The rising prominence of Huawei's Ascend accelerator series has generated plenty of news articles over the past couple of months. Earlier in the week, reports suggested that the Chinese technology giant had circumvented international trade sanctions. Many observers will forget that parts of Huawei's portfolio contain "less ambitious" products; namely notebook PCs. Despite navigating tricky conditions in 2024, the manufacturer still has access to a supply of Intel-produced processors—but a contingency plan is reportedly in place. TP Huang (aka tphuang) reckons that a completely in-house formulated "domestic commercial PC" is due for release in China. The technology soothsayer's social media post mentioned a "7 nm Kunpeng-920 CPU" clocked at 2.6 GHz. Previously, Huawei's Kunpeng proprietary processor designs were deployed in server (2024) and desktop (2020) environments.

According to Huawei Central, company representatives have disclosed that the mysterious Kunpeng-920-powered AI laptop will be added to their "Qingyun" enterprise-oriented product range. TP Huang elaborated with additional leaked details—the forthcoming model is tipped to feature: "fully domestic power management integrated circuits (PMIC); including Unisoc CT100 replacing EPSON in clock generator. It is expected to run on HarmonyOS PC version and a localized dev tool chain." An extension of Huawei and DeepSeek's current relationship is anticipated to trickle down to portable devices—insiders believe that the new AI notebook will support DeepSeek's model on edge. TP Huang predicts an April launch window.

According to Digital Chat Station—a repeat leaker of unannounced Qualcomm hardware—unnamed Android smartphone manufacturers are considering an eSIM-only operating model for future flagship devices. Starting with the iPhone 14 generation (2022), Apple has continued to deliver gadgets that are not reliant on "slotted-in" physical SIM cards. According to industry insiders, competitors could copy the market leader's homework—Digital Chat Station's latest Weibo blog post discusses the space-saving benefits of eSIM operation; being "conducive to lightweight and integrated design." Forthcoming top-tier slimline Android mobile devices are tipped to utilize Qualcomm's rumored second-generation "Snapdragon 8 Elite Gen 2" (SM8850) chipset.

Digital Chat Station reckons that: "SM8850 series phones at the end of the year are testing eSIM. Whether they can be implemented in China is still a question mark. Let's wait and see the iPhone 17 Air. In order to have an ultra-thin body, this phone directly cancels the physical SIM card slot. Either it will be a special phone for the domestic market, or it will get eSIM." The phasing out of physical SIM cards within the Chinese mobile market could be a tricky prospect for local OEMs, but reports suggest that "traditionally-dimensioned" flagship offerings will continue to support the familiar subscriber identity module standard. Physical SIM card purists often point out that the format still provides superior network support range.

Software redesigns are often a polarizing affair. No matter how a company approaches the task, the final product will always please some people, while displeasing others. However, this fact has hardly ever hindered tech giants from tinkering - or 'innovating', as they call it - and the same is expected for this year's iOS 19 and macOS 16 operating systems. As multiple sources have pointed out, including the reliable Mark Gurman, iOS 16 is likely to receive a major redesign this year, which will include design elements borrowed from Apple's VisionOS - the operating system that powers the Vision Pro.

The same applies to the upcoming macOS 16, internally code-named "Cheer", which is also borrow inspiration from VisionOS. We have witnessed hints of this in iOS 18 already, which includes round icons in the control center, as opposed to its predecessor's square icons with rounded edges. Interestingly, Gurman did not confirm whether all app icons will switch to a round shape, which will be quite interesting since older macOS and OS X versions, all the way to Catalina, boasted round app icons. Of course, design changes are often cyclic, especially in the case with Apple. Moreover, increased translucency should also be prevalent throughout macOS 16 and iOS 19 - yet another design choice that is likely to be divisive in nature. Either way, both iOS 19 and macOS 16 are expected to debut in the summer at WWDC 25, which means we don't have much waiting to do before being granted a sneak peek into the future of macOS and iOS.

According to a Reuters technology report, Meta's engineering department is engaged in the testing of their "first in-house chip for training artificial intelligence systems." Two inside sources have declared this significant development milestone; involving a small-scale deployment of early samples. The owner of Facebook could ramp up production, upon initial batches passing muster. Despite a recent-ish showcasing of an open-architecture NVIDIA "Blackwell" GB200 system for enterprise, Meta leadership is reported to be pursuing proprietary solutions. Multiple big players—in the field of artificial intelligence—are attempting to breakaway from a total reliance on Team Green. Last month, press outlets concentrated on OpenAI's alleged finalization of an in-house design, with rumored involvement coming from Broadcom and TSMC.

