• AMD's Helios beats Vera Rubin NVL72 on memory at rack level, with 31TB of HBM4 against roughly 20.7TB
• AMD also claims it beats the competition by 15% on FP4 compute on a 'per-GPU' basis even as Nvidia comes out ahead on published 'per-rack' figures
• AMD claims Helios also offers 30% more tokens per dollar spent versus the competition

AMD has launched its Helios rackscale offering, outlining five comparisons in which it claims wins over what it calls "the leading competitive solution."

While the company skipped naming Nvidia, the market leader's Vera Rubin-based NVL72 rack-scale solution is the only real competitor to Helios and the one it continues to compare itself against.

While its memory claims hold, its GPU FP4 claim might fall short when comparing rack to rack, and many of its calculations are based on peak performance rather than Nvidia's published numbers, making Helios an interesting "win" but one that does encourage potential adopters to look more closely.

A numbers game that continues to grow complex even as AMD ekes out some wins

While AMD claims a 15% win versus Nvidia's Rubin on a per-GPU basis for FP4 compute, its 72-GPU rack-scale solution falls short of Nvidia's published rack-level numbers: 2.9 exaflops versus 3.6. AMD does not specify whether it is counting Nvidia's individual dies or its two-die packages, and the distinction matters: against dies the gap runs in AMD's favor by far more than 15%, while against packages AMD trails.

The two figures also use different formats, AMD's MXFP4 against Nvidia's NVFP4, so they are not measuring identical arithmetic.

This might, however, be indicative of a very real situation that hampers both vendors' headline numbers: real-world FP4 workloads rarely reach hardware peak ratings due to memory movement constraints, scheduling overheads, and software kernel efficiency. AMD conceded as much at its own event, putting measured FP4 throughput at roughly half its peak rating.

Nvidia may sustain more of its peak thanks to its custom Vera CPU, a mature NVLink 6 software stack and a larger pool of what it calls fast memory, 75 TB per rack once 54 TB of LPDDR5X is counted alongside 20.7 TB of HBM4. AMD's 31 TB is all HBM, which is better suited to models that must be held entirely in high-bandwidth memory, and Helios offers higher capacity and bandwidth per accelerator at 432 GB and 23.3 TB/s.

On raw scale-up fabric, the two are level, both delivering 3.6 TB/s per accelerator and 260 TB/s per rack.

Despite this, Helios is an exceptionally strong product on paper, and it may be the first time AMD has produced a credible rack-scale answer to Nvidia since the AI race began. Seventy-two MI455X accelerators, 18 EPYC Venice CPUs, 31 TB of HBM4, UALink over Ethernet inside the rack and Ultra Ethernet out, on an OCP Open Rack Wide chassis with merchant Broadcom switch silicon, is a serious response to a company that had a two-year head start on the form factor.

More importantly, its fabric specifications are publicly available, enabling hyperscalers to build customized variants that meet their requirements. Nvidia's platform offers no equivalent latitude.

AMD frames openness as the platform's central advantage, with Vamsi Boppana, senior vice president of AI at AMD, saying that Helios "brings together leadership compute, high-performance networking and open software in a unified rackscale platform."

The more important question for AMD, however, might be memory supply. Nvidia has had Vera Rubin in full production since Q1 with partner availability this half, while AMD's first Helios deployments are not due until Q4. AMD may be further gated by HBM4 supply, much of which is reported to be already committed to hyperscalers, which could keep its deployment volumes well below Nvidia's this year.

• AMD EPYC 9006 "Venice" offers mammoth 256-core, 512-thread configuration that incorporates up to 1GB of L3 cache and PCI-E 6.0 and 16 DDR5 memory channels under the hood
• It's the first time AMD's EPYC lineup sees high core count CPUs push past the 5GHz barrier
• AMD is claiming a win by stating that its newest EPYC CPU offers the "most agents per watt, per dollar, per rack"

AMD's newly-announced 6th Gen EPYC 9006 "Venice" seems to be a genuine silicon milestone (the first x86 server CPU on a 2nm-class node, up to 256 cores and 512 threads) which sees the company aim to secure its position in the AI data center CPU market.

The launch comes at a time when AMD has caught Intel on the back foot but finds its position threatened in turn by Nvidia's Vera CPU, which it touts as a faster, more efficient alternative to traditional x86 offerings, including AMD's own EPYC lineup.

With a staggered launch in place for the four CPUs AMD showcased at the AAI event, the stack is expected to be fully available to consumers sometime in 2027, even as its beefy 256-core CPU targets a Q4 2026 launch.

A massive upgrade on multiple fronts

AMD's Venice is a serious generational step. It is built on TSMC's N2 process, the foundry's first move from FinFET to gate-all-around nanosheet transistors, and Team Red says the family is the first high-performance computing product in the industry to reach volume production on the node.

AMD's recently announced flagship model, the 256-core, 512-thread EPYC 9996, features its new SP7 socket, offering 16-channel DDR5 memory, PCIe 6.0 support, and boost clocks that push past 5GHz.

This effectively doubles what is currently on offer for AMD's current-gen EPYC CPUs: memory moves to 16 channels of DDR5-8000, or MRDIMM at 12,800 MT/s, for roughly 1.6 TB/s per socket.

AMD goes one step further with the 9006X lineup (expected 2H 2027), which replaces the cores it cuts down on with a much larger L3 cache per core, a design that it touts makes it much better than rivals at workflows including HPC, technical computing, and data-intensive workloads, which benefit from the larger cache (and memory bandwidth) on offer.

“6th Gen AMD EPYC delivers workload-optimized compute for the full spectrum of modern data center needs; from running more agents and feeding GPUs across rack-scale AI infrastructure to powering the business-critical applications and services organizations rely on every day," noted SVP & GM of Compute and Enterprise AI at AMD, Dan McNamara.

AMD argues that the shape of the data center is changing into a three-tier agentic architecture, with each tier creating distinct CPU demand, which results in workflows centering around agent sandbox execution, AI host nodes, and general-purpose servers, all of which its EPYC CPUs are designed to handle.

The EPYC 9006 series also powers AMD's rack-scale Helios, as it aims to take on Nvidia's Vera Rubin NVL72 rack-scale offering with its own solution.

With AMD also offering a more mainstream 9006 SP8 and the EYPC 9006 LP as an AI host node CPU that uses LPDDR memory, it is aiming to capture all ends of the AI data center CPU market and cement its current lead in a segment that is sure to be increasingly competitive in the days to come.

• AMD's MI400 press release publishes exactly one performance figure
• 288 TFLOPS of hardware FP64 on the MI430X, a GPU it positions as its go-to for sovereign AI and HPC
• Unlike the MI455X, which it launched alongside, the MI430X will have researchers and sovereign AI clients waiting longer, with AMD targeting early 2027 for its first shipments

AMD has launched its latest Instinct MI400 series, unveiling two new chips which bring performance gains, larger, faster HBM4 memory, and a continued promise to support its open-source ROCm stack.

While both chips come with 432GB of HBM4 memory, they both have very different use cases: the MI455X is AMD's attempt to dominate AI data centers and run training and inference for frontier models, while the MI430 is positioned as a high-performance compute (HPC) alternative that offers superior precision thanks to it being tuned for superior FP64 performance at a hardware level.

Both GPUs are based on AMD's CNDA 5 architecture, but the MI430X offers FP64 performance that dwarfs both the MI455X and Nvidia's Rubin-based offerings.

Frontier AI demand comes first

The launch, which was effectively on paper, still showcased where AMD's focus lies: the Instinct MI455X. There is a simple reason for that: AI data centers prioritize faster FP4 and FP8 performance for most tasks, such as inference, and the MI455X delivers on that front.

AMD is simply responding to market demand by aiming to ship the MI455X first, even as it has sovereign AI buyers lining up, both directly and indirectly, including the U.S. Department of Energy, Oracle, and even the UK government, to name a few.

AMD stated that the MI455X delivers up to 34 times the token throughput of the MI355X at 18 times lower cost, showcasing a significant performance and efficiency upgrade. It is also arguably AMD's most heavily drummed-up release to date, even as it continues to play catch-up to Nvidia's market dominance in the AI data center space.

The MI455X ships with the recently announced Helios, AMD's first rack-scale AI system that launched alongside the new Instinct chips. One of these has 72 MI455X GPUs and 18 sixth-gen EPYC server CPUs, configured in a 1:6 ratio per compute node.

The MI430X, on the other hand, seems to be competing specifically to score buyers who require higher precision, such as science researchers or even government departments where correctness is required at the edges, and memory limitations apply even with the MI series accelerator's 432GB of HMB4 in play.

“The next generation of AI will span frontier AI, sovereign AI and scientific computing, and each requires infrastructure optimized for its unique demands," said Vamsi Boppana, SVP AI at AMD.

"The AMD Instinct MI400 Series extends our AI portfolio with purpose-built solutions optimized for the full spectrum of AI and HPC deployments, built on an open software foundation that gives customers the flexibility to innovate at every scale.”

AMD also has an important buyer for the MI455X in its Helios rack-scale form that will also grant it further exposure to it pre-IPO: Anthropic. The company has agreed to invest $5 billion in the AI giant, with the latter agreeing to deploy 2 GW of AMD's MI450 accelerators, even though its AI head wasn't exactly enthused recently by a perceived drop in performance on Claude Code in early 2026.

• Nvidia's Olympus core architecture prioritizes single-thread IPC over frequency, using a 10-wide decode front end, deep out-of-order mid-core, and a graph prefetcher tuned for agentic AI's branch-heavy, pointer-heavy workloads
• Vera trades chiplet-style core density for a monolithic 88-core die and single-NUMA-per-socket design, a deliberate bet on agentic AI workloads that Nvidia's own engineers admit comes at the expense of legacy workload performance
• Self-reported SPEC CPU 2026 results by Nvidia paint a per-core advantage figure of anywhere between 70% and 80% versus AMD's EPYC 9755 server CPU

Nvidia's Vera CPU has a lot to prove, representing the company's first real attempt at the AI server CPU market, where traditional vendors Intel and AMD, along with third-party Arm-based providers, are all gunning for a piece of an increasingly lucrative data center pie.

To that end, Nvidia has published its most detailed technical account yet of the chip, promising substantial performance gains over the competition. Vera is the company's first server processor built around a fully in-house core design, a departure from the stock Arm cores that powered its Grace predecessor.

As Vera heads toward general availability in the second half of 2026, Nvidia is making its case on a specific front: not raw core count, but sustained per-core performance under load, the metric it argues matters most for agentic AI.

What's actually under the hood of Nvidia's Vera CPU?

At the core of every Vera CPU is Olympus, Nvidia's first custom server core, built to the Armv9.2 instruction set but designed in-house rather than derived from Arm's stock Neoverse designs, as its predecessor, Grace, was.

Nvidia's second-generation data center CPU is a completely reworked design built around a single goal: to lead in agentic workload performance.

Rather than following Intel and AMD down the chiplet route, Nvidia packs all 88 cores (176 threads) onto a single monolithic die, with a dual-socket configuration delivering 176 cores and 352 threads in one system.

Nvidia has detailed the core design extensively, and several publications have since dug into the microarchitecture. The front end runs a 10-wide decode engine paired with a neural branch predictor that can resolve up to two taken branches per cycle, designed to handle the large instruction footprints and irregular control flow of interpreters, compilers, and agent runtimes.

The mid core combines a wide rename-and-allocation engine with a large reorder buffer and dependency-breaking techniques including memory renaming and value prediction.

The execution engine schedules dynamically across integer, vector, floating-point, cryptographic, load, and store resources, while the cache subsystem adds a graph prefetcher targeting the pointer-chasing access patterns that defeat conventional streaming prefetchers.

That design has not gone unanswered. AMD has countered with estimated figures for its 256-core Zen 6 "Venice" part, claiming a 3.3x rack-level advantage over Vera, though those numbers are extrapolated rather than measured.

AMD responding first isn't surprising. Within x86, it continues taking ground from Intel, reaching 33.2% of x86 server CPU shipments in Q1 2026 per Mercury Research, up from 27.2% a year prior.

With Arm reporting that its architecture now accounts for roughly 50% of CPU compute among top hyperscalers, competition is building from several directions at once. Vera may well set a new bar for agentic AI.

Intel is one of the most ubiquitous companies in the world, let alone the technology industry, with its technologies powering many of the computers most people have used for generations. Central to its success was a philosophy from its former CEO Andy Grove, who was a part of the company from day one and ran it from 1987 to 1998.

Inside Intel

Andy Grove was setting out the foundational theme for his management book 'Only the Paranoid Survive', which he wrote toward the latter years as CEO of Intel.

Grove, who is widely recognised as one of the most successful tech executives of the late 20th century, advocated strongly for putting paranoia at the heart of business. Worrying about various risks that can affect your business — from new competitors to macroeconomic shocks — means you can prepare for them.

But what's worse, as he described it, are 'strategic inflection points' that signal a massive shift in fundamentals that could either lead to staggering growth or a grave decline by an order of magnitude. It's fair to say, in the years since he left Intel, that the company has endured its ups and downs, putting it mildly.

Sliding doors

Although Intel enjoyed incredible success in the 20th century and into the early 2000s, the company stuttered in recent years with rival chipmakers including Nvidia and AMD gaining ground and eating into its market share.

Ironically, from the outside, it appears that the company had succumbed to the sort of complacency that its former CEO had been warning against.

That said, the company is in the midst of a remarkable comeback under the leadership of new CEO Lip-Bu Tan, who has overseen a dramatic surge in the company's stock price that underlines its improved overall outlook as well as new ventures to support the AI buildout.

• AMD's Versal Premium Gen 2 Memory on Package launch formally ends HBM in its adaptive SoC lineup
• The move represents a ~65% bandwidth cut from the discontinued Versal HBM's 840 GB/s, clocking in at just 288 GB/s
• AMD is painting this as a win, citing better availability, a smaller form factor, efficiency wins, and guaranteed memory supply for the next 15 years

The AI boom may have just eaten AMD's own lunch, thanks to an HBM shortage that forced it to resort to lower-bandwidth LPDDR5x for its Versal Premium Gen 2 Memory on Package offerings.

AMD recently announced the Versal Premium Gen 2 Memory-on-Package family, which leverages up to 32GB of LPDDR5x memory directly onto the chip package.

The move, which effectively cuts bandwidth by 65% compared with previous-generation Versal offerings that use HBM, is seen by many as the need of the hour as memory supplies dwindle amid overwhelming demand.

A needs-based move to lower-bandwidth memory

AMD's move is calculated, even if it comes at a significant performance cost to the chip designer. HBM memory supply is strained, and even within AMD's own lineup, its more profitable (and more demanding) Instinct datacenter GPUs have priority for HBMs' current and future iterations.

AMD is therefore both a beneficiary and a casualty of the same AI demand wave that creates opportunities on one end but constrains supply for other segments, such as its consumer-grade hardware, gaming, and SoC divisions.

AMD's Versal lineup stems from Xilinx, which it acquired in 2022. Xilinx shipped its first on-package memory FPGAs, the Virtex UltraScale+ HBM parts, in 2018 with up to 16 GB of first-generation HBM. The follow-up Versal HBM series, a variant of the Versal Premium line, supported up to 32 GB of HBM2e with 840 GB/s bandwidth.

The problem for AMD isn't just sourcing supply for its new Versal FPGAs, something it calls an Adaptive SoC (System-on-Chip), but the fact that these are extremely long-tail products. In other words, support, dedicated supply, and accessories must remain available to consumers for a long time if they are to embrace and continue working with a particular FPGA class, which complicates matters.

AMD discontinued its last-generation Versal lineup in September 2025, citing HBM2E supply constraints rather than any issues with the chips themselves, and offered no alternatives to customers, stating only that "final orders (LTB) for Adaptive SoC parts will be accepted until June 30, 2026, subject to material availability."

The new Versal lineup effectively addresses this gap, stating that it has a 15-year lifecycle and citing "memory longevity" as the reason for its pivot to LPDDR5X.

AMD's move does bring it some other advantages despite the obvious bandwidth chokepoint: LPDDR5X has better availability than HBM for the foreseeable future, and it also operates at industrial temperatures, whereas HBM tends to be stacked in ways that require advanced cooling. Not only does LPDDR5X run cooler, often passively so in most configurations, but thanks to sporting only 4 memory chips onboard, it is over 60% smaller than comparable FPGAs.

The newer FPGAs will be considerably cheaper to manufacture than their HBM alternatives in current market conditions, and with Chinese memory suppliers such as CXMT also eyeing the same market, Versal Gen 2 might be the long-stay that its predecessor was initially intended to be in a rapidly changing market.

• AMD is reportedly preparing a 10% price hike on graphics memory kits for Radeon GPUs
• This means retailers could increase prices of Radeon GPUs from the likes of Sapphire or Asus
• This case is yet another example of just how hard the RAM crisis is hitting the PC market

Gaming is becoming an expensive hobby for both console and PC gamers, and it's all because of the ongoing RAM crisis. Unfortunately, another manufacturer's GPU components may be the next victim of price hikes.

As reported by VideoCardz, AMD is planning a 10% price increase on graphics memory kits for its Radeon GPUs this July, according to a new post on a Board Channels forum. This means that AMD's AIB partners, such as Sapphire and Asus, will pay more for components, which might result in higher retail prices for Radeon GPUs.

The Board Channels post states that these price hikes are a result of 'tight global supply of graphics memory chips,' including the noticeable trend of 'continuously rising market prices'.

Of course, there's no confirmation from AMD on the matter just yet, but if true, it doesn't help AMD's position in the GPU market up against its fierce rival, Nvidia. Frankly, both are in a tough spot due to higher memory costs across the board, which is well established for effectively every PC manufacturer at this stage in 2026.

While most of Team Green's RTX GPUs are fairly expensive for consumers, the advantage of better upscaling technologies and raw performance when using ray tracing in games makes them more appealing, in comparison to newer Radeon GPUs.

While we await confirmation from AMD, consumers should start buying new hardware now. Both AMD and Nvidia GPUs are still expensive across several retailers, and it's a tough spot for any gamer to be in when searching for an upgrade.

However, if you don't own a gaming PC at all, it's worth keeping an eye out for several deals — and fortunately, Best Buy's 4th of July sale is already underway, and this iBuyPower Slate Intel Arc B570 gaming PC is only $899.99 on sale.

It also comes with 16GB of DDR5 RAM (which would be expensive to buy separately), and most importantly, it's using a GPU that's perfectly capable of great gaming performance at 1080p and 1440p resolutions.

When acknowledging the messy state of the PC hardware market and the prospect of more price hikes, it's hard to ignore a deal such as this, especially when Intel's XeSS upscaling has improved over the last few years — so if it were me, I wouldn't hesitate to purchase.

GMKtec EVO-T2: 30-second review

Mini PCs rarely arrive with fanfare. The GMKtec EVO-T2 is an exception. It debuted at CES 2026 with Intel CEO Lip-Bu Tan visiting the booth, testing the unit in person and signing a prototype. His signature wasn’t on my test version, and you can imagine my disappointment.

That kind of endorsement is unusual for a compact desktop from a Chinese OEM, and it signals something real. This is not a routine refresh.

At its core sits Intel's newest Panther Lake architecture. The Core Ultra X7 358H is built on Intel's own 18A process node, making the EVO-T2 one of the first consumer products to ship that fabrication technology at volume. The question every reviewer has to answer is whether the real-world performance justifies a price tag that sits above £1,500 for the standard retail configuration.

The review unit supplied by GMKtec uses the Core Ultra X7 358H with 64GB of LPDDR5X RAM and a 1TB NVMe SSD, but some SKUs offer the Core Ultra X9 388H for those willing to pay extra.

What GMKtec has here is cutting-edge Intel technology that’s ideal for power users, AI LLM fans, and creatives, all in a remarkably compact package that doesn’t cost a fortune.

With so few brands offering products with Panther Lake silicon, this is easily one of the best Mini PC machines available today.

