Intel spent months building up to the Xeon 6+ launch, its answer to AMD’s data-center CPU gains and the chip meant to prove Intel 18A could compete on Intel’s home turf: the server room. Instead, the reveal landed on the same Computex 2026 stage where Nvidia unveiled its first PC processor, and nobody outside the trade press noticed Intel’s news at all. Three months later, the fallout is still shaping how Wall Street prices the entire chip sector, and Intel’s enterprise pitch has barely gotten a hearing.
On June 1, 2026, shares of Intel and AMD each fell roughly 4% to 6% in premarket trading after Nvidia used its Computex keynote to announce the RTX Spark superchip (also referred to as N1X), an Arm-based processor built with Microsoft to run as the primary chip inside upcoming Windows PCs, according to Yahoo Finance and Tom’s Hardware. Buried in the same news cycle was Intel’s own Xeon 6+ data-center launch, headlined by the 288-core Xeon 6990E+ built on the Intel 18A process node, with Intel confirming commercial availability for June 2, 2026, according to Firstpost’s coverage of the rollout. It was, by any normal week, a significant enterprise hardware story. Instead it became a footnote to Nvidia’s PC ambitions.
Don't miss new tech stories on Google
Add Tech Insider once in the Google app and our stories appear in your news suggestions.
Nvidia’s RTX Spark Superchip: What Actually Shipped at Computex
Nvidia’s announcement marked the company’s first processor built to serve as the main chip inside a Windows PC, rather than a discrete GPU sitting alongside an Intel or AMD CPU. CNBC described Nvidia, which “dominates the market for AI chips in data centers,” as “expanding its prowess to chips that will serve as the main processor for personal computers,” entering a category long controlled by Intel, AMD, Qualcomm, and Apple, according to CNBC’s reporting on the keynote.
Tom’s Hardware characterized the RTX Spark superchip as “a powerful new Arm platform designed to transform Windows 11 into an ‘agentic AI operating system,'” built to run AI agents locally on the device rather than routing every workload to a cloud GPU. That framing matters for how investors read the announcement: this was not pitched as a niche AI PC add-on, it was pitched as infrastructure for a new category of software that runs directly on the chip Nvidia now controls.
Nvidia said the RTX Spark chip will appear in Windows PCs this fall from Microsoft, Dell, HP, ASUS, Lenovo, and MSI, per Yahoo Finance and CNBC. Coverage from the keynote did not detail core counts, clock speeds, or pricing for the chip, a gap that has left analysts guessing at how directly it will compete with Intel’s mainstream laptop silicon versus how narrowly it will target AI-heavy workstations. What the market reacted to wasn’t a spec sheet. It was the symbolism of Nvidia, a company that already sets the terms for AI training and inference hardware, walking into the one PC market segment it had never touched.
Intel’s Xeon 6+ Launch Got Buried the Same Day
Intel’s Xeon 6+ family, built around the Clearwater Forest architecture on the Intel 18A node, was supposed to be Intel’s headline moment at Computex 2026, and Futurum Group’s analysis of the June 2026 launch confirmed the lineup scales from 144 up to 288 Darkmont cores per socket across its stack. The flagship SKU, Xeon 6990E+, packs 288 Darkmont cores spread across as many as 12 compute tiles connected with Foveros Direct 3D packaging, 576 MB of L3 cache, 12-channel DDR5-8000 memory support, and 96 PCIe 5.0 lanes, and remains compatible with existing LGA 7529 socket platforms, according to Tom’s Hardware’s Computex wrap-up, Futurum Group, and Let’s Data Science. Intel positioned the chip directly against AMD’s 192-core Epyc 9965, claiming an average 30% improvement in performance-per-thread and up to 30% better power efficiency versus that AMD part.
That is a serious claim in a data-center CPU market where AMD’s Epyc line has steadily taken server share from Intel over the past several years. Tom’s Hardware framed the launch in blunt terms: “Intel won’t allow AMD to dominate the data center market, hence the arrival of Xeon 6+.” Under normal circumstances, a 30% generational performance claim against AMD’s flagship server chip would dominate hardware headlines for days. Instead, it competed for attention with Nvidia’s PC announcement on the same Computex floor, and lost.
