AMD – Earlybirds Invest https://earlybirdsinvest.com Latest Crypto News Tue, 15 Jul 2025 13:35:13 +0000 en-US hourly 1 https://wordpress.org/?v=6.9.8 https://i0.wp.com/earlybirdsinvest.com/wp-content/uploads/2024/12/cropped-New-Project-2024-12-17T235703.455.png?fit=32%2C32&ssl=1 AMD – Earlybirds Invest https://earlybirdsinvest.com 32 32 240146708 Allnodes Among First to Launch Bare Metal Servers powered by AMD Threadripper 9000 Series https://earlybirdsinvest.com/allnodes-among-first-to-launch-bare-metal-servers-powered-by-amd-threadripper-9000-series/ https://earlybirdsinvest.com/allnodes-among-first-to-launch-bare-metal-servers-powered-by-amd-threadripper-9000-series/#respond Tue, 15 Jul 2025 13:35:12 +0000 https://earlybirdsinvest.com/allnodes-among-first-to-launch-bare-metal-servers-powered-by-amd-threadripper-9000-series/

July 15th, 2025 – Los Angeles, USA


Allnodes, a leading platform for blockchain infrastructure, is among the first in the world to offer pre-orders for hosting on new AMD Ryzen Threadripper PRO 9000 Series processors as part of its bare-metal server lineup.

The newest hardware upgrade enables Allnodes to deliver even more powerful and efficient infrastructure to its users, meeting the highest performance demands of any blockchain.

“At Allnodes, we work hard to bring our users the latest and most powerful infrastructure available,” said Konstantin Boyko-Romanovsky, Founder and CEO at Allnodes. “The launch of the 9000 Series provides a natural next step, offering significantly improved architecture efficiency, critical for handling the increasing computational standards across today’s blockchain networks.”

The new AMD Ryzen Threadripper PRO 9000 Series processors, available through Allnodes’ bare-metal hosting solutions, include the 9965WX and 9975WX models, featuring up to 32 cores, 64 threads, 5.4 GHz Max Boost, and up to 1024GB of memory.

With up to 22% higher multi-core performance compared to the previous generation of Threadripper series, it is an ideal upgrade for customers who prioritize maximum single-core performance with more than 256GB of memory.

About Allnodes

Allnodes delivers high-performance bare metal servers tailored for Web3 infrastructure, along with secure, non-custodial solutions for node hosting and staking. With over $3.1 billion in hosted node value, support for 119 blockchains, and more than 30,000 active nodes, Allnodes is trusted by individuals and institutions worldwide. The platform also provides reliable RPC endpoints for developers and teams building in the decentralized ecosystem.

https://www.allnodes.com/hosting

Contact

Support
press@allnodes.com

This content is sponsored and should be regarded as promotional material. Opinions and statements expressed herein are those of the author and do not reflect the opinions of The Daily Hodl. The Daily Hodl is not a subsidiary of or owned by any ICOs, blockchain startups or companies that advertise on our platform. Investors should do their due diligence before making any high-risk investments in any ICOs, blockchain startups or cryptocurrencies. Please be advised that your investments are at your own risk, and any losses you may incur are your responsibility.

Follow Us on X Facebook Telegram

Check out the Latest Industry Announcements
 

 

]]>
https://earlybirdsinvest.com/allnodes-among-first-to-launch-bare-metal-servers-powered-by-amd-threadripper-9000-series/feed/ 0 47778
Did Nvidia Just Say "Checkmate" to AMD? https://earlybirdsinvest.com/did-nvidia-just-say-checkmate-to-amd/ https://earlybirdsinvest.com/did-nvidia-just-say-checkmate-to-amd/#respond Fri, 27 Jun 2025 22:03:29 +0000 https://earlybirdsinvest.com/did-nvidia-just-say-checkmate-to-amd/ Nvidia’s market share gains in the artificial intelligence (AI) accelerator market continue to trounce the competition.

Over the last two years, semiconductor powerhouse Nvidia (NVDA 1.74%) has emerged as the biggest force fueling the artificial intelligence (AI) revolution. The company’s industry-leading graphics processing units (GPU) and CUDA software platform have helped Nvidia build a substantial lead over its competition in the chip market.

While Advanced Micro Devices has carved out an impressive pocket for itself in the AI data center landscape, recent reporting suggests that the company is still far behind Nvidia.

Let’s explore the dynamics behind Nvidia’s lead over AMD, and assess if the king of the chip realm just made its checkmate move against its top rival.

