, which will streamline the deployment process for our customers and partners,” Intel Vice President Lisa Spelman said in a post on the company’s website. “Based on this, we now expect Sapphire Rapids to be in production in the first quarter of 2022, with ramp beginning in the second quarter of 2022.”
Intel shares declined 1.3% to $56.75 at the close of New York trading while AMD gained 2.8% to $89.52. Intel’s slide limited overall gains by chip stocks. The Philadelphia Stock Exchange Semiconductor Index rose less than 1%.
Given the delays to the 10-nanometer manufacturing process and holdups in bringing its replacement online, the latest announcement will cause concern about further decays in Intel’s competitive positioning, according to Sanford C. Bernstein analyst Stacy Rasgon.“A less charitable explanation, however, is that the company’s competitive positioning is worsening, and that the downside associated with delaying Sapphire Rapids (both for the products and, it should be said, for the stock) is perceived by management to be less that what would be experienced by launching earlier with a less competitive product,” he wrote in a note to clients. “Recent executive changes in the datacenter business may feed into this as well.”
Intel’s data center unit until recently had more than 99% of the market for chips that are the heart of corporate networks and internet infrastructure. Such processors can sell for the price of a compact car and are the company’s most profitable offerings. Regaining a foothold in that market has been a key part of the turn around in AMD’s fortunes under Chief Executive Officer Lisa Su.
The smaller company is projected to report a 50% increase in revenue this year following a 45% surge in 2020. Intel is one of the few major chipmakers that will suffer shrinking sales, according to analysts’ estimates.
(Updates with closing share prices in the fifth paragraph.)
More stories like this are available on bloomberg.com
Subscribe now to stay ahead with the most trusted business news source.
The PHLX Semiconductor Sector Index has generated solid gains so far this year, despite the chip shortage and the sell-off in tech stocks earlier in 2021. However, not all of the index's components have been on fire.
While NVIDIA(NASDAQ:NVDA) stock has roared higher on the back of a strong showing from its gaming and data center businesses, Advanced Micro Devices(NASDAQ:AMD) stock price has failed to sustain its terrific momentum from 2020 and headed south. Intel(NASDAQ:INTC) stock, meanwhile, has pulled back over the past couple of months as its April quarterly report gave rise to a bunch of red flags.
Does this mean NVIDIA is the one you should be buying? Or will AMD's and Intel's fortunes turn around in the second half of the year, making them potentially better buys than the richly valued NVIDIA? Let's find out.
NVIDIA: Benefitting from long-term growth catalysts
It is safe to say that NVIDIA is firing on all cylinders, as almost all the company's business segments registered eye-popping growth in the first quarter of fiscal 2022. Barring the automotive business, which saw a 1% revenue decline, the gaming, data center, professional visualization, and OEM (original equipment manufacturer) segments recorded varying degrees of impressive growth during the quarter.
As a result, NVIDIA's total Q1 revenue jumped 84% year over year to $5.66 billion, while non-GAAP earnings more than doubled to $3.66 per share. The company's guidance was also a solid one, with its estimated second-quarter revenue of $6.3 billion expected to increase nearly 63% year over year at the midpoint of its guidance range.
Image source: Getty Images.
NVIDIA can sustain such high rates of growth beyond 2021 as the data center and gaming markets, which account for nearly 85% of the total revenue, are sitting on long-term catalysts. The data center segment's revenue hit $2 billion for the first time last quarter, recording 79% growth over the year-ago period. NVIDIA credits the segment's outstanding growth to strong demand from hyperscale customers deploying its GPUs (graphics processing units) for artificial intelligence (AI) applications.
NVIDIA says that its data center GPUs are "deployed across all major hyperscale and cloud service providers globally and we see strengthening demand in the coming quarters." That's not surprising, as the demand for data center GPUs could grow at a compound annual growth rate (CAGR) of 42% through 2027, hitting $20.6 billion in revenue. NVIDIA is in a solid position to tap this opportunity, as it supplies its GPUs to the top cloud service providers and has been taking steps to diversify its revenue streams.
Meanwhile, NVIDIA's 80%-plus market share of gaming graphics cards will be another big tailwind for the company. Its video gaming revenue more than doubled last quarter to $2.76 billion. Jon Peddie Research estimates that the market for discrete gaming GPUs could jump from $23.6 billion last year to more than $54 billion by 2025. NVIDIA's dominant market share means that it could win big from the additional revenue opportunity.
Advanced Micro Devices: Widening its gap over Intel
AMD stock's retreat on the market isn't justified, as the chipmaker's impressive financial growth has continued this year after a strong 2020. AMD is on track to record 50% revenue growth in 2021, though it can do better as it was originally anticipating a 37% increase when the year began. The company's market share gains in the client and server CPU (central processing unit) markets, as well as the growing demand for graphics cards, have encouraged it to raise the guidance.
However, AMD can finish the year on a much stronger note, as it recorded 93% year-over-year revenue growth in Q1 to $3.4 billion, which was ahead of the market's expectations. The company is pulling the right strings to ensure that it sustains its high levels of growth, such as consistently winning market share from Intel.
Popular video game distribution platform Steam reports that AMD's share of the client CPU market has now exceeded 30%, a massive improvement from three years ago when it held just over 16% of the market. AMD's gains have been driven by a superior manufacturing process based on a 7-nanometer manufacturing node, while Intel's competing chips are expected to be out later this year.
But AMD seems all set to widen the gap over Chipzilla, as the former's foundry partner Taiwan Semiconductor Manufacturing is working on smaller 5nm and 6nm nodes. Such a move is likely to give AMD's Ryzen CPUs another performance boost and ensure that they continue to sell like hotcakes.
On the other hand, the massive increase in sales of gaming graphics cards, as discussed earlier, will also be a tailwind for AMD. That's because AMD plays second fiddle to NVIDIA in the GPU market, and the additional revenue opportunity over there could substantially add to its revenue in the long run. Throw in other catalysts such as the new console cycle that's giving AMD a significant lift, and it becomes easier to see why the company's fast growth is here to stay.
Intel: Trouble regaining its mojo
Unlike its rivals, Intel is in a rut. The chip giant has been unable to regain its mojo since losing its competitive edge. That was evident from Intel's Q1 results, as non-GAAP revenue of $18.6 billion was flat on a year-over-year basis. Gross margin shrank 6.1 percentage points year over year while operating margin fell 6.7 percentage points. As a result, Intel's net income was down 6% over the year-ago period.