GMKtec EVO-T2: Price and availability

• How much does it cost? $1899/£1521/€1,700
• When is it out? Available now
• Where can you get it? Currently, this machine can be obtained directly from GMKtec

The GMKtec EVO-T2 is available direct from the GMKtec US and GMKtec UK websites. I'm also seeing a configuration on Amazon.com.

There appears to be plenty of confusion about the pricing of these products on the GMKtec website, and pricing on Amazon seems only to compound things.

The USA pricing for the X7 is $1,899 for the standard model, but oddly $3,299.99 for the 853GB model. That second price is obviously a mistake, and I’d assume that’s the dollar pricing for the X9 version.

What I can say with some certainty is that on the GMKtec UK website, there are four SKUs: two X7 and two X9 models. And, the X7 358H design with 64GB of RAM and 1TB of storage, as per my review, the hardware is £1,520.99.

There is another X7 version with an 853GB drive, which I’m going to assume is a special partition for running LLMs, and it sells for £1,629. There are two corresponding X9 388H models, which are both priced at £2999.99, curiously. That seems like a ridiculous markup for only a few hundred MHz extra across a couple of clock settings, but that’s what they’re asking.

As a slight sweetener, UK customers are being offered a code that gets them a further £60 off at the time of writing.

European prices are €1.699,99 for the X7, with no X9 stock currently available.

Comparing these prices to anything else is a fraught exercise, since this is the first Mini PC to use this platform, even if Asus has a new NUC 16 Pro planned that uses this chip. And, Minisforum has a new AI X1 Pro model in the pipeline that uses an advanced AMD chip using its Gorgon Point core.

What I might point out is that the GMKtec EVO-T1, which uses the Arrow Lake Core Ultra 9 285H, is only $1,279.99 with 32GB of RAM and 1TB of storage, but it can also be bought barebones with no memory or storage for close to $900.

From a performance perspective, the GMKtec EVO-X2 Ryzen AI Max+ 395 is more powerful, and that, with 64GB of RAM and 1TB of storage, is only another $100 more.

On that basis, the EVO-X2 seems to be a better value than the EVO-T2. And, GMKtec offers a 128GB version of the EVO-X2, for those wishing to blow $3,299.99 on one.

The big issue here is patently the cost of memory, and these platforms all need the faster soldered modules that are currently 400% more than they were only months ago.

I’ve given this a score of 4 out of 5 since you can’t get this hardware anywhere else at this time, and when Asus and Minisforum do come to market, they’re unlikely to undercut GMKtec's price.

• Value: 4 / 5

GMKtec EVO-T2: Specs

GMKtec EVO-T2: Design

• Same enclosure as EVO-T1
• Easy internal access
• No memory upgrades

The case of the EVO-T2 is oddly familiar, as it seems relatively unchanged from the EVO-T1 that I reviewed in August of 2025.

The EVO-T2 continues GMKtec's established squared, compact form factor. The chassis uses a precision surface finish and houses what GMKtec calls its Ice Storm 3.0 cooling system, an active solution with a visible RGB fan. It sits comfortably on a desk or mounts via VESA behind a monitor.

Port placement is well considered. The front panel carries the USB4 port with 100W Power Delivery alongside two USB-A 3.2 connections and one USB-A 2.0. These are the ports users reach for most often, and they are exactly where they should be. The rear houses the second USB4, dual HDMI 2.1, DisplayPort 1.4, dual Ethernet, the OCuLink port and the audio jack.

That rear panel is dense. It rewards deliberate cable planning rather than improvisation. Quad 4K display support is a genuine differentiator at this size. Creative professionals and multi-monitor users will appreciate having that headroom without needing an external dock.

This hints that the EVO-T2's connectivity specification is exceptional for the category.

Dual Ethernet with 2.5GbE and 10GbE ports on a machine this small opens it to homelab, network-attached storage and professional networking roles that most mini PCs cannot fill. The 10GbE port alone makes this worth serious consideration for anyone who moves large files regularly.

Wi-Fi 7 and Bluetooth 5.4 cover wireless needs for the foreseeable future. The front-panel USB4 port carries 40Gbps bandwidth and supports Power Delivery at 100W, which is genuinely practical for desk setups where the EVO-T2 acts as a hub. DisplayPort Alt Mode via USB4 adds display flexibility beyond the two HDMI outputs and the dedicated DisplayPort at the rear.

The OCuLink Gen4x4 port provides a direct PCIe pathway to external GPU enclosures without the bandwidth constraints of USB4. For users who plan to add a discrete GPU later or need GPU-accelerated compute for specific workloads, this is a meaningful long-term expansion option, though not all systems include it.

Internal access requires removing the feet and then four screws. It’s not difficult, but it might have been easier. Inside, you can access the three M.2 slot positions, two already being occupied by the Wi-Fi 7 adapter and the 1TB NVMe drive. One oddity about the two M.2 slots allocated for storage is that the one GMKtec filled with the supplied Gen4 drive is the PCIe 5.0 slot, leaving the PCIe 4.0 slot free. That seems silly, but this isn’t the only mini PC maker doing these things.

For testing, I moved the provided drive to the Gen4 slot and put a Gen5 drive in a slot where it works best. And, if anyone buys one of these, I’d recommend doing that and cloning the drive to the Gen5 module. I’d also suggest you get a Gen5 drive with a heatsink or add one before installation.

There are no memory upgrade options, because all the memory here is soldered to the mainboard.

• Design: 4 / 5

GMKtec EVO-T2: Features

• Intel Core Ultra X7 358H
• Soldered memory
• 180 TOPS AI

Panther Lake is easily Intel's most architecturally interesting mobile platform in years. The Core Ultra X7 358H uses a chiplet design built across three separate tiles. The compute tile is manufactured on Intel's 18A process and incorporates RibbonFET gate-all-around transistors alongside backside power delivery using PowerVia technology. Intel states this reduces voltage drop by around 30 per cent and improves transistor density over the prior generation.

I have a not-unreasonable feeling that this was what the 100 series was meant to be from the outset, and not what actually arrived.

The processor carries 16 cores across three tiers. Four Cougar Cove performance cores handle peak single-threaded workloads and boost to 4.8 GHz. Eight Darkmont efficiency cores manage sustained parallel tasks. Four low-power efficiency cores handle background activity.

It is a sensibly layered approach, and the trickle-down from mobile laptop silicon means the EVO-T2 benefits from extensive driver and platform maturity work done for notebook OEMs.

The gap between the X7 358H and the flagship X9 388H is modest in practice. The 388H adds 0.3 GHz of peak boost. For most productivity and AI workloads, the difference will be within the margin of thermal variation. The X7 is the right choice for a machine where value matters.

The soldered RAM configuration is the most significant design decision to understand before purchasing. Retail units ship with 64GB of LPDDR5X-8533. The review unit carries 16GB. Neither configuration can be upgraded after purchase. Buyers who anticipate needing more capacity should look at the GMKtec EVO-X2, which uses AMD Ryzen AI Max+ 395 with up to 128GB of unified memory.

Soldered LPDDR5X enables the higher bandwidth speeds that integrated graphics depend on heavily. The Arc B390 needs fast, wide memory access to perform at its best. The trade-off is permanent capacity. For most professional workflows and AI tasks at the 64GB retail tier, the ceiling should be workable for several years.

Storage is considerably more flexible. The primary M.2 slot runs PCIe 5.0, offering sequential read speeds above 10 GB/s with compatible drives. The secondary slot is PCIe 4.0, still capable of strong throughput for a second drive or overflow storage. Combined capacity can reach 16TB, which opens the machine to NAS-adjacent roles that most mini PCs cannot consider.

GMKtec's headline claim is 180 TOPS of combined AI performance across the CPU, GPU and NPU. The NPU alone contributes 50 TOPS, which represents a meaningful step up from prior Intel generations. The remaining compute performance is divided between the Arc B390 and the CPU cores themselves.

The 180 TOPS figure is a heterogeneous combined total. Real-world AI workload distribution depends heavily on the framework and the model. Not all applications can efficiently split inference across three compute blocks simultaneously. The NPU handles fixed-function acceleration well. More general local inference typically leans more on the GPU or CPU, depending on quantisation level and context window size.

The aiDAPTIV+ AI SSD technology developed with Phison is worth mentioning. It extends available memory by intelligently paging model data between DRAM and NVMe storage. GMKtec claims this allows the EVO-T2 to run models with up to 70 billion parameters locally. For a machine with 64GB of RAM, that is an extraordinary claim. The paging mechanism will introduce latency penalties for data not resident in DRAM, and the practical throughput impact under sustained inference loads has not been verified by me, since I don’t have an aiDAPTIV+ AI SSD.

One thing that did make me wonder about the design of the EVO-T2 was that GMKtec claims that the USB4 port on the front can push out 100W for recharging a laptop. As the PSU is only rated at 148.2W and the system could take 60W, there appears to be a voltage shortfall in that calculation.

Overall, the X7 358H is a dramatic uplift from the 200 series chips that came before it, although AMD has such a significant lead with its Ryzen AI 395 series that it's perhaps too much to ask in one generational change.

• Features: 4.5 / 5

GMKtec NucBox EVO-T2: Performance

Normally, I compare AMD and Intel Mini PCs against their ilk. But as this is a flagship design of a new silicon generation, and the best Intel has to offer, I thought it would be useful to compare it to the top-of-the-line AMD platform.

For those who like the fine details, the EVO-T2 was set into ‘Performance’ mode in the BIOS, rather than balanced or silent. According to GMKtec, there should be a mode above this that consumes 80W, but it was missing from my BIOS. Therefore, it might be possible to get even more out of the EVO-T2, and the BIOS certainly has plenty of features for overclocking some aspects.

The Arc B390 GPU is the headline graphics story in Panther Lake. It uses twelve Xe3 cores built on TSMC's N3E process and clocks up to 2.5 GHz. Intel positions it as delivering 76 per cent more performance than the Arc 140T from the previous Arrow Lake generation and 82 per cent more than AMD's Radeon 890M found in Ryzen AI HX 370 platforms. As usual, Intel ignores the Radeon 8060S in the Ryzen AI 395 Max, which is its true competitor.

And, as the results show, B390 might make Arc 140T and Radeon 890M look poor, but it still can’t run with the Radeon 8060S, not even close.

Obviously, you could use the OcuLink port to add a discrete GPU, but that’s the only way that the EVO-T2 would compete with a Bosgame M5 AI.

And, while all those extra cores in the Intel Core Ultra X7 358H CPU do boost the multi-processing considerably, it still falls short of the AMD Ryzen AI 395 Max.

It did well in the PCMark score, and it also had a better SSD than the Bosgame machine. I inserted a Crucial P510 Gen5 drive into this hardware, and it delivered 10,587MB/s reads and 8,977MB/s writes in the Gen5 M.2 slot, so it could be dramatically better with a small investment.

Overall, Intel is moving in the right direction, but AMD don’t need to worry yet.

• Performance: 4.5 / 5

GMKtec EVO-T2: Final verdict

The EVO-T2 is an impressive piece of equipment that exploits the dramatically better silicon Intel has released with Panther Lake. However, given the modest differences between the X7 358H and the X9 388H, I wouldn’t be rushing to spend the extra on the top model unless money isn't a concern for you.

Where this rig excels is in providing a punchy platform for software development, video editing and running AI LLMs. It would have been nice to have seen Thunderbolt 5 ports on this PC, but the cost of adding this feature appears to be putting mini PC makers off.

At least it has USB4 and an OCuLink port if you want more GPU power.

But the focus here is foremost AI, with GMKtec even deploying a version of OpenClaw to the desktop along with Herdman. While there are free tokens to be had, these are a portal to monthly subscription packages for those who want to fully exploit AI agents.

That seems slightly at odds with the premise of having hardware like the EVO-T2 that can run powerful local AI at no other cost than electricity, and your time to get it trained in your workflows. But agents will always come with extra costs, regardless of where they are launched from.

Getting back to the hardware, this is an exceptional piece of gear that nearly brings Intel back up to AMD’s current level. Although with new AMD hardware soon to launch, this might prove to be a false dawn for them.

What it comes down to here is the price, since all memory and storage costs are artificially inflated.

For those who, for whatever reason, don’t trust AMD, then this is quite clearly the best Intel mini PC technology available, especially for those running local LLMs.

Should you buy a GMKtec EVO-T2?

Buy it if...

Don't buy it if...

Also Consider

For more professional hardware, we've reviewed the best business computers.

• AMD Zen 5-based EPYC 9965 is on sale on eBay for just under $6,000
• With 192 cores and 384 threads in tow, AMD's offering can handle as much as 6TB of RAM per processor for AI-centric workloads
• The CPU not only offers the highest core count in all x86 CPUs to date, but also holds its own against the competition

When AMD unveiled its new EPYC CPUs in June 2024, one particular chip caught everyone's eye, given how outrageously powerful it was compared to the competition.

The AMD EPYC 9965 CPU was also the company's most ambitious server-grade offering to date, and one that has held its own against the competition with relative ease thanks to its yet-unmatched 192-core, 384-thread x86 configuration.

At a suggested MSRP of $14,813 (now $11,988) at launch, it is also one of AMD's most expensive products across the board, with only a few of its server-grade GPUs pushing noticeably higher.

Why the eBay listings are somewhat hard to digest

Despite being unveiled over two years ago, the EPYC 9965 CPU remains largely unchallenged by the competition, thrashing most benchmarks with relative ease. It remains the flagship server chip for AMD-based systems worldwide, with Team Red shipping it globally, including China.

The chip itself is a genuine monster. The EPYC 9965 packs 192 Zen 5c cores and 384 threads onto a single SP5 package, with a 2.25 GHz base, a 3.35 GHz all-core, and a 3.7 GHz maximum boost, 384 MB of L3 cache, twelve channels of DDR5, 128 lanes of PCIe 5.0, and a 500 W default TDP, AMD's highest yet. It is the densest part in AMD's 9005 stack, built for cloud providers aiming to cram the maximum number of virtual machines into a rack.

All in all, it is designed primarily to serve hyperscalers and cloud providers, with a focus on low-power, efficient cores that dominate such setups.

Even at its reduced reference price of just under $12,000, eBay listings selling the same CPU for just under $6,000 seem inexplicable, offering a sub-50% discount over current advertised prices and over 60% off its original launch price.

This is even lower than the last time we tracked this CPU in 2025, but there might be an explanation of sorts for what would otherwise be seen as an anomaly. Most server-grade CPUs trade at prices below MSRP, with OEM and negotiated discounts that quickly add up to significant savings for large-scale deployments.

The other explanation that often applies to older server-grade CPUs that often get cycled out or decommissioned by data centers does not apply here, however; the EPYC 9965 does not have a successor yet, and it is hard to imagine a currently available CPU replacing AMD's most capable chip yet in a data center just yet

Other reasons prices are low could include excessive inventory, canceled orders, or simply someone liquidating hardware in an industry that is increasingly GPU- and memory-focused to meet its AI needs.

Most industry players prefer a direct OEM solution, especially given AMD's support for PSB-locking (secure boot) on their CPUs for certain vendors, which makes them unwilling to purchase from 3rd-party sellers. This also raises the risk profile for buyers, who could unwittingly end up purchasing a vendor-locked CPU, making for a rather annoying testing and return process.

With a shift in spending towards GPUs and memory becoming prohibitively expensive, hardly helping matters, while enterprise consumers are still buying up the EPYC 9965, prices might reflect a market reality: they are unlikely to be scouring eBay for their next 192-core monster CPU purchase, even assuming they have a compatible motherboard and memory modules to match.

• AMD has 'relaunched' three processors that use legacy cores to address demand at the lowest end of its mobile and desktop CPU offerings
• The CPUs range from 2 to 8 cores and leverage its older Zen+ and Zen 2 core technology as the chip designer grapples with memory and storage shortages to serve an increasingly cash-strapped budget segment
• All three processors support Microsoft Windows 11 and leverage cheaper DDR4 memory as manufacturers aims to 'hold the fort' in the face of rising AI demand creating challenges for entry-level PC buyers

AMD's latest Ryzen CPU offering might not be one that most of its performance-centric consumers would appreciate in 2026.

The company is looking to reintroduce three budget-centric mobile CPUs that were last seen over half a decade ago, using legacy Zen+ and Zen 2 cores.

The move comes at a time when assemblers and consumers alike are grappling with increasing chip, memory and storage costs, even as lower-end options continue to grow increasingly limited for price-conscious users.

A Dual-core CPU in 2026?

AMD is reintroducing the Ryzen 3 3100U, Ryzen 5 3501U, and Ryzen 7 4700LE to consumer markets, according to reports. The CPUs offer dual, quad, and octa-core options.

While the Zen 2-based desktop-class Ryzen 7 4700LE has a listed launch date of 25th March 2026, the other two SKUs were launched more recently, in June 2026, according to AMD's own listings.

The Ryzen 3100U and the 3501U both use AMD's legacy Zen+ architecture on a 12nm process node, making them less power-efficient than the 7nm-based Ryzen 7 4700LE.

In an emailed statement to PC World, AMD stated, "The Ryzen 3100U and Ryzen 3501U are additional SKUs based on AMD’s existing Picasso architecture that were developed to support specific OEM requirements in the value segment. These processors are intended to address targeted customer demand for lower-cost solutions and will be available in limited volumes through select OEMs".

As OEM components are made directly available to laptop assemblers, AMD does not disclose prices for these models, even as rival chipmaker Intel seems to be taking a similar route.

What might rankle some users in particular is not the fact that the 3100U and the 3501U are rebadged CPUs that will still lag behind pretty much even the most basic laptop CPUs available in the last few years, but the fact that AMD might be thinking it is acceptable to offer a 2-core, 4-thread CPU in 2026 as a baseline for new, entry-level laptops.

The Ryzen 3 3100U is getting considerable ire, as Redditors are already drawing unfavorable comparisons to current-generation Athlon CPUs. While it offers the same iGPU (Vega 8), TDP (15W), and memory configuration (DDR4-2400), along with the same socket (FP5) as its quad-core sibling, it is noticeably weaker than most, if not all, recently launched mobile CPUs.

Unsurprisingly, it does make the cut for Windows 11 by offering support for fTPM, allowing for secure boot, and just pulling ahead of Microsoft's stated requirements for its OS (1 gigahertz (GHz) or faster with 2 or more cores), even as most users will struggle to do anything but the bare minimum on a laptop sporting a CPU essentially rebadged from 2019.

AMD's move, however, is predicated on the fact that rising RAM, SSD, and cutting-edge silicon prices all create a void that must be addressed for the foreseeable future, even as its target audience seems to be getting an increasingly short end of the stick as part of the solution.

AMD has seen its Instinct GPUs continue to get traction in an increasingly competitive marketplace, as it continues to take data center market share from new and existing players and ekes out wins with gaming-centric CPUs in the consumer market.

Its most recent acquisition of MEXT, an AI-centric startup that currently deploys software that allows users to treat NAND flash as DRAM at an operating system level.

AMD says Santa Clara-based MEXT a "pioneer in AI-driven memory optimization technology".

SSD storage to DRAM for data centers?

The idea that MEXT builds on is hardly a new one, but one that it seems to have refined considerably, making it an important acquisition at a time when hyperscalers continue to struggle with limited DRAM availability, even as an even worse SSD crisis looks to be on the horizon.

MEXT's Predictive Memory is essentially a tiering engine that monitors which memory pages applications tend to access, treating regularly accessed sections as "hot" working sets kept in DRAM while offloading "cold" or less frequently accessed sections to SSDs.

This allows for a far lower performance offset than using all of one's flash memory as DRAM, with the latter being an order of magnitude faster for access, even as speed becomes a driving factor for newer chips that are increasingly memory-bound.

There is also an important economic factor at play here: DRAM is nearly 50 times more expensive than the corresponding NAND flash, making cost and scalability key considerations for most data centers looking to avoid an already expensive DRAM market that is slated to only get worse over time.