One of the Reuters industry moles believes that Meta has signed up with TSMC—supposedly, the Taiwanese foundry was responsible for the production of test batches. Tom's Hardware reckons that Meta and Broadcom were working together with the tape out of the social media giant's "first AI training accelerator." Development of the company's "Meta Training and Inference Accelerator" (MTIA) series has stretched back a couple of years—according to Reuters, this multi-part project: "had a wobbly start for years, and at one point scrapped a chip at a similar phase of development...Meta last year, started using an MTIA chip to perform inference, or the process involved in running an AI system as users interact with it, for the recommendation systems that determine which content shows up on Facebook and Instagram news feeds." Leadership is reportedly aiming to get custom silicon solutions up and running for AI training by next year. Past examples of MTIA hardware were deployed with open-source RISC-V cores (for inference tasks), but is not clear whether this architecture will form the basis of Meta's latest AI chip design.

A curiously-specced Acer Nitro N50 (N50-656) pre-built gaming PC system was spotted and then reported online, courtesy of momomo_us's diligent eye—trained on all manner of international retail and e-tail webstores. EvoPC—a small indie outfit, located in Nancy, France—has inadvertently hinted about a potential upcoming launch of NVIDIA's unannounced GeForce RTX 5060 model. VideoCardz believes that Team Green will be unveiling the lower end of its gaming-oriented "Blackwell" GPU product stack.

An imminent announcement—reportedly marked down for this week—could include another GB203-based variant; the GeForce RTX 5060 Ti—in addition, they reckon that NVIDIA will reveal a next-gen "entry-mainstream" offering; in the shape of GeForce RTX 5050. Comprehensive details leaked onto the internet over the past weekend. EvoPC's webshop advertises the "currently unavailable" Acer Nitro N50-656 system with a steep €1589 (~$1735 USD) price tag (subject to change). Their basic rundown of internal parts lists: an Intel Core i7-14700F processor, 32 GB (2 x 16 GB) DDR5 RAM, and NVIDIA's GeForce RTX 5060 GPU with 8 GB of GDDR7 SDRAM.

In an exclusive report, Windows Central has claimed that Microsoft is collaborating with an unnamed "PC gaming OEM" on an Xbox handheld console. Jez Corden has heard from several insiders; they reckon that a "later in 2025" launch is possible, if development goes smoothly enough. A leak from Winter 2024 indicated that Microsoft's gaming division was "targeting a handheld gaming experience," but Phil Spencer has freely disclosed his ambitions for his team's portable project. Last November, he confirmed that something was in pipeline—with a view to take on very visible competition: Steam Deck, ASUS ROG Ally and Lenovo Legion Go. The latest Windows Central investigative piece divulges some unprecedented details about: "codename 'Keenan,' this gaming handheld will look unmistakably 'Xbox' we're told, complete with an official Xbox guide button, and Xbox design sensibilities. Given that this is a partner device, similar to Lenovo's SteamOS partnership with Valve, I expect this handheld to be more PC-oriented. Keenan is almost definitely running full Windows, putting the Microsoft Store and PC Game Pass front and center, alongside the ability to install things like Steam."

Based on inside track knowledge, Corden anticipates operating system innovations: "I expect the handheld will test new Windows 11 'device aware' capabilities, while reducing third-party OEM bloatware that are typical of devices like the Lenovo Legion Go and ASUS ROG Ally. I expect Microsoft will leverage widgets on the Xbox Game Bar on PC for controlling things like TDP and fan speed, while hopefully having a more streamlined OS experience for controller use. Of course, existing PC gaming OEMs will benefit from these efforts as well—but the research will also help with Microsoft's further-out plans." Almost a month and a half ago, Spencer outlined a sort of wishlist for "innovative" future Xbox hardware—given that Sony is rumored to be devising a "standalone PlayStation handheld," he was likely envisioning next-gen home consoles. Windows Central's sources have provided additional insights into potential Xbox Series X|S successors.

Supply chain insiders believe that NVIDIA's "Blackwell Ultra" GB300 AI chip design will get a formal introduction at next week's GTC 2025 conference. Jensen Huang's keynote presentation is scheduled—the company's calendar is marked with a very important date: Tuesday, March 18. Team Green's chief has already revealed a couple of Blackwell B300 series details to investors; a recent earnings call touched upon the subject of a second half (of 2025) launch window. Industry moles have put spotlights on the GB300 GPU's alleged energy hungry nature. According to inside tracks, power consumption has "significantly" increased when compared to a slightly older equivalent; NVIDIA's less refined "Blackwell" GB200 design.

A Taiwan Economic Daily news article predicts an upcoming "second cooling revolution," due to reports of "Blackwell Ultra" parts demanding greater heat dissipation solutions. Supply chain leakers have suggested effective countermeasures—in the form of fully liquid-cooled systems: "not only will more water cooling plates be introduced, but the use of water cooling quick connectors will increase four times compared to GB200." The pre-Christmas 2024 news cycle proposed a 1400 W TDP rating. Involved "Taiwanese cooling giants" are expected to pull in tidy sums of money from the supply of optimal heat dissipating gear, with local "water-cooling quick-connector" manufacturers also tipped to benefit greatly. The UDN report pulled quotes from a variety of regional cooling specialists; the consensus being that involved partners are struggling to keep up with demand across GB200 and GB300 product lines.