Things are about to get worse for Intel in the second quarter. Revenue and adjusted earnings are expected to drop 10% and 15% year over year, respectively, while full-year non-GAAP revenue is anticipated to drop 7% to $72.5 billion. So, Intel's poor financial performance and outlook put it behind the likes of NVIDIA and AMD, which are delivering irresistible growth. As such, growth-oriented investors may drop Chipzilla from their watchlist of stocks to buy.
The verdict
With Intel out of the race, investors looking to buy a fast-growing semiconductor stock can choose either NVIDIA or AMD. From a valuation standpoint, AMD looks like the better buy as it trades at 36 times trailing earnings, compared to NVIDIA's multiple of nearly 90 times. AMD's price-to-sales ratio of nine is also much lower than NVIDIA's multiple of nearly 25 times.
Given that AMD's pace of growth was better than NVIDIA's in Q1 and is likely to keep up that momentum for the remainder of the year, it looks like a steal right now. AMD also looks like a better buy for those looking for a growth stock at a reasonable valuation. However, investors with a higher appetite for risk may also consider buying NVIDIA, as it can remain a top growth stock and justify its rich valuation by delivering terrific growth thanks to huge end-market opportunities.
This article represents the opinion of the writer, who may disagree with the “official” recommendation position of a Motley Fool premium advisory service. We’re motley! Questioning an investing thesis -- even one of our own -- helps us all think critically about investing and make decisions that help us become smarter, happier, and richer.
The post goes over justification for the system requirements, but also says that Microsoft is open to expanding the lists that cut off at a handful of Qualcomm processors, 8th Gen Intel and AMD's second-gen Ryzen chips.
"As we release to Windows Insiders and partner with our OEMs, we will test to identify devices running on Intel 7th generation and AMD Zen 1 that may meet our principles," the post reads. "We’re committed to sharing updates with you on the results of our testing over time, as well as sharing additional technical blogs."
Microsoft is justifying the new system requirements by stating that Windows 11 is far more secure, including potentially reducing malware by 60% and featuring better encryption and biometric support. A notable security feature is using virtualization-based security, which should pass TPM through to virtualized machines.
Microsoft's three justifications for Windows 11's new requirements.(Image credit: Microsoft)
Additionally, Microsoft is opting for processors that work with the latest Windows drivers and work with the minimum specs for Office and Teams. Part of the security requirement comes from Microsoft requiring TPM 2.0.
The company specifies "fundamentals" of speeds over 1 GHz, dual-core processors, 4GB of memory and 64GB of storage, which still includes a number of PCs that are years old and are not on the current list of supported processors.
It's good to see that Microsoft is considering including more processors, as a number of perfectly capable PCs still use those chips. (Including a number of Microsoft's Surface devices that are still on sale right now!) When Microsoft debuted the list of chips, it angered many enthusiasts who still have perfectly capable PCs, and who suggested that it would create e-waste and force people to get new computers in the middle of a global component shortage.
This testing will come from Windows insider builds, so it's up to the early-adopter community to see how it fares. The first Windows Insider preview for Windows 11 is live now.
Microsoft is temporarily removing the PC Health Check app that let you know if your Windows 10 PC could handle Windows 11.
"Based on the feedback so far, we acknowledge that it was not fully prepared to share the level of detail or accuracy you expected from us on why a Windows 10 PC doesn’t meet upgrade requirements," the post reads. "We are temporarily removing the app so that our teams can address the feedback."
The post promises that tool will be back online before general availability in the fall.
Chip maker Intel Corp. is delaying production of one of its newest chips to improve performance, the first significant product setback under new Chief Executive Pat Gelsinger as he seeks to rebuild the company’s competitiveness.
Intel now is planning to start producing the next generation of central processing units for servers—the brains of those machines—in early 2022 after previously saying it would be ready late this year, Lisa Spelman, the company’s corporate vice president, who manages the server-chip business, said in a Tuesday blog post.
The additional time, Ms. Spelman wrote, would allow Intel to improve the chips’ performance, in particular around the highly prized metrics of data handling and artificial-intelligence processing. Production is now set to begin in next year’s first quarter and ramp up in the second quarter, she wrote.
The delay of the new chips is the first under Mr. Gelsinger, who became chief executive in February following major delays in chip-making advances under his predecessor, Bob Swan. Intel almost a year ago said the following generation of even more advanced chips with super-small transistors wouldn’t be ready until late next year, about a year later than initially expected.
Mr. Gelsinger has vowed to make Intel more reliable in producing new chips. At his first shareholder meeting as the company’s CEO in May, he said Intel was aiming to deliver a “steady cadence of leadership products that our customers can depend upon.”
HP launched its HP Pavilion Aero 13 laptop that will start at $749, feature four colors, have a 90% screen-to-body ratio and weigh in at less than a kilogram.
The HP Pavilion Aero 13 is an all-AMD machine with the latest AMD Ryzen 7 5800U processor with AMD Radeon Graphics. The device has Wi-Fi 6, up to 10.5 hours of battery life and a screen with a 16:10 aspect ratio and 2.5k resolution.
According to HP, Pavilion Aero 13 will be available this month at HP.com and other US retailers in the fall. Key details include:
A full magnesium aluminum chassis with 4-sided narrow bezel.
Colors of Pale Rose Gold, Warm Gold, Ceramic White and Natural Silver.
400 nits of brightness for sunlight viewing.
Manufactured with post-consumer recycled and ocean bound plastics.
Windows 10 with an upgrade available to Windows 11.
Along with the HP Pavilion Aero 13, HP rolled out the M Series monitor line announced at CES 2021. The company's M Series line includes a 24- and 27-inch displays with the HP M24fwa FHD Monitor and HP M27fwa FHD Monitor, respectively. Both monitors feature streamlined design and cable management and integrated dual speakers with availability in July. The 24-inch monitor starts at $229.99, and the 27-inch monitor starts at $289.99.
Stocks rose Tuesday and the S&P 500 set a record intraday high as equities built off the gains that kicked off the week.
TheStreet founder Jim Cramer said that among other things he was keeping an eye on Micron Technology (MU) - Get Report, which is scheduled to report earnings on Wednesday and Nvidia (NVDA) - Get Report.