The move itself is not AMD's first foray into the storage segment, with its consumer-focused StoreMi offering essentially allowing a faster SSD to work as a cache, making up for slower drives on one's system by essentially creating a copy of files that regularly need to be loaded or accessed on the fastest possible storage solution.

Its lesser-known (and since abandoned) Radeon RAMdisk offering allows users to do the exact opposite of what MEXT is offering: creating a very fast virtual disk on existing system memory. even as enthusiasts have replicated the idea on AMD's ultra-fast 3D V-Cache tech.

AMD's purchase makes sense given how deeply embedded its hardware is expected to be in datacenters over the next decade, and one could argue that MEXT's team, which offers expertise in AI infrastructure and memory systems, could be a much more prized acquisition than the underlying technology it offers.

AI and chip talent have become increasingly difficult to lock in, with companies splurging to attract some of the biggest names in both segments, and MEXT's acquisition could help both AMD's short-term and long-term goals in the data center segment.

• Nvidia is pitching its upcoming Vera CPUs to Chinese clients while advertising availability soon
• Move into data center CPUs pits Nvidia against traditional rivals Intel and AMD, which currently control the bulk of the market in China and other regions
• Nvidia's move comes at a time when the Chinese government continues to advocate for home-grown chip solutions, and the US's chip controls have effectively reduced its share to 0% of a lucrative Chinese data center market

Nvidia is apparently pushing to win Chinese customers for what its CEO regards as the next multi-billion dollar frontier for the company: data center CPUs.

The firm has spent the past two years watching the 2nd-most important chip market in the world effectively cut it out with a mix of consumer-grade chips and homegrown solutions such as Huawei's Ascend offerings, backed by a Chinese government push for self-reliance.

While Chinese officials have held the line with soft barriers - no official restrictions on Nvidia's chip exports to China exist in the mainland - Nvidia is seemingly betting on a reset in relations when it comes to its Vera CPUs for the data center.

Why CPUs and why now for Nvidia?

Nvidia's Vera CPU is more than just another competitor in the market. It threatens to upend the existing status quo, with Intel and AMD chips dominating the market, by adopting an AI-first approach to its design.

Nvidia is presenting Vera as a CPU that is up to 1.8 times faster than current x86 CPUs from Intel and AMD in certain workloads, offers 4 times the memory bandwidth, and delivers up to a 50% increase in performance versus traditional rack-scale CPUs.

According to a Reuters report, at least one major unnamed Chinese cloud company intends to purchase 300 servers, each containing 2 Vera CPUs. The CPUs themselves are estimated to cost upwards of $20,000 before bulk discounts kick in.

However, whether this results in an order remains to be seen. Chinese regulators seem to be making a stronger push for self-sufficiency in their chip sector, prompting many of their AI startups and giants alike to opt for local chip options, such as Huawei's Ascend and T-Head's Hanguang.

Nvidia's salvation however, might come from an unexpected place, as battle lines might be different this time around: the same AI export controls that crushed its Chinese business might work in its favor now.

Not only are CPUs considerably less regulated by US export rules, but the Chinese market is also reeling from a squeeze on server CPUs, with Intel pushing delivery lead times to as much as 6 months in some cases, even as AMD noted that the CPU market remains tight, with demand outstripping supply.

If Nvidia can navigate past the politics and the incumbent x86 architecture's software advantage, as well as the ecosystem built around it, it might carve out an important piece of the lucrative Chinese data center market, even without the CUDA 'stickiness' that makes its GPUs so sought-after in this segment.

Whether that is enough to overturn the obvious advantage that both well-placed chipmakers (Intel and AMD) have in terms of an ecosystem, as well as the edge the domestic champion, Huawei, enjoys in terms of government-level backing, however, remains to be seen.

• AMD has a new marketing campaign criticizing the MacBook Neo
• The main thrust of the argument is that the Neo fails as a gaming laptop
• That's a highly confusing angle to take, seeing as this isn't what MacBooks are about – and especially not the budget-friendly Neo

AMD appears to be getting defensive about its Windows 11 laptops and the threat posed by Apple's MacBook Neo.

Tom's Hardware noticed that AMD has a new marketing campaign that intends to take the Neo down a notch or two, promoting the benefits of Ryzen AI CPUs instead.

AMD insists that: "The competition made compromises. You don't have to." It then adds that: "Everything MacBook Neo leaves out, built in with AMD Ryzen AI processors."

The main thrust of the compromise messaging is that the MacBook Neo isn't up to scratch for gaming. AMD notes that 15 of the top 20 PC games don't run natively on a MacBook Neo, and that you need 'workarounds' (emulation) to play them. The games are picked by Team Red itself, although it's a fair enough selection of esports classics and more modern efforts such as Cyberpunk 2077, Battlefield 6, and Black Ops 7.

It's not just about games, of course, but that's the lead item here. AMD also points out that Windows 11 laptops with its Ryzen AI CPUs can have touchscreens (which the MacBook Neo doesn't), and a better port selection than the MacBook Neo's two USB-C connectors.

AMD then highlights benchmarks of the Ryzen 5 220 against Apple's A18 processor as seen in the Neo, where both laptops have 8GB of RAM, but the Ryzen is up to 57% faster for multitasking, and 38% faster for content creation as measured in Blender and Cinebench, among others.

Analysis: confusion over compromises

This is an odd one, really, because what plenty of confused people on Reddit have been saying – echoing my own first thought – is: "I didn't buy a Mac to play games. That's not really what they're for."

Someone else on that thread agrees: "Ok, but no one is buying a MacBook for gaming, right?"

And a further Redditor observes: "Imagine if 'Macs don't game' is the best you've got, lol. I'd fire my marketing team."

These folks are, of course, quite right. You don't buy a MacBook for gaming, and moreover, you especially don't buy a MacBook Neo for PC games. This isn't just an Apple laptop, but a low-end one with a mobile chip designed entirely around being affordable (for students, mainly) and able to cope with daily computing workloads – and that's all.

Okay, so you might want to do a spot of casual gaming, but the MacBook Neo can cope with that just fine. What you aren't going to be looking to do on a Neo is play the top PC games, not unless you're very misguided, and okay, probably a few people out there do fall into this category.

So why AMD has led with this gaming angle in its Ryzen AI laptop marketing, versus MacBook Neo marketing, really is a head-scratcher. It isn't like the kind of Windows 11 laptops AMD is using as comparisons to the Neo, in a similar price bracket, can run those top PC games all that well, anyway.

Of course, AMD has chosen its 'top games' very carefully, which goes without saying. But still, the Radeon 760M integrated graphics (used in the comparative benchmarking) can't come close to running some of these games with any kind of fluidity, even with low details (at the Full HD resolution in which the testing was conducted).

The same cherry-picking applies to the productivity and creativity benchmarks, and the thing about the MacBook Neo is that, as we've seen at TechRadar, despite only having 8GB of RAM, it runs smoothly and responsively in everyday tasks. That's not something you can always say about a Windows 11 laptop with 8GB of RAM in the same price bracket.

Granted, Microsoft is busy working on making Windows 11 better with lower RAM loadouts, but Apple isn't standing still either, providing quite a performance boost with the recently launched beta of macOS 27.

AMD should've thought more about the positioning of this marketing piece, especially seeing as it's the Neo the company is going after. The truth is that, as far as gaming is concerned, you shouldn't expect much out of any budget-friendly laptop – even if what you get from a Windows 11 machine is obviously a better deal, thanks to the mentioned game compatibility with no need for emulation tricks.

• Researcher Paul found RCE via MITM in AMD’s auto‑updater, but bounty denied
• AMD imposed extended embargo, later changed disclosure rules after criticism
• Security community pushed back, saying new policy discourages transparency and undervalues researchers

A security researcher discovered a remote code execution (RCE) vulnerability in an AMD product, but the company allegedly denied him the bug bounty it promised for such findings.

In February 2026, a researcher called Paul discovered a potential RCE flaw via a man-in-the-middle attack (MITM) in AMD’s auto-updated software. He reported it to AMD and published a blog post about his findings.

However, AMD said MITM attacks are not covered by the bounty (despite this being an RCE flaw) and asked the researcher to pull the blog offline, which he did.

Google files a lawsuit

The company asked for a 100-day embargo on breaking the news, since additional tools were allegedly vulnerable as well. That embargo later ended up being 124 days, significantly longer than the usual 90-day window.

In its writeup, Tom's Hardware argues this alone merits reconsideration over denying the $10,000 bounty reserved for such flaws.

AMD addressed the issue by reengineering the download code in the autoupdater, but then another issue arose: the updater was actually broken and unable to update itself.

To make matters worse, after news broke that it denied the researcher the bounty, AMD allegedly updated its bug bounty disclosure rules to extend the non-disclosure requirements to cover bugs deemed out of scope. According to TechSpot, critics “immediately pointed out it appeared to be a direct response to the public criticism rather than a pre-existing policy.”

The same publication also said that the security community “pushed back hard”, since the change effectively “tells future researchers that even if a bug falls outside bounty scope, they cannot immediately disclose it publicly, removing one of the only tools researchers have to pressure companies into taking their findings seriously.”

On Reddit, the community discusses if AMD “values the researchers who bring it critical vulnerabilities”.

• AMD's RX 9070 XT GPU is finally in the Steam hardware survey
• As of May 2026, it's suddenly the most popular AMD GPU
• It has a 1.33% market share, but Nvidia is still way ahead of AMD overall

While Nvidia's current focus certainly isn't on its gaming graphics cards, it still has the desktop GPU market firmly in its grasp — although things might be looking up for rival AMD.

As reported by TweakTown, the AMD Radeon RX 9070 XT is suddenly the most popular AMD GPU, according to Valve's latest Steam Hardware Survey. As of May 2026, the 9070 XT has leapt to 1.33% of Steam's market share, placing it above the likes of Nvidia's GeForce RTX 4070 Ti (whereas previously the AMD card failed to register at all in the survey).

As always, it's worth noting that Valve's Steam surveys are only a limited perspective on the whole market. Not everyone runs Steam, of course, and the survey is limited to a relatively small slice of gamers on the platform. That said, Valve's survey results are an interesting glimpse into the state of hardware and software popularity.

The RX 9070 XT's popularity shouldn't come as a huge surprise, considering its raw performance capabilities, with the additional benefit of FSR 4 (FSR Redstone) providing solid image quality and performance when upscaling.

What's more surprising is why this GPU wasn't present in Steam's GPU rankings at all before now. TweakTown theorizes that the sudden appearance of the 9070 XT could be down to recent updates to the survey regarding how GPUs are classified, which is one possibility.

Still a one-sided battle

Since the GPU market is still in disarray due to the RAM crisis and its impact on video RAM, both the Radeon RX 9070 XT and 9060 XT (which is also now present among Valve's May survey results) are arguably the most value-packed GPUs available. They're not as cheap as Nvidia's low-end RTX 5000 series GPUs, but these Radeon graphics cards are still affordable and provide much better performance.

Still, it's hard to see Nvidia's GeForce GPUs falling out of favor anytime soon, and they remain clearly dominant in the desktop arena. Team Green's DLSS upscaling technology remains superior to AMD and Intel's equivalents, and it really impresses on the image quality and performance fronts.

Nvidia's RTX 3060 is still leading as the most popular GPU, as it has for years in the Steam survey results, but it's good to see Team Red's latest GPU line-up putting up a fight finally — even if the competition is still very one-sided.

• AMD's David McAfee expects DDR5 RAM prices to 'recover slowly in the future'
• However, the exec added that DDR5 won't return to 'normal' pricing levels for some two years
• Meanwhile chatter from the rumor mill claims Lenovo is going to implement another sizeable round of price hikes with its PCs

An AMD exec has told us that DDR5 RAM pricing isn't likely to normalize until 2028, and in other news, fresh from the memory and wider component crisis, Lenovo seemingly has more price hikes in the works for its PCs.

Let's start with the AMD story. VideoCardz flagged up the interview that 4Gamers conducted at Computex 2026, in which the Taiwanese site spoke with David McAfee, who is VP and general manager of Ryzen CPU and Radeon Graphics.

When questioned on the memory shortage (bearing in mind translation issues with the article), McAfee said he expects prices to 'recover slowly in the future' but that DDR5 RAM won't return to normal pricing levels for about another two years.

VideoCardz also highlighted a post on X from leaker Harukaze5719, which in turn points to a report from Sina Finance in China that tells a separate tale of the Lenovo price hikes.

Obviously, this is just whispers from the rumor mill, but the theory is that Lenovo plans to jack up prices in China by the equivalent of $150 in July. This applies to its PCs and laptops, and, in fact, to every product line Lenovo sells, we're told (though common sense would indicate that cheap peripherals, like a basic Lenovo mouse, won't get a $150 price hike).

While this is a (rumored) move over in Asia, it would clearly make sense that mirroring cost increases in the same order (perhaps even a bit more) would be implemented in other regions.

Analysis: more urgency for laptop buyers?

The AMD exec has essentially thrown his lot in with the forecasters who see the RAM crisis ebbing away during 2028. While that still means it has a fair way to go, this is actually a relatively optimistic prediction compared to some more recent thoughts – notably from Nvidia's CEO. Jensen Huang said the memory crisis will last "quite a few years", suggesting that we may have to be patient until 2029 or 2030 to see the worst of this over. Others have guessed that we'll be sailing rough waters with RAM pricing until 2030, too.

The sad truth is that at this point, if I could have a guaranteed end-in-2028 outcome for these RAM woes (somehow), I'd take that prospect – and run. Of course, there's the question of what McAfee means exactly by returning to 'normal levels', complicated by the fact that this is a translated interview. Broadly speaking, I assume this means that prices will drop back down to more palatable levels relative to now – though I don't see them ever returning to pre-crisis levels, frankly.

It's a bleak old outlook, for sure, although there is some hope in Chinese memory chip makers tackling the RAM shortfall to some extent with increased production from these Asian giants. McAfee alludes to this when mentioning Changxin Memory in China increasing its DDR5 production capacity, but others – notably Huang – seem distinctly unswayed by such arguments.

A lot, of course, could rest on what happens with AI going forward. However, with predictions of increased memory demands one way or another in that sphere – and no remote indication of the AI 'bubble' bursting, or clever tech innovations helping out as much as some folks think they might – I'm not optimistic on the whole.

On top of this, the news from Lenovo is a further blow. If you recall, the computer giant already hiked PC prices earlier this year, and so this is another helping of RAM crisis-related cost rises for laptops and desktops (and other hardware besides, apparently).

As I've been discussing recently, after the Nvidia CEO's recent end-of-decade prediction for the extent of the memory crisis, now really does seem to be the right time to buy a laptop. The news that Lenovo may be about to pull the trigger on yet more substantial PC price rises only puts more urgency into that theory, because this surely won't happen in isolation among PC vendors.

• AMD CEO Dr Lisa Su praises AI capabilities and promise at London Tech Week 2026
• ‘We're here to use technology to solve some of the world's most important problems, to do things we never thought were possible’, she says
• But Su also cautions, “we are still so so early in the process”

AMD CEO Dr Lisa Su has hailed the early progress made by AI in just a few years, but also warned the technology still has some way to reach true usefulness for everyone.

"I love talking about AI, because it has so much capability and promise,” Su told attendees at London Tech Week 2026, “if you look at the last few years there has been so much progress, and we’ve clearly seen AI go from what was research initiatives to something that everybody is using.”

“But what I like to say is that we are still so so early in the process...I think we're early in the research and development phase, as we think about new models and where they’re going, we’re early in AI for science…in terms of what AI can do to really push the frontiers of science, and we’re still very early in what AI can do for enterprise - I can tell you that every month we’re seeing in enterprise, that we’re taking AI from something that we experiment with, to something that actually changes the way we do business.”

"Compute actually equates to intelligence"

Su was speaking at London Tech Week, where she also revealed AMD would be “substantially” increasing its investment in the UK.

The company plans to spend up to £2billion over the next five years, including support across the fields of advanced computing, scientific research and workforce development, as it looks to support what Su called the “incredibly vibrant ecosystem” in the UK.

“Overall, we want to invest in the UK as it's good for business overall for AMD,” she declared.

Elsewhere, Su noted how AMD's goal is, "to build the highest performance chips", and how the company was using AI extensively through our own research and development to achieve this.

“What I’m very passionate about is ensuring that the AI infrastructure is enabling capability that everybody in the world has access to,” she noted, “we’re in this phase where compute is actually the foundation - I like to say compute actually equates to intelligence - and so when I sit in a room with entrepreneurs and dev people, everyone’s like, ‘you know, I’d like more compute’!”

“I think compute is a foundation - but I think what’s important, and what we’re learning, is that there is no one type of compute that will satisfy every AI application - in fact, we need a whole host of compute, whether you’re talking about the latest accelerators, you’re talking about Agentic AI, the CPUs, or general AI infrastructure in terms of networking - that’s what this is about, because you need the entire ecosystem to come together and collaborate across all of those foundational elements - that’s what we really believe at AMD.”

“My view is we're here to use technology to solve some of the world's most important problems, to do things we never thought were possible,” Su added.

“AI is the tool that helps each of us become a better researcher, or a better engineer, or a better entrepreneur…technology is only as useful as the problems we solve.”

“It's an exciting time, but it's important for us to realise it's also an early time.”

Computex 2026 happened this week, and the computing extravaganza brought with it a raft of interesting hardware revelations, questions and commentary. As well as the shiny new gadgets, some of the speeches or interviews given by key players in the tech industry made us all sit up and take notice — and below we've picked out the five most pertinent quotes that tell us about the future of computing.

From Qualcomm's boss through to Nvidia's leather jacket-toting chief, we heard some weighty utterances, and there was no shortage of talk around one particular subject. AI wasn't the only topic mentioned regarding the future of computing, though — thankfully — as you'll see, but there's no denying it was a very prominent subject indeed.

1. "Resistance is futile" (Cristiano Amon, Qualcomm)

No, the Borg didn't turn up at Computex 2026 this year (though sources tell me there's a mysterious cube-shaped booth booked for 2027), but rather it was Cristiano Amon, the chief executive at Qualcomm, who told us that "resistance is futile".

What should we not waste our energy resisting, then? Well, predictably enough it's the inevitable advance of AI, and more specifically, Amon sees the irresistible force that'll pervade our lives as AI agents.

As The Register reported, the Qualcomm CEO believes that in the future, these agents – fancy AI bots that will automate tasks, do things for us, and organize our lives – will have their tendrils everywhere, in all our devices.

From your phone to your earbuds to your smart glasses (if big tech's campaign to make the latter an acceptable gadget succeeds), all of this hardware will feed data of one kind or another to the AI agent.

Amon explained: "The agent isn't tied to the device, it actually moves with the user. It's there with the user, regardless of the device that you have. Once you understand that change, you understand how the whole mobile industry is going to change."

In short, we're all going to turn into walking sensors and cameras, with 6G connectivity enabling all this to be piped back to AI which will gorge itself on this data – all for our benefit, of course. (Ahem – what could possibly go wrong?)

The 'camera' reference isn't mine, by the way, as Amon explicitly stated: "If you have smart glasses, they see what you see, so the connectivity needs to enable a very fast uplink. 6G is going to make all of us into walking cameras in this world."

Don't want your every waking moment, and probably your sleeping moments too, to be shadowed by AI? Hate this idea with every fiber of your being? Well, tough, because Amon explains that this vision of the future – which is seemingly inevitable, as this is where the "resistance is futile" remark was made – can only work by leveraging the power of all our devices.

These sophisticated agents can't be run only in the data center, and so future devices— with ever-more powerful CPUs, and of course NPUs for local AI acceleration — will take the load off with on-device processing in various degrees. And unsurprisingly, Qualcomm's hardware, in various forms, will figure in this dystopian-sounding future of omnipresent surveillance.

2. "Our goal is to deliver unmetered intelligence to every home and every desk with Windows" (Satya Nadella, Microsoft)

Qualcomm's chief exec wasn't the only CEO to be heavily pushing AI, and to no one's surprise, Satya Nadella got in on the act here. As part of Computex 2026 – or GTC 2026 Taipei therein, strictly speaking – Microsoft and Nvidia got together for the big reveal of Team Green's RTX Spark Arm-based 'superchip' which is set to drive the reinvention of the Windows PC, no less.