Nvidia continues to dominate the competition

During the early phases of the AI megatrend, Nvidia benefited from having a first-mover advantage over other semiconductor companies when it comes to GPUs specifically. While being a first mover can help companies experience outsized growth relative to the competition or form strategic partnerships with leaders in adjacent industries, there’s no guarantee that these businesses can sustain their leads.

In the case of Nvidia, however, trends suggest that company’s lead over AMD may only be getting bigger.

Beth Kindig, who serves as a technology research analyst and CEO of I/O Fund, recently shared a data point from SemiAnalysis that pointed out that Nvidia’s market share in AI accelerators increased by roughly two points during the first quarter — now hovering around 88%. By contrast, AMD’s share shrunk by about one point, now comprising roughly 4% of the market.

AI data center chip powering a GPU cluster.

Image source: Getty Images.

Why these trends could spell trouble for AMD

The chart illustrates AMD’s revenue and operating income by reportable segment during the first quarter. One of the more notable takeaways is that AMD’s data center operation is its fastest-growing business, all while remaining highly profitable.

AMD revenue by segment.

Data source: AMD investor relations.

However, a more subtle idea is that sales from the data center business shrunk by 5% quarter over quarter. To be fair, there could be a number of reasons for this.

First, the semiconductor industry is cyclical — which makes quarterly trends tougher to predict and gauge when it comes to the overall health of the business. In addition, AMD’s latest accelerator architectures are expected to ship later this year. This timeline could be playing a role in the slight deceleration of the data center business compared to the fourth quarter.

While these ideas may make some sense in theory, I find them hard to believe, given Nvidia actually gained ground in the AI accelerator market during the first quarter.

Since AI developers raced to buy Nvidia’s newest Blackwell chips, factors such as cyclicality or new competitive chipsets from AMD didn’t seem to be enough to persuade customers from waiting on AMD’s products over those of Nvidia.

In the long run, these dynamics could spell trouble for AMD. Despite the company’s ability to win impressive data center customers such as Oracle, Meta Platforms, and Microsoft, AMD’s innovative efforts do not appear to be enough to outmaneuver Nvidia at this time.

Is this a checkmate move by Nvidia?

As of this writing (June 25), shares of AMD have risen by 18% so far in 2025. Per the chart, these gains are slightly ahead of Nvidia stock’s increase.

NVDA Chart

NVDA data by YCharts

I think it’s hard to justify AMD’s gains over Nvidia, given the company’s lack of market share and apparent decelerating growth (at least for now).

While I suspect AMD’s growth profile could turn around following the release of new chipsets during the second half of the year, I also think it will be challenging for the company to gain any meaningful momentum back from Nvidia — as Nvidia also has new architectures releasing later this year, too.

To me, Nvidia may have put AMD in a checkmate position for the time being. I think these figures reported could imply that Nvidia will remain the leader in AI data center chips and could be on its way to a new, prolonged, and sustained wave of growth.

Randi Zuckerberg, a former director of market development and spokeswoman for Facebook and sister to Meta Platforms CEO Mark Zuckerberg, is a member of The Motley Fool’s board of directors. Adam Spatacco has positions in Meta Platforms, Microsoft, and Nvidia. The Motley Fool has positions in and recommends Advanced Micro Devices, Meta Platforms, Microsoft, Nvidia, and Oracle. The Motley Fool recommends the following options: long January 2026 $395 calls on Microsoft and short January 2026 $405 calls on Microsoft. The Motley Fool has a disclosure policy.

]]>
https://earlybirdsinvest.com/did-nvidia-just-say-checkmate-to-amd/feed/ 0 44498
AMD Ryzen 5 9600X3D leak hints at mid-range push with 3D V-Cache https://earlybirdsinvest.com/amd-ryzen-5-9600x3d-leak-hints-at-mid-range-push-with-3d-v-cache/ https://earlybirdsinvest.com/amd-ryzen-5-9600x3d-leak-hints-at-mid-range-push-with-3d-v-cache/#respond Mon, 23 Jun 2025 03:01:31 +0000 https://earlybirdsinvest.com/amd-ryzen-5-9600x3d-leak-hints-at-mid-range-push-with-3d-v-cache/

Something to look forward to: AMD is preparing to expand its mid-range CPU lineup with the Ryzen 5 9600X3D, a chip that would bring 3D V-Cache tech to a broader audience of gamers and PC builders. The existence of this new chip wasn’t revealed through a formal announcement, but rather via a support list for AMD’s recently launched Radeon AI R9700 graphics card, where the processor appeared alongside other unreleased models.