Apple's recent unveiling of refreshed Mac Studio models—in "mismatched" M3 Ultra and M4 Max forms—was greeted with a lukewarm reception from press and public. The absence of an M4 Ultra option has disappointed many folks within the high-end Mac buying populace—rumors of a delayed development of Apple's "Mac Studio M4 Ultra model" emerged online last October. The M3 Ultra processor serves as a somewhat dissatisfying stopgap—prior to last week's official announcement, insiders were still actively questioning the existence of said chip. Last week, Apple representatives reportedly informed Ars Technica and Numerama about their "Ultra" tier not reaching "every chip generation." Follow-up articles have suggested that the M4 Max chip design does not feature an UltraFusion connector; thus cutting off a main path to potential M4 Ultra routes.

Based on previous-gen history, Mac-specialist news sites propose the upcoming M3 Ultra chipset being—in effect—the result of two M3 Max chips joined together via Apple's UltraFusion connection system. Further speculation points to the company's engineering department having to start with a blank canvas; involving a speculative monolithic die design. Noted Apple leaker—Mark Gurman—has disclosed additional theories via his paywalled Bloomberg "Power On" newsletter. As interpreted by MacRumors: "Apple is reluctant to develop an M4 Ultra chip from scratch due to production challenges, costs, and the relatively small sales volume of its desktop computers, like the Mac Studio. So, that seems to rule out the only other way in which Apple could have released an M4 Ultra chip." Several media outlets posit that Apple will skip a generation, and instead focus on getting UltraFusion connections working with next-gen "M5" processors. A refreshed Mac Pro lineup is reportedly on the cards; mid-January reports linked the next-gen workstation series to a very powerful "Hidra" chip design.

AMD has officially stated that its portfolio of RDNA 4 generation graphics cards consists of Radeon RX 9070 XT, RX 9070, and RX 9060 XT models. Last December, a longer list of GPUs appeared online—this pre-Christmas 2024 leak revealed Radeon RX 9060, RX 9050 and RX 9040 identifiers. Team Red seems to be working on populating the lower end of its Radeon RX 9000 family—similarly, Team Green is reportedly expanding its "Blackwell" GPU generation into the "entry-mainstream" segment with their RTX 5050 card design.

Late last week, a Mexican PC hardware retailer inadvertently added Radeon RX 9060 and 9050 categories (not product listings) to its webstore. Noted graphics card watchdogs picked up on the shop's accidental adding of search filters; prompting the swift removal of NDA-busting material (by the time of writing). Resultant VideoCardz news coverage has spread the word of possible forthcoming launches of much cheaper RDNA 4 GPU models. AMD and involved board partners are likely aiming to get lower-mid-range Radeon RX 9060 XT cards out of the door by the second quarter of 2025. Recent leaks indicate Radeon RX 9060 (non-XT) and RX 9050-class products potentially arriving at retail closer to summertime, with speculated "Navi 44" GPUs onboard.

Last week, Acer registered a glut of new graphics card SKUs—an extensive EEC filing revealed several unannounced Nitro and Predator BiFrost Radeon RX 9060 XT models. The AMD board partner seems to be preparing two memory configurations for its lower-end RDNA 4 custom cards: 16 GB and 8 GB (GDDR6). VideoCardz reached out to AIB industry moles, in order to establish whether the aforementioned specification points were concrete. A past weekend report "reconfirms" incoming 16 GB and 8 GB variants. Going back to RDNA 3, Team Red released its Radeon RX 7600 XT GPU in 16 GB form, while the cheaper Radeon RX 7600 made do with a VRAM pool of 8 GB. An unnamed board partner's list of launch day offerings reportedly contains basic technical specs (see screenshot below).

Leaked Radeon RX 9060 XT details show 16 GB GDDR6 and 8 GB GDDR6 variants, with 128-bit memory interfaces. VideoCardz believes that 20 Gbps memory chips will be utilized, as featured on recently launched Radeon RX 9070 series cards. The leaked product list indicates that required juice will be delivered through a lone 8-pin power connector; a minimum PSU requirement suggestion is 500 W, although this could jump up to 550 W for factory overlocked SKUs. A speculative "Navi 48 LE" GPU variant is present within TechPowerUp's GPU database entry for the Radeon RX 9060 XT. VideoCardz reckons that a "Navi 44" GPU would be more appropriate for this class; perhaps derived from a mobile part. The leak suggests the presence of a single HDMI 2.1 port, as well as two DP 2.1 ports—a smaller GPU would have trouble driving more than that. During an official RDNA 4 launch event—held in Beijing, late last month—a Team Red representative announced a loose Q2 2025 launch window for Radeon RX 9060 XT graphics cards.