Micron Technology: 'The Most Important One'
Cramer said that he thinks there is still strength in DRAM, still strength in strength in NAND.
"I think Micron is the most important one because the semiconductors are really hanging by a thread," he said.
TST Recommends
In March, Micron reported fiscal-second-quarter revenue and adjusted earnings that exceeded Wall Street estimates.
Cramer also discussed Nvidia (NVDA) - Get Report, noting that three of the world's largest chipmakers went against other big tech companies and blessed the graphics-chip specialist's proposed acquisition of U.K.-based Arm Holdings.
Nvidia: 'I'd Like to See Some News Here'
Broadcom (AVGO) - Get Report, Marvell (MRVL) - Get Report and Taiwan's MediaTek have become the first Arm customers to publicly support the $40 billion proposed takeover.
Cramer said "this is very, very important."
"I'd like to see some news here," he said.
Cramer also said he'd like to see Advanced Micro Devices (AMD) - Get Report close the deal with Xilinx (XLNX) - Get Report.
Earlier this month, Dealmaker reported that the European Commission hasn’t raised objections to the semiconductor titan’s proposed $35 billion takeover of Xilinx.
Advanced Micro Devices, Nvidia, Broadcom and Marvell are holdings in Cramer's Action Alerts PLUS charitable portfolio.
Want to be alerted before Jim Cramer buys or sells the stocks?Learn more now.
When Microsoft released the documentation for Windows 11, it was clear that this is going to be the first time in over a decade that the minimum requirements for Windows have been significantly raised. The minimum RAM and storage have been increased to 4GB and 64GB, respectively. And for CPUs, there’s no more 32-bit support, nor is there support for single-core chips.
But moreover, Microsoft drew a hard line on CPU generations that are supported. For Intel, you need eighth-gen or newer, and for AMD, you need Ryzen 3000 or newer. There was a bit of outrage over this. Fortunately, Microsoft announced today that it might just open things up to Intel’s seventh-generation chips and AMD Zen 1.
In a blog post explaining the new system requirements, it said that Windows 11 is meant to be a set of features, so it needs to be installed on chips that can support those features. That includes eighth-gen and newer Intel, AMD Zen 2, and Qualcomm 7 and 8 series. It also said that Intel sixth-gen and older, and AMD pre-Zen simply can’t meet the criteria for Windows 11. However, Intel’s seventh-generation chips and AMD Zen 1 might be OK. It’s going to test them out with Insiders.
Specifically, the blog post said, “As we release to Windows Insiders and partner with our OEMs, we will test to identify devices running on Intel 7th generation and AMD Zen 1 that may meet our principles. We’re committed to sharing updates with you on the results of our testing over time, as well as sharing additional technical blogs.”
While TPM 2.0 is a requirement for Windows 11, that’s not the only thing that the new chips are required for. There are other security features to be considered, such as virtualization-based security (VBS), hypervisor-protected code integrity (HVCI), and Secure Boot. Plus, the CPU needs to have adopted the new Windows Driver model.
This explains why the minimum requirements for Windows 11 when it ships are different from the requirements for testing it out in the Windows Insider Program. You’ll be able to enroll a PC with a seventh-gen Intel CPU, such as the very newest Microsoft Surface Studio 2, but you might not be able to run Windows 11 when it comes out.
One other thing that Microsoft noted is that it’s pulling the PC Health app, which was designed to help you know if your PC is compatible with Windows 11. It’s going to work on making it even better, and then it will re-release the app. Meanwhile, third-party tools already do a better job at telling you why your PC can’t run Windows 11, though those will need to be updated in light of today’s clarification on the minimum requirements.
Update: Microsoft has already removed the part of its blog post that said that Intel sixth-gen and earlier chips, and AMD pre-Zen processors definitely don’t meet its criteria.
Advanced Micro Devices (NASDAQ:AMD) stock has been on a tear since last year, with a slew of positive developments, strong revenue growth, and a robust outlook.
Source: Joseph GTK / Shutterstock.com
Despite its stellar performance, AMC stock has moved sluggishly in the past few months in comparison to its peers.
The delay in the approval of its Xilinx deal in China and Europe seems to be holding its stock price back.
Hence AMC stock is trading at a fairly reasonable valuation and should break out once it overcomes its regulatory hurdles.
AMD is coming off one of its strongest years in recent memory, with its year-over-year revenue growth at over 57%.
Moreover, its year-over-year EBITDA growth is at an incredible 118%. It enjoys a technical edge over its competitors, such as Intel (NASDAQ:INTC), who have been forced to play catch-up for most of the year.
AMD currently trades at roughly seven times forward sales, which is relatively expensive. However, given its business momentum, robust financials and technical superiority, its valuation is more than justified at this point.
Glittering Financials
AMD recently reported its first-quarter results, where its revenues rose to $3.5 billion, representing a 93% increase from the prior-year period. Consequently, GAAP net income rose shot up to $555 million, representing a 243% growth from year-ago levels.
More importantly, net income dropped 41% during the period, mainly due to a restructuring charge of $2.2 billion. Additionally, its revenues rose 8% last year compared with 2019.
The technical edge has mainly driven AMD’s success in CPUs. A few years ago, many had written off the company, but under the leadership of Lisa Su, its fortunes have changed completely.
A series of game-changing decisions have driven share gains against Intel in the CPU department.
Furthermore, AMD has made some great strides as far as free cash flows are concerned. It generated an incredible $832 million in the quarter compared to Intel’s $1.6 billion.
However, considering that Intel generates six times as much revenue as AMD, its performance is lackluster.
AMD/Xilinx Merger
AMD agreed to acquire semiconductor company Xilinx in a $35 billion all-stock transaction late last year. The aim was for both companies to complement each other’s product portfolios, create synergies, and expand AMD’s revenue base.
In addition, the acquisition will give AMD’s data center business a massive boost. Both companies have similar competencies in areas such as artificial intelligence and systems-on-a-chip.
The merger provides access to new markets in the industrial and automotive realms for AMD. Xilinx’s data center business will effectively complement AMD’s robust server capabilities and will enable it to leverage its EPYC technology.
However, the merger had run into regulatory roadblocks in Europe and China. The primary concern of these markets relates to anti-competitive behavior, with the company becoming remarkably powerful.
As it stands, the EU’s top antitrust regulator states it has no objections to the merger. The UK’s antitrust regulator is evaluating the merger proposal and will decide by next month.