This leak has fueled speculation that AMD is looking to strengthen its position in the competitive mid-range segment, where it faces stiff competition from Intel’s Core i5-14600K and Core Ultra 5 245K. Just earlier this week, we reported that Intel’s Core Ultra 7 265KF had hit a record-low price of $229, or about 40% off its launch price less than a year ago.

The Ryzen 5 9600X3D is expected to feature 6 cores and 12 threads, built on AMD’s Zen 5 microarchitecture and “Granite Ridge” silicon. The standout feature is the inclusion of second-gen 3D V-Cache in a mainstream CPU, stacking an additional 64 MB of cache on top of the standard 32 MB on-die cache, resulting in a total of 96 MB of L3 cache.

Clock speeds for the new processor are anticipated to closely match those of the standard Ryzen 5 9600X, with boost frequencies likely reaching up to 5.4 GHz. The chip is expected to maintain a 65-watt TDP, striking a balance between performance and energy efficiency that should appeal to users seeking a capable yet cost-effective upgrade. The 9600X3D will also support DDR5-5600 memory and fit into AMD’s AM5 socket, ensuring compatibility with the latest motherboards and platforms.

With its six-core configuration, the Ryzen 5 9600X3D is positioned as a more affordable alternative to higher-end X3D models, such as the Ryzen 7 9800X3D, which offer more cores but come at a significantly higher price.

This strategy mirrors AMD’s previous approach with the limited-release Ryzen 5 7600X3D. However, the growing trend of modern games recommending or requiring 8-core CPUs raises questions about the long-term viability of 6-core chips for future gaming titles.

Still, AMD’s V-Cache technology provides a unique advantage in cache-sensitive scenarios, potentially offsetting the lower core count in many current games.

Industry observers expect the Ryzen 5 9600X3D to debut first in pre-built systems from major integrators, with retail availability for DIY builders anticipated later in the year, possibly during the holiday season.

Pricing details remain unconfirmed, but the chip is widely expected to undercut the cost of higher-tier X3D models, making it an attractive entry point for those seeking next-generation gaming performance on a budget.

]]>
https://earlybirdsinvest.com/amd-ryzen-5-9600x3d-leak-hints-at-mid-range-push-with-3d-v-cache/feed/ 0 43576
Intel Slashes Arrow Lake Prices Amid Tough Competition From AMD https://earlybirdsinvest.com/intel-slashes-arrow-lake-prices-amid-tough-competition-from-amd/ https://earlybirdsinvest.com/intel-slashes-arrow-lake-prices-amid-tough-competition-from-amd/#respond Thu, 08 May 2025 11:07:43 +0000 https://earlybirdsinvest.com/intel-slashes-arrow-lake-prices-amid-tough-competition-from-amd/

Intel (INTC 2.16%) launched its Arrow Lake family of desktop central processing units (CPUs), officially the Core Ultra 200 series, in late 2024. The company outsourced most of the manufacturing to TSMC, moved to a chiplet-based architecture, and managed to improve energy efficiency substantially, compared to its previous-generation chips. For productivity tasks, Arrow Lake performed well.

However, there were two problems. First, Arrow Lake’s gaming performance fell flat, losing to Intel’s last-gen chips. Some software fixes have improved the situation, but for those looking to squeeze every last frame per second out of their gaming PC, Arrow Lake isn’t the answer.

Second, pricing was on the high side. Not only were Arrow Lake CPUs expensive on their own, but they also required a new motherboard, making all upgrade paths pricier.

Arrow Lake CPUs have been selling below Intel’s original suggested retail pricing for a while, but the company is now officially slashing prices on one of its chips. The 265K, a mid-range part that was originally priced at $399, now has a suggested retail price of $299. Suggested retail prices for the higher end 285K and the lower end 245K are staying put, although actual retail prices vary.

While Intel’s price cutting is limited, the 265K is in the sweet spot for many potential customers, with most of the performance of the 285K for much less money. Featuring eight performance cores and 12 efficiency cores, the 265K is a solid all-arounder that falls a bit short in gaming and makes a lot more sense at $299 than it did at $399.

A CPU being placed onto a motherboard.

Image source: Getty Images.

A sign of things to come?

Intel replaced its CEO in March with Lip-Bu Tan, a veteran of the semiconductor industry and a critic of Intel’s sluggish pace and bureaucratic nature. Tan’s strategy revolves around cutting a bloated cost structure, putting out better products faster, and listening to customers. More aggressive pricing could be part of the equation as Intel looks to make Arrow Lake more competitive.