A former Respawn Entertainment employee has revealed the recent cancellation of an "unannounced incubation project." The ex-team member's LinkedIn profile post/job-seeking alert (now removed) was widely reported on by games media outlets; leading to speculation regarding the mystery title's origins and identity. The Los Angeles, California-based company created a "new game incubation team" not too long ago—recruitment information from last year pointed to a role description that demanded: "significant professional experience designing multiplayer FPS experiences." Respawn Entertainment operates as a multi-team and multi-location development studio—with branches working on Apex Legends, Star Wars and mystery IPs. The Electronic Arts-owned firm is masterminded by Vince Zampella; a veteran developer of popular first-person shooter franchises. Based on recent official account disclosures, the EA executive is currently tasked with the challenging revival of Battlefield's fortunes.

In 2023, Zampella indicated that he was open to revisiting Respawn's long-dormant Titanfall intellectual property. The second entry in the series underperformed sales-wise when it first launched back in October 2016. Despite a fantastic critical reception and vocal support from Titanfall 2's audience, EA leadership decided to put the franchise on ice. Last March, insiders started to make speculative noise about a revival—Steve Fukuda was mentioned in leaks. His CV lists the helming of development on both Titanfall 1 and 2; as game director. The latest talk of a canceled project was soon linked back to Respawn's rumored "unannounced Titanfall IP" prototype. Respawn's website still advertises the existence of their "new game incubation team," but links lead to recruitment posts for other "active" projects. EA officially announced the shelving of the studio's "Star Wars FPS Action game" around early 2024. Respawn and EA are likely focusing on core franchises for the foreseeable future; their Star Wars Jedi: Survivor sequel seems to be a safe bet.

Earlier in the week, online chatter pointed to a possible delay in the production of Panther Lake silicon. Well-known industry analyst—Ming-Chi Kuo—has kept tabs on the inner workings of several big semiconductor players. A previous insider tale revealed NVIDIA's allegedly revised "Blackwell" architecture roadmap. Kuo's latest insight focused on Intel and their 18A node process; rumored setbacks have reportedly pushed the launch of next-gen Panther Lake (PTL) mobile processors into 2026. Team Blue leadership has already reacted to the relatively fresh allegations—earlier in the week, John Pitzer sat down with Morgan Stanley Semiconductor Research's Joe Moore. During their conference fireside chat, Intel's Corporate Vice President of Investor Relations addressed recent internet whispers.

When asked about 18A being developed on schedule, Pitzer responded with: "yes, it is. I mean, I tend to wake up every morning trying to fish through rumors that are coming across on social media about Intel 18A. I want to be very clear. Panther Lake is on track to launch in the second half of this year. That launch date has not changed. We feel really good about the progress that we are making. In fact, if you look at where our yields are on Panther Lake today, they're actually slightly ahead at a similar point in time to Meteor Lake, if you look at the development process for Meteor Lake. I think a couple of weeks ago, there was a technical paper out that actually looked at our SRAM density on Intel 18A that compared well with TSMC's N2. Lots of different metrics you can compare technologies on. I think in general, we think about Intel 18A being an N3 type/N2 sort of comp with the external peers." Panther Lake is set to become the company's first product family that will utilize its own Foundry's 18A node process. Mid-way through February, we heard about the importance of PTL with Intel's portable gaming strategy.

AMD's Ryzen 9 9950X3D and 9900X3D processors are reportedly due to launch mid-way through next week (March 12)—this "unofficial" release date was revealed by JD.com listings. Yesterday's report focused on the popular Chinese e-commerce platform's apparent leaking of Team Red's mid-March schedule. Last month, a local tipster—Golden Pig Upgrade—also alluded to a possible March 12 rollout of 16 and 12-core 3D V-Cache-equipped Zen 5 chips. In a surprise move, JD has started selling Ryzen 9 9950X3D and 9900X3D units. Earlier today, ITHome spotted updated product pages—JD disclosed that it was prepping an earlier than expected sale: "limited time and limited quantity available at 20:00 (local time) on March 7."

Officially, AMD has only teased a loose March launch window for its latest Ryzen Niners. VideoCardz has kept track of the company's recent announcements, and opines that the Ryzen 9000X3D release strategy is strange one. Embargoes for reviewer and influencers are tipped for lifting on March 11, so JD's premature sale of Ryzen 9 9950X3D and 9900X3D processors clashes with this schedule. JD's unusual 20:00 release time was highlighted by VideoCardz; they reckon that the 22:00 hour is a more appropriate kick-off time. It is possible that the limited quantity/short period sale was a mislabeled promotion for pre-orders—something could be lost in translation (see primary screenshot below), but similar events were reported in recent history.