However, the approval process in China will take significantly longer than in Europe. The average review period for such deals in the industry is more than 300 days.
Nonetheless, the merger’s approval in the EU and UK will be a major catalyst in itself and should push AMD stock’s price higher.
Bottom Line on AMD Stock
AMD is coming off one of its most successful years, but its stock price momentum has been sluggish overall.
AMD’s merger with Xilinx has stalled in Europe and China, which has slowed down the growth in its stock price. However, these roadblocks are likely to fade away soon, removing any impediments for AMD stock.
AMD stock is a definite buy at this stage, with an incredible outlook ahead and its stellar financial positioning.
On the date of publication, Muslim Farooque did not have (either directly or indirectly) any positions in the securities mentioned in this article. The opinions expressed in this article are those of the writer, subject to the InvestorPlace.com Publishing Guidelines.
If you've been holding on to your Odyssey card from Intel, now's the time to use it. The chipmaker has announced via its Twitter account that cardholders can now register their cards to receive some swag.
The microsite asks you to fill out some basic information to get in on the freebies. More importantly, Intel confirmed that its Xe-HPG DG2 announcement is imminent. The chipmaker wrote, "We are soon heading toward a milestone moment, the pending release of the Xe HPG microarchitecture from Intel."
Intel stated earlier today at International Supercomputing (ISC) 2021 that the company has already started sampling its DG2 lineup of discrete gaming graphics cards. Although Intel didn't put a date next to the DG2, it's clear that DG2 is nearing the finish line.
Do you still have your special event #JoinTheOdyssey cards? Now’s the time to use it! Fill in the details on the form here, and some swag may soon be coming your way… https://t.co/BuKBRk6aqb #XeHPG pic.twitter.com/CSLOQOjW3WJune 28, 2021
See more
A couple of days ago, a couple of DG2 benchmarks emerged on Geekbench 5 courtesy of Insyde, a big-name company specializing in UEFI development. The sole benchmark found the DG2 performing similarly to Nvidia's five-year-old GeForce GTX 1050. It should be noted that early benchmarks can be misleading, so we should wait until the retail product is on the market before passing judgment.
The DG2 will likely come in both desktop and mobile formats. Intel's DG1 was a budget graphics card and geared primarily towards OEMs. DG2, on the other hand, may feature a more elaborate design. On the mobile end, DG2 is rumored to accompany Intel's 12th Generation Alder Lake chips. There is some truth in the rumors, as we've already seen Insyde's test platform consisting of the DG2 and a 14-core Alder Lake-P processor.
AMD may be about to launch two new Navi 23 graphics cards, according to a new leak. Drivers for the RX 6600 and RX 6600XT have been found in a recent update, suggesting that AMD is likely preparing to release these two GPUs soon.The information comes from a leaker known on Twitter as @KOMACHI_ENSAKA, who posted screenshots of AMD’s recent drivers, including both of the new RX 6000-series cards.
It’s possible that these are the graphics cards AMD was talking about earlier this year at CES 2021, an annual trade show organized by the Consumer Technology Association. During that event, AMD announced that it would be releasing a single-fan GPU in the future. It now seems that the cards in question may in fact have been the Radeon RX 6600 and 6600XT.
We don’t know the full specifications of the upcoming cards just yet, although some information has already been revealed. Part of that knowledge stems from the fact that both of these cards are housing a Navi 23 GPU. Navi 23 comes with a 128-bit memory bus and 2,048 stream processors. The GPU measures 237mm2 and has 11.06 billion transistors. It also utilizes 10 teraflops (TFLOPs) and runs on up to 100 watts of power.
Previous rumors suggest that the AMD Radeon RX 6600XT graphics card will provide 8GB of GDDR6 memory. Assuming this turns out to be true, the 6600 might come with 4GB VRAM. However, there’s a chance that even the RX 6600 will be decked out with the full 8GB. Both models are likely to come with a 32MB AMD Infinity Cache.
It’s possible that the RX 6600 will come with 28 compute units, while the RX 6600XT will have 32. Both the cards will have a 7nm process node. Most of the other specs are pure guesswork at this point. Rumors suggest a 128-bit memory bus for the RX 6600 and a 192-bit for the RX 6600XT, as well as a 16Gbps memory clock with 256GB/s of bandwidth for both models.
AMD
These graphics cards are likely to be the first Navi 23 GPUs for AMD when it comes to the gaming sector. The company already has Navi 23 cards out on the market, but they are found inside mobile devices and high-end workstations. They can also be found inside Tesla’s Model S vehicles. According to Elon Musk, the CEO of Tesla, the performance of that system will be powerful enough to rival that of Sony’s Playstation 5 consoles.
Like any leak, this should be taken with skepticism, but it does come from a reputable source with a proven track record. Another leak has also recently confirmed that Gigabyte is about to release six custom GPUs from the RX 6600 XT series, further signaling that the cards are almost ready. Given that, as well as the hints from AMD, we are looking at a possible release shortly. All the information put together suggests a simultaneous launch within the next few months.
As part of today’s International Supercomputing 2021 (ISC) announcements, Intel is showcasing that it will be launching a version of its upcoming Sapphire Rapids (SPR) Xeon Scalable processor with high-bandwidth memory (HBM). This version of SPR-HBM will come later in 2022, after the main launch of Sapphire Rapids, and Intel has stated that it will be part of its general availability offering to all, rather than a vendor-specific implementation.
Hitting a Memory Bandwidth Limit
As core counts have increased in the server processor space, the designers of these processors have to ensure that there is enough data for the cores to enable peak performance. This means developing large fast caches per core so enough data is close by at high speed, there are high bandwidth interconnects inside the processor to shuttle data around, and there is enough main memory bandwidth from data stores located off the processor.
Our Ice Lake Xeon Review system with 32 DDR4-3200 Slots
Here at AnandTech, we have been asking processor vendors about this last point, about main memory, for a while. There is only so much bandwidth that can be achieved by continually adding DDR4 (and soon to be DDR5) memory channels. Current eight-channel DDR4-3200 memory designs, for example, have a theoretical maximum of 204.8 gigabytes per second, which pales in comparison to GPUs which quote 1000 gigabytes per second or more. GPUs are able to achieve higher bandwidths because they use GDDR, soldered onto the board, which allows for tighter tolerances at the expense of a modular design. Very few main processors for servers have ever had main memory be integrated at such a level.