Intel also used the event to confirm that its next-generation Xeon 7 platform, codenamed Diamond Rapids, remains on track for 2027 on the more advanced Intel 18A-P node, according to Tom’s Hardware. That roadmap detail barely registered in financial media coverage of the day, which focused almost entirely on what Nvidia’s move meant for Intel and AMD’s PC businesses rather than Intel’s data-center roadmap.
The Premarket Selloff: How Sharp Was It, Really
Different outlets clocked the June 1 premarket move at slightly different levels, depending on the exact timestamp and whether they measured from the steepest intraday dip or a later, partially recovered print. Yahoo Finance reported Intel and AMD each fell about 4.2% in Monday’s premarket session. A Reuters-style brief carried by Stockopedia put INTC down 4.4% and AMD down roughly 4%, while Qualcomm, whose Snapdragon chips also compete in Windows laptops, tumbled about 8%. TradingNews described Intel getting “smoked,” sliding nearly 6% ahead of the open, framing the move as traders reading Nvidia’s arrival as “a direct shot at the x86 architecture that’s owned the Windows world for four decades.”
TradingView’s Invezz coverage noted Intel shares had fallen more than 4% and AMD more than 3% at the low, with both stocks partially recovering from steeper intraday losses by the time regular trading opened. The spread between a 4.2% figure and a nearly-6% figure isn’t a contradiction between outlets so much as a reminder that “premarket move” is a moving target on thin volume, measured differently minute to minute. What’s consistent across every outlet: Intel fell harder than AMD, and both fell hard enough to wipe out billions in combined market value before the opening bell.
The selloff didn’t stop after one session. Intel shares fell nearly another 2% in premarket trading the following Tuesday, June 2, extending the prior day’s losses as investors continued to digest Nvidia’s PC ambitions, per Stocktwits reporting. By June 5, Yahoo Finance tracked a second consecutive down day for the sector, with INTC off 2.85% and AMD off 2.33%. Intel’s stock had effectively absorbed a compounding hit across nearly a full trading week, all triggered by a single Computex keynote slide that never mentioned Intel by name.
Stock Reaction Timeline: June Crash to August Rebound
| Date | Event | INTC Move | AMD Move | Source |
|---|---|---|---|---|
| May 31 – Jun 1, 2026 | Nvidia unveils RTX Spark/N1X at Computex; Intel’s Xeon 6+ overshadowed | -4.2% to -6% | -3% to -4% | Yahoo Finance, TradingNews, Tom’s Hardware |
| Jun 2, 2026 | Selloff extends into second session | -2% | Not specified | Stocktwits |
| Jun 5, 2026 | Chip stocks fall for a second consecutive week | -2.85% | -2.33% | Yahoo Finance |
| Jun 16, 2026 | Follow-up coverage on Nvidia’s PC and AI push rattling traders | ~-6% cumulative, briefly >-5% premarket | Not specified | Timothy Sykes / INTC coverage |
| Jul 22, 2026 | Broader semiconductor pullback stalls a nascent rebound | -2.7% | -1.8% | Stocktwits |
| Late Jul 2026 | Chip-stock rout tied to OpenAI-deal concerns | Not specified | Not specified | Stocktwits (NVDA -1%) |
| Aug 27, 2026 | Chip stocks jump premarket after Nvidia’s blowout earnings report | +3% | +1.8% | Stocktwits |
The pattern that emerges across three months is not a straight line. Intel and AMD took the sharpest hit on the day Nvidia announced RTX Spark, kept bleeding for roughly a week, got caught up in a broader semiconductor pullback in late July tied to worries about OpenAI’s compute spending commitments, and then rallied alongside the rest of the chip sector in late August after Nvidia posted a blowout earnings report that Stocktwits said sent Intel up 3%, Broadcom and AMD up about 1.8% each, and SK Hynix up 4% in premarket trading. Nvidia’s results, in other words, both triggered the initial Intel and AMD selloff in June and helped lift both stocks back up in August. The same company has now moved the same rivals’ share prices in opposite directions within a single quarter.