A mysterious GameSir wireless gaming controller appeared online late last year, courtesy of video footage uploaded by an unidentified tester. The presence of a Steam logo on the leaked device's central "home/guide" button sent speculation activities into overdrive. The so-called GameSir "Steam Deck Controller" was linked back to a teased Xbox-licensed G7 Pro wireless controller—both designs seem to share a white/gray aesthetic and control layout. At CES 2025, TechPowerUp staffers were fortunate enough to experience hands-on time with GameSir's upcoming Xbox-branded flagship model. Curious visitors did not spot any sign of Steam/Valve-related products at the company's Las Vegas booth. Game Tech Talk—a YouTube channel that covers the latest in handheld gadgets—managed to conduct an interview with Betta; GameSir's CEO. Their virtual discussion was uploaded two (or so) days ago; the 28-minute video explores current and forthcoming products.

When asked about the unannounced "G7 Pro Valve Edition" (aka unofficial Steam Deck Controller), Betta admitted that he was behind last year's leak. This guerilla marketing tactic produced a lot of reactions and feedback from the hardcore gaming controller community; all taken onboard by the young entrepreneur. Betta said that his personal prototype was "hacked together" in order to provide proof of concept to Valve, although he states that GameSir only has a loose relationship with the House of Half-Life. The G7 Pro controller served as a basic foundation for the Steam-oriented variant—with added motion controller capabilities, capacitive stick technology, independently addressable back buttons, and integration into Valve's operating "ecosystem." The fledgling project awaits an official signing-off—with agreements over licensing, branding, software support, etc.—but a firm connection between the two parties is not yet established. Apparently, there were brief interactions during January's CES event. Betta believes that: "people at Valve are pretty hard to reach." GameSir leadership is "figuring out" a plan; they hope to form a direct connection to Valve HQ in the near future.

The Chiphell forum has provided a steady flow of AMD RDNA 4-related leaks—going back to early December; members believed that "Radeon RX 8800 XT" GPUs were imminently entering into a mass production phase. Since then, Team Red and board partners have officially revealed a full deck of Radeon RX 9070 XT and RX 9070 models—complete with a "modernized" naming scheme. By Christmas (2024), insiders appeared to have working units in their clutches—denizens of Chiphell have continued to dole out pre-release info; even deep into launch week. UNIKO's Hardware picked up on the latest signals; with owners of unnamed custom Radeon RX 9070 XT cards: "calling out bad cooling on GDDR6 VRAM."

As highlighted by Wccftech, NDA-busting disclosures have alluded to commendable GPU thermal measurements—when driven at full load—but several leakers have noted less than stellar results from VRAM temperature readings. Chiphell-sourced GPU-Z screenshots indicate a maximum recorded VRAM temperature of 94℃, with the involved GPU's hotspot hitting a top temp of 79°C. This leaked candidate seems to be a 329 W TBP-rated model. Wccftech observed inconsistencies with the other evaluated sample: "under full load, the GPU temperature at the hot spot reaches a max of 63°C but the memory temperature touches 88°C. The surprising thing to note in the first case is that the TBP is only 237 W, which seems weird considering the Radeon RX 9070 XT (reference spec) starts at a TBP of 260 W. This might be a bug...The user has confirmed that the first one isn't the RX 9070 as one would think, but it is the RX 9070 XT as well." The "guided" upper limit for newer VRAM standards is 95°C, so one of the anonymous custom cards is dancing dangerously in close proximity to the proverbial flame. Certain hardware news outlets reckon that GDDR6X memory will succumb to damage once a 120°C ceiling is hit. Hopefully, these issues are limited to a handful of review samples—a couple of AMD's trusted board partners have opted for Honeywell PTM7950 thermal pads and robust heatpipe formations. Stay tuned to TechPowerUp for W1zzard's incoming RDNA 4-related verdicts.

The "Medusa" or "Medusa Point" codename started to appear online over the past couple of months. These mysterious AMD projects were linked to next-generation "Zen 6" Ryzen desktop and mobile processor families (respectively). Initially, insiders reckoned that Team Red had selected an RDNA 4-based graphics solution for integration their futuristic new-gen laptop APU design. Two days ago, Golden Pig Upgrade weighed in with a different theory—the veteran leaker believes that provisions have regressed on the "Medusa Point" iGPU front.