Intel Xeon Phi 'KNL' with 8 MCDRAM Pads in 2015
One of the processors that used to be built with integrated memory was Intel’s Xeon Phi, a product discontinued a couple of years ago. The basis of the Xeon Phi design was lots of vector compute, controlled by up to 72 basic cores, but paired with 8-16 GB of on-board ‘MCDRAM’, connected via 4-8 on-board chiplets in the package. This allowed for 400 gigabytes per second of cache or addressable memory, paired with 384 GB of main memory at 102 gigabytes per second. However, since Xeon Phi was discontinued, no main server processor (at least for x86) announced to the public has had this sort of configuration.
New Sapphire Rapids with High-Bandwidth Memory
Until next year, that is. Intel’s new Sapphire Rapids Xeon Scalable with High-Bandwidth Memory (SPR-HBM) will be coming to market. Rather than hide it away for use with one particular hyperscaler, Intel has stated to AnandTech that they are committed to making HBM-enabled Sapphire Rapids available to all enterprise customers and server vendors as well. These versions will come out after the main Sapphire Rapids launch, and entertain some interesting configurations. We understand that this means SPR-HBM will be available in a socketed configuration.
Intel states that SPR-HBM can be used with standard DDR5, offering an additional tier in memory caching. The HBM can be addressed directly or left as an automatic cache we understand, which would be very similar to how Intel's Xeon Phi processors could access their high bandwidth memory.
Alternatively, SPR-HBM can work without any DDR5 at all. This reduces the physical footprint of the processor, allowing for a denser design in compute-dense servers that do not rely much on memory capacity (these customers were already asking for quad-channel design optimizations anyway).
The amount of memory was not disclosed, nor the bandwidth or the technology. At the very least, we expect the equivalent of up to 8-Hi stacks of HBM2e, up to 16GB each, with 1-4 stacks onboard leading to 64 GB of HBM. At a theoretical top speed of 460 GB/s per stack, this would mean 1840 GB/s of bandwidth, although we can imagine something more akin to 1 TB/s for yield and power which would still give a sizeable uplift. Depending on demand, Intel may fill out different versions of the memory into different processor options.
One of the key elements to consider here is that on-package memory will have an associated power cost within the package. So for every watt that the HBM requires inside the package, that is one less watt for computational performance on the CPU cores. That being said, server processors often do not push the boundaries on peak frequencies, instead opting for a more efficient power/frequency point and scaling the cores. However HBM in this regard is a tradeoff - if HBM were to take 10-20W per stack, four stacks would easily eat into the power budget for the processor (and that power budget has to be managed with additional controllers and power delivery, adding complexity and cost).
One thing that was confusing about Intel’s presentation, and I asked about this but my question was ignored during the virtual briefing, is that Intel keeps putting out different package images of Sapphire Rapids. In the briefing deck for this announcement, there was already two variants. The one above (which actually looks like an elongated Xe-HP package that someone put a logo on) and this one (which is more square and has different notches):
There have been some unconfirmed leaks online showcasing SPR in a third different package, making it all confusing.
Sapphire Rapids: What We Know
Intel has been teasing Sapphire Rapids for almost two years as the successor to its Ice Lake Xeon Scalable family of processors. Built on 10nm Enhanced SuperFin, SPR will be Intel’s first processors to use DDR5 memory, have PCIe 5 connectivity, and support CXL 1.1 for next-generation connections. Also on memory, Intel has stated that Sapphire Rapids will support Crow Pass, the next generation of Intel Optane memory.
For core technology, Intel (re)confirmed that Sapphire Rapids will be using Golden Cove cores as part of its design. Golden Cove will be central to Intel's Alder Lake consumer processor later this year, however Intel was quick to point out that Sapphire Rapids will offer a ‘server-optimized’ configuration of the core. Intel has done this in the past with both its Skylake Xeon and Ice Lake Xeon processors wherein the server variant often has a different L2/L3 cache structure than the consumer processors, as well as a different interconnect (ring vs mesh, mesh on servers).
Sapphire Rapids will be the core processor at the heart of the Aurora supercomputer at Argonne National Labs, where two SPR processors will be paired with six Intel Ponte Vecchio accelerators, which will also be new to the market. As part of this announcement today, Intel also stated that Ponte Vecchio will be widely available, in OAM and 4x dense form factors:
Sapphire Rapids will also be the first Intel processors to support Advanced Matrix Extensions (AMX), which we understand to help accelerate matrix heavy workflows such as machine learning alongside also having BFloat16 support. This will be paired with updates to Intel’s DL Boost software and OneAPI support. As Intel processors are still very popular for machine learning, especially training, Intel wants to capitalize on any future growth in this market with Sapphire Rapids. SPR will also be updated with Intel’s latest hardware based security.
It is highly anticipated that Sapphire Rapids will also be Intel’s first multi compute-die Xeon where the silicon is designed to be integrated (we’re not counting Cascade Lake-AP Hybrids), and there are unconfirmed leaks to suggest this is the case, however nothing that Intel has yet verified.
The Aurora supercomputer is expected to be delivered by the end of 2021, and is anticipated to be the first official deployment of Sapphire Rapids. We expect a full launch of the platform sometime in the first half of 2022, with general availability soon after. The exact launch of SPR-HBM is unknown, however given those time frames, Q4 2022 seems fairly reasonable depending on how aggressive Intel wants to attack the launch in light of any competition from other x86 vendors or Arm vendors.
In a blog post today, Microsoft is making another attempt at clarifying the minimum requirements for Windows 11, which have confused many, especially regarding what generation of CPUs the new OS will support. The post is clear that Intel 6th Gen Skylake and earlier CPUs, along with non-Zen AMD processors, will not meet “principles around security and reliability and minimum system requirements for Windows 11.” That lines up with the company’s original statement that Windows 11 would require Intel 8th Gen Coffee Lake or Zen 2 CPUs and up, but there is some hope for Intel 7th Gen Kaby Lake and Zen 1 users.
Where things get confusing again is the news that its first Insider build of Windows 11, which is available today, won’t require TPM 2.0 or specific CPUs. Microsoft will use data from testers to inform potential adjustments to the minimum requirements. Microsoft will be paying extra attention to users with Intel 7th Gen CPUs and AMD Zen 1 CPUs to see if they’re getting the performance that Microsoft is looking for when it comes to reliability and security.