I'll come on to Nvidia's big quote on this project next, but Microsoft had two telling soundbites for us, one of which was from Nadella, who observed: "Our goal is to deliver unmetered intelligence to every home and every desk with Windows. RTX Spark marks a real breakthrough towards that vision."

Unmetered intelligence refers to AI, and elsewhere, Microsoft's head of Windows, Pavan Davuluri, talked about a new chapter for Windows PCs, and said: "We are building toward a future where Windows provides a unified foundation for AI, from the device in your hands to the infrastructure behind it."

So, while Microsoft has promised to cut back on AI as part of the big drive to fix Windows 11, that may mean that some Copilot options are pared back from various menus – but it absolutely doesn't mean that Microsoft is minimizing AI within the OS. As we see from these quotes, when it comes to AI agents – with agentic AI being something Microsoft has been pushing in a big way of late – Windows is very much being developed with these firmly in mind.

Much like Qualcomm's Amon, Nadella envisions AI being present in every home — and on every desk — thanks to Windows devices. (And again, are you distinctly not stoked about that idea? Well, don't worry, AI will surely be happy to listen to your feedback).

3. "We're really focused on doing something that is just such a big deal, reinventing the PC after 40 years" (Jensen Huang, Nvidia)

Naturally, Nvidia's boss had some weighty quote material to throw around at Computex 2026, and the highlight for me pertained to the Arm-based RTX Spark CPU (as mentioned above).

Huang said: "We're really focused on doing something that is just such a big deal, reinventing the PC after 40 years."

Of course, this is a collaborative venture between Nvidia and Microsoft, who according to the joint press release for the project, want to "reinvent Windows PCs for the age of personal AI".

It is, of course, a big move for Nvidia to launch a new processor for the PC, the gravity of which is reinforced by the kind of changes Microsoft is making to Windows 11 for this chip. We're talking about specific optimizations in Windows 11 for Nvidia RTX Spark, and not small moves either – sizable changes, some of which advance the cause of all Arm CPUs (like Qualcomm's) considerably with the OS. That includes a drive to support more software with Arm-based Windows 11 devices, and even to improve game compatibility (with some notable anti-cheat stumbling blocks being resolved).

As it happens, Huang's above quote actually came as part of a response to a question about whether a gaming handled might be made with the RTX Spark CPU. As PC Gamer reported, the full quote was: "If somebody wants to do it [a gaming handheld], you know, we'll work with them on it. But right now we're really focused on doing something that is just such a big deal, reinventing the PC after 40 years."

So, the idea of a handheld was brushed aside, as what this is all about, really, is that 'reinvention' of the PC around AI.

As Huang further enthused: "Can we help reinvent the PC? And, boy, if you get a chance to reinvent the single most important instrument, the single most important tool of humanity, what you and I grew up with defined just about everything about our lives, and we have an opportunity after 40 years to go reinvent it for the age of AI. Wow, we're not going to sit around, not let it get done."

When you put this together with Nadella's comment, and Amon's vision of the AI-centric future, there was a clear vibe at Computex 2026: agentic AI is coming, and you might as well accept this now.

Of course, more than most, Huang has a vested interest in promoting this line of thinking. Nvidia's done quite well out of AI so far, I believe.

4. "We know the pressures that are there in building systems aren't going away anytime soon" (David McAfee, AMD)

Not everything about the future of computing at Computex 2026 revolved around AI, of course, and there was some commentary on the dire situation around PC component price hikes — and what might be done about that going forward.

When questioned about why AMD had just released a last-gen 3D V-Cache processor as an alternative budget offering — as a reaction to rising PC build costs — rather than a current-gen Zen 5 model, AMD's VP and general manager of Ryzen CPU and Radeon Graphics, David McAfee, had something pertinent to say.

McAfee teased: "I think as we go through the rest of this year, I mean we're always looking for ways to, you know, create as many options as we can, and that [Zen 5 X3D CPU] may become… maybe something that we look at doing as a runway product later this year, simply because you know we know the pressures that are there in building systems aren't going away anytime soon."

So, we may see a Ryzen 5 9600X3D CPU (or similar) emerge as an affordable engine for a gaming rig later this year.

More broadly, this is a signal that AMD is actively thinking about how to help ease the demands that price hikes on RAM, storage, CPUs and GPUs (and more besides) are exerting on the wallets of consumers. Although it also serves as a depressing acknowledgement that pricing woes aren't going away for quite some time yet.

It was also good to see AMD underline its commitment to the AM5 chipset with support guaranteed through to 2029 at Computex, for those who aren't planning on building a new PC anytime soon. Waiting out the worst of the pricing crisis may well be the best approach, frankly, if that's a viable option in your circumstances.

5. "Longer term, I think something has to give, right? The over-inflation, we will have to keep an eye out" (Nish Neelalojanan, Intel)

Like AMD, Intel also had thoughts on the current PC component pricing nastiness which it shared at Computex 2026.

Specifically, in a chat with Tom's Hardware, Intel's Nish Neelalojanan, who is senior director of product management for client computing, said of the RAM supply crisis and spiking costs: "Longer term, I think something has to give, right? The over-inflation, we will have to keep an eye out."

Neelalojanan elaborated on this, noting that Intel was adapting its strategy to help frustrated consumers in terms of maintaining support for last-gen (DDR4) memory and older (more affordable) CPUs. The exec noted: "We do have products that support DDR4 on both desktop and mobile. Raptor Lake [13th-gen CPUs that first emerged in 2022], we're not end-of-life-ing any of them; they're there. We'll continue to make sure that there are products which can take care of older memory technologies if they're available and cheap."

With laptops, Neelalojanan pointed to Intel's new Wildcat Lake mobile silicon, of which the exec promised "we are validating lower configs", meaning lower-end notebooks starting at 8GB RAM.

Neelalojanan said: "Wildcat Lake is a single-channel product, so there are products which can leverage low memory and give reasonably good performance."

This range of mobile silicon does indeed seem like a promising addition for more affordable laptops, and so it's not all bleak news for the future of PCs in terms of the undoubtedly painful price hikes we've all sadly become accustomed to.

• AMD's David McAfee has hinted that a Ryzen 5 9600X3D could be in the cards for later this year
• This follows the launch of the Ryzen 7 7700X3D as a response to pricing pressures
• McAfee said that "we know the pressures that are there in building systems aren't going away anytime soon"

AMD appears to be considering making a new addition to the family of current-gen Zen 5 CPUs in the form of a wallet-friendly 3D V-Cache (X3D) processor.

Tom's Hardware spoke to AMD's VP and general manager of Ryzen CPU and Radeon Graphics, David McAfee, at Computex, and questioned the exec on why the company decided to release a past-gen X3D chip rather than one from the Zen 5 (Ryzen 9000) range.

The backdrop here is the launch of the Ryzen 7 7700X3D, which has just happened, and Tom's is essentially wondering why Team Red didn't go for a Ryzen 5 9600X3D.

McAfee responded that the launch of such a contemporary CPU "may be something that we look at doing… later this year."

The exec elaborated: "I think as we go through the rest of this year, I mean we're always looking for ways to, you know, create as many options as we can, and that may become… maybe something that we look at doing as a runway product later this year, simply because you know we know the pressures that are there in building systems aren't going away anytime soon."

Analysis: under pressure

Obviously this is far from a confirmation that AMD will produce a new budget X3D chip for Zen 5, but the admission that it's possible the CPU maker could start rolling such a product off the assembly lines is a hopeful nugget that Team Red is still looking to further ease the pressure on PC builders.

Right now, building a new computer is a depressing task consisting of shopping for various components, most of which have had substantial price increases applied. (And the price hikes for certain parts, notably RAM and storage, are positively astronomical in some cases). A wider choice of more affordable processors obviously won't hurt for those carrying out a DIY build, and for that matter, it'll help with pre-built computers too, as another option for PC makers looking at building cheaper gaming rigs.

Of course, McAfee's comments don't come without an element of concern, namely the casual acknowledgement that "we know the pressures that are there in building systems aren't going away anytime soon". In other words, don't expect PC component pricing pressures to ease for quite some time, and that's a prediction that has been widely made elsewhere. Most notably in recent times by the major RAM chip manufacturers – with one especially bleak forecast.

Indeed, AMD's CEO, Lisa Su, recently predicted that demand is going to wane on the consumer and gaming side for Team Red due to "higher memory and component costs" in the second half of 2026. Adding to that is further uncertainty around the global shipping situation, as continued disruption in the Strait of Hormuz is now affecting the RAM supply chain in other ways.

All in all, it's not a pretty picture, and we'll take what relief we can in terms of the prospect of more affordable components.

• AMD has announced it will be supporting Socket AM5 for CPUs through 2029
• It's following in the same footsteps of its decade-long support for Socket AM4
• The AMD Ryzen 7 5800X3D has also returned for the AM4 socket platform

AMD has consistently supported its Socket AM4 for CPUs over the last decade, with new additions to its Ryzen CPU lineup — and, fortunately, its current CPU socket platform is receiving the same treatment.

Announced at Computex 2026, AMD is set to extend its support of the Socket AM5 platform through 2029, after the launch of the new Ryzen 7 7700X3D processor. This comes after Team Red has offered a decade of support for AM4, as the manufacturer adheres to longevity.

It's also worth noting that the popular AMD Ryzen 7 5800X3D has returned as a '10-Year AMD AM4 Anniversary Edition', serving as AMD's way of celebrating its lengthy AM4 support — and frankly, it's one of the best gaming CPUs available, and the best AM4 gaming CPU.

"For gamers, the support extension represents more time to build, upgrade, and improve a system around the same platform," AMD says in its blog post.

"A longer platform roadmap can reduce the need for full system replacements and gives PC builders more confidence when choosing AMD Ryzen processors, AM5 motherboards, and compatible memory."

Both announcements are great news for AM4 and AM5 CPU users; not only is AM5 confirmed to have at least three more years of support, but it's highly likely we'll see irregular cases of new CPUs launch for Socket AM4.

The latter means PC gamers, like myself, using AM4 motherboards can continue doing so comfortably, especially since AMD's AM6 platform doesn't seem to be launching any time soon — and it's probably better if it doesn't release for another few years, at least until this RAM crisis disappears.

Although AMD has been at the heart of the recent components craze, with its stock surging by more than 3,000% in the last few years, the company hasn't been as essential to the AI buildout as Nvidia. However, its leadership in the form of Lisa Su has, for a long time, been attuned to the changing landscape of the industry.

Overcoming bottlenecks

The world of computing was a little different in 2013, when Lisa Su delivered remarks at the International Solid-State Circuits Conference (ISSCC) about the challenges in continuing to scale up processing power.

CPUs were still the dominant component then, but Su, who was then AMD's senior vice president and general manager for its global business units, hinted at a future in which other breeds of computing architectures would come to the fore.

The solution that Su proposed — known as heterogeneous computing — essentially involved combining CPUs with GPUs and specialized accelerators to offload processors onto chips or components that are best suited to efficiently carry them out. With these systems, shared memory pools would allow the different processors to work in harmony.

Building for the AI supercycle

If that system sounds familiar, it's because it's exactly the sort of arrangement that goes into designing systems like AMD's Instinct MI400 family of components or Nvidia's Vera Rubin platform. These are superchips that are integral to the ongoing AI buildout.

Although the rise of AI in the way that it ensued wasn't on the cards in 2013, Su, who was appointed as AMD's chief in 2014 shortly after this presentation, articulated a vision that has been realized with full force more than ten years on.

The traditional era of computing, in which CPUs are the dominant technological force within a computing system, has given way to that more heterogeneous environment — not only in data centers but in consumer devices too — with GPUs, NPUs, accelerators, and other components, all cooperating (and often pooling memory) to run as efficiently as possible.

• AMD has announced its Q1 results, with booming revenue driven by AI
• There's bad news for the gaming division, though, due to 'higher memory and component costs, ' AMD's CEO Lisa Su observed
• AMD's CFO has forecast 'gaming revenue to decline by more than 20%' in the second half of 2026 compared to the first half of the year

AMD just revealed its latest financial results, with good news for investors in the form of a major surge in revenue, but bad news for consumers, with more RAM-related worries looming on the horizon.

Tom's Hardware reports that AMD's Q1 2026 fiscal results witnessed a new record for data center revenue, as the AI boom drove further growth, but CEO Lisa Su warned of PC component price spikes going forward.

Su predicted that demand is going to wane with its client and gaming businesses – essentially the consumer side of AMD's hardware – in the second half of 2026 due to "higher memory and component costs".

So, yes, that means AMD's chief executive believes that after next month, as we head into Q3, RAM and other components are only going to get pricier.

With gaming, the damage done by price hikes could be quite considerable, as AMD's Chief Financial Officer, Jean Hu, observed: "We expect second half [of 2026] demand in gaming to be impacted by higher component and memory cost. We now expect second half gaming revenue to decline by more than 20% compared to the first half."

Analysis: Radeon price hikes – or indeed pricier consoles?

In other words, compared to the first half of the year (of which less than two months now remain – with time flying by, as ever), the second half of 2026 is going to be considerably more sluggish for AMD's gaming revenue. The expectation isn't just a 20% drop, but a more than 20% fall, so that could be a quarter less money raked in, or maybe more, up towards 30%, even.

This would seem to indicate that AMD's Radeon graphics cards are going to be in shorter supply in Q3 and Q4, and that there may be further price hikes on RX 9000 models. Clearly, AMD is expecting things to slow down with these graphics cards as 2026 rolls on, but its gaming revenue isn't just about Radeon, of course – Team Red also makes the semi-custom GPUs for the PlayStation and Xbox consoles.

Sales of those consoles are softening naturally, mind you, given that they're in the later stages of their expected lifespan now, so we're reaching saturation levels for would-be buyers. What could also be factored in here is the price hikes for console hardware, causing further unwillingness to buy – or possibly, AMD is anticipating further PS5 or Xbox price rises later this year, compounding the misery.

That's just guesswork, but clearly the outlook isn't great for the second half of 2026, and this is the latest in a quickfire round of pessimistic RAM crisis predictions, two of which have come from memory chipmakers themselves. Micron has warned of growing AI demand and more pressure on RAM supply, while Samsung has observed that 'significant shortages' of memory will continue to plague us through 2027 as a best-case scenario.

There isn't much faith out there in Reddit-land that RAM pricing will recover anytime soon, or indeed that prices will ever reach the levels we saw last year, before the memory hikes started coming thick and fast.

As one Redditor put it in reaction to this news from AMD: "My prediction is that over 2027 prices will drop from an insane 400% [price increase] to a nice and summerly 200%. But they'll never fall lower than that ever again. The RAM economy will permanently change and data centers will be a nuisance for the rest of recorded time."

• AMD frames openness as essential for multi-vendor space mission architectures
• Vendor lock-in risks increase significantly in long-duration orbital deployments
• Modular systems improve flexibility across complex, multi-supplier mission environments

AMD has outlined its vision for AI in space, claiming to be a champion of open platforms and modular design in a market where monolithic solutions dominate.

The company argues no single vendor can or should dictate the full solution for space missions, which are often the work of multiple firms.

"Space missions are assembled from many specialized suppliers, and no single vendor can (or should) dictate the full solution," AMD stated in a recent announcement.

AMD pushes open platforms for space

AMD’s argument rests on the structure of the space industry itself. Missions typically combine hardware, software, and subsystems from multiple contractors.

This makes interoperability a requirement rather than a preference, as components must work together across different vendors.

In that context, proprietary platforms risk introducing dependencies that can limit flexibility or complicate long-term operations.

The company is leaning into open standards and modular design to reduce that friction.

Its strategy centers on enabling partners to integrate and validate systems across vendors without being tied to a single ecosystem.

That includes investment in open approaches to security, interconnects, and infrastructure, alongside its ROCm software stack for AI and high-performance computing.

ROCm is intended to give developers a pathway from low-level kernels to full applications on AMD accelerators.

More importantly, it represents an alternative to tightly controlled software ecosystems that dominate AI development today.

Why space amplifies the need for modular systems

AMD ties its openness strategy directly to the realities of operating in orbit. Space systems face strict power and thermal limits, intermittent communication with Earth, and long mission lifecycles.

These constraints make adaptability and resilience more critical than in most terrestrial deployments.

In such environments, reliance on a single vendor can introduce risk. If a component becomes obsolete or unsupported, replacing or upgrading it is far more complex than in ground-based systems.

AMD’s position is that modular, interoperable architectures allow mission designers to swap, upgrade, or validate components more easily over time.

The same logic extends to onboard AI. With limited bandwidth and communication windows, spacecraft increasingly need to process data locally.

AMD argues that open platforms make it easier to deploy and evolve these capabilities across heterogeneous hardware, rather than locking missions into a fixed stack from launch.

Openness alone may not be enough

The challenge for AMD is that the space market has historically rewarded proven reliability over architectural philosophy.

Competitors already have deep relationships with space agencies and, in some cases, purpose-built hardware designed specifically for radiation-heavy environments.

AMD points to its existing track record, including contributions to image processing for NASA missions.

However, extending that experience into large-scale AI infrastructure in orbit is a different step.

For now, AMD is making its case early, framing openness as not just a design preference but a requirement for resilience in space.

Whether that argument translates into contracts will depend less on philosophy and more on execution in an environment where failure is not easily tolerated.

• Most companies are planning, piloting or deploying AI PCs now
• High-performance NPUs are "critical" – performance and security benefits are key
• The most popular use cases show AI PCs being most useful for knowledge workers

New data from AMD claims AI PC adoption has increased rapidly lately, with over four in five organizations either planning, piloting or deploying AI PCs in some form.

This shows a shift from experimentation to real deployment, with enterprises set to embed AI into everyday workflows beginning with the right hardware that's capable of running some workloads locally.

Already, 70% report faster performance and reduced latency when they're using dedicate AI PCs, with 59% now seeing high-performance NPUs as "critical."

Are AI PCs the foundation to good AI deployment?

According to AMD, two in every three companies noted increased employee productivity from using AI PCs as computers shift to become execution platforms rather than just simple productivity tools.

And this has never been more critical than today, in the age of always-on AI through agentic capabilities. AMD described the latest generations of PCs as a "local execution later" for real-time, secure task processing, citing lower latency, better privacy and security, and the possibility of reducing cloud costs as key benefits.

Still, respondents saw PCs acting as an interface for cloud agents (51%) and as a secure local execution hub (47%) in near-equal measure.

Moreover, 81% of the study's participants claimed meaningful knowledge of AI PCs. As a result, while the desire to futureproof devices is a part-driver in the reason to upgrade, most companies see productivity gains, innovation, a competitive advantage and security improvements as higher priorities.

As for real-world uses (27% have already deployed them at scale), AI PCs are proving most popular among knowledge workers, helping boost efficiency across tasks like document, presentations, spreadsheets, meeting transcription and summarization. In fact, coding assistance ranks lower than image and video generation, likely due to workers preferring online Copilot tools that are cloud-first.

It's clear from AMD's research that AI PCs are shifting from a 'nice-to-have' to being a business staple. However, AI PCs aren't set to replace traditional cloud-based AI as we enter a hybrid era of local and cloud processing for the best balance of performance and security.

When it comes to system performance, stability and compatibility, it is essential to understand how gear modes work. It is gear modes that are responsible for defining the speed ratio between a CPUs integrated memory controller (IMC) and system RAM.

Intel has its own gear modes to manage super-fast memory speeds but, in this article, we’re looking specifically at how AMD gear modes work and why DDR5-6000 is often called the sweet spot.

AMD Gear Modes are the clock speed ratios of the integrated memory controller (IMC) inside the CPU to the clock speed of the memory modules.

Essentially, they dictate how fast the CPU's memory controller communicates with the RAM. For users for whom speed and performance are essential, this enables greater flexibility in achieving higher memory speeds and compatibility with various memory kits.