Previous reports have suggested that the "Medusa Point" processor's iGPU aspect will utilize up to 16 compute units (CU), based on a theorized count of eight workgroup processors (WGPs) from leaked imagery. The latest insider tip points to the utilization of a non-specific "RDNA 3.x" branch, instead of conjectured RDNA 4 graphics technology. Industry watchdogs hold the belief that AMD will be sticking with RDNA 3.5 for a while—as featured on their current-gen "Zen 5" mobile-oriented Strix Point, Strix Halo and Krackan Point chips. As pointed out by Notebookcheck, Team Red leadership disclosed that RDNA 4 is exclusive to discrete card families (for the time being). RDNA 3.5-equipped APUs have—so far—received a warm welcome; AMD engineers could be reserving development resources for a distant future project.

Samsung's foundry operation has experienced many setbacks over the past six months, according to a steady feed of insider reports. Last November, industry moles leaked details of an apparent abandonment of the company's 3 nm Gate-All-Around (GAA) process. Significant yield problems prompted an alleged shift into 2 nm territories, with a next-gen flagship Exynos mobile processor linked to this cutting-edge node. According to a mid-week Chosun Daily article, Samsung and its main rival—TSMC—are in a race to establish decent yields of 2 nm wafers, ahead of predicted "late 2025" mass production kick-offs. The publication's inside track points to the Taiwanese foundry making the most progress (with an estimated 60%), but watchdogs warn that it is too early to bet against the South Korean competitor.

Despite murmurs of current 20 - 30% yields, the Samsung's Hwaseong facility is touted to make "smooth" progress over the coming months. Chosun's sources believe that Samsung engineers struggled to get 3 nm GAA "up to snuff," spending around three years on development endeavors (in vain). In comparison, the making of 2 nm GAA is reported to be less bumpy. A fully upgraded "S3" foundry line is expected to come online by the fourth quarter of this year. An unnamed insider commented on rumors of better than anticipated forward motion chez Samsung Electronics: "there are positive aspects to this as it has shown technological improvements, such as the recent increase in the yield of its 4 nm process by up to 80%." Recent-ish reports suggest that foundry teams have dealt with budget cuts, as well as mounting pressure from company leadership to hit deadlines.

AMD's proper introduction of first wave RDNA 4 graphics cards has focused on Radeon RX 9070 XT and RX 9070 models; officially due for retail release next week (March 6). Lower-end options popped up—via naming scheme material—during CES 2025, albeit with little fanfare. According to the latest reports emerging from China's tech press, AMD teased an upcoming Radeon RX 9060 Series launch. TechPowerUp's well-maintained GPU database has listed a speculative Radeon RX 9060 XT 12 GB model since early January; a fresh update has revealed the "theorized" existence of a Team Red "Navi 48 LE" GPU.

As reported earlier today, AMD rolled out a "surprise" in-person Chinese RDNA 4 event—in contrast, Western audiences have just watched an official virtual presentation. According to ITHome and VideoCardz, a Team Red representative slipped in a brief mention of the forthcoming Radeon RX 9060 Series. This announcement was made right at the tail end of the company's presentation (in China); teasing a second quarter 2025 launch window. AMD did not go into great detail—ITHome divulged the company's key goals for forthcoming Radeon RX 9060 cards: "mid-range market, focusing on cost-effectiveness, and aiming to provide high-performance/low-power solutions for 1080p and 1440p gamers." Local insiders reckon that the lower-end of AMD's RDNA 4 lineup will be downgraded with fewer computing units and a pool of 8 GB GDDR6 VRAM. ITHome's sources believe that Radeon RX 9060 TBPs are expected to be somewhere "between 150 W and 200 W." Speculated MSRP info indicates a range of $349 to $449 (USD) pricing.

According to South Korea's Money Today, SK hynix is currently engaged in the development of yet another variation of LPDDR5. The mega supplier of DRAM and flash memory chips has publicly disclosed its LPDDR5 Turbo (T) design—going back to late 2023; this iteration was advertised as the "world's fastest mobile memory standard." The first public demonstration of LPDDR5T (10533) was performed at last February's IEEE Solid State Circuit Conference. Currently, the familiar LPDDR5X standard is prevalent throughout commercial channels. Insiders believe that a proposed new "LPDDR5M" design will be released as a lower power alternative to LPDDR5X.

Insiders reckon that the unannounced LPDDR5M standard operates at lower voltages (reportedly 0.98 V), when compared to current offerings (X: 1.05 V). Given the nature of its acronym—Low Power Double Data Rate—this memory type was first devised with efficient operations in mind; ideal for mobile applications. An industry mole proposes that internal company discussions have highlighted a key percentage difference: "at maximum speed, LPDDR5M is ~8% more power efficient than LPDDR5X." The recent Money Today SK news article mentions that older LPDDR4 standards are classed as "legacy products" by company leadership. In contrast, LPDDR5 variants are (allegedly) categorized as "high value-added products." The rumored addition of LPDDR5M is viewed—by regional memory industry watchdogs—as a fortification (and diversification) of SK hynix's strategy; that already encompasses LPDDR5X and LPDDR5T. Tipsters posit that LPDDR5M memory is destined to feature inside next-gen smartphone devices with on-board AI capabilities.