This news seems to confirm people’s fears that Windows 11 won’t support older CPUs, some of which can still keep up with the newer models on a pure performance basis. In the blog post, Microsoft claims that the reason Windows 11 won’t run on those CPUs is to maintain a higher level of security, stability, and reliability.
Microsoft will also be “temporarily removing” the PC Health Check App, which was designed to tell people running Windows 10 whether they’d be able to update their computers to Windows 11. The company says the app “was not fully prepared to share the level of detail or accuracy” that users expected, but it said the app would be available again for Windows 11’s launch in the fall (likely October).
Sapphire Rapids with built-in HBM raises performance bar; Intel’s GPU, networking and storage capabilities enhance HPC toolbox.
NEWS HIGHLIGHTS
Latest 3rd Gen Intel® Xeon® Scalable processors will power the next generation of supercomputers and high-performance computing systems.
The next generation of Intel Xeon Scalable processors (code-named "Sapphire Rapids") will integrate High Bandwidth Memory (HBM).
Intel’s Xe-HPC-based Ponte Vecchio GPU has powered-on, is in system validation, and will include OAM form factor and subsystems.
Intel announces High Performance Networking (HPN) with Ethernet, extending Intel Ethernet into HPC applications.
Intel introduces commercial support for DAOS (distributed application object storage).
SANTA CLARA, Calif., June 28, 2021--(BUSINESS WIRE)--At the 2021 International Supercomputing Conference (ISC) Intel is showcasing how the company is extending its lead in high performance computing (HPC) with a range of technology disclosures, partnerships and customer adoptions. Intel® processors are the most widely deployed compute architecture in the world’s supercomputers, enabling global medical discoveries and scientific breakthroughs. Intel is announcing advances in its Xeon processor for HPC and AI as well as innovations in memory, software, exascale-class storage, and networking technologies for a range of HPC use cases.
"To maximize HPC performance we must leverage all the computer resources and technology advancements available to us," said Trish Damkroger, vice president and general manager of High Performance Computing at Intel. "Intel is the driving force behind the industry’s move toward exascale computing, and the advancements we’re delivering with our CPUs, XPUs, oneAPI Toolkits, exascale-class DAOS storage, and high-speed networking are pushing us closer toward that realization."
Advancing HPC Performance Leadership
Earlier this year, Intel extended its leadership position in HPC with the launch of 3rd Gen Intel Xeon Scalable processors. The latest processor delivers up to 53% higher performance across a range of HPC workloads, including life sciences, financial services and manufacturing, as compared to the previous generation processor.
Compared to its closest x86 competitor, the 3rd Gen Intel Xeon Scalable processor delivers better performance across a range of popular HPC workloads. For example, when comparing a Xeon Scalable 8358 processor to an AMD EPYC 7543 processor, NAMD performs 62% better, LAMMPS performs 57% better, RELION performs 68% better, and Binomial Options performs 37% better. In addition, Monte Carlo simulations run more than two times faster, allowing financial firms to achieve pricing results in half the time. Xeon Scalable 8380 processors also outperform AMD EPYC 7763 processors on key AI workloads, with 50% better performance across 20 common benchmarks. HPC labs, supercomputing centers, universities and original equipment manufacturers who have adopted Intel’s latest compute platform include Dell Technologies, HPE, Korea Meteorological Administration, Lenovo, Max Planck Computing and Data Facility, Oracle, Osaka University and the University of Tokyo.
Integration of High Bandwidth Memory within Next-Gen Intel Xeon Scalable Processors
Workloads such as modeling and simulation (e.g., computational fluid dynamics, climate and weather forecasting, quantum chromodynamics), artificial intelligence (e.g., deep learning training and inferencing), analytics (e.g., big data analytics), in-memory databases, storage and others power humanity’s scientific breakthroughs. The next-generation of Intel Xeon Scalable processors (code-named "Sapphire Rapids) will offer integrated High Bandwidth Memory (HBM), providing a dramatic boost in memory bandwidth and a significant performance improvement for HPC applications that operate memory bandwidth-sensitive workloads. Users can power through workloads using just High Bandwidth Memory or in combination with DDR5.
Customer momentum is strong for Sapphire Rapids processors with integrated HBM, with early leading wins such as the U.S. Department of Energy’s Aurora supercomputer at Argonne National Laboratory and the Crossroads supercomputer at Los Alamos National Laboratory.
"Achieving results at exascale requires the rapid access and processing of massive amounts of data," said Rick Stevens, associate laboratory director of Computing, Environment and Life Sciences at Argonne National Laboratory. "Integrating high-bandwidth memory into Intel Xeon Scalable processors will significantly boost Aurora’s memory bandwidth and enable us to leverage the power of artificial intelligence and data analytics to perform advanced simulations and 3D modeling."
Charlie Nakhleh, associate laboratory director for Weapons Physics at Los Alamos National Laboratory, said: "The Crossroads supercomputer at Los Alamos National Labs is designed to advance the study of complex physical systems for science and national security. Intel’s next-generation Xeon processor Sapphire Rapids, coupled with High Bandwidth Memory, will significantly improve the performance of memory-intensive workloads in our Crossroads system. The [Sapphire Rapids with HBM] product accelerates the largest complex physics and engineering calculations, enabling us to complete major research and development responsibilities in global security, energy technologies and economic competitiveness."
The Sapphire Rapids-based platform will provide unique capabilities to accelerate HPC, including increased I/O bandwidth with PCI express 5.0 (compared to PCI express 4.0) and Compute Express Link (CXL) 1.1 support, enabling advanced use cases across compute, networking and storage.
In addition to memory and I/O advancements, Sapphire Rapids is optimized for HPC and artificial intelligence (AI) workloads, with a new built-in AI acceleration engine called Intel® Advanced Matrix Extensions (AMX). Intel AMX is designed to deliver significant performance increase for deep learning inference and training. Customers already working with Sapphire Rapids include CINECA, Leibniz Supercomputing Centre (LRZ) and Argonne National Lab, as well as the Crossroads system teams at Los Alamos National Lab and Sandia National Lab.
