Pat Gelsinger has quite bravely stepped into the belly of the beast this week. The former Intel boss was an invited guest at NVIDIA's GTC 2025 conference; currently taking place in San Francisco, California. Technology news outlets have extracted key quotes from Gelsinger's musings during an in-person appearance on Acquired's "Live at GTC" video podcast. In the past, the ex-Team Blue chief held the belief that NVIDIA was "extraordinarily lucky" with a market leading position. Yesterday's panel discussion provided a repeat visit—where Gelsinger repeated his long-held opinion: "the CPU was the king of the hill, and I applaud Jensen for his tenacity in just saying, 'No, I am not trying to build one of those; I am trying to deliver against the workload starting in graphics. You know, it became this broader view. And then he got lucky with AI, and one time I was debating with him, he said: 'No, I got really lucky with AI workload because it just demanded that type of architecture.' That is where the center of application development is (right now)."

The American businessman and electrical engineer reckons that AI hardware costs are climbing to unreasonable levels: "today, if we think about the training workload, okay, but you have to give away something much more optimized for inferencing. You know a GPU is way too expensive; I argue it is 10,000 times too expensive to fully realize what we want to do with the deployment of inferencing for AI and then, of course, what's beyond that." Despite the "failure" of a much older Intel design, Gelsinger delved into some rose-tinted nostalgia: "I had a project that was well known in the industry called Larrabee and which was trying to bridge the programmability of the CPU with a throughput oriented architecture (of a GPU), and I think had Intel stay on that path, you know, the future could have been different...I give Jensen a lot of credit (as) he just stayed true to that throughput computing or accelerated (vision)." With the semi-recent cancelation of "Falcon Shores" chip design, Intel's AI GPU division is likely regrouping around their next-generation "Jaguar Shores" project—industry watchdogs reckon that this rack-scale platform will arrive in 2026.

Graphics card price watchers have highlighted refreshing downward motion in Europe, apparently affecting three out of the four GeForce RTX 50 series graphics cards. VideoCardz received a couple of email tip-offs from its pan-European audience, prompting the publication of a short investigative piece. NVIDIA's slight adjustment of official pricing for GeForce RTX 5090, RTX 5080, and RTX 5070 models is the result of a strengthened Euro. The US dollar's value has dropped by roughly 3.9 %; according to recent detective work, focusing on German trends. Team Green's "generous" reductions have arrived roughly two weeks after a stabilization of the USD-EUR exchange rate.

Curiously, the GeForce RTX 5070 Ti is an outlier here—NVIDIA did not reduce its German guide price (€879 + VAT) for this upper-mid-range offer. A Founders Edition does not exist at this GPU level, so Team Green has tasked its board partners with the creation of so-called "MSRP conformant" alternatives. One of VideoCardz's tipsters has observed various GeForce RTX 50 series models simply "rotting on shelves," due to potential buyers balking at unreasonable retailer-implemented price hikes. NVIDIA's minor changes (4.3 to 4.6 %) are unlikely to make a noticeable impact across the Euro zone.

"Second wave" ASUS price hikes were documented online over the past weekend; affecting air-cooled premium ROG Astral and mid-tier TUF Gaming GeForce RTX 5090 models. Looking at the company's North American webshop, visitors noticed a freshly adjusted price for the ROG Astral GeForce RTX 5090 32 GB OC Edition—going from a previous level of $3079.99 up to $3359.99. Curiously, the asking price of a liquid-cooled sibling was not adjusted—remaining at a "first wave" point of $3409.99. The "cheapest" model—TUF Gaming GeForce RTX 5090 (non-OC)—experienced a $460 (representing 20%) price hike, bringing total cost of ownership up to $2759.99. As a reminder, NVIDIA's baseline MSRP guideline was $1999—as announced at CES 2025—but ROG Astral and TUF Gaming designs demand a premium or two for fancier feature sets. VideoCardz has fervently explored worrying market trends in the recent past; several of NVIDIA's big board partner players have jacked up asking prices for GeForce RTX 50-series graphics cards. Availability of stock is still a major sore point for potential buyers, who were not able to secure launch day wares. Despite a driving up of costs, the ASUS US webstore has absolutely zero stock of GeForce RTX 5090 SKUs—at the time of writing.