PowerColor's Radeon RX 9070 XT Hellhound model is a known property—in terms of its visual presentation—due to a public unveiling at CES 2025, as well as renders appearing online via official product pages. The Taiwanese manufacturer has decided to update its custom graphics card designs for AMD's incoming RDNA 4 generation—for example; the flagship Red Devil family is also refreshed for 2025. So far, PowerColor has showcased "standard black" Red Devil, Hellhound and Reaper designs. Rumors of additional Spectral White variants were circulating online earlier in the week.

Chinese market-exclusive PowerColor Radeon RX 7650 GRE Reaper models were launched very recently, complete with a Spectral White option. Following this official launch, VideoCardz picked up on inside track information; pointing to possible pale variants of forthcoming RDNA 4-based Red Devil cards. Days later, another PowerColor leak has unearthed an unannounced Spectral White SKU, albeit in Hellhound flavor. This product family usually offers a nice balance of high-end features and favorable pricing; we hope to see concrete details tomorrow. The leaked Hellhound card seems to feature an almost all-white aesthetic; extending to its PCB design and I/O bracket. No major surprises were disclosed in VideoCardz's report, but they noted a potential absence of RGB lighting zones. Previous-gen Hellhounds sport "fixed-color" schemes; enabling blue or purple lighting.

A formerly dormant member of China's Chiphell discussion board has resurfaced with a major GALAX leak; they claimed ownership of a pre-release GeForce RTX 5090D HOF OC LAB edition graphics card. The Hong Kong-based AIB did not update its "Hall of Fame" series with flagship NVIDIA "Blackwell" GPU silicon in time for last month's launch. Their highest-end offering arrived in the form of a GeForce RTX 5080 HOF OC LAB Plus-X SKU; reportedly an over-specced juggernaut. According to the latest reports, GALAX's top GeForce RTX 5090D candidate is in the pipeline—early details indicate an impressive feature set. The manufacturer is well known for its selective methodology; GPU industry watchdogs believe that the incoming flagship utilizes only the very best pre-binned GB202 GPU dies and GDDR7 memory modules.

Despite a fitting of elite-tier parts, the lucky owner reported issues with their allegedly sanction box-ticking card: "this time the overclocking was more troublesome. The bracket and backplate of the graphics card gun had to be remade. Finally, it was done. I won't say more. I'm going to fill it with liquid nitrogen." VideoCardz believes that GALAX has sent preview samples to local overclocking specialists, minus finalized air-cooled solution attachments. As disclosed by the Chiphell member
(laoshu1919), early evaluators appear to be experimenting with liquid cooling. A photo of GALAX's GeForce RTX 5090D HOF OC LAB PCB design was shared on the forum; showing a grand total of 38 high-quality power phases. The GPU die is surrounded by 26 of these units, while 10 are positioned closer to the card's VRAM. Another 2 are located nearby to a single 16-pin power connector, with enough extra room to accommodate two more. Judging from laoshu1919's photo, this prototype seems capable of hosting another 16-pin power connector. The sample's I/O bracket likely sports GALAX's signature "1-Click OC" button.

Sapphire's premium NITRO+ Radeon RX 9070 XT graphics card model was the subject of several leaks in the recent past—unsurprisingly, gaming GPU detectives gathered evidence of a very high speculative price point. The AIB's top "champagne gold" offering is due for a full unveiling; we expect to see examples tomorrow—AMD will broadcast a special Radeon RX 9070 Series presentation. Within the past 24 hours, VideoCardz received a compelling tip-off—the GPU news specialist was pointed in the direction of TechPowerUp. Currently, the site's VGA BIOS Collection hosts a downloadable "Sapphire 9070 XT 16 GB BIOS (Nitro+)" ROM.

The VGA BIOS build date is listed as "2024-12-13," with the file becoming available to download roughly two days ago (February 25). VideoCardz highlighted interesting "BIOS Internals" information; namely the mentioning of a "Navi48 XTX" GPU variant. Since CES 2025, AMD and its board partners have kept quiet about finer RDNA 4 details, but insiders and leakers have noted the existence of a generic "Navi 48" GPU. TechPowerUp's GPU database listings of AMD Radeon RX 9070 XT 16 GB and RX 9070 16 GB (non-XT) are freshly updated; reflecting new information sourced from pre-launch VGA BIOS listings. The incoming flagship seems to utilize a "Navi 48 XT" GPU variant (note: not 48 XTX), while its non-XT sibling is (supposedly) based on "Navi 48 XL." VideoCardz has heard whispers of a next-gen "Navi XTXH," industry moles have linked this sub-model to a supposed "higher-end" 32 GB RDNA 4 card. TPU's GPU database also alludes to an AMD Radeon RX 9060 XT GPU; just updated with a "Navi 48 LE" designation.

iBUYPOWER and AVGPC appear to be preparing PC pre-built systems featuring upcoming AMD Radeon RX 9070 XT and RX 9070 graphics cards; Newegg listings were detected by Everest/Olrak29 during yesterday evening's sleuthing session. System integrators have likely been sitting on distributed RDNA 4 stock for a while, but finalized builds seem to be prepped and ready for launch next month. As reported by VideoCardz, an AVGPC "Whirwlind" system—priced at $1799 (with free shipping)—is based on Team Red hardware; namely a Ryzen 7 9700X CPU and a "Radeon RX 9070 XT 16G" card.