AMD‘s latest Ryzen 9000 series CPUs didn’t get great reviews, and retail pricing quickly dropped to reflect muted demand. But the Ryzen 9000 series looks a lot better relative to Arrow Lake, especially with Arrow Lake’s sky-high initial pricing.

AMD’s gaming-centric X3D variants, which feature ultra-fast cache memory capable of boosting gaming performance, are the undisputed kings of gaming CPUs. AMD noted in its first-quarter report that there was strong demand for its newest Ryzen chips, likely at the expense of Intel.

Intel also cut the prices of its Granite Rapids server CPUs in January, although there was no official announcement. The Granite Rapids family is the best set of server CPUs that Intel has put out in years, and it largely caught up to AMD in terms of core counts, performance, and efficiency. But like Arrow Lake, pricing skewed high.

There was a time not long ago, before AMD’s comeback, when Intel faced no real competition in either the PC or server CPU markets. The company’s dominance afforded it the ability to price its products high and generate impressive profit margins. With AMD now highly competitive and stealing market share, that era is over. With price cuts for Granite Rapids and Arrow Lake, Intel appears to finally recognize that it needs to compete on price to win back market share.

Intel has a long road ahead as it attempts to turn itself around. If recent price cuts are any indication, more aggressive pricing will likely be part of Tan’s strategy as he looks to stop the bleeding and regain some lost market share from AMD.

Timothy Green has positions in Intel. The Motley Fool has positions in and recommends Advanced Micro Devices, Intel, and Taiwan Semiconductor Manufacturing. The Motley Fool recommends the following options: short May 2025 $30 calls on Intel. The Motley Fool has a disclosure policy.

]]>
https://earlybirdsinvest.com/intel-slashes-arrow-lake-prices-amid-tough-competition-from-amd/feed/ 0 35054
1 Artificial Intelligence (AI) Semiconductor Stock to Buy on the Dip Hand Over Fist Right Now (Hint: It's Not Nvidia or AMD) https://earlybirdsinvest.com/1-artificial-intelligence-ai-semiconductor-stock-to-buy-on-the-dip-hand-over-fist-right-now-hint-its-not-nvidia-or-amd/ https://earlybirdsinvest.com/1-artificial-intelligence-ai-semiconductor-stock-to-buy-on-the-dip-hand-over-fist-right-now-hint-its-not-nvidia-or-amd/#respond Sat, 08 Mar 2025 00:28:45 +0000 https://earlybirdsinvest.com/1-artificial-intelligence-ai-semiconductor-stock-to-buy-on-the-dip-hand-over-fist-right-now-hint-its-not-nvidia-or-amd/ Chip stocks are sliding, and one critical player looks too cheap to ignore.

Chipsets known as graphics processing units (GPUs) are perhaps the most important hardware in generative AI development right now. For the last couple of years, investing in semiconductor stocks has generally been a great idea — as you’re nearly guaranteed some form of exposure to GPUs or data centers.

However, 2025 hasn’t gotten off to the best start for chip stocks.

Whether it was drama brought on by Chinese start-up DeepSeek, U.S. President Donald Trump’s new tariffs, or lofty investor expectations, many names in the chip realm haven’t fared so well this year. From a macro perspective, the VanEck Semiconductor ETF has dropped 4% so far in 2025 (as of March 3). When it comes to specific companies, take Nvidia and Advanced Micro Devices, which have seen their stocks decline by 7% and 17%, respectively, so far this year.

While many investors can’t seem to look away from Nvidia or AMD, there’s another stock that’s been caught up in broader selling in the semiconductor landscape — and I think it’s worth buying the dip right now.

Let’s explore why now looks like a lucrative opportunity to buy Taiwan Semiconductor Manufacturing (TSM 0.71%) stock hand over fist.

Don’t underestimate Taiwan Semi’s influence in the chip realm

When it comes to brand recognition in the chip market, investors don’t need to look much further than Nvidia and AMD. These two juggernauts lead the charge in the GPU revolution. Meanwhile, Broadcom plays an integral role in outfitting data centers with advanced chipware, while Micron Technology‘s high bandwidth memory storage solutions are increasingly important as AI data workloads get bigger and more complex.

With so many other names dominating headlines and talking points, I wouldn’t be surprised if you aren’t even aware of Taiwan Semi, or TSMC. The thing is that many leaders in the chip space — including Nvidia, AMD, and Broadcom — should credit Taiwan Semi for much of their success.

TSMC specializes in foundry solutions, which is basically a fancy term that means it actually manufactures chips and integrated systems for semiconductor companies. In other words, without TSMC, Nvidia’s chip architecture would be more of an idea than a tangible product.