In addition, VideoCardz and other PC hardware media outlets noted price hikes affecting the manufacturer's stable of recently launched AMD Radeon RX 9070 Series TUF Gaming and PRIME models. In the absence of AMD-built (MBA) reference card designs, board partners were tasked with the providing of baseline "MSRP" conformant custom cards. The ASUS PRIME Radeon RX 9070 XT OC and RX 9070 OC Editions were readied as $599 and $549 options (respectively). Weekend sleuthing work put the spotlight on newly adjusted price points of $719.99 and $659.99 (respectively)—representing further cases of plain 20% elevations over baseline. AMD's debut batch of RDNA 4 cards was met with unprecedented demand earlier on in March, but secondary/tertiary stock shipments face unclear market conditions—Team Red GPU enthusiasts have (similarly) voiced their collective displeasure about elevated prices at retail. Mid-way through last week, the PC hardware community heard about ASUS leadership considering a new pricing strategy. The company is reportedly accelerating its manufacturing exodus from China.

Over the past weekend, PC hardware news outlets spent time analyzing NVIDIA GeForce RTX 50-series price fluctuations. One keen market watcher—VideoCardz—has consistently stuck to a main theory of Team Green AIBs implementing last minute price hikes/market manipulations; coinciding with product launch periods. Almost two weeks ago, the online publication directed ire at ASUS and MSI—noted as very high profile board partners. The current GeForce RTX 5070 Ti graphics card lineup is populated by custom designs only; a Founders Edition was not made available within this tier. VideoCardz and Hardware & Co. have observed worrying price trends with AIB-produced models that are supposed to conform to NVIDIA's baseline MSRP of $749 (USD). Their latest reports singled out MSI's North American webstore—already a source of some contention.

Hardware & Co. (a French outlet) observed movement at the bottom-end of the manufacturer's "Blackwell" GPU lineup: "in the case of the GeForce RTX 5070 Ti launched on February 20, 2025, it took nine days...for a "big" brand to officially turn its back on NVIDIA MSRP. On Saturday (March 1), MSI has just updated its RTX 50-series catalog on its official American website with new prices for the RTX 5070 Ti. From now on, the cheapest reference is $820, $70 more than MSRP." VideoCardz is steadfast in its belief that GeForce RTX 50-series "MSRPs are a joke" at this point in time. MSI's VENTUS 3X and (newer) SHADOW 3X models are barebones packages that are designed as alternatives to basic first-party solutions (i.e. Founders Editions)—but VideoCardz has accused the manufacturer of becoming its own "scalper," with (apparently) little intervention coming from NVIDIA. At the time of writing, MSI's US webstore has updated its GeForce RTX 5070 Ti VENTUS and SHADOW listings. Prices have (temporarily?) reverted to original figures (refer to the third screenshot below); likely in reaction to recent "constructive" criticism levied by popular hardware news sites. As evidenced by a sea of "notify me" tags, the official North American store appears to have zero stock in their warehouse(s).

TrendForce's latest findings reveal that weaker consumer demand has persisted through 3Q24, leaving AI servers as the primary driver of memory demand. This dynamic, combined with HBM production displacing conventional DRAM capacity, has led suppliers to maintain a firm stance on contract price hikes.

Smartphone brands continue to remain cautious despite some server OEMs continuing to show purchasing momentum. Consequently, TrendForce forecasts that Q4 memory prices will see a significant slowdown in growth, with conventional DRAM expected to increase by only 0-5%. However, benefiting from the rising share of HBM, the average price of overall DRAM is projected to rise 8-13%—a marked deceleration compared to the previous quarter.