The cheapest "leaked" iBUYPOWER Y40 PRO build contains a Ryzen 7 9700X processor, and a "Radeon 9070 16 GB" graphics card. This system's listed price was $1749.99 (shipping is gratis), prior to Newegg's scrubbing. A $50 upcharge grants access to a Ryzen 7 7800X3D-equipped Y40 PRO model, featuring the same non-XT card. The most expensive iBUYPOWER desktop option—going for $1949.99—houses a Ryzen 9 9900X CPU and a "Radeon 9070 XT 16 GB" discrete graphics solution. Via crafty deduction, VideoCardz reckons that the iBUYPOWER price differences indicate a $130 gulf between non-XT and XT siblings. Naturally, this could be subject to change—we expect to absorb nitty-gritty MSRP details tomorrow, during AMD's special RDNA 4 event. As noted by everyone's favorite source of GPU news; a single (not scalped) NVIDIA GeForce RTX 5090 graphics card seemingly costs roughly the same as an entire RDNA 4-based pre-build.

A Valve-designed next-gen VR headset is a compelling prospect, but not many details have emerged online since the company's (2022) teasing of new developments. Press outlets noticed a September 2023 registration of a mysterious device in South Korea; following this discovery, many anticipated a steady flow of leaks. Relative silence ensued; brewing speculation that Valve Corporation had shelved another top secret project. Fast-forward to the present day; Gabe Follower—a self-confessed Valve/Half Life 3 tipster—believes that all systems are go for a codenamed "Deckard" VR headset. The amusingly-named leaker (referencing Gabe Newell) claims to have an inside track: "several people have confirmed that Valve is aiming to release new standalone, wireless VR headset (codename Deckard) by the end of 2025. The current price for the full bundle is set to be $1200. Including some "in-house" games (or demos) that are already done. Valve want to give the user the best possible experience without cutting any costs."

The "Index" was Valve's first attempt at cornering the VR gaming market, but the buying public (back in 2019) largely favored rival models. A long-gestating follow-up would need to really "hit it out of the park," when placed against a new generation of competing hardware. Valve's alleged Blade Runner-themed device, was linked to a set of previously leaked VR controllers; codenamed "Roy." Additionally, Gabe Follower reckons that Valve is prepared to make financial sacrifices, in order to establish a foothold within an extremely competitive market: "even at the current price, it will be sold at a loss. A few months ago, we saw leaked models of controllers—Roy—in the SteamVR update. It will be using the same SteamOS from Steam Deck, but adapted for virtual reality. One of the core features is the ability to play flat-screen games that are already playable on Steam Deck, but in VR on a big screen without a PC. The first behind closed doors presentations could start soon."

XFX has not generated much official RDNA 4 noise, ahead of this Friday's special event. The long-term AMD board partner unveiled a pair of mysterious Radeon RX 9070 series cards at CES 2025, but no marketing/hype campaign was launched in the interim. VideoCardz and its insider networks have kept tabs on XFX activities; mostly from retail sources. Their latest report concentrates on custom Radeon RX 9070 XT options; claiming that a grand total of nine SKUs are destined for launch next month. Around mid-February, a Canadian e-tailer inadvertently published a comprehensive selection of unannounced XFX MERCURY, SWIFT, and QUICKSILVER models. Days later, the South Korean and Canadian branches of Amazon uploaded packaging imagery.

VideoCardz has gathered product renders and basic spec information from various retail leaks, as well as inside sources. They believe that XFX has (wisely) dropped its old/odd naming schemes—e.g.: SWFT and QICK—in favor of traditional spellings. The high-end MERCURY population count is (allegedly) four; consisting of two Magnetic Air variants, and two normal editions (sporting non-modular fans). Reportedly, an "M" tag adorns Magnetic Air model retail boxes (refer to images below)—VideoCardz believes that XFX's red-ringed fans indicate detachable functionality, but this color coding is only visible on the black version. As previously reported, XFX's upper-tier Radeon RX 9070 XT custom options are touted to sport the "greatest" factory-set overclock: 3.1 GHz. Press outlets have noted the presence of three 8-pin power connectors in leaked MERCURY renders.