Intel Xe-HPC GPU (Ponte Vecchio) Powered On
Earlier this year, Intel powered on its Xe-HPC-based GPU (code-named "Ponte Vecchio") and is in the process of system validation. Ponte Vecchio is an Xe architecture-based GPU optimized for HPC and AI workloads. It will leverage Intel’s Foveros 3D packaging technology to integrate multiple IPs in-package, including HBM memory and other intellectual property. The GPU is architected with compute, memory, and fabric to meet the evolving needs of the world’s most advanced supercomputers, like Aurora. Ponte Vecchio will be available in an OCP Accelerator Module (OAM) form factor and subsystems, serving the scale-up and scale-out capabilities required for HPC applications.
Extending Intel Ethernet For HPC
At ISC 2021, Intel is also announcing its new High Performance Networking with Ethernet (HPN) solution, which extends Ethernet technology capabilities for smaller clusters in the HPC segment by using standard Intel Ethernet 800 Series Network Adapters and Controllers, switches based on Intel® Tofino™ P4-programmable Ethernet switch ASICs and the Intel® Ethernet Fabric suite software. HPN enables application performance comparable to InfiniBand at a lower cost while taking advantage of the ease of use offered by Ethernet.
Commercial Support for DAOS
Intel is introducing commercial support for DAOS (distributed application object storage), an open-source software-defined object store built to optimize data exchange across Intel HPC architectures. DAOS is at the foundation of the Intel® Exascale storage stack, previously announced by Argonne National Laboratory, and is being used by Intel customers such as LRZ and JINR (Joint Institute for Nuclear Research).
DAOS support is now available to partners as an L3 support offering, which enables partners to provide a complete turnkey storage solution by combining it with their services. In addition to Intel’s own data center building blocks, early partners for this new commercial support includes HPE, Lenovo, Supermicro, Brightskies, Croit, Nettrix, Quanta, and RSC Group.
More information about Intel’s participation at ISC 2021, including a full list of talks and demos, can be found at https://hpcevents.intel.com.
About Intel
Intel (Nasdaq: INTC) is an industry leader, creating world-changing technology that enables global progress and enriches lives. Inspired by Moore’s Law, we continuously work to advance the design and manufacturing of semiconductors to help address our customers’ greatest challenges. By embedding intelligence in the cloud, network, edge and every kind of computing device, we unleash the potential of data to transform business and society for the better. To learn more about Intel’s innovations, go to newsroom.intel.com and intel.com.
Inside a largely HPC and server-focused announcement today, Intel slyly dropped word of its upcoming gaming graphics card, DG2. According to the presentation slides (PDF warning), DG2 is already sampling out to Intel's partners now, and that's a good sign of things to come for Intel's venture into discrete pixel pushers.
Intel DG2 is based on the Intel Xe architecture that we've already had arrive in Tiger Lake chips and the low-power DG1 chip. However, Intel's discrete gaming aspirations have been largely waiting on DG2's arrival, which uses the gaming-specific Intel Xe-HPG arch.
It looks as if that could happen in good time, too. Intel has promised Xe discrete cards this year, and the announcement of DG2 sampling suggests its well on its way to doing just that.
Sampling will mean early versions of Intel's reference DG2 GPUs are in the hands of Intel's partners, the people who will finally turn that Xe HPG silicon into the graphics cards we buy. That doesn't just mean distribution is going swimmingly, it also likely means Intel's drivers are in a fairly decent shape to offer partners some idea of actual Xe DG2 performance.
And that all speaks well to Intel's gaming focused graphics card actually arriving on the shelves sometime soon to salve the ongoing GPU shortage woes. If it's going to have a chance of being a genuine third way against Nvidia and AMD then Intel needs to launch DG2 before GeForce and Radeon cards are in plentiful supply again.
SANTA CLARA, Calif., June 28, 2021 — At this year’s International Supercomputing 2021 digital event, AMD (NASDAQ: AMD) is showcasing momentum for its AMD EPYC processors and AMD Instinct accelerators across the High Performance Computing (HPC) industry. The company also outlined updates to the ROCm open software platform and introduced the AMD Instinct Education and Research (AIER) initiative. The latest Top500 list showcased the continued growth of AMD EPYC processors for HPC systems. AMD EPYC processors power nearly 5x more systems compared to the June 2020 list, and more than double the number of systems compared to November 2020. As well, AMD EPYC processors power half of the 58 new entries on the June 2021 list.
“High performance computing is critical to addressing the world’s biggest and most important challenges,” said Forrest Norrod, senior vice president and general manager, data center and embedded systems group, AMD. “With our AMD EPYC processor family and Instinct accelerators, AMD continues to be the partner of choice for HPC. We are committed to enabling the performance and capabilities needed to advance scientific discoveries, break the exascale barrier, and continue driving innovation.”
AMD HPC Momentum Continues
With the recent launch of the AMD EPYC 7003 Series processor, which provides industry leading performance for HPC workloads1, AMD is continuing to enable its partners and customers to deploy all sizes of clusters, across key research areas including manufacturing, life sciences, financial services, climate research and more.
A 2020 Intersect360 perception study of HPC users’ impressions of CPUs showed that AMD EPYC processors had a 78 percent favorable impression amongst respondents, growing from 36 percent in 20162. In a 2021 study from Intersect360 asking HPC institutions about AMD EPYC penetration within their sites, 23 percent of respondents said they have broad usage of AMD EPYC processors, and an additional 47 percent said they are testing or using AMD EPYC processors at some level3.
Recent systems utilizing AMD HPC solutions include:
Durham University’s COSMA8supercomputer with Dell EMC PowerEdge C6525 servers powered by 2nd and 3rd Gen AMD EPYC processors.
Microsoft Azure supercomputers for UK Met Office using 3rd Gen AMD EPYC processors and next generation AMD EPYC processors in an HPE Cray EX supercomputer, delivering an expected 60 petaflops of advanced supercomputing capabilities for weather and climate research.
The Perlmutter supercomputer from the National Energy Research Scientific Computing Center (NERSC) and Lawrence Berkeley National Laboratory features nodes that include 3rd Gen AMD EPYC processors in an HPE Cray EX supercomputer that provides four times the computational power currently available at NERSC.
AMD is introducing its new AMD Instinct Education and Research (AIER) initiative, designed to help scientists, researchers and academics accelerate the performance of their code on AMD Instinct Accelerators. The AIER initiative, based on member requirements, offers remote access to AMD Instinct technologies, the AMD ROCm Learning Center, and ROCm software and support as well as access to technical guidance on AMD software and hardware solutions. In addition to regional Solution Partners, AIER Global Solution Partners include Dell Technologies, Gigabyte, Hewlett Packard Enterprise, and Supermicro.