Given how much demand there’s been for GPUs over the last couple of years, it shouldn’t come as a surprise that Taiwan Semi’s revenue and profits are soaring. With that said, I think the company’s growth is just beginning to kick into gear.

Many of the “Magnificent Seven” companies, such as Microsoft, Amazon, Alphabet, and Meta Platforms, are exploring custom silicon as a strategy to migrate from an overreliance on Nvidia’s chipware. These big tech giants, as well as ChatGPT maker OpenAI, are reportedly collaborating with TSMC to help bring their visions to life.

TSM Revenue Estimates for Current Fiscal Year Chart

TSM Revenue Estimates for Current Fiscal Year data by YCharts.

Although TSMC has already acquired nearly two-thirds of the foundry market opportunity, I think the advent of more custom silicon — in addition to new architectures from Nvidia and AMD over the next couple of years — will further strengthen the company’s leadership position and lead to a prolonged phase of revenue and profit acceleration.

Robot arm manufacturing chips in a factory.

Image source: Getty Images.

TSMC shares are priced to perfection

Despite TSMC’s strong market position and robust financial outlook, shares of the chip stock are shockingly cheap.

TSM PE Ratio (Forward) Chart

TSM PE Ratio (Forward) data by YCharts.

Right now, the average forward price-to-earnings (P/E) multiple for the S&P 500 is about 21. As the chart above illustrates, Taiwan Semi’s forward P/E is roughly 19. To me, this disparity suggests that investors may see an investment in the S&P 500 as less risky than TSMC — and one that potentially carries more upside, too.

In my eyes, the two main risks revolving around an investment in TSMC are the following:

  1. The semiconductor industry being cyclical.
  2. Geopolitical tensions between China and Taiwan.

While I can understand those points in an academic sense, I think any fears around those topics are overblown. Chip demand isn’t expected to slow down anytime soon, as the market is forecast to increase tenfold over the next decade and reach a size of nearly $1 trillion.

On top of that, TSMC’s operations are not exclusive to Taiwan. In fact, the company just announced in early March that it will be investing an additional $100 billion to expand its manufacturing footprint in the U.S. This seems like a logical decision given big tech is planning to spend more than $300 billion in AI infrastructure in 2025 alone.

I think TSMC stock is a bargain right now. Long-term investors may want to consider buying this stock hand over fist, before the company’s manufacturing operation witnesses even further scale as the AI revolution continues to move full steam ahead.

John Mackey, former CEO of Whole Foods Market, an Amazon subsidiary, is a member of The Motley Fool’s board of directors. Randi Zuckerberg, a former director of market development and spokeswoman for Facebook and sister to Meta Platforms CEO Mark Zuckerberg, is a member of The Motley Fool’s board of directors. Suzanne Frey, an executive at Alphabet, is a member of The Motley Fool’s board of directors. Adam Spatacco has positions in Alphabet, Amazon, Meta Platforms, Microsoft, and Nvidia. The Motley Fool has positions in and recommends Advanced Micro Devices, Alphabet, Amazon, Meta Platforms, Microsoft, Nvidia, and Taiwan Semiconductor Manufacturing. The Motley Fool recommends Broadcom and recommends the following options: long January 2026 $395 calls on Microsoft and short January 2026 $405 calls on Microsoft. The Motley Fool has a disclosure policy.

]]>
https://earlybirdsinvest.com/1-artificial-intelligence-ai-semiconductor-stock-to-buy-on-the-dip-hand-over-fist-right-now-hint-its-not-nvidia-or-amd/feed/ 0 23886
Should You Forget AMD and Buy 2 Tech Stocks Instead? https://earlybirdsinvest.com/should-you-forget-amd-and-buy-2-tech-stocks-instead/ https://earlybirdsinvest.com/should-you-forget-amd-and-buy-2-tech-stocks-instead/#respond Thu, 13 Feb 2025 00:43:46 +0000 https://earlybirdsinvest.com/should-you-forget-amd-and-buy-2-tech-stocks-instead/ Broadcom and TSMC might be more balanced chipmaking investments.

AMD was once a hot chipmaker, but it lost nearly 40% of its value over the past 12 months. It lost its momentum as its sluggish sales of gaming chips partly offset its stronger sales of PC and data center chips, and investors started to question the long-term growth potential of its closely watched artificial intelligence (AI) accelerators.

AMD’s data center revenue still surged 69% year over year in the fourth quarter of 2024, but that marked a significant slowdown from its 122% growth in the third quarter and 80% growth in the second quarter. That deceleration dampened the bullish hopes that AMD would loosen Nvidia‘s iron grip on the AI market with its cheaper AI accelerators. In the PC market, AMD’s share of the discrete GPU market also shrank against Nvidia’s.