There are three key elements in AMD systems that determine memory performance:

1. Memory Clock (MCLK) – The speed of the memory where active data is temporarily stored for the processor.

2. Internal Memory Controller Clock (UCLK) – The speed of the processor’s memory controller that manages the communication between memory and the CPU.

3. Infinity Fabric Clock (FCLK) – The speed of the interconnect that links and coordinates data flow across the entire processor.

Any intensive computing workload, whether it’s gaming, video editing, 3D rendering, or AI processing, will depend heavily on how the three AMD clocks stay in sync. To manage this, AMD uses two operating modes: Gear 1 and Gear 2.

These modes define the relationship between the Memory Clock (MCLK) and the Memory Controller Clock (UCLK). If users have a clear understanding of how the two gears interact, this allows them to select the memory kit speed that will deliver the best performance for their needs.

Explaining clock speeds

MCLK – this is the internal clock speed of the memory, measured in megahertz (MHz). Today’s memory technology is all based on DDR (Double Data Rate), which transfers data twice per clock cycle. This means an internal clock frequency of 3000MHz translates to a memory speed of 6000MT/s (Megatransfers per second), or DDR5-6000.

UCLK – the memory controller’s clock speed. It runs either at the same speed as MCLK in Gear 1 or at half the MCLK speed in Gear 2.

FCLK – the clock speed of AMD’s Infinity Fabric interconnect, which typically defaults to 2000MHz unless manually adjusted. One important point for users is that unless they change their settings manually, the system’s BIOS will automatically select the appropriate parameters for the selected memory speed.

Gear 1 (1:1 ratio)

- IMC and the memory operate at the same frequency.

- Infinity Fabric operates at 2000MHz

• Best for gaming or latency-sensitive workloads
• By keeping the memory and IMC ratio at 1:1, you’re able to minimize latency
• DDR5-6000 is the sweet spot for AM5 systems since it is generally the highest speed that can achieve Gear 1 out of the box without adjusting additional overclock parameters
• ex: DDR5-6000 (3000MHz) → IMC runs at 3000MHz → Fabric runs at 2000MHz

Gear 2 (2:1 ratio)

- IMC operates at half the memory frequency

- Infinity Fabric operates at 2000MHz

• Best if you need memory bandwidth for AI, video editing, 3D rendering or any applications where throughput is more important than responsiveness
• Required for higher-speed memory (DDR5-6200+)
• 2:1 ratio allows for more stability, but may introduce additional latency
• Generally, DDR5-6400 to DDR5-7200 speeds will offer higher bandwidth (allow more data to move to and from the memory and processor at a time), but increased latency offsets the gains
• DDR5-7600 and higher is where we start to see the bandwidth overcome the latency penalty
• Example: DDR5-8000 (4000MHz) → IMC runs at 2000MHz → Fabric runs at 2000MHz

Benchmark results

The following AIDA64 benchmark results show how different memory speeds are affected by Gear Modes:

- Benchmark system

• Motherboard: ASUS ROG Crosshair Hero X870E (BIOS v1512)
• Processor: AMD Ryzen 9 9950X

Benchmark Results (All capacities: 32GB / 16GB × 2)

1. 6000 MT/s, CL30, Gear 1

• Read: 77,703 MB/s
• Write: 77,631 MB/s
• Copy: 67,837 MB/s
• Latency: 77.7 ns

2. 6400 MT/s, CL32, Gear 2

• Read: 80,231 MB/s
• Write: 84,374 MB/s
• Copy: 71,409 MB/s
• Latency: 79.9 ns

3. 7600 MT/s, CL38, Gear 2

• Read: 82,931 MB/s
• Write: 93,378 MB/s
• Copy: 75,115 MB/s
• Latency: 78.5 ns

4. 8000 MT/s, CL38, Gear 2

• Read: 85,302 MB/s
• Write: 94,818 MB/s
• Copy: 77,864 MB/s
• Latency: 75.7 ns

Important to note is that an emerging new sweet spot for systems using the Ryzen 9000 series is 6400MT.

As memory speed increases, so does memory bandwidth, but users should be aware that there’s a trade-off. When switching to Gear 2, latency rises noticeably. In fact, latency remains worse than a DDR5-6000 kit until you reach speeds around DDR5-8000, where the extra bandwidth finally offsets the higher latency.

Bringing it all together

For most gamers and everyday users, DDR5-6000 in Gear 1 offers an excellent balance of bandwidth and low latency, translating into jitter-free gameplay, higher FPS and better overall responsiveness.

However, for users whose workload involves bandwidth-heavy tasks like video editing, 3D rendering or AI processing, higher-speed memory running in Gear 2 will be significantly more advantageous. A good rule of thumb for those requiring maximum bandwidth and the lowest-possible latency, is to aim for DDR5-7600 or higher.

By clearly understanding how AMD Gear Modes affect performance, users are empowered to choose the right memory configuration and unlock the full potential from their system.

We've featured the best business computers.

This article was produced as part of TechRadarPro's Expert Insights channel where we feature the best and brightest minds in the technology industry today. The views expressed here are those of the author and are not necessarily those of TechRadarPro or Future plc. If you are interested in contributing find out more here: https://www.techradar.com/news/submit-your-story-to-techradar-pro

• Codasip sale fuels speculation about GlobalFoundries expanding RISC-V processor capabilities
• Studio licensing strengthens case for customized silicon alongside manufacturing services
• Growing RISC-V portfolio raises possibility of future competition with Arm and AMD

Codasip, a Munich and Bristol-based semiconductor design company that develops RISC-V processor cores, is selling part of its business to an unnamed public U.S. semiconductor company, with speculation that the buyer could be none other than GlobalFoundries, AMD's former foundry arm.

Codasip confirmed the company will divest its low-end RISC-V processor design unit while directing future work toward security-focused chip architectures and system-level products.

That lower-end division includes processor cores used in embedded electronics, automotive systems, and industrial hardware, where efficient and customizable designs are widely used.

A three-part structure

GlobalFoundries is seen as the most likely buyer because it has spent the past year acquiring RISC-V processor technology through multiple deals.

The company already owns MIPS and has agreed to acquire ARC-V processor intellectual property from Synopsys, expanding its reach into higher-performance and specialized processor designs.

Adding Codasip’s entry-level RISC-V cores would extend that portfolio into simpler embedded designs, creating coverage across both low-power and more advanced processor tiers.

The agreement also includes a broad license to Studio, Codasip’s processor development software that allows customization of processor instructions.

That capability allows GlobalFoundries, which began as AMD’s manufacturing arm before becoming an independent semiconductor manufacturer, to support tailored chip development, giving customers the ability to modify processor behavior instead of relying only on fixed designs licensed from companies such as Arm.

Arm has long dominated embedded processor markets through royalty-based licensing, collecting fees from companies that build chips using its architecture.

A manufacturer that offers both fabrication services and tailored processor designs creates an alternative path, especially for companies developing automotive, industrial, and edge computing systems.

Continued expansion into processor technology could eventually place GlobalFoundries into competition with established chip designers.

Codasip described the sale as part of a broader change in direction toward what it calls cyber-resilient semiconductor architectures.

Its future development work centers on CHERI, short for Capability Hardware Enhanced RISC Instructions, a technology focused on limiting software exploits by enforcing stricter memory access rules directly in hardware.

“Cyber-resilience has become a strategic requirement for governments, infrastructure operators, and technology providers worldwide,” said Ron Black, Chief Executive Officer of Codasip.

“Traditional approaches inefficiently bolt security onto systems after the fact. Our focus is on enabling partners to build security into the fundamental architecture of compute systems from the beginning.”

The transaction is expected to close within about a month, and we'll know then whether GlobalFoundries is indeed the buyer of Codasip's low-end RISC-V processor design unit.

• AMD introduces dual V-cache processor aimed at heavier mixed workloads
• Zen 5 architecture, paired with expanded cache, drives a new performance focus
• AMD Ryzen 9 9950X3D2 Dual Edition records modest gains across creative software and compilation tasks

AMD has introduced its most advanced desktop processor so far, the Ryzen 9 9950X3D2 Dual Edition, with cache size now forming a central part of its performance strategy.

Jack Huynh, the Senior Vice President and General Manager of the Computing and Graphics group at AMD, claims this is “the world’s first dual AMD 3D V-Cache technology enabled processor.”

The processor is built on the Zen 5 architecture and integrates 16 cores with a total cache of 208MB.

Dual cache and real workload impact

AMD’s 3D V-Cache design stacks additional memory directly on top of the processor die, reducing the distance data must travel.

This approach is intended to improve latency and throughput in demanding applications.

With both chiplets now using this design, the company claims a measurable shift in responsiveness rather than a marginal upgrade.

According to AMD, the chip delivers between 5% and 10% over its predecessor in workloads such as DaVinci Resolve, Blender, and large code compilation projects.

These gains are not dramatic on paper, but they suggest incremental refinement rather than a complete redesign.

“Expanded on-chip memory helps reduce latency and keep demanding workflows responsive, from complex timelines and large datasets to parallel builds and simulations,” AMD said in a YouTube video.

However, the launch of this processor raises questions about how close desktop chips are getting to workstation-class products like Threadripper and EPYC.

Those platforms have traditionally been defined by higher core counts and broader scalability, but they also come with higher costs and platform requirements.

By combining high core counts with large cache reserves, AMD appears to be reducing the practical differences for certain users.

Tasks such as rendering, simulation, and software builds are increasingly feasible on a mainstream desktop platform without moving to specialized hardware.

Alongside the processor itself, AMD continues to build around the AM5 platform, maintaining its emphasis on long-term compatibility across multiple generations.

This approach reduces the need for full system replacements when upgrading processors, extending a pattern that was already established during the AM4 era.

The new chip fits within that same framework, offering an upgrade path that does not require a complete rebuild.

This may carry practical importance for users maintaining systems over longer cycles, as AMD describes this processor as an effort to reduce trade-offs between gaming performance and productivity workloads, particularly with the introduction of dual-chiplet 3D V-Cache.

The chip is set to launch on April 22, 2026, when these performance claims will be tested against real-world use.

• PlayStation Pro architect Mark Cerny has confirmed that FSR Frame Generation will be coming to PlayStation "at some point"
• Cerny says he's "very happy with how that work is progressing"
• He adds that Sony has no more major releases planned for this year

Sony has confirmed that PlayStation will soon receive AMD's FSR (FidelityFX Super Resolution) Frame Generation technology.

That's according to PlayStation 5 and PlayStation 5 Pro console architect Mark Cerny, who discussed Sony's Project Amethyst collaboration with AMD with Digital Foundry and shared that the company's FSR tech will be coming to "PlayStation platforms".

"Just to clarify a few things about the collaboration with AMD, the new PSSR uses the same core co-developed algorithm as FSR Redstone's Upscaling (to avoid confusion, I'll use the new names today rather than FSR4)," Cerny said.

"FSR Frame Generation is also based on co-developed technology (or as my good friend Jack Huynh puts it, 'co-engineered technology'). I’m very happy with how that work is progressing, and an equivalent frame generation library should be seen at some point on PlayStation platforms."

When asked what these platforms would be, Cerny couldn't say, nor could he determine which machine-learning-based innovations would come to PS5 Pro, or which could arrive with the PS6. However, he did confirm no more major console update will arrive in 2026.

"Great questions, particularly considering that FSR Frame Generation is technology that was co-developed between SIE and AMD, we're intimately familiar with it," Cerny said. "All I can say is that we have no more releases planned for this year. And that I look forward to discussing this more in the future!"

In the same interview, Cerny revealed that Sony’s updated PSSR (PlayStation Spectral Super Resolution) AI upscaling technology is faster than its original incarnation.

Cerny said that the new rendition is around "100 microseconds faster" than the original and allowed Sony to introduce an improved image quality option to "force-upgrade all PSSR-supported games."

The new PSSR tech is now available for a range of games, including Resident Evil Requiem, Silent Hill f, Monster Hunter Wilds, Alan Wake 2, Final Fantasy 7 Rebirth, and more. It will also be available for Crimson Desert, Cyberpunk 2077, and Assassin’s Creed Shadows soon.

• Agentic AI enables PCs to autonomously execute multiple tasks in parallel
• Persistent local AI reduces reliance on cloud computing for sensitive workflows
• Professionals can delegate urgent tasks and wake to completed project summaries

The personal computer has been central to work and creativity for four decades, allowing users to write, build, design, and analyze with professional-grade tools - but PCs have largely remained tools operated directly by humans, opening apps and performing tasks manually.

However, AMD now suggests that Agentic AI could be the killer app for even decades-old PCs, transforming them into systems that autonomously execute tasks and manage workflows.

Unlike traditional software, these AI agents can operate persistently, moving through information, drafting messages, compiling data, and summarizing key insights without constant human intervention.

Agents transform task management

In practical terms, professionals can wake up to find urgent communications addressed, briefing documents assembled, and project summaries ready for review — work that previously required hours of human attention.

Agent Computers differ from standard PCs because they are not directly operated for each action.

A user delegates responsibilities, and the agents execute tasks in parallel, continuously, and autonomously.

Dropping a request into Slack or sending a message triggers the agent to perform follow-up work, research information, and compile results in real time.

AMD argues this approach amplifies human capability rather than replacing it, allowing creators, developers, and professionals to focus on higher-value work while the machine manages logistics.

The combination of local AI persistence and autonomous execution is essential for everyday usability, particularly where privacy, cost efficiency, and control over sensitive data are required.

Running AI agents effectively demands high-performance components capable of supporting parallel workloads.

AMD’s Ryzen AI Max+ processors, including the AI Max+ 395, are designed for persistent local models and multi-agent environments.

These processors provide the bandwidth, memory capacity, and compute efficiency necessary for a system that is always on and capable of handling multiple simultaneous AI tasks.

Systems like the Framework Desktop or AMD Ryzen AI Halo are cited as practical examples of machines built to serve as Agent Computers.

The Agent Computer functions differently from regular PCs — it works for the user, not simply with them.

AMD sees this as a new evolution in personal computing where AI agents operate independently yet locally, ensuring tasks are executed efficiently while maintaining data privacy.

However, widespread adoption of this system will depend not only on hardware capability but also on software reliability, cost, and user trust in persistent AI systems.

The Agent Computer concept makes AI agents central to modern computing, requiring high-performance processors, fast memory, and always-on systems.

AMD argues that combining such hardware with agentic software extends traditional PCs into autonomous, persistent task management for users and organizations.

• Huawei Atlas 950 and TaiShan 950 SuperPoDs unveiled globally at MWC 2026
• Atlas 950 links up to 8,192 Ascend NPUs into a single logical AI system
• Huawei enters AI infrastructure race with full-stack platform to rival Nvidia and AMD

At MWC 2026, Huawei unveiled its Atlas 950 and TaiShan 950 SuperPoDs to a global audience for the first time, expanding its largest AI computing clusters beyond the Chinese market.

AI models are now measured in trillions of parameters, and agentic systems are beginning to run inside real production environments. Scaling those workloads by simply adding more servers is becoming inefficient, with coordination overhead and latency limiting performance in very large clusters.

Atlas 950 is built around up to 8,192 Ascend NPUs connected through Huawei’s UnifiedBus interconnect. Instead of operating as thousands of loosely linked accelerators, the system is engineered to behave as a single logical computer, reducing communication delays between processors during large training runs.

TaiShan 950 SuperPoD

At full configuration, the system is rated for up to 8 exaflops of FP8 performance and 16 exaflops in lower-precision formats.

It spans roughly 160 cabinets across close to 1,000 square meters, supports more than a petabyte of memory, and delivers 16.3PB/s of interconnect bandwidth.

That level of scale targets large model training and high-throughput inference workloads.

The TaiShan 950 SuperPoD, also shown off at MWC 2026, extends the same architectural approach to general-purpose computing, targeting enterprise data center workloads beyond dedicated AI training.

TaiShan 500 and TaiShan 200 servers round out the portfolio at lower performance tiers.

The global debut places Huawei in direct competition with Nvidia’s DGX SuperPOD and NVL platforms, as well as AMD’s forthcoming MegaPod systems built around Instinct accelerators (see how they compare here).

Nvidia’s advantage lies in its long-established CUDA software platform and GPU clusters that are already widely deployed in research labs and enterprise data centers.

Atlas 950 runs on Huawei’s Ascend AI chips and works with CANN, its open-sourced compute architecture that supports frameworks such as PyTorch and Triton.

That combination gives developers a way to build and run AI workloads without relying on Nvidia’s CUDA platform, offering an alternative path for large-scale AI systems.

By bringing Atlas 950 to MWC’s global stage, Huawei is presenting itself not just as a chip designer, but as a builder of complete AI computing systems competing at the highest end of the data center market.

• AMD reveals its Ryzen AI 400 series at MWC 2026 as the successor to the Ryzen 8000G range
• These chips have been refocused on AI, with an NPU that hits 50 TOPS
• It means a desktop PC will qualify for Windows 11's Copilot+ features

Over at MWC 2026 AMD has revealed new Ryzen AI processors designed to bring a powerful NPU – and Copilot+ (AI) features – to desktop PCs.

As TechSpot flagged up, the Ryzen AI 400 series (which comes alongside Pro variants for businesses) is built with Zen 5 CPU cores (as with Ryzen 9000), along with an RDNA 3.5 integrated GPU and an XDNA 2-powered NPU for accelerating on-device AI tasks.

The idea is to provide a relatively wallet-friendly all-in-one processor for more budget-oriented PCs which don't have a discrete graphics card, bringing in the ability to use AI features with an NPU that's rated at 50 TOPS. That's fast enough to qualify as a Copilot+ PC and to access Windows 11's AI trappings therein, such as Windows Studio Effects for improving video chats.

The flagship model is the Ryzen AI 7 450G, which offers eight cores (16 threads), 24MB of cache, and boosts up to 5.1GHz, packing Radeon 860M graphics. The latter is RDNA 3.5 and has eight Compute Units (CUs).

There are also Ryzen AI 5 chips, the 440G and 435G, which have six cores (12 threads) and boost up to 4.8GHz and 4.5GHz respectively (with 22MB and 14MB cache). They step down to a Radeon 840M integrated GPU with four CUs, but all chips have the full-fat NPU with (up to) 50 TOPS.

The TDP of these processors is 65W, but they also come in low-power versions (GE models) that only use 35W.

According to TechSpot, AMD (and other sources) have said that these Ryzen AI 400 desktop models won't be sold as boxed standalone products, at least not initially – they'll be for PC makers (OEMs) instead.

In other words, you'll likely only be able to buy prebuilt machines with these CPUs, with those PCs expected to debut in Q2 2026. However, later on, these chips should be available directly on retailer's shelves.

Analysis: It's all about that AI

As mentioned, these processors are targeted for budget builds with no GPU, or mini PCs, or indeed office computers (there are Ryzen AI Pro 400 versions as mentioned). If you're wondering where they fit in AMD's range of silicon, they're replacements for the old Ryzen 8000G offerings.

So, they could theoretically be Ryzen 9000G chips, but due to the new AI slant – and that much more powerful NPU for the desktop – AMD has rejigged the naming to reflect this.

These chips are mainly designed for efficiency (particularly the power-sipping GE variants) and the ability to run AI tasks swiftly, or indeed running (modest) LLMs (Large Language Models) locally.

That isn't an exciting prospect for many folks, though, and a good deal of the early online feedback very much reflects this. The Ryzen AI 7 450G flagship will make a decent enough effort at casual gaming, but there's been some disappointment around the chip's performance levels in that respect, too. Mainly because the integrated GPU has fewer CUs than its flagship Ryzen 8000G predecessor, and so isn't much of a step forward. (It's architecturally more advanced, of course, being built on RDNA 3.5, which means a refreshed RDNA 3).

Would you rather have had the extra chip space put into a powerful NPU so you can have Copilot+ offerings in Windows, or would you have rather had the effort put into a juicier integrated GPU for a machine that could handle a bit more gaming as well as office work?