The Taiwanese Ministry of Economic Affairs (MOEA) managed to herd government representatives and leading Information and Communication Technology (ICT) industry figures together for an important meeting, according to DigiTimes Asia. The report suggests that the main topic of discussion focused on an anticipated growth of Taiwan's ICT industry—current market trends were analyzed, revealing that the nation absolutely dominates in the AI server segment. The MOEA has (allegedly) determined that Taiwan has shipped 90% of global AI server equipment—DigiTimes claims (based on insider info) that: "American brand vendors are expected to source their AI servers from Taiwanese partners." North American customers could be (presently) 100% reliant on supplies of Taiwanese-produced equipment—a scenario that potentially complicates ongoing international tensions.

The report posits that involved parties have formed plans to seize opportunities within an evergrowing global demand for AI hardware—a 90% market dominance is clearly not enough for some very ambitious industry bosses—although manufacturers will need to jump over several (rising) cost hurdles. Key components for AI servers are reported to be much higher than vanilla server parts—DigiTimes believes that AI processor/accelerator chips are priced close to ten times higher than general purpose server CPUs. Similar price hikes have reportedly affected AI adjacent component supply chains—notably cooling, power supplies and passive parts. Taiwanese manufacturers have spread operations around the world, but industry watchdogs (largely) believe that the best stuff gets produced on home ground—global expansions are underway, perhaps inching closer to better balanced supply conditions.

NVIDIA is reported to be setting up an AI-focused semi-custom chip design business unit, according to inside sources known to Reuters—it is believed that Team Green leadership is adapting to demands leveraged by key data-center customers. Many companies are seeking cheaper alternatives, or have devised their own designs (budget/war chest permitting)—NVIDIA's current range of AI GPUs are simply off-the-shelf solutions. OpenAI has generated the most industry noise—their alleged early 2024 fund-raising pursuits have attracted plenty of speculative/kind-of-serious interest from notable semiconductor personalities.

Team Green is seemingly reacting to emerging market trends—Jensen Huang (CEO, president and co-founder) has hinted that NVIDIA custom chip designing services are on the cusp. Stephen Nellis—a Reuters reporter specializing in tech industry developments—has highlighted select NVIDIA boss quotes from an incoming interview piece: "We're always open to do that. Usually, the customization, after some discussion, could fall into system reconfigurations or recompositions of systems." The Team Green chief teased that his engineering team is prepared to take on the challenge meeting exact requests: "But if it's not possible to do that, we're more than happy to do a custom chip. And the benefit to the customer, as you can imagine, is really quite terrific. It allows them to extend our architecture with their know-how and their proprietary information." The rumored NVIDIA semi-custom chip design business unit could be introduced in an official capacity at next month's GTC 2024 Conference.

The World AI Cannes Festival (WAICF) is set to be the epicenter of artificial intelligence innovation, where the globe's top 200 decision-makers and AI innovators will converge for three days of intense discussions on groundbreaking AI strategies and use-cases. Against the backdrop of this premier event, GIGABYTE has strategically chosen to participate, unveiling its exponential growth in the AI and High-Performance Computing (HPC) market segments.

The AI industry has witnessed unprecedented growth, with Cloud Service Providers (CSP's) and data center operators spearheading supercomputing projects. GIGABYTE's decision to promote its GPU server portfolio with over 70+ models, at WAICF is a testament to the increasing demands from the French market for sovereign AI Cloud solutions. The spotlight will be on GIGABYTE's success stories on enabling GPU Cloud infrastructure, seamlessly powered by NVIDIA GPU technologies, as GIGABYTE aims to engage in meaningful conversations with end-users and firms dependent on GPU computing.

Huawei's Ascend 910B AI GPU is reportedly in high demand in China—we last learned that NVIDIA's latest US sanction-busting H20 "Hopper" model is lined up as a main competitor, allegedly in terms of both pricing and performance. A recent Reuters report proposes that Huawei is reacting to native enterprise market trends by shifting its production priorities—in favor of Ascend product ranges, while demoting their Kirin smartphone chipset family. Generative AI industry experts believe that the likes of Alibaba and Tencent have rejected Team Green's latest batch of re-jigged AI chips (H20, L20 and L2)—tastes have gradually shifted to locally developed alternatives.