An illustration of an AI chip.

Image source: Getty Images.

For 2025, analysts still expect AMD’s revenue and adjusted earnings per share (EPS) to grow 24% and 43%, respectively. Those seem like robust growth rates for a stock that trades at just 23 times forward earnings. However, those estimates might drift lower if its data center business continues to cool off and it falls further behind Nvidia in the gaming market.

AMD isn’t doomed yet, but investors might consider investing in more balanced and diversified chipmakers that don’t directly compete against Nvidia. Two of those stocks are Broadcom (AVGO 0.56%) and Taiwan Semiconductor Manufacturing (TSM -1.13%).

1. Broadcom

Broadcom operates two main businesses: its chipmaking division, which sells a wide range of chips for the mobile, data center, networking, wireless, and storage markets, and its infrastructure software division, which provides enterprise software, security services, and cloud-based services. It significantly expanded both businesses through some big acquisitions (including Vmware in 2023) over the past decade.

Broadcom’s chip and software sales are cyclical, but the growth of the AI market is driving more data centers to ramp up their purchases of its networking chips and custom XPU accelerators. In fiscal 2024 (which ended last November), Broadcom’s sales of AI-oriented chips surged 220% to $12.2 billion and accounted for 41% of its semiconductor revenue.

The rapid expansion of that business, along with the stronger growth of its non-AI chip and software businesses in a warmer macro environment, should drive Broadcom’s revenue and profits higher over the next few years.

From fiscal 2024 to fiscal 2027, analysts expect Broadcom’s revenue to grow at a compound annual growth rate (CAGR) of 17% as its EPS rises at a CAGR of 75%. Its stock might seem pricier than AMD’s at 43 times next year’s earnings, but its broader diversification and more balanced growth justify that higher valuation.

2.Taiwan Semiconductor Manufacturing

Taiwan Semiconductor Manufacturing, also known as TSMC, is the world’s largest and most technologically advanced contract chipmaker. All of the world’s leading fabless chipmakers, including AMD and Nvidia, outsource the production of their smallest, densest, and most power-efficient chips to TSMC’s fabs.

Nvidia’s soaring sales of AI GPUs have been driving the growth of TSMC’s high-performance computing (HPC) market. In 2024, its HPC revenue surged 58% and accounted for 51% of its top line. Its smartphone chip revenue, which accounted for another 35% of its top line, also rose 23% for the year as new handset sales warmed up again.

This year, TSMC aims to widen its lead against its two closest competitors, Intel (NASDAQ: INTC) and Samsung, by ramping up its production of its smallest 2 nm chips. It also continues to expand its overseas plants in the U.S., Germany, and Japan to offset the geopolitical risks for its most advanced foundries in Taiwan.

From 2024 to 2026, analysts expect TSMC’s revenue and EPS to grow at a CAGR of 23% and 26%, respectively, as its core markets expand again. Its stock still looks cheap at 19 times forward earnings, but that’s probably because its valuations are being squeezed by some near-term concerns regarding the tighter export curbs for AI chips, higher tariffs, and the escalating tensions between Taiwan and China. If those headwinds wane, it should command a higher valuation and rally even higher.

Leo Sun has no position in any of the stocks mentioned. The Motley Fool has positions in and recommends Advanced Micro Devices, Intel, Nvidia, and Taiwan Semiconductor Manufacturing. The Motley Fool recommends Broadcom and recommends the following options: short February 2025 $27 calls on Intel. The Motley Fool has a disclosure policy.

]]>
https://earlybirdsinvest.com/should-you-forget-amd-and-buy-2-tech-stocks-instead/feed/ 0 19108
Guide to AMD Ryzen AM5 Motherboard Chipsets https://earlybirdsinvest.com/guide-to-amd-ryzen-am5-motherboard-chipsets/ https://earlybirdsinvest.com/guide-to-amd-ryzen-am5-motherboard-chipsets/#respond Thu, 06 Feb 2025 06:44:01 +0000 https://earlybirdsinvest.com/guide-to-amd-ryzen-am5-motherboard-chipsets/

When selecting a motherboard for your newly purchased Ryzen CPU, you will encounter many options: X870, X670, B650, A620, and the newest additions, B850 and B840. But what do these names mean? How do X870E, X670E and B650E boards differ from their non-E counterparts? Which should you choose for your system? This guide will provide all the answers and more.