We'll have to see how pricing shakes out with the prebuilt PCs that turn up with Ryzen AI 400 processors inside, but obviously the likes of Lenovo, HP and Dell are going to struggle keeping a lid on price tags with the RAM crisis still very much making life miserable for anyone looking at buying (or upgrading) a computer.

TechRadar is on the show floor for this year’s MWC (Mobile World Congress) in Barcelona, Spain, and we’ll be covering the latest news from some of the biggest names in mobile, computing, fitness and more.

• AMD expands aggressively into virtualized 5G infrastructure and edge deployments
• AMD Sorano now delivers 84 cores for demanding telecom network workloads
• Improved LDPC decoding efficiency directly affects overall network capacity scaling

AMD has introduced its latest 8005-series Epyc processors, codenamed Sorano, with a clear focus on telecom and edge infrastructure.

The new chips raise core counts to as many as 84 Zen 5 cores, marking a notable jump from the earlier Siena generation.

Its power consumption is rated at up to 225 watts, while lower thermal envelopes may follow based on previous designs.

Expanding the Zen 5 Epyc lineup for edge workloads

Sorano is built for virtualized radio access network deployments, where operators increasingly rely on standard server hardware instead of proprietary systems.

In this environment, both CPU throughput and predictable latency matter more than peak clock speed.

AMD says the architecture includes a full 512-bit data path for vector instructions, reflecting broader Zen 5 changes already disclosed by the company.

A central claim around Sorano involves improvements to low-density parity check decoding, a core requirement in 5G networks.

According to AMD, greater efficiency in LDPC handling allows operators to free compute capacity for additional Layer 1 and Layer 2 processing, which could translate into more network functions running per server inside a data center or edge facility. The company is also emphasizing energy efficiency alongside higher core density.

If Sorano mirrors Siena’s lower power variants, telecom operators may see configurations under 100 watts for specific deployments.

In edge scenarios, where thermal limits and environmental tolerances are stricter, the balance between performance and consumption carries financial implications.

AMD’s latest move does not occur in isolation. Intel continues to develop its own telecom-focused processors, including the Xeon 6E and Xeon 6 SoC lines.

The Xeon 6700E can scale up to 144 efficiency cores, trading advanced instruction features for density and lower power draw.

Meanwhile, the Xeon 6 SoC integrates accelerators aimed at vRAN workloads, alongside high-speed networking and support for AI and media tasks often handled by a GPU in broader deployments.

Companies such as Ericsson and Nokia continue to deploy Intel-based platforms in commercial networks, showing that long-term partnerships still influence procurement decisions.

AMD will need to show measurable gains beyond core counts to shift entrenched vendor relationships.

Sorano may represent the last major Zen 5 Epyc release before the arrival of Venice, AMD’s next-generation server CPU planned for 2026.

Whether this iteration materially changes telecom infrastructure economics remains uncertain — core increases alone rarely determine purchasing cycles in a conservative industry.

• Meta commits to six gigawatts of AMD Instinct GPU hardware
• Initial one-gigawatt deployment scheduled for the second half of 2026
• Custom MI450 silicon engineered specifically for Meta AI workloads

AMD and Meta have agreed to deploy up to 6GW of AMD Instinct GPUs over several generations, marking one of the largest disclosed infrastructure commitments in the AI sector.

The scale of the plan suggests a major expansion of Meta’s internal computing capacity as demand for training and running advanced AI models continues to grow.

The first phase of shipments is expected to begin in the second half of 2026 and will support an initial 1GW deployment.

First deployment timeline takes shape

That rollout will rely on a custom GPU derived from AMD’s MI450 architecture and integrated within the company’s Helios rack-scale platform.

The systems will also use sixth-generation EPYC processors, code-named Venice, alongside ROCm software.

The agreement builds on an existing relationship in which Meta has already deployed large numbers of EPYC CPUs and earlier Instinct GPUs across its global data center footprint.

As AI workloads increase in complexity, the balance between CPUs and GPUs has become more critical, particularly for orchestration, scheduling, and efficiency across massive clusters.

Beyond the hardware supply deal, the companies are co-engineering a custom GPU optimized for Meta’s AI workloads.

The custom MI450-based design is expected to support large-scale AI training and inference tasks across Meta’s services.

Meta says AI infrastructure is central to delivering services to billions of users, including recommendation systems, generative tools, and future applications that rely on intensive computation.

“We’re excited to form a long-term partnership with AMD to deploy efficient inference compute and deliver personal superintelligence,” said Mark Zuckerberg, Founder and CEO of Meta.

“This is an important step for Meta as we diversify our compute. I expect AMD to be an important partner for many years to come.”

For AMD, the agreement deepens its presence in hyperscale data centers at a time when competition in AI accelerators remains intense.

“We are proud to expand our strategic partnership with Meta as it pushes the boundaries of AI at unprecedented scale,” said Dr. Lisa Su, chair and CEO, AMD.

“This multi-year, multi-generation collaboration across Instinct GPUs, EPYC CPUs, and rack-scale AI systems aligns our roadmaps to deliver high-performance, energy-efficient infrastructure optimized for Meta’s workloads, accelerating one of the industry’s largest AI deployments and placing AMD at the center of the global AI buildout.”

The partnership also includes a performance-based warrant allowing Meta to acquire up to 160 million AMD shares, with vesting tied to shipment and stock milestones.

The first portion vests after the initial gigawatt of GPU shipments, while additional tranches depend on scaling purchases toward the full 6GW commitment.

Such structures link commercial success directly to equity participation, aligning incentives as infrastructure expands.

“We expect this partnership to drive substantial multi-year revenue growth and be accretive to our non-GAAP earnings per share,” said Jean Hu, EVP, CFO, and treasurer, AMD.

“The performance-based structure also tightly aligns AMD and Meta around execution and long-term value creation.”

• Micron has unveiled faster new GDDR7 video memory
• It runs at 36Gbps compared to 32Gbps for the initial incarnation of GDDR7
• This memory also comes in 3GB modules rather than 2GB, which is an important step forward for graphics card VRAM capacity

Micron has released a new type of GDDR7 memory for graphics cards, which is the firm's fastest such RAM yet.

Tom's Hardware reports that Micron issued a press release detailing the new GGDR7 memory, which offers a speed of 36Gbps. For perspective, the first GDDR7 modules that hit the market — and arrived on consumer GPUs with the RTX 5000 series from Nvidia — were 32Gbps modules. (Although they were run slower than that, at 28Gbps in most cases, for better thermals and stability).

Another key step forward with Micron's new GDDR7 is that it uses 3GB modules rather than 2GB (and a 24Gb density). You can only fit so many memory chips on any given graphics card — based on the bus width of the board, the lanes that link the memory to the GPU chip itself — so going from 2GB to 3GB modules allows for loading up more video RAM and a higher overall capacity.

As Micron notes: "Modern games are pushing GPU architectures harder than ever."

The firm expands on that: "As game environments expand and visual assets grow, memory capacity becomes critical to maintaining seamless, artifact-free experiences. Micron's new 24Gb density enables up to 96GB of graphics memory, giving GPUs significantly more space for high-resolution textures, expansive worlds, and advanced visual effects."

Essentially, the maximum video RAM loadout with these chips on a consumer graphics card with a 512-bit bus (as seen on the RTX 5090) is now 96GB, rather than 64GB with 2GB modules. (Note that this maximum involves using chips front and back of the board in a 'clamshell' configuration, a more complex and costlier endeavor, which is why Nvidia ran with a spec of 32GB with just modules on the front.)

Analysis: looking to the future

Right here and right now, this doesn't mean a whole lot to gamers, of course. Currently we're all more worried about the scarcity of existing forms of video RAM pushing up the prices of graphics cards slowly but surely.

And while it may seem odd to see 96GB of VRAM mentioned in relation to gaming – and to be fair, it is odd — Micron is really just promoting the benefits of 3GB modules more broadly. It's a boast of 'up to' 96GB, and the effects will be felt across whole GPU product stacks, including the likes of 128-bit affordable GPUs that have 8GB now, but could be configured with 12GB using 3GB modules. That's an important upgrade which avoids the aforementioned complications of a 'clamshell' design to increase video RAM capacity.

Micron should have angled its press release better in this respect, but instead wanted to go with an eye-catching 96GB figure, I guess — marketing is often a 'numbers game' in that respect.

At any rate, for the mentioned reasons, it's good news to see a bolstered form of GDDR7 from another memory chip maker. And I say another, because there are two other big forces in this arena, Samsung and SK Hynix, and both already have faster takes on GDDR7 that run at up to 42.5Gbps and 40Gbps respectively.

However, 36Gbps is still a more than respectable speed boost, and as I already mentioned, Nvidia hasn't even pushed to get the top speed out of the initial GDDR7 incarnation, which offered 32Gbps. As noted, all the initial Blackwell GPUs that used GDDR7 ran at 28Gbps, and just the RTX 5080 differs – and even that graphics card only pushed to 30Gbps. (A move that was implemented to make up for its much leaner memory bus compared to the RTX 5090.)

It is also noteworthy that Micron spends some time considering gaming angles in its press release, and this new GDDR7 could well be what we see in Nvidia's next-gen graphics cards. That could be true not just for RTX 6000 GPUs, but also for AMD's next-gen graphics cards. (Team Red stuck with GDDR6 in RDNA 4 this time around).

There is another possibility, though, and that's the rumored RTX 5000 Super refreshes employing this Micron VRAM. I wouldn't rule that out, as these GPUs are supposedly really beefing up memory loadouts. On the subject of these cards, the latest speculation insists that Nvidia's Blackwell Super revamps won't arrive this year (as previously hoped). So, whatever the case, it's likely we're quite a long way out from seeing this faster GDDR7 memory in consumer graphics cards.

Mind you, in the current climate, with the RAM crisis making matters difficult for graphics card makers, talk of scope for expanding VRAM loadouts feels rather pointless at best. Hopefully this isn't a situation that'll persist for too long, but the general indicators right now aren't good – certainly not for this year.

• Crimson Desert will feature both AMD's FSR Redstone and Nvidia's DLSS 4
• Redstone's Ray Regeneration will also be available
• The open-world game is being optimized to run natively, with upscaling as a secondary consideration

Pearl Abyss is working diligently to refine Crimson Desert, ensuring the open-world action adventure performs optimally for both PC and console players ahead of its March 19 launch date – and that now includes FSR Redstone support.

As reported by VideoCardz, AMD's FSR Redstone will be available in Crimson Desert at launch for RDNA 4 GPU users, providing enhanced performance and image quality via upscaling. The game will also support FSR Ray Regeneration, another part of the Redstone package designed to boost ray-traced details in-game.

While Crimson Desert is an FSR Redstone game, it's worth noting that Nvidia's DLSS 4 will also be available at launch as well.

However, Pearl Abyss has made it clear that it's not optimizing the game with upscaling in mind, but rather the developer is focusing on running natively, as noted by PR director Will Powers. In other words, Pearl Abyss isn't relying on upscaling for good performance, but instead views upscaling tools as bonuses once the game's base optimization at native resolution is complete.

There's no denying the benefits of upscaling via FSR or DLSS, as it's often handy for lower-end hardware, but having a game optimized without using either is great news for everyone. And frankly, it's a positive sign that Crimson Desert will run well on both console and PC.

Digital Foundry previously observed that Crimson Desert was running at 4K native resolution seemingly at around 40 to 50fps on a Radeon RX 9070 XT GPU at CES 2026, and that was supposedly in an older build of the game. If that's anything to go by, we should be in for a treat in terms of performance.

It's all possible thanks to the BlackSpace Engine

Pearl Abyss wanted to build its own proprietary engine for a game like Crimson Desert, and its high level of detail in aspects such as the draw distance and weather cycle is only possible thanks to that BlackSpace Engine.

Will Powers reiterated this and highlighted how Crimson Desert wouldn't be able to run on Unreal Engine 5, considering how much depth and detail are included in the visuals. The fact that Pearl Abyss is focusing hard on native frame rate optimization is a big statement.

There's still no official information on performance targets for either PC or console, but the marketing so far all sounds very promising.

The proof will be in the actual playing of the game, of course, but I've got a hunch that we could be looking at a well-optimized title at launch – and that doesn't happen very often.

• ComputerBase ran a comparison between DLSS, FSR and native 4K
• Readers watched videos of all three and voted for the best image quality
• Nvidia's DLSS came out top by far, with FSR falling behind it, and native rendering too — but we need to be careful about what conclusions we draw

Which is best for image quality: Nvidia's DLSS, AMD's FSR, or not using any upscaling at all, and running your games at native 4K resolution? If you thought native was the best choice, well, think again — because a vote held by a tech site has crowned Nvidia the clear winner here.

Tom's Hardware flagged up the intriguing test conducted by ComputerBase, with the readers of the German website being presented with three side-by-side videos.

These showed off DLSS 4.5, FSR 4 (Redstone), and native 4K, and viewers were asked to vote for the video that offered the best image quality. Both upscaling techs were running in 'quality' mode (rather than 'performance'), and native 4K had TAA applied (temporal anti-aliasing, which smooths out jagged edges).

Six games were involved here, with votes registered over two weeks. This was a blind test — meaning the videos were presented unlabelled, so biases towards AMD or Nvidia could be set aside — and readers had to choose which they thought looked best.

This was purely judged on image quality, and you could only pick a winner (no second places). However, if you couldn't tell any real difference between the choices, you could vote to say it was a tie and they were all equivalent.

The end result was a big victory for Nvidia, with DLSS snagging 48.2% of the total vote (6,700 opinions were registered, by the way). Native rendering was in second place with 24% of respondents preferring that, with FSR lagging considerably behind on 15%.

Around 12.8% of those who took the test effectively abstained, as they couldn't see any meaningful difference between the three.

The following games were tested: Anno 117, Arc Raiders, Cyberpunk 2077, Horizon Forbidden West, Satisfactory, and The Last of Us Part II.

Breaking down the results for the individual games showed some clear wins for Nvidia, which notably secured 60.9% of the vote in Satisfactory, and 56.3% in Horizon Forbidden West.

Nvidia won with every single game, although the worst result for DLSS, which was in Cyberpunk 2077, still beat out native rendering (just). Here Nvidia picked up 34.4% of the vote versus 32.4% for native 4K, with AMD hitting its lowest percentage at just 10.6%.

Interestingly, Cyberpunk 2077 was something of an outlier in that it was the only game to have considerable doubt around respondents' assessments of the best quality, with 22.6% being unable to make a call, and voting them all equivalent. With all the other games, the abstainers were in the 8% to 12% ballpark, meaning roughly one in 10 — but in the case of Cyberpunk 2077 approaching one in four gamers were unable to tell.

AMD's best result was for The Last of Us Part II where FSR captured 25.3% of the vote, but it was still in last place here, falling just behind native rendering on 25.9%, with Nvidia winning with 40.9% of the vote (its weakest showing aside from Cyberpunk 2077).

Analysis: a measure of just how good upscaling has become

This is a really interesting set of stats, and it shows just how much upscaling has supercharged contemporary GPUs in terms of producing a better-looking image than native rendering at 4K — and of course a frame rate boost, too. (Although 'quality' obviously doesn't provide the same boost as 'performance' for DLSS or FSR).

It also reflects the broader sentiment you'll find online, which is that DLSS is the reigning monarch of the upscalers. However, AMD has received considerable acclaim for the strides it has taken forward with FSR 4, but that doesn't come across clearly here.

As ComputerBase points out, though, we need to be careful about concluding that AMD FSR is worse than native rendering based on these results, as only one pick was made — for the best quality — and second or third place weren't taken into account. Having a full picture of rankings in that respect could have changed the overall findings.

It's also worth noting that the videos weren't simply uploaded to YouTube, but ComputerBase readers had to download them from the site and watch them via Nvidia's ICAT player. This was to ensure a higher level of quality for the footage and avoid YouTube's various compression antics, which would have watered down the comparison here.

This is clearly a big win for Nvidia, then, and a healthy prod for AMD in terms of needing to catch up more than Team Red has managed so far with the release of Redstone.

From browsing various online forums, you'll see that there are already a lot of gamers sold on the benefits of Nvidia DLSS over and above native rendering — but this test underlines just how good Team Green's tech is in terms of providing a more detailed, superior image quality.

If you're wondering where Intel's XeSS got to in this comparison, it was likely ruled out due to running the tests at 4K — and the lack of an appropriate high-end Intel Arc GPU in that regard — plus the fact that a fourth solution would've complicated matters considerably for ComputerBase (and the viewers judging). Discrete Arc graphics cards are, of course, very much a niche proposition anyway.

• A new report estimates GPU prices are up by 15% globally over the past three months
• Higher-end Nvidia graphics cards have been hit the hardest by price hikes
• The US is suffering from some of the most eye-watering price increases

If you needed any confirmation that GPU prices are rising, a new report shows that graphics cards are definitely getting more expensive across the globe – particularly for some Nvidia models.

As VideoCardz highlighted, TechSpot compiled some statistics which involved tracking 14 GPUs (from AMD, Intel and Nvidia) across 10 regions globally, using local price comparison websites to find the cheapest products from retailers in those countries. Only prices for models in stock and available to buy were used, as you might imagine (sold-out cheap GPUs aren't any good to anyone).

TechSpot collected the initial data on pricing in November 2025, before graphics card pricing started to spike – a side-effect of RAM shortages, which hit video memory too, of course. The second set of data was gathered this month, providing a roughly three-month snapshot of the rising costs of GPUs across the globe.

Overall, there's been a 15% increase across all models and regions on average. To put that another way, a graphics card that was $300 a few months ago now costs $345.

Nvidia graphics cards are clearly the hardest hit by price inflation here. It probably comes as no surprise to you that the RTX 5090 is the worst offender, as we've already observed that the asking price for this flagship Blackwell GPU has gone through the roof since the start of 2026 (and it was already pricey).

TechSpot estimates the overall increase for the RTX 5090 globally to be 31%, although in the US it's particularly hefty at 40% (it's actually 50%, or just over, in India and Poland).

The situation with Nvidia's RTX 5080 is almost as bad, with a 25% increase, the same as the RTX 5070 Ti. Again, in the US, the inflation of those two GPUs is around 40%, a good deal heftier than in many other countries.

Lower-end Nvidia models aren't faring quite so badly with inflation. RTX 5060 models are only up 10 to 11%, which isn't so terrible (especially not when compared with the RAM price rises out there, of course). The RTX 5070 is up 14%, but there's a notable outlier: the RTX 5060 Ti 16GB is up by 22% globally.

Away from Team Green, the increases with AMD GPUs are not nearly as bad. The worst offender on the Radeon side of the fence is the RX 9060 XT 16GB, which is up by 15% worldwide, certainly an unwelcome jump. RX 9070 models are only up by 7 or 8%, though, which is relatively palatable.

There are some more uncomfortable shifts in the US retail market, mind, with the 9070 XT up by 21% in the last three months, and the 9060 XT 8GB has jumped by 20%. So that's quite brutal in both cases.

As for Intel, the Arc B580 has seen its price rise by 11% globally, but the B570 has only witnessed inflation to the tune of 4%.

Analysis: a worrying future?

Theories have been floating around about Nvidia prioritizing AI GPUs over gaming (GeForce) models since the supply of VRAM got shakier, and this evidence appears to back up those notions. Obviously, we can't read too much into this, but whichever way you slice and dice these stats, Nvidia's graphics cards are very much hit by the worst of the increases, particularly at the higher-end of the GPU scale.

The RTX 5090 remains at a ludicrous price in the US, frankly, with the cheapest model at Newegg currently maintaining an asking price of $3,600. (Meaning that buying a whole prebuilt Alienware PC with an RTX 5090 inside is only about 25% more expensive, and you're getting a whole lot of matching high-end kit for the additional premium, including some seriously expensive RAM, naturally).