Why the Market Reacted So Hard to a Product With No Public Specs
None of the outlets covering the RTX Spark launch published core counts, clock speeds, or a price for the chip. Investors weren’t repricing Intel and AMD based on a benchmark. They were repricing based on precedent. Nvidia already controls the AI training and inference hardware market outright, and TradingView’s Invezz coverage framed the RTX Spark launch as directly “challenging Intel and AMD’s long-standing dominance in the PC chip market” while simultaneously increasing pressure on Qualcomm, which has spent several years building its own Snapdragon-based Windows-on-Arm push.
That is the core of why traders moved first and asked questions about specs later. Nvidia doesn’t need to beat Intel’s mainstream laptop chips on raw CPU performance to change the competitive calculus. It only needs Microsoft, Dell, HP, ASUS, Lenovo, and MSI treating RTX Spark as a credible third or fourth platform choice for AI-capable Windows PCs, which is exactly what those five OEM names committing to fall 2026 launches signal. Intel’s x86 architecture has held functionally unchallenged control of mainstream Windows computing for roughly four decades, broken only at the margins by Arm-based Windows laptops running Qualcomm silicon since 2023. Nvidia entering that market with Microsoft as a co-development partner is a different order of threat than Qualcomm’s incremental Arm push, because Nvidia brings its own AI software ecosystem, CUDA relationships, and a brand that already means “AI leader” to enterprise buyers and consumers alike.
The Xeon 6+ vs. Epyc 9965 Fight Nobody Watched
Intel’s Xeon 6990E+ claims deserve more scrutiny than they got in June, if only because the data-center CPU market is where Intel has bled the most share to AMD over the past five years. A 30% average performance-per-thread gain and up to 30% better power efficiency against AMD’s 192-core Epyc 9965, if it holds up in independent benchmarks, would be Intel’s strongest generational claim against AMD’s server line in years, especially set against Futurum Group’s finding that Xeon 6+ boosts last-level cache roughly 5x over the prior generation while spanning a wide 330-450W TDP range across SKUs. The Xeon 6+ family’s move to the Intel 18A node is also the first real proof point for Intel’s foundry roadmap delivering a flagship server part on schedule, which matters enormously for Intel’s broader turnaround narrative under its foundry strategy.
None of that made it into the stock-moving headlines on June 1. The Xeon 6+ launch got covered thoroughly by hardware trade press like Tom’s Hardware, but financial media coverage of the day was almost entirely about Nvidia’s PC ambitions and what they meant for Intel and AMD’s client-computing businesses. That’s a real cost for Intel: the Xeon 6+ launch was arguably the more consequential enterprise hardware story of the day, and it barely moved the stock in either direction because the market’s attention was entirely elsewhere.
Comparing the Three Chips at the Center of the Story
| Chip | Maker | Target Market | Key Spec | Node/Architecture |
|---|---|---|---|---|
| RTX Spark (N1X) | Nvidia | Windows PCs / AI PCs | Arm-based main PC processor; specs not disclosed | Co-developed with Microsoft |
| Xeon 6990E+ | Intel | Data center / hyperscale servers | 288 Darkmont cores, 576 MB L3 cache | Intel 18A (Clearwater Forest) |
| Epyc 9965 | AMD | Data center / hyperscale servers | 192 cores (Intel-cited comparison baseline) | AMD’s current-generation Epyc line |
The table above underscores an odd asymmetry in the story: Intel published detailed, specific numbers for its data-center chip, while Nvidia’s PC chip announcement generated the far bigger stock reaction with almost no published specs at all. That’s a reminder that in this market, narrative and platform commitments from OEMs like Microsoft and Dell can move share prices more than a spec sheet, at least in the short term.
Historical Context: Nvidia’s Long Path Into CPUs
Nvidia’s PC processor push didn’t come out of nowhere. The company has spent years building CPU credibility through its Grace data-center CPU line and Grace-Blackwell superchips aimed at AI training clusters, establishing that Nvidia could design competitive CPU silicon, not just GPUs. Nvidia also disclosed a $5 billion investment in Intel in 2025, a deal that at the time was read by some investors as a sign of cooperation between the two companies rather than escalating rivalry. The RTX Spark launch complicates that reading: Nvidia is now simultaneously an Intel investor and a direct competitor in the PC processor category, an unusual position for two companies to occupy at once.