The biggest difference between motherboard types is the chipset – the chip (or two chips) that connects the CPU to other components in the system. In the past, the chipset consisted of a northbridge chip, which connected the CPU to the RAM and graphics card, and a southbridge chip, which connected the northbridge to the rest of the components. Nowadays, however, the functionality formerly attributed to the northbridge is integrated into the CPU, which connects directly to many components.

Chipset Processor PCIe lanes Graphics card NVMe SSD & additional GPP lanes Total PCIe lanes Max usable PCIe 5 lanes USB 5 Gbps USB 10 Gbps USB 20 Gbps USB4 Max PCIe 3.0 or SATA ports
X870E PCIe 5.0
(1×16 slot or
2×8 slots)
PCIe 5.0 1 x4 PCIe 5.0
4x PCIe GPP
44 24 2 12 2 Yes 8
X870 PCIe 5.0
(1×16 slot or
2×8 slots)
PCIe 5.0 1 x4 PCIe 5.0
4x PCIe GPP
36 24 1 6 1 Yes 4
B850 PCIe 4.0
(1×16 slot or
2×8 slots)
PCIe 4.0 1 x4 PCIe 5.0 36 4 1 6 1 opt 4
B840 PCIe 4.0
(1×16 slot)
PCIe 4.0 1 x4 PCIe 4.0 34 0 2 2 0 opt 4
X670E PCIe 5.0
(1×16 slot or
2×8 slots)
PCIe 5.0 1 x4 PCIe 5.0
4x PCIe GPP
44 24 2 12 2 opt 8
X670 PCIe 4.0
(1×16 slot or
2×8 slots)
PCIe 4.0 1 x4 PCIe 5.0
4x PCIe GPP
44 8 2 12 2 opt 8
B650E PCIe 5.0
(1×16 slot or
2×8 slots)
PCIe 5.0 1 x4 PCIe 5.0
4x PCIe GPP
36 24 1 6 1 opt 4
B650 PCIe 4.0
(1×16 slot or
2×8 slots)
PCIe 4.0 1 x4 PCIe 4.0
[PCIe 5.0 opt]
36 0 1 6 1 opt 4
A620 / A620A PCIe 4.0
(1×16 slot)
PCIe 4.0 1 x4 PCIe 4.0 32 0 2 2 0 opt 4

Ryzen 7000 and 9000 processors connect to high-bandwidth components through 28 PCIe lanes and support PCIe 5.0 speeds. Of these lanes, 16 are reserved for a graphics card (or two, with each utilizing eight lanes), and four are designated for an M.2 drive. Another four lanes are used to connect to the chipset. The remaining four lanes can be employed for elements such as another M.2 slot or a USB4 adapter that also connects to the CPU’s integrated graphics to enable DisplayPort tunneling.

Ryzen 8000G processors only have eight PCIe lanes for a graphics card (20 in total), and support PCIe 4.0 speeds. Additionally, Ryzen CPUs support four USB 3.1 (or “3.2 Gen 2”) ports and a single USB 2.0 connector.

Take a shortcut to your next Ryzen build…

The current chipsets on the AM5 platform support either PCIe 4.0 speeds, which limits the total bandwidth of the devices connected to them at approximately 7GB/s, or PCIe 3.0 speeds, which sets the bandwidth at about 3.5GB/s. These devices may comprise storage drives utilizing either the NVMe or SATA interface, USB devices, network cards, and other internal components such as soundcards.

Some motherboards feature more connectors than their CPU and chipset support, either to allow you to choose between some of them or because they employ splitters to share their bandwidth. With the basics covered, let’s move on to the chipsets themselves…

Old Is New Again: B840

AMD B840 motherboards use the same Promontory 19 chip that was previously used on B550 motherboards with AM4 processors. Unlike the Promontory 21 used in other AM5 motherboards, it only supports PCIe 3.0 speeds, but it adds 10 PCIe lanes, and supports four SATA, two USB 3.1, two USB 3.0 (or “3.2 Gen 1”), and six USB 2.0 ports.

In addition to the slower chipset, B840 motherboards omit several enthusiast-oriented features. They don’t support PCIe 5.0 speeds, even for devices connected directly to the CPU. They don’t allow CPU overclocking, although with today’s CPUs it’s more of an enthusiast hobby than a recommended practice. They also only support a single graphics card, but the era of multi-GPU gaming has essentially come to an end so that’s not a big one.

Only a few B840 motherboards are available at the time of writing, and quite surprisingly, they aren’t cheap. Despite their limitations, they are quality boards with several heatsinks and many slots. That raises the question: whom are they meant for?