The rumor that supply is dwindling with the RTX 5070 Ti – which has 16GB of VRAM – and also the RTX 5060 Ti with that same video memory loadout, appears to be backed up by what TechSpot observes here. Those GPUs are up strongly in price (25% and 22% respectively), suggesting that stock could be getting leaner – which would cause prices to rise due to demand.

All of this is grist for the rumor mill theory that Nvidia's graphics cards with higher VRAM loadouts are suffering as AI GPUs (with whopping memory pools) need to take priority, as they make a lot more in the way of profit for Team Green.

The worry here is what the future holds, and whether this kind of rise is going to be sustained – or indeed spread more to AMD's RDNA 4 GPUs. With no signs of the RAM pricing crisis abating, the knock-on effects for graphics card makers are likely to continue to be felt perhaps even more keenly as this first half of 2026 progresses.

Building a PC has certainly become a miserable business, with the huge hike in the cost of system RAM, along with storage following a similar upwards pricing trajectory, and now GPUs are causing a bit of extra chaos in the mix.

• Ayaneo announced the Next 2 handheld's specifications and price
• The new gaming handheld will cost up to $4,299, and its early bird price is $3,499
• The Next 2 will be available via Ayaneo's Indiegogo campaign this February

Gaming PC hardware is taking a huge hit thanks to the ongoing RAM crisis, with rising prices and even the delaying of the highly anticipated Steam Machine. Unfortunately, it appears as though things are only just getting started.

Handheld brand Ayaneo has revealed a wealth of new information about its upcoming Next 2 handheld gaming PC including the fact that it will cost $4,299 / around £3,160 / AU$6,130 for a 128GB RAM and 2TB storage model. That's more than an RTX 5090!

Its early bird price is set at $3,499 / around £2,570 / AU$4,990, for consumers who want to save a little by backing the Indiegogo campaign.

The Ayaneo Next 2 is a beefy handheld gaming PC, equipped with AMD's Ryzen AI Max+ 395 processor, a 116WH battery, and a 9-inch OLED display. Those specifications are almost equivalent to those of a budget gaming laptop, as the Max+ 395's performance is similar to that of an RTX 4060.

While the Ayaneo Next 2 was always expected to be an expensive handheld – since all Ryzen AI Max+ 395-powered handhelds are – the $4,299 retail price tag could be a step too far, even for consumers willing to pay high prices for handheld devices, and it's largely due to the RAM crisis.

It's not just the 128GB model either, as the base model matches the RTX 5090's $1,999 MSRP, equipped with the less powerful Ryzen AI Max 385 processor and 32GB of RAM.

A model with the AI Max+ 395 has 64GB of RAM, which isn't needed for handheld gaming (and certainly not 128GB), but consumers will have to pay $2,699 to get their hands on it.

If RAM prices continue to skyrocket, there's a high chance Ayaneo's listed retail prices won't stick once it officially launches too.

Analysis: 128GB of RAM isn't necessary for handhelds

The RAM crisis isn't going anywhere any time soon, and Ayaneo hasn't made it any easier by using high RAM configurations. Most mainstream handhelds run perfectly fine with 32GB of RAM (some even with 16GB), which is why the push for 64GB and 128GB seems like complete overkill.

The AI Max+ 395 APU, 9-inch OLED display, and 116WH battery are already enough to make the Next 2 a niche device, with a high price, but the addition of 128GB of RAM makes it unattainable, even for consumers willing to spend big bucks for portable gaming.

Nvidia's RTX 5090 is hardly available at any retailer store, and the third-party options are all far above the $1,999 / £1,799 / AU$4,039 MSRP, with some almost costing $5,000 on Best Buy. That same amount can net a consumer a brand-new desktop gaming PC, which is a better option over both the Next 2 handheld and RTX 5090 GPU.

It wouldn't surprise me if new handhelds in 2026 end up creating 8GB RAM configurations, because the PC hardware market looks like it's only going to get worse at this rate.

• AMD quarterly revenue surges despite mounting pressure across the global PC market
• Memory and SSD prices are making new PC builds increasingly unrealistic
• AMD expects PC demand to decline even while its own sales rise

Memory and SSD prices have continued to rise sharply, creating major hurdles for building new PCs, as in some cases, RAM kits now cost four times more than recent lows - but despite these pressures, AMD expects its PC business to grow in 2026.

"Even in that environment, with the PC market down, we believe we can grow our PC business…I think the PC market is an important market. Based on everything we're seeing today, we're probably seeing the PC [Total Addressable Market] down a bit," AMD CEO Dr. Lisa Su said announcing the company's most recent financial results.

"Our focus areas are enterprise... and just continuing to grow at the premium, you know, higher end of the market."

PC market challenges remain

AMD announced $10.3 billion in revenue for its most recent financial quarter, marking a 34% year-over-year increase and bringing total annual revenue to $34.6 billion.

The results include $440 million in inventory of Instinct MI308 accelerators, of which export controls cleared about $360 million.

Total MI308 revenue from sales to China reached about $390 million during the quarter. Gross margin stood at 54%, with net income totaling $1.5 billion.

For the full year, AMD reported a 52% non-GAAP gross margin and $6.8 billion in net income, reflecting steady financial performance despite broader market uncertainty.

AMD’s data center business is now its largest segment and generated $5.4 billion for the quarter, up 39% year-over-year.

Full year data center revenue reached $16.6 billion, representing 32% growth, although it remains smaller than Nvidia’s data center revenue of $51.2 billion.

The client and gaming segment posted $3.9 billion for the quarter, a 37% year-over-year increase, with full year revenue reaching $14.6 billion, up 51%.

Gaming-only revenue rose 50% to $843 million, driven by semi-custom consoles such as the PlayStation 5 and handheld devices like the Steam Deck, along with Radeon GPUs.

AMD’s Ryzen CPUs fall under the client business, and demand for GPUs and AI tools continues to shape revenue trends.

The embedded segment generated $950 million for the quarter, up 3% year-over-year, although full year revenue declined 3% to $3.5 billion.

AMD has said that growth in premium enterprise products and client upgrades could offset slower performance in smaller segments, particularly as RAM and SSD prices remain high.

• A Chinese manufacturer, Seaviv, has revealed a new all-in-one PC
• The AideaOne R27 is a premium-looking all-in-one like Apple's iMac, but with a big difference
• It's built around AMD's powerful Ryzen AI Max+ 395 processor, backed with an eye-opening amount of RAM

AMD's flagship Strix Halo chip is the powerful engine of a new all-in-one PC that's been revealed over in Asia.

As reported by VideoCardz, Chinese manufacturer Seaviv has unveiled the AideaOne R27 all-in-one PC, which is built around a 27-inch 4K screen and AMD's Ryzen AI Max+ 395 processor (which has found a home in plenty of mini PCs and handhelds).

The hardware components are in the monitor's stand, a different approach to situating them behind the screen as with a typical all-in-one device.

The AideaOne R27 is a premium-looking PC and is listed at 21,999 yuan in China, which is around $3,170 in the US. That's a big step up from the Apple iMac M4 starting at $1,299 / £1,299 / AU$1,999, but the R27 packs some seriously powerful components to justify the price.

Seaviv's all-in-one PC doesn't just have that beefy 16-core Ryzen chip, which incorporates a Radeon 8060S integrated GPU, but it's also packed with 128GB of RAM – which in itself costs a small fortune these days – as well as a 2TB SSD.

The RAM crisis has led to an unprecedented hike in prices for memory kits, so all in all, the $3,000 price tag is not that surprising. After all, this is an all-in-one PC that can tackle demanding tasks for productivity, and gaming with performance results similar to an Nvidia RTX 4060 GPU.

It's a niche product, though, obviously enough – and not one the average consumer will be able to stomach the price of, certainly not in comparison to Apple's iMac.

Analysis: the OLED factor

There's a notable weak spot with the AideaOne R27, and it's not just the availability of the device outside of Asia (which we don't have any info on). This downside is that the all-in-one PC has a rather pedestrian screen in the form of a regular 60Hz IPS monitor.

That may seem like something of a mismatch given the rest of the high-powered spec, but it's likely a reflection of the intended usage – meaning productivity, and those working with AI locally (given that massive lump of RAM inside).

It'll do gaming, and facilitate creative work, sure, but an OLED screen – which are gradually becoming more affordable and common for PC monitors – would be much better for those types of usage.

OLED offers much better contrast and color accuracy than a regular LED, of course, while delivering a punchier, more immersive image. There's no backlight bleed with OLED, either, as the tech relies on individual self-lit pixels (which allows for a truly deep black level).

Maybe in the future, we'll see an all-in-one built around the powerful Ryzen AI Max+ 395 with an OLED screen to give Apple something to really think about. Price would, of course, remain a question mark over such an all-in-one, which definitely wouldn't come cheap.

I admire niche products, like this AideaOne R27 PC, which are trying to do something different, with its peppy hardware and the approach of building the components into the stand, rather than within the display.

• Valve has said the Steam Machine is effectively delayed
• The device is now expected to arrive in the first half of 2026, so by June
• The pricing is also being revisited due to the RAM crisis, which has thrown a spanner in the works for the Steam Machine

No sooner had AMD's CEO Lisa Su said the Steam Machine was on track for its original release timeframe, Valve has told us that it's delayed – although not directly.

It's clear enough that there's been a delay, although Valve hasn't quite worded it as such in a new blog post which is framed as a kind of mini-FAQ (which was spotted by Eurogamer).

Let's look at the key part of the statement from Valve first, which addresses the two points that everyone is wondering about the most: namely the launch date and cost of the Steam Machine (as well as the Steam Frame, and Steam Controller).

Valve says: "When we announced these products in November, we planned on being able to share specific pricing and launch dates by now. But the memory and storage shortages you've likely heard about across the industry have rapidly increased since then.

"The limited availability and growing prices of these critical components mean we must revisit our exact shipping schedule and pricing (especially around Steam Machine and Steam Frame).

"Our goal of shipping all three products in the first half of the year has not changed. But we have work to do to land on concrete pricing and launch dates that we can confidently announce."

Analysis: goalposts moved, wallets now feared for, gamers dismayed

There's something I need to take issue with immediately, and that's the stated "goal of shipping all three products in the first half of the year" not having changed. I beg to differ, because Valve said to expect these hardware releases "early in 2026" which I (and many others) took to mean Q1 2026.

Okay, so there's a reason Valve said "early" this year, rather than a more specific timeframe, and that's to avoid being pinned down to Q1 and a more exact launch target. As I've said before, April 2026 could still technically be regarded as relatively early in the year – there's a bit of room for maneuver here.

However, saying the "first half of 2026" is moving the goalposts completely. That could mean June, and in no way can that month be regarded as early in the year – it's the middle of 2026 (obviously). So, this is a delay, even though Valve has tried to couch it in terms of 'nothing having changed' here.

Clearly, things have changed – and rapidly – regarding memory pricing, as Valve states, and this has led to it revisiting the "exact shipping schedule and pricing". Indeed, Valve also told us here that the plan was to have already shared a launch date and pricing by now.

We all knew the price was always going to be a big issue here, and that Valve would likely struggle to keep a lid on it, especially given the meteoric rise of the cost of RAM and, to a lesser extent, SSDs.

So, what we have now, predictably enough, is a great deal of disappointment. Online comments lament that once again we are on 'Valve time' (meaning launch dates being pushed back), and there's a lot of skepticism about whether the first half of 2026 is even a realistic target. I can't say I blame the doubters, either.

Price-wise, the pessimistic cloud that's descended is even thicker than the murk surrounding the release timeframe. Commenters point out that Valve was already coy about sharing anything around the pricing, and we know that there'll be no subsidizing the PC to take the sting out of the MSRP that way, potentially. And now it feels like Valve is priming us to expect the worst with talk of 'revisiting' that costing exercise.

Previously, the guesswork was that the Steam Machine could be pitched starting at $800 or so. If we factor in RAM and storage price spikes, it feels like we could be looking at $900 or more likely closer to $1,000 for the recommended price at the entry level. Some even anticipate the living room PC could weigh in at more than a grand.

As one Redditor, Keraunos01, put it succinctly: "I cannot see this being affordable for most people at this rate..."

What's also worrying here is the indication that Valve isn't sitting on a bunch of component inventory for these PCs bought at previously lower (volume) prices, as we have a clear signal that the RAM price hikes will indeed be a heavy load to bear for the Steam Machine. Obviously, Valve must have some preexisting parts bought, but not nearly enough to carry any kind of launch volume through in 2026 (presumably).

In summary, then, we've moved from a Q1 to Q2 launch (in all likelihood), and from a hopeful $800 to more like looking at $1,000 for the base Steam Machine (though that remains pure guesswork, it should be noted).

A key question then becomes: would you buy the Steam Machine at these kind of asking prices rather than a cheaper PlayStation 5 Pro? (Although admittedly, game consoles may face RAM-pressured price hikes this year, too, but Sony could have a lot of room to maneuver if these Valve price predictions are anywhere near the mark).

Maybe Valve could look at alternative strategies for base-level pricing, like a barebones Steam Machine sold without RAM or storage, so some potential buyers might be able to cannibalize memory from an old PC (or grab it second-hand), for example, to achieve a more palatable bottom-tier price.

For now, though, disappointment reigns with this latest info drop from Valve. More updates via blog posts are promised from the company, and let's hope they ignite a bit more positivity around the Steam Machine again.

• AMD's GPUs are rumored to be getting more price hikes
• This follows price increases that were seemingly applied last month
• A separate claim is that AMD will also focus more on 8GB graphics cards rather than 16GB models

The GPU grapevine is again carrying chatter about AMD's plans for its Radeon graphics cards, which seemingly includes another price hike, and an adjustment of production strategy to favor 8GB products.

VideoCardz reports that these two rumors come from different sources, but they were both aired via the Board Channels over in China, which is a font of speculation from the supply chain over in Asia.

Firstly, let's look at the talk of price rises, where the claim is that either later this month or in March, some AMD graphics card makers may bump up the cost of their products, following a 5% to 10% rise that happened in January.

The extent of the increase this time round isn't known, but the expectation is that it will bring these Radeon GPUs in line with the price of comparable Nvidia models.

The second source floats the idea that AMD is going to modify its production levels of Radeon graphics cards to favor 8GB models more, but that any adjustment will be more of a tweak than a big shift on the assembly lines.

The further claim is that this new strategy will push two 8GB GPUs more to the fore with higher production levels: the RX 9060 XT 8GB (there's also a 16GB model of this card) and RX 7650 GRE.

Analysis: 8GB priority fully makes sense for both AMD and Nvidia

In case you were wondering what the RX 7650 GRE is, this is an Asia-only GPU from AMD, so you can't get it at retailers elsewhere (although you could have it shipped from Hong Kong or similar via a third-party marketplace seller, but I wouldn't recommend that for a number of reasons, such as returns and customer support for starters). It's essentially a slightly faster version of the RX 7600, although there's not a lot of difference.

As mentioned, there are two flavors of the RX 9060 XT, one with 16GB of video RAM and the other with 8GB, so gamers will doubtless be groaning to hear that the latter might be favored for stock levels.

There's long been a debate over whether 8GB of VRAM is enough for modern gaming, and I won't get into that again here, but there's certainly a lot of skepticism over this – and more to the point for me, the level of future-proofing you're getting with this amount of video memory.

Of course, with the RAM pricing crisis also affecting video memory, it only makes sense that AMD might pursue a course which shifts priorities away from more (relatively) affordable GPUs that pile up the VRAM like the RX 9060 XT 16GB. Nvidia is rumored to be doing the same thing, notably with the RTX 5060 Ti which has 8GB and 16GB spins, too.

As for the rumored price rises, AMD's GPUs are already creeping up in that respect, which may reflect the mentioned previous price bumps (which have been rumored going back some way) to an extent. Team Red has even admitted that asking prices may not 'remain flat' across its GPUs as we progress into 2026.

If we look at the RX 9060 XT 16GB, the cheapest I can see this for at the time of writing on Newegg is $440 in the US, which is almost $100 more expensive than its launch price ($349) last year. While it was good value at that release pricing, it's not looking so clever currently.

Interestingly, the RTX 5060 Ti 16GB – which is faster than the 9060 XT 16GB, but not hugely so (ray tracing aside) – has maintained the same price gap with the AMD GPU compared to their respective launch prices. They're $80 apart now (for the cheapest models on Newegg), and those recommended prices were also $80 apart. This seems to back up the idea that AMD Radeon prices will track Nvidia's mark-ups, which again, only makes sense.

We may well see GPU prices creeping up further over the next couple of months as predicted on the Board Channels – but that's hardly a shocker given the RAM crisis, which, despite some signs of it levelling out currently, is still expected to get worse throughout this first quarter of 2026.

If you're thinking of buying one of these more affordable 16GB GPUs, it's likely that now is the time to do so. Either that, or be prepared to wait out the crisis, which could last through this year and next, or at least that's what some analysts are predicting.

• AMD's CEO mentioned the Steam Machine in an earnings call
• Lisa Su said: "Valve is on track to begin shipping its AMD-powered Steam Machine early this year."
• This underlines what Valve has promised in the past in terms of an early-2026 arrival, and the device should hopefully debut in Q1

AMD's chief executive, Lisa Su, has told us that Valve is on course to release its Steam Machine early in 2026.

As Notebookcheck.net noticed, over on X Wario64 flagged up what the CEO had said during AMD's earnings call (following its Q4 fiscal report), as reported by Gamesbeat.

Su observed that: "From a product standpoint, Valve is on track to begin shipping its AMD-powered Steam Machine early this year."

She also mentioned that the next-gen Xbox is on target for a 2027 release, too, featuring an AMD chip as the engine inside.

If you recall, Valve promised that its compact living-room gaming PC would debut early in 2026, so Su is underlining that this is still the case for the Steam Machine.

Analysis: March it is, then?

So, apparently Valve still plans to launch the Steam Machine early this year, which could mean it arrives later this month, possibly, or in March. For me, the use of the word 'early' indicates Q1 of 2026, because once we move into Q2, that's technically not the beginning of the year anymore, although I guess you could argue that April is relatively early in the year at a pinch.

What makes me a bit hesitant about embracing the idea of an imminent arrival of the Steam Machine is the fact that we've heard little about it since the new year rolled in – rumors have gone pretty quiet. Still, there's not much more to know about the device, in fairness, except for that all-important price tag.

Speaking of which, you might also believe that the RAM crisis being in full swing now could give Valve pause for thought, as this is clearly going to make pitching said launch price a trickier business.

The value proposition is crucial here, sure enough, but what can Valve do about the RAM situation exactly? Nothing really (beyond what's already been done previously in terms of securing volume purchasing), and it's not like waiting for the RAM crisis to calm down is realistically an option. If anything, pricing is forecast to go up throughout this quarter, and the situation around the lean supply, and spiking pricing, with system memory may not change until next year – or even 2028 going by some assertions.

It's also interesting that Su uses the term 'shipping' in her quote, as in the Steam Machine will actually be going out to gamers within this timeframe, meaning that an announcement – and pre-orders going live before shipping starts – really could be due pretty soon. (Remember the Steam Deck handheld went on pre-order quite some time before it was shipped).

Obviously all this is just my guesswork, but the fact that Lisa Su has specifically mentioned that the release is still on course is definitely a good sign that we might see the Steam Machine launch happen in Q1, most likely next month. This is going to be a closely watched product, that's for sure, and it could be a pivotal moment to drive forward Valve's ambitions with PC gaming (and Linux to boot).

PC gaming has had its ups and downs over the years, with many games launching in a poorly optimized state. It's an issue that has been more prevalent recently, and Nvidia's DLSS and AMD's FSR upscaling have both been implemented across multiple titles to help improve performance.

Nvidia's latest iteration, DLSS 4.5, is an incredible upscaling tool that significantly improves image stability and clarity, even when using aggressive upscaling methods (namely, the DLSS Performance mode) for better frame rate results.

The DLSS Frame Generation feature plays a vital part in the tool's functionality, theoretically boosting frame rates by up to four times the base fps, leading to faster and smoother gameplay experiences (with the significant caveat that it's naturally affected by your base frame rate, meaning you'll need at least 50 to 60 fps to really benefit).