Qualcomm’s Snapdragon X series had already established that Arm-based chips could run Windows competently, breaking Intel and AMD’s multi-decade x86 monopoly on mainstream Windows PCs starting in 2023. Nvidia’s entry effectively validates that Arm-on-Windows is now a permanent third lane in PC computing rather than a Qualcomm-only experiment, and raises the stakes for Microsoft’s own investment in making Windows run natively and efficiently across x86 and Arm architectures alike.
Market Impact: Who Actually Loses Business Here
The immediate stock reaction overstates the immediate business impact. RTX Spark PCs don’t ship until fall 2026, and no pricing or performance benchmarks have been published, so no OEM has actually shifted volume away from Intel or AMD silicon yet. What did shift immediately was investor expectations about the next several years of PC CPU market share, and that expectation shift alone was enough to erase billions in combined Intel and AMD market capitalization within a single trading session.
The more durable damage may be to Intel’s enterprise narrative. Xeon 6+ needed a clean launch to reinforce the story that Intel 18A is back on track and that Intel can still win data-center sockets against AMD’s Epyc line on merit. Instead, the launch got buried in the same news cycle as an existential threat to Intel’s other core business, client computing. Investors and IT buyers evaluating both product lines this summer got a confused signal: is Intel losing ground everywhere, or winning in servers while losing in PCs? The muddled coverage made it harder to tell.
Competitive Landscape: Four Companies, One Windows PC Market
- Intel: Defending four decades of x86 dominance in Windows PCs while simultaneously trying to prove Intel 18A can deliver competitive data-center silicon against AMD.
- AMD: Caught in the same premarket selloff as Intel despite AMD having no direct product conflict with RTX Spark’s Arm architecture, reflecting how tightly AMD and Intel now trade as a pair on PC-market sentiment.
- Qualcomm: The company with the most direct architectural overlap with RTX Spark, since both are Arm-based Windows chips; Qualcomm’s shares fell further than Intel’s or AMD’s in the initial premarket reaction, per Stockopedia and TradingView.
- Nvidia: Entering PC CPUs from a position of AI-market dominance, backed by Microsoft co-development and five major OEM commitments for fall 2026.
Apple sits outside this competitive set entirely, since it designs its own Apple Silicon chips for Macs and doesn’t compete for Windows OEM sockets, a distinction CNBC’s coverage of the keynote specifically noted when listing the incumbents Nvidia is now challenging.
What Enterprise Buyers Should Actually Watch
For IT buyers evaluating server refresh cycles, the Xeon 6+ story matters more than the stock chart, and the ecosystem has already started moving: Supermicro announced 12 new Xeon 6+ server platforms on May 31, 2026, giving buyers immediate hardware options rather than a paper launch. A 30% performance-per-thread claim against Epyc 9965, if independently verified, would be a meaningful reason to re-run total-cost-of-ownership models before committing to an all-AMD data-center refresh, particularly for AI-heavy deployments now that Intel’s companion E835 Ethernet adapter reaches 200 GbE for AI networking, per Intel’s May 2026 announcement. The fact that Xeon 6+ remains compatible with existing LGA 7529 socket platforms also matters operationally, since it means data centers already running earlier Xeon 6 generations can upgrade without a full platform redesign.
For consumer and enterprise PC buyers, the practical guidance is simpler: RTX Spark machines won’t be available to test or buy until fall 2026, and no independent benchmarks exist yet. Anyone making PC purchasing decisions this quarter is choosing between known Intel and AMD x86 performance and an unproven Arm platform with strong OEM backing but zero public reviews.
Predictions: Where This Goes Through 2027
- RTX Spark’s fall 2026 launch window will draw outsized media coverage that further overshadows Intel’s Xeon 7 “Diamond Rapids” roadmap, which Tom’s Hardware reports remains on track for 2027 on the Intel 18A-P node.
- Independent benchmarks of Xeon 6990E+ against Epyc 9965 will become a bigger story once neutral labs publish numbers, testing whether Intel’s 30% claims hold up outside company-sponsored comparisons.
- Qualcomm faces the most direct competitive pressure from RTX Spark given the architectural overlap, and its response, likely an accelerated Snapdragon X refresh, is the item most worth watching for anyone tracking the Arm-on-Windows race.
- Chip-sector stocks will likely keep swinging on Nvidia-linked news even when the news isn’t about Intel or AMD directly, as the June-through-August pattern already shows with the OpenAI-deal-driven July pullback and the earnings-driven August rally.