The ROCm open software platform continues to gain industry support and momentum with a growing list of applications, third-party libraries, and frameworks supporting AMD accelerators. The HPC community has embraced HIP as a heterogenous programming model, that developers use to write or adapt their codes to for acceleration on AMD’s GPUs including Gromacs, TensorFlow, and GridTools.
Additionally, PyTorch for ROCm is now available as an installable Python package and includes full capability for mixed-precision and large-scale training using AMD’s MIOpen and RCCL (communications) libraries. This innovation provides a new option for data scientists, researchers, students, and others in the community to get started with accelerated PyTorch using AMD GPUs. Most recently, CuPy, an open-source array library with Python, has expanded its traditional GPU support with the introduction of version 9.0 that now offers support for the ROCm stack for GPU-accelerated computing.
Commitment to Advancing Research
Last year, AMD announced its HPC Fund for COVID-19 research, which included a donation of systems powered by AMD EPYC processors and AMD Instinct accelerators to provide researchers with petaflop-scale compute power to fight the pandemic. With over 12 petaflops of capacity awarded to-date, AMD has since delivered HPC capabilities to 23 institutions across seven countries including Massachusetts Institute of Technology (MIT), New York University (NYU), Rice University, University of Texas at Austin and University of Toronto.
“From COVID-19 research to analyzing genome architectures, AMD CPU and GPU processors help the Center for Theoretical Biological Physics to run some codes from 1,000 to 10,000 times faster than before,” said Peter Rossky, the Harry C and Olga K. Wiess Professor of Natural Sciences. “The ability to run parallel simulations on this game-changing equipment enables the full analysis of one chromosome in as little as 20 minutes – research that used to take up to a month to complete. These accomplishments come just months after adding the high-performance computing capabilities from AMD, and we expect adoption across our researcher community to grow exponentially.”
For more than 50 years AMD has driven innovation in high-performance computing, graphics and visualization technologies ― the building blocks for gaming, immersive platforms and the datacenter. Hundreds of millions of consumers, leading Fortune 500 businesses and cutting-edge scientific research facilities around the world rely on AMD technology daily to improve how they live, work and play. AMD employees around the world are focused on building great products that push the boundaries of what is possible. For more information about how AMD is enabling today and inspiring tomorrow, visit the AMD (NASDAQ: AMD) website.
AMD silicon lies at the heart of the Xbox Series X, Xbox Series S, and PlayStation 5, which means the red team is hard at work to produce as much RDNA 2 and Zen 2 silicon as it can to meet demand. In doing so, it is likely amassing quite the collection of broken chips, those with a few (or a few million) transistors off the norm. The chips that don't quite make the cut may end up in this desktop kit before you: the AMD 4700S.
The AMD 4700S is nearly the complete package for a PC (via Videocardz). Under the included cooler lies an AMD Zen 2-powered CPU, an eight-core chip, which puts it in line with the latest consoles and some Ryzen CPUs. Alongside that is the tell-tale sign of this chip's console origin: it comes with either 8GB or 16GB of GDDR6 memory.
Modern consoles use shared memory pools, namely GDDR6 today, which isn't something you'd find on any gaming PC. While GDDR6 offers plenty of bandwidth for GPUs, DDR4 is much better suited to the task of general system memory thanks to low latencies. So there's no real reason for AMD to produce a desktop kit with GDDR6 memory unless it already had the chips lying around.
From what I can tell eyeing up images from Disclosuzen on Twitter (via TechPowerUp), the memory must be located on the underside of the motherboard, beneath the extended rear heatsink. That doesn't give the memory the optimal cooling solution, although this machine is hardly at the forefront of PC performance anyways.
With the memory likely located on the underside of the PCB, this looks the part of the PS5 SoC. Twitter leaker rogame also spots matching traces between the 4700S and the PS5, suggesting this chip may have once been destined for Sony's console. Perhaps AMD and Sony's chipmaking agreement has some stipulation in it whereby AMD can use or purchase chips Sony has no use for, but we can't say for sure.
I'm fairly sure it's a PS5 APU. Just check the PCB traces for the memory. https://t.co/QvZhX4wBMc pic.twitter.com/fS7iQfYfTGJune 28, 2021
See more
Note the lack of Ryzen branding here too—this isn't technically a Ryzen product despite sharing the Zen 2 architecture with AMD's Ryzen 3000 chips. AMD's 4700S page does feature the painted circle logo I had assumed to be the Ryzen logo, but perhaps it's representative of the Zen architecture instead. Either way, AMD is clearly trying to avoid the Ryzen brand here.
To build out the entire kit, the CPU and memory are paired with a fairly basic motherboard. It features two SATA ports, a 1Gbps LAN port, eight USB ports, basic audio out from the I/O, and a couple of headers for extra USB connections. There's no M.2 NVMe port, however, and only one PCIe 2.0 x4 slot for a single GPU.
You'll have to provide one of those yourself, though. This kit doesn't come with any form of GPU. If this were a chip ripped right out of a console, it would feature a working RDNA 2 GPU with a beefy compute unit count—56, 36, or 20, depending on the console—yet here that's been entirely disabled, meaning a graphics card of some description is required.
It's a shame really, an APU with console graphics would be immense.
A PS5 GPU up close. Photo by Fritzchens Fritz.(Image credit: Fritzchens Fritz)
There are only a few graphics cards listed with official compatibility with the 4700S: AMD's RX 500-series and Nvidia's GT 710, 730, and GTX 1050, 1050 Ti, and 1060. Further compatibility is not specified, but it's likely the board's low PCIe bandwidth is to blame for the GPU restrictions. You can read up on the system's specs within its installation and warranty booklet (PDF warning).
Not that you'd want to build a gaming PC around this board anyways. We suspect its performance wouldn't be quite up to scratch with the latest gaming machines.
In graphics cards, wafers that don't quite make the grade are generally why we end up with cards like the GeForce RTX 3060 Ti, built with a slightly less capable version of the GA104 GPU which would otherwise go into an GeForce RTX 3070. Microsoft will employ the same tactic with chips that don't making the cut for the Xbox Series X, instead finding a home in the Xbox Series S. But if you ever wondered what happens to all the disused chips and non-functional parts with no clear use in console-land, the AMD 4700S should give you some idea of where they end up.