Previous iterations of DLSS have also been monumental in allowing lower-end RTX GPUs to perform at reasonable frame rates – while DLSS 4.5 is a bigger task for RTX 3000 and 4000 series cards to handle, DLSS 4 is still a viable solution.

So then, what's the issue? Why are DLSS 4.5 and other upscaling methods (like AMD's FSR and Intel's XeSS) a detriment to gaming? Well, the blame isn't on Nvidia, AMD, or Intel, but rather game developers, and here's why.

Frame generation isn't an effective solution

In several recent cases, games have launched with noticeably poor performance due to a lack of optimization, and frame generation is often touted as a solution to improve frame rates.

If you weren't aware, using frame generation in a game with a low base frame rate (or even worse, one with consistent stutters due to poor frame pacing) leads to a choppy and unresponsive experience.

Unfortunately, Capcom's Monster Hunter Wilds is a prime example, as it launched with very poor performance on both low- and high-end hardware. In the game's system requirements (which you can find on Steam), an additional note states, 'this game is expected to run at 1080p, 60 fps with Frame Generation enabled under the Medium graphics setting'.

As a Monster Hunter fan, I've played hundreds of hours of Wilds on an RTX 4080 Super, and I can attest that playing the game since its launch without DLSS Frame Generation was a nightmare. The game failed to maintain a standard 60 fps in many cases, even when actively using DLSS super resolution to upscale from 720p.

Now, a year after launch, Capcom's recent major performance patch fixed many of the performance woes by reducing stutters and improving 1% lows (that's the average framerate during the slowest 1% of gameplay). Let me make that clear: it took an entire year of relying on DLSS Frame Generation before Capcom could get Wilds to perform at an acceptable standard.

I don't think I can stress enough how untenable that is, considering many PC gamers aren't using high-end CPUs or GPUs – as evidenced by the Steam Hardware Survey – and when frame generation itself struggles, it leaves users with lower-end hardware with even bigger performance problems.

It's not just Monster Hunter Wilds, either. We've seen the same overreliance on DLSS upscaling with Borderlands 4, to the point that Gearbox Interactive's CEO, Randy Pitchford, lashed out at consumers for criticisms over the game's poor optimization, urging them to use DLSS 4 and its Frame Generation for better performance. Never mind the fact that some people literally don't have that option...

DLSS 4.5 could make matters worse

Again, I have little to no complaints with the regular resolution upscaling functionality of DLSS 4.5, as it's made PC games easier to run on low-end hardware, and further improves image quality at high resolutions on beefier PC configurations.

What concerns me is that the more DLSS improves, the more game developers may feel incentivized to rely on upscaling as a shortcut rather than spending more time and effort on optimization.

DLSS 4.5 Dynamic Multi-Frame Generation is on its way to RTX 5000 series GPUs soon, and it sounds like an amazing tool that automatically shifts between different frame gen multipliers (2x, 3x, and so on), depending on what's needed – in other words, how graphically demanding a sequence is, or high the graphical settings are.

With DLSS 4.5 6x Multi-Frame Generation also on its way, Nvidia seems intent on getting new games running at exorbitantly high frame rates. However, based on the evidence we have on developers relying on DLSS's Frame Generation feature, I'm worried that the prevalence of DLSS will make matters worse for users who don't have access to a brand-new GPU.

Fortunately, some upcoming games have given me positive signs, notably Pearl Abyss' Crimson Desert.

The open-world action-adventure game is being built using Pearl Abyss' proprietary engine, and PR director Will Powers has indicated that the developers are focused on optimizing the game natively, with tools like DLSS viewed as an optional additional boost to be applied at the player's discretion.

Unfortunately, I doubt this will be the case for most game developers, and I'm expecting multiple game titles to launch this year using upscalers and frame-gen as a stopgap fix for performance woes.

DLSS 4.5 is supposed to be a major benefit to PC game performance, but history tells me it's going to be massively misused – and I have to admit that this time around, Nvidia isn't really the one to blame.

• Sony is encouraging PS4 users to upgrade to PS5 via console messages
• A new analyst report suggests that Sony wants to extend the PS5 lifecycle to retain its current playerbase
• It's also suggested that the PS6 won't arrive until after 2028

Sony is sending messages to PlayStation users still using their PS4, suggesting that they finally switch to the current-generation PS5, and it could be tied to a PS6 launch delay.

As PushSquare reports, PS4 owners are now receiving messages from Sony, encouraging them to upgrade to a PS5, along with a QR code that takes them to PlayStation's official website, where more information about pricing and deals can be found.

"Whether you’re catching up with hits from 2025 like Clair Obscur: Expedition 33, ARC Raiders, and Ghost of Yotei or getting ready for the most anticipated games of 2026, including SAROS, 007 First Light, Nioh 3, and so much more, now’s the perfect time to upgrade your PS5 console," the message reads.

Notably, all these games are only available on PS5 and other current-gen hardware, so it sounds like Sony is persuading players to finally take the plunge.

However, a new analyst report may have shed some light on the situation.

According to a recent Sandstone Insights Japan report published by senior analyst at MST, David Gibson, Sony may push the launch of the PS6 to after 2028 in order to extend the PS5's lifecycle (via IGN).

The report notes that Sony’s earnings for the third quarter are expected to exceed previous predictions, with first and third-party titles continuing to drive sales, and game and network services, including PSN subscriptions, are expected to remain strong.

In addition, Q3 sales for Sony are forecasted for around 1.8 trillion yen total (approx. $11.6 billion), with operating income at 160 billion yen (approx. $1 billion).

With the PS5 hardware and software continuing to push big numbers, it's suggested that Sony will want to expand the life of the console, therefore delaying the launch of the PS6.

"Sony expects the PS5 lifecycle to be longer than that of previous console generations," Gibson stated, adding that the PS6's launch likely won't occur until after 2028.

Gibson also suggested that Sony is more focused on retaining existing PS5 users over boosting console sales, as well as noting that PSN services and active PS5 users are at an all-time high.

"PS5 user activity continues to set all-time record highs according to usage data," Gibson said. "Sony is focusing more on retaining active users than expanding hardware sales."

The tech and games industry has suffered price hikes over the past year or so; an issue we've seen recently is due to memory shortages and the ongoing rise in tariffs.

For now, it seems like PS5 consoles aren't going to be hit with another price increase like last August, although Gibson believes it could happen eventually.

Gibson explained that "rising memory prices will not impact short-term performance thanks to Sony’s existing inventory", but said the increased memory costs could affect the company in the fiscal year, adding, "Sony might pass future cost increases onto consumers."

• AMD Ryzen 9850X3D bundles have been spotted in China
• They bundle the new CPU with 32GB of DDR5 RAM and a cooler
• This could be a way to give PC builders a break on the price of memory, and might be a plan AMD brings to other regions with any luck

AMD could have a plan to tackle the pain of component price hikes – most notably the rocketing cost of RAM – and that's to sell bundles with its imminent new Ryzen X3D CPU, or at least there's some evidence to indicate this in Asia.

Tom's Hardware highlighted a post on X from Uniko's Hardware which shows a bundle that AMD is reportedly launching in China, consisting of a processor plus cooler and system RAM.

What buyers get in this package is the new Ryzen 9850X3D CPU, which was recently revealed – and is about to go on sale, looking a great addition to AMD's gaming lineup – along with a Cooler Master cooler for this chip, and 32GB of V-Color memory (a pair of 16GB DDR5 sticks to be precise).

We don't have pricing yet, only a photo of this bundle as you can see in the post above. However, what's interesting here is that this appears to be an official collaboration between AMD and V-Color alongside Cooler Master, rather than a retailer simply throwing together its own bundle, hence the logos on the box.

There's another telling difference in that the motherboard isn't bundled here. Typically you'd get a CPU and memory plus motherboard in a bundle box, but here the latter is replaced by said cooler.

Analysis: a hopeful hint in a deeply pessimistic market

Presumably, the motherboard has been dispensed with to keep the cost down, as obviously a bundled cooler in its place will ensure an overall lower price (while still making sense as a partner for a high-end gaming CPU). And the main point of this bundle is, of course, to provide DDR5 RAM at a (relatively) palatable price alongside pushing AMD's new processor.

It is, of course, likely that the V-Color DDR5 RAM comprises of affordable 16GB memory sticks, so there'll be no high-end, speedy RAM here (as that would defeat the point of an affordable bundle). As observed by Uniko's, it's likely DDR5-4800 memory (meaning it runs at a speed of 4800MT/s, which is the slowest performance for DDR5).

As Tom's further notes, what's important in that regard is that AMD has been busy making it clear that the Ryzen 9850X3D doesn't need fast RAM. In fact, there's barely any difference in performance between DDR5-4800 and faster DDR5-6000 RAM with the 9850X3D, with a less than 1% difference in frame rates (across an average of 30 games – see the above slide courtesy of VideoCardz).

In other words, whatever DDR5 RAM you're using won't make any noticeable difference with the Ryzen 9850X3D. This is true of X3D processors in general, as their beefier cache (3D V-Cache which the naming convention 'X3D' refers to) means the CPU has to tap the system RAM less often, so the speed of that memory isn't such a factor in determining overall gaming performance.

This is all well and good, but will these bundles arrive outside of Asia? That's the key question here – this and exactly where AMD might pitch pricing. On the latter point, if the bundle isn't a considerable saving on the separate parts, there would hardly be much point.

As to whether the US, or Europe, or anywhere else will see Ryzen 9850X3D plus RAM bundles, we obviously don't know. But the fact that this is happening in China is certainly an indication AMD is mulling strategies to keep RAM pricing down for PC builders – given this is an official collaboration as noted – so I don't see any reason why that strategy might not apply to the wider global marketplace. Well, except for supply worries, naturally, but this is at least a hopeful hint that similar bundles could be in the works.

Meanwhile, standalone DDR5 RAM remains at ridiculous prices, even if it's discounted as we saw yesterday.

• AMD desktop shipments rose from 3 million to over 35 million units quarterly
• Intel’s server share fell from 97% in 2019 to 72% by 2025
• AMD’s revenue from EPYC grew from under $100 million to $3.5 billion (3400%) in 8 years

Intel’s market share in the desktop segment has declined since 2017, coinciding with the introduction of AMD’s first Ryzen CPUs.

Although Intel regained some ground with its 12th Gen Alder Lake and 13th Gen Raptor Lake chips, reports indicate that performance and thermal issues in later generations pushed many DIY enthusiasts and OEMs to switch to Ryzen processors.

AMD has steadily increased its desktop share to over 30%, while Intel now holds around 60%.

AMD’s adoption trends continue to rise

Between 2017 and 2025, AMD’s desktop unit shipments rose from approximately 3 million to over 35 million per quarter.

In the notebook segment, AMD’s growth has remained limited to approximately 20% share due to competitive ARM-based alternatives.

However, Apple and AMD have indirectly benefited from Intel’s weaker positioning in this market.

Intel’s decline also appears in the server segment, where EPYC processors have driven major market shifts.

AMD launched the first EPYC family, Naples, in 2017, and early adopters reacted with surprise at its performance and efficiency.

Initial EPYC adoption accounted for roughly 5% of new server deployments in 2017 and grew to 28% by 2025.

The processors gained traction among enterprise customers and cloud hosting providers that sought higher core counts and improved performance per watt.

Intel’s server unit share dropped from 97% in early 2019 to around 72% in 2025, with revenue share falling to roughly 61%.

AMD’s EPYC now approaches 30% of the server market, up from less than 2% in 2018, according to Mercury Research.

Its revenue share also grew from under $100 million in 2017 to over $3.5 billion in 2025.

Successive Zen architecture updates and the introduction of X3D chips helped drive AMD’s rise.

Cloud storage and cloud hosting environments increasingly favor EPYC processors due to their scalability and energy efficiency.

Intel continues to compete with offerings like the 5th Gen Xeon for AI workloads and claims advantages in certain optimized scenarios.

Yet adoption data shows that its overall unit and revenue shares continue to decline across both client and server markets.

Intel’s server shipments fell from approximately 12 million units in 2019 to under 8.5 million in 2025.

Although Intel remains the largest vendor in both segments, its declining market share points to growing pressure from AMD.

Client CPU markets show slower recovery prospects, while servers and cloud infrastructure continue to shift toward high-core, high-efficiency alternatives.

Observers note that ongoing adoption of EPYC processors in cloud hosting and enterprise servers could further accelerate AMD’s growth, leaving Intel’s long-term lead increasingly uncertain.

Analysts estimate that by 2026, AMD could capture over 35% of the server market if current growth trends continue.

Via Wccf Tech

• AMD has announced that its Ryzen 9850X3D CPU will go on sale on January 29
• The new gaming chip is priced at $499 in the US
• That's a mild bump in cost of around 5% over its predecessor, with a 400MHz jump in boost speed making it a compelling new option

AMD has announced that its Ryzen 7 9850X3D processor – a chip that PC gamers are excited about as a step on from the 9800X3D –will be out in less than a week – and we've got a confirmed price.

VideoCardz reports that AMD's David McAfee, Corporate VP of the Client Channel Business, revealed on X that the "world's most advanced gaming processor just got faster" and that the Ryzen 9850X3D will retail at $499 in the US.

The CPU goes on sale on January 29, and pricing will be in line with that in other regions. To put that US price in perspective, it's just shy of a 5% increase on the MSRP that was pinned to the 9800X3D upon launch, and which is widely regarded as the current top gaming chip on the market.

The Ryzen 9850X3D offers an increase in boost speed to 5.6GHz, meaning it's 400MHz faster than its predecessor (which ran at 5.2GHz), which is a considerable bump. Other than that, though, it's the same spec as the 9800X3D.

Analysis: the price is right – assuming no stock woes or inflation

Some folks who have bought a Ryzen 9800X3D in the recent past may feel their noses have been put a bit out of joint. Especially considering that chip has held its value – it is still selling at close to MSRP – and the 9850X3D doesn't cost much more. At least not in theory, anyway, unless stock and availability issues inflate the price.

Ultimately, those who bought the 9800X3D have still got an excellent gaming CPU for their money, and the 9850X3D is essentially a higher-binned version (meaning a top-performing sample, capable of having the max boost speed pushed further, with no other changes in the spec elsewhere as noted). In fact, if you got lucky with your 9800X3D, it might even be close to the 9850X3D in gaming performance terms, anyway.

So, let go of any buyer's remorse, but that said, the Ryzen 9850X3D is a welcome option to have in terms of that faster boost. And with the price tag not being nearly as high as some rumors had us wondering about – some European retailers had pre-release pricing on their product listings as high as €700 (around $820, £600 or AU$1,200) – this is a great release for gamers who are after a fast CPU.

Although always bear in mind that the GPU is the more important part of the equation for gaming, and the processor only comes more to the forefront with certain games (or lower resolutions).

• Asus is investigating reports of issues with Ryzen 9800X3D CPUs
• These chips are failing in its 800-series motherboards, according to posts on Reddit
• This follows previous problems with the Ryzen 9800X3D in ASRock (and other) motherboards

Asus has issued a statement to acknowledge that it's investigating reports of problems where Ryzen 9800X3D CPUs are failing with its 800-series motherboards.

VideoCardz noticed the statement, which comes following a cluster of recent reports on Reddit detailing issues with these processors and Asus boards.

Asus informs us: "We are aware of recent reports concerning AMD Ryzen 7 9800X3D CPUs and Asus AMD 800-series motherboards, and we have initiated an immediate internal review.

"Our teams are conducting preventive checks on product compatibility and performance, working closely with AMD to validate reported cases and ensure ongoing stability and quality. We are looking to provide timely solutions to ensure our products and services meet expected standards."

Obviously, it's good to hear that the worries here are being directly addressed. What's happening with these CPUs is that the reports (like this one on Reddit) indicate that they are dying and the system is failing to boot (naturally) with an error 'Q-Code 00'.

In the meantime, pending the results of the review, Asus advises those who own an 800-series motherboard to update to the latest BIOS to "help ensure system stability".

That goes for all motherboards, of course – the latest BIOS should provide the most reliable experience running your PC components (not just the CPU). Although don't confuse the 'latest BIOS' with any advance beta release – these aren't fully tested, and may cause more problems than they solve.

Stick to the latest full version, avoiding beta software, especially with something like the BIOS which is a sensitive area to be deploying technically still unfinished releases.

Analysis: a long-running saga

If all this sounds familiar, it's because there are long-running reports of problems with Ryzen 9800X3D processors falling over in this manner. In the past, these have been more of an issue with ASRock motherboards, although other brands have been hit – it's just there's a notable uptick right now with complaints around Asus boards.

Eventually, the problems with ASRock motherboards were put down to a "memory compatibility issue" with earlier BIOS versions, and that issue was fixed. Even so, after that fix, there have still been some 50 recorded cases of 9800X3D CPUs failing in the Reddit megathread about the problems with (mainly) ASRock boards. (A few of these past cases were Asus motherboards, too).

For some 9800X3D owners, the controversy continues to rage across social media posts, and there's some unhappiness (to say the least) about pinning down the root cause.

Although the truth is this could be a complex issue with multiple causes. And don't forget, the 9800X3D is the kind of gaming chip that's going to be tinkered with and overclocked – perhaps substantially by enthusiasts – which could certainly make these kinds of failures more likely to happen. (Other Ryzen 9000 chips have been affected by this problem, I should make clear, but it's mostly hitting the popular 9800X3D – perhaps partly due to its positioning with enthusiasts, as noted).

We can but hope that the investigation Asus is carrying out will shed some further light on the matter, especially seeing as there's a newcomer taking the baton from the 9800X3D now. This is the even faster, higher clocked, Ryzen 9850X3D, which AMD has just announced is coming out next week – with a more tempting price tag than expected.

• AMD Ryzen AI Halo delivers extreme compute power in a device smaller than most hands
• Developers gain full connectivity without relying on adapters or external docking stations
• Supports massive AI models locally with 128GB memory access

AMD used its CES 2026 keynote to introduce Ryzen AI Halo, a compact developer kit intended for local AI workloads rather than general consumer computing.

The system is built around processors from the Ryzen AI Max 300 series, also referred to as Strix Halo, and is scheduled for availability during the second quarter.

Close-up images now reveal how small the device actually is, which reflects AMD’s decision to emphasize form factor alongside compute density.

Extreme compactness without compromise

The Ryzen AI Halo is small enough to fit entirely in the palm of an adult hand, with a footprint smaller than most compact office desktops and a height that barely exceeds a stack of two thick external hard drives.

Its dimensions are noticeably smaller than those of typical mini PCs, which makes it one of the most compact development systems offered.

Despite this small form factor, the device includes a full complement of ports, with four USB Type-C connectors, HDMI, and Ethernet providing the connectivity normally expected from desktops many times its size.

Users do not need additional adapters or docking stations to connect displays, storage, or networking, which is unusual for a device of this volume.

The Ryzen AI Halo ships without Windows or any pre-installed operating system, which makes clear that it is a mini PC aimed at developers rather than general-use desktops.

This choice reinforces its role for running AI workloads locally under Linux or other supported environments rather than serving as a plug-and-play office device.

The casing also shows subtle attention to detail, with a checkered grid pattern on the top surface and a small AMD logo visible on close inspection.

In addition to the Strix Halo processor, which integrates CPU, GPU, and NPU resources, the device supports up to 128GB of LPDDR5 memory across four channels.

It also features a unified memory architecture that allows the GPU to access a large portion of system memory.

The market for portable PCs is crowded, with over 35 Strix Halo devices already available, and hardware alone is no longer enough to stand out.

AMD appears aware of this, which explains why the Ryzen AI Halo emphasizes compact size, a full set of connectors, and flexibility for users to choose their own operating system.

Despite its familiar mini PC appearance, AMD does not market this device as a desktop replacement or office system.

As of the time of writing, there is no official pricing information, although Strix Halo devices typically average around $1500 and can reach as high as $2500.

Via Hot Hardware | Heise | PC Watch