- Intel’s dual identity as both an Nvidia investor (via the 2025 $5 billion stake) and a direct RTX Spark competitor will keep generating mixed market signals until one relationship clearly takes precedence over the other.
The Bigger Picture for the Chip Sector
Three months on from Computex 2026, the RTX Spark announcement has done more to reshape how investors price Intel and AMD than it has to actually change what’s sitting inside Windows PCs, since no RTX Spark hardware has shipped yet. What it has changed is the baseline assumption that x86 dominance in Windows computing is permanent. Once a company with Nvidia’s AI credibility and Microsoft’s OEM relationships commits publicly to a PC CPU roadmap, every subsequent Intel and AMD product announcement gets read partly through that lens, including a genuinely strong data-center chip like Xeon 6+ that deserved a bigger moment than it got.
Frequently Asked Questions
What is Nvidia’s RTX Spark chip?
RTX Spark, also referred to as N1X, is Nvidia’s first processor designed to serve as the main chip inside a Windows PC rather than a companion GPU. It’s an Arm-based platform co-developed with Microsoft and set to appear in PCs from Microsoft, Dell, HP, ASUS, Lenovo, and MSI starting fall 2026, according to CNBC and Yahoo Finance.
Why did Intel and AMD stock fall after Nvidia’s announcement?
Investors read Nvidia’s entry into PC processors as a long-term threat to Intel and AMD’s decades-long control of the Windows CPU market. Shares fell roughly 4% to 6% in premarket trading on June 1, 2026, per Yahoo Finance, Stockopedia, and TradingNews, even though no RTX Spark hardware, pricing, or benchmark data had been published.
What is Intel’s Xeon 6+ and how is it different from Xeon 6?
Xeon 6+ is Intel’s updated data-center CPU family built on the Clearwater Forest architecture and the Intel 18A process node. Its flagship chip, the Xeon 6990E+, has 288 Darkmont cores and 576 MB of L3 cache, and remains compatible with existing LGA 7529 server platforms, according to Tom’s Hardware’s Computex 2026 coverage.
How does Xeon 6990E+ compare to AMD’s Epyc 9965?
Intel claims the Xeon 6990E+ delivers an average 30% improvement in performance-per-thread and up to 30% better power efficiency compared with AMD’s 192-core Epyc 9965, per figures reported by Tom’s Hardware and Yahoo Finance. These are Intel’s own comparisons; independent third-party benchmarks were not available as of this writing.
Has the chip stock selloff continued since June 2026?
No, it has not moved in a straight line. After the initial June 1 drop and a second down day on June 5, chip stocks fell again in a broader pullback around July 22, 2026, tied partly to concerns about OpenAI’s compute spending. By August 27, 2026, Intel, AMD, and other chip stocks jumped in premarket trading after Nvidia posted a blowout earnings report, per Stocktwits.
Is Nvidia still an investor in Intel while competing against it?
Yes. Nvidia disclosed a $5 billion investment in Intel in 2025. That relationship now coexists with Nvidia’s RTX Spark launch, which directly competes with Intel’s client-computing business, creating an unusual dynamic where Nvidia is simultaneously a shareholder and a rival.
When will RTX Spark PCs actually be available to buy?
Nvidia and its OEM partners have targeted a fall 2026 launch window for the first wave of RTX Spark-powered Windows PCs, according to CNBC and Yahoo Finance. No specific release date, pricing, or independent benchmark results had been published as of late August 2026.
What is Intel’s Xeon 7 “Diamond Rapids” and when is it coming?
Diamond Rapids is Intel’s next-generation Xeon 7 data-center platform, planned to follow Xeon 6+ on the more advanced Intel 18A-P process node. Tom’s Hardware reports it remains on track for a 2027 launch.
Related Coverage
- Nvidia RTX Spark Sinks Intel, AMD Stock 4.4% [2026]
- Atari Gamestation Go vs Super Pocket: 288-Game Gap [2026]
- Logitech vs Razer vs ROG Mice: 8K vs 4K Polling Gap [2026]
- How to Upgrade a Graphics Card: 12 Steps, 90 Min [2026]
- DLSS 5 Patched to Run on RTX 4000: 39% FPS Drop [2026]