Huawei leadership is seemingly keen to seize these growth opportunities—their Ascend 910B is supposedly ideal for workloads "that require low-to-mind inferencing power." Reuters has spoken to three anonymous sources—all with insider knowledge of goings-on at a single facility that manufacturers Ascend AI chips and the Kirin smartphone SoCs. Two of the leakers claim that this unnamed fabrication location is facing many "production quality" challenges, namely output being "hamstrung by a low yield rate." The report claims that Huawei has pivoted by deprioritizing Kirin 9000S (7 nm) production, thus creating a knock-on effect for its premium Mate 60 smartphone range.

TrendForce has issued its latest findings, indicating that the global SSD market has rectified its supply and demand dynamics in 2022, following a resolution in the shortage of master control ICs that had hampered the market in 2021. Despite the normalization of supply, global SSD shipments witnessed a decline, with only 114 million units shipped in 2022—a 10.7% decrease from the prior year.

The top three SSD shipment leaders of 2022 were Kingston, ADATA, and Lexar, with Kingston and ADATA maintaining solid advantages and experiencing growth in market share over 2021. Lexar's growth was attributed to an aggressive push for revenue in anticipation of going public. Kimtigo, in 2022, made significant strides in expanding into industrial control and OEM markets, which in turn boosted its shipment volume and market share. Netac maintained its competitive edge in the SSD market alongside securing several government orders in the enterprise SSD sector, keeping its market share and ranking consistent with the previous year.

The price of GDDR6 memory has continued to fall sharply - over recent financial quarters - due to an apparent decrease in demand for graphics cards. Supply shortages are also a thing of the past—industry experts think that manufacturers have been having an easier time acquiring components since late 2021, but that also means that the likes of NVIDIA and AMD have been paying less for VRAM packages. Graphics card enthusiasts will be questioning why these savings have not been passed on swiftly to the customer, as technology news outlets (this week) have been picking up on interesting data—it demonstrates that spot prices of GDDR6 have decreased to less than a quarter of their value from a year and a half ago. 3DCenter.org has presented a case example of 8 GB GDDR6 now costing $27 via the spot market (through DRAMeXchange's tracking system), although manufacturers will be paying less than that due to direct contract agreements with their favored memory chip maker/supplier.

A 3DCenter.org staffer had difficulty sourcing the price of 16 Gb GDDR6 VRAM ICs on the spot market, so it is tricky to paint a comparative picture of how much more expensive it is to equip a "budget friendly" graphics card with a larger allocation of video memory, when the bill-of-materials (BoM) and limits presented by narrow bus widths are taken into account. NVIDIA is releasing a GeForce RTX 4060 Ti 16 GB variant in July, but the latest batch of low to mid-range models (GeForce RTX 4060-series and Radeon RX 7600) are still 8 GB affairs. Tom's Hardware points to GPU makers sticking with traditional specification hierarchy for the most part going forward: "(models) with double the VRAM (two 16 Gb chips per channel on both sides of the PCB) are usually reserved for the more lucrative professional GPU market."

Analyst firm Yole Group has predicted that SSD sales revenues will grow to $67 billion in the year 2028, generated by 472 million unit sales - indicating a very healthy outlook in the long term. However, their predictions for market performance in 2023 appear to be less cheerful for manufacturers of NAND flash memory. The SSD market dynamic was positive in 2021 and the starting months of 2022, but demand has dropped sharply since then due to a number of factors including global inflation, geopolitical tensions, and inventory digestion at electronics manufacturers. Sales revenues in 2022 totaled $29 billion (352 million units), down from $34 billion (400 million units) in 2021 - demonstrating a 14% year-to-year decline.

The continued weakening of global demand in 2023 will have an effect on SSD sales revenues, and the Yole Group has foreseen troublesome outcomes for manufacturers. The average selling price of NAND memory and solid-state drive units has been on the decline in the recent quarters, caused by sluggish demand and a surplus of stock. Despite the grim outlook in the short term, the research body is predicting a compounded annual growth rate (CAGR) of about 15% between 2022 and 2028 for the overall size of the SSD market.