If you want to upgrade an old system with PCIe 3.0 devices, then B840 motherboards may offer a good value, as they are cheaper than otherwise comparable B850 boards. Otherwise, feel free to ignore them.

A Needless Compromise: A620

A620 motherboards use the same Promontory 21 chip used in B650, B850 and X870 motherboards, but with some of its connectors disabled or limited. Like the Promontory 19, it adds support for two USB 3.1, two USB 3.0 (or “3.2 Gen 1”), and six USB 2.0 ports. It also incorporates eight PCIe 3.0 lanes, of which two or four can be employed for SATA ports.

The real issue with A620 motherboards isn’t the chipset itself; rather, it’s that they are designed to be affordable boards with inexpensive voltage regulator modules (VRM) and like B840 boards, do not support PCIe 5.0 speeds. Worse, A620 boards not only disallow CPU overclocking, but they may also officially support CPUs with a TDP of up to only 120W, or even 65W.

On paper, A620 motherboards are a good fit for Ryzen 8000G processors, which are meant for compact systems with fewer components and lower power consumption. In practice, decent A620 boards cost about the same as B650 ones, so we can only recommend them when they are on sale.

The Best Value: X870 vs. B650E vs. B850 vs. B650

B650, B650E, B850 and X870 motherboards all use a single, fully enabled Promontory 21 chip. It adds eight PCIe 4.0 lanes, allowing an M.2 drive to fully utilize the chipset’s bandwidth on its own. Like in A620 boards, it also includes four PCIe 3.0 lanes that can be used for up to four SATA ports.

The chipset provides support for six USB 2.0 and four USB 3.1 ports, as well as bandwidth that can be allocated for an additional two USB 3.1 ports or a single USB 3.2 (“Gen 2 x 2”).

Read our review: Most Affordable AMD B650 Boards Tested

If you were wondering about the difference between B650 and B650E motherboards, the answer is that B650E boards (E = short for “Extreme”) support PCIe 5.0 for both the graphics card and at least one M.2 drive. X870 motherboards are basically the same as B650E, except they are required to use four PCIe lanes for USB4. However, because they are newer, the Wi-Fi cards they come with support Wi-Fi 7 rather than 6E.

While B850 motherboards are required to support PCIe 5.0 on at least one M.2 drive, regular B650 boards may still have a PCIe 5.0 M.2 slot. Other than that, these boards don’t limit the CPU in any way and will support any Ryzen processor until 2027.

We recommend B650, B850, B650E and X870 motherboards for gamers and others who want to get everything that AM5 motherboards have to offer, but don’t need to connect many drives to their PC simultaneously. Another advantage of these boards is their ability to fit all connectors into a compact micro-ATX form factor, which is useful if you’re seeking a compact case to place on your desk.

More is More: X870E vs. X670E vs. X670

X670, X670E and X870E motherboards feature a Promontory 21 chip connected to another one through four PCIe 4.0 lanes, providing the second chip with the same bandwidth as the one connected to the CPU.

Combined, the chipset adds twelve PCIe 4.0 lanes, allowing each M.2 drive to saturate the chipset’s bandwidth individually, and eight PCIe 3.0 lanes for up to eight SATA ports. With SATA now becoming a legacy connector, it’s common for some of those lanes to be allocated to network cards, freeing up all PCIe 4.0 lanes for faster interfaces.

The chipset also offers support for twelve USB 2.0 ports and one of three options: twelve USB 3.1 ports; ten USB 3.1 ports and one USB 3.2 port, or eight USB 3.1 ports and two USB 3.2 ports.

Read our full reviews: The Best AMD X670E Motherboards: 22 Board Roundup & AMD X870/X870E Motherboard Roundup: 21 Motherboards Tested

Unlike B650, all X670 motherboards feature a PCIe 5.0 M.2 slot, and of course, X670E boards support PCIe 5.0 graphics cards. If you purchase such a card when they become available, it probably won’t even require all 16 lanes. So, if your board has another 16-lane PCIe 5.0 slot, you’ll be able to use it for up to two M.2 drives with an adapter (some boards actually come with one). With X870E you can be sure you’ll get USB4 and Wi-Fi 7.

X670, X670E and X870E motherboards are the ideal choice for content creators who constantly feel the need to add another drive to their system. With numerous ways to utilize the dual chipset, particularly, you shouldn’t buy a motherboard based solely on its name; instead, make sure it has all the connectors you require.

]]>
https://earlybirdsinvest.com/guide-to-amd-ryzen-am5-motherboard-chipsets/feed/ 0 17725