Instructions per clock is a metric used to define and compare CPU architecture performance usually. However, enthusiast colleagues at ComputerBase had an idea to test the IPC improvement in GPUs, comparing it across current and past generations. NVIDIA's Blackwell-based GeForce RTX 50 series faces off against the Ada Lovelace-based RTX 40 generation, while AMD's RDNA 4-powered Radeon RX 9000 lineup challenges the RDNA 3-based RX 7000 series. For NVIDIA, the test used RTX 5070 Ti and 4070 Ti SUPER, aligning ALU counts and clock speeds and treating memory bandwidth differences as negligible. For AMD, the test matched the RX 9060 XT to the RX 7600 XT, both featuring identical ALUs and GDDR6 memory. By closely matching shader counts and normalizing for clock variations, ComputerBase isolates IPC improvements from other hardware enhancements. In rasterized rendering tests across 19 popular titles, NVIDIA's Blackwell architecture delivered an average IPC advantage of just 1% over the older Ada Lovelace.

This difference could easily be attributed to normal benchmark variance. Ray tracing and path tracing benchmarks showed no significant IPC uplift, leaving the latest generation essentially on par with its predecessor when normalized for clock and unit count. AMD's RDNA 4, by contrast, exhibited a substantial IPC leap. Rasterized performance improved by around 20% compared to RDNA 3, while ray-traced workloads enjoyed a roughly 31% gain. Path tracing results were even more extreme, with RDNA 4 delivering nearly twice the FPS, a 100% increase over its predecessor. These findings suggest that NVIDIA's performance improvements primarily stem from higher clock speeds, increased execution unit counts, and enhanced features. AMD's RDNA 4 represents a significant architectural advance, marking its most notable IPC gain since the original RDNA launch.

An AMD research team has introduced a game-changing approach to procedural tree creation that runs entirely on the GPU, delivering both speed and flexibility, unlike anything we've seen before. Showcased at High-Performance Graphics 2025 in Copenhagen, the new pipeline utilizes DirectX 12 work graphs and mesh nodes to construct detailed tree models on the fly, without any CPU muscle. Artists and developers can tweak more than 150 parameters, everything from seasonal leaf color shifts and branch pruning styles to complex animations and automatic level-of-detail adjustments, all in real-time. When tested on an AMD Radeon RX 7900 XTX, the system generated and pushed unique tree geometries into the geometry buffer in just over three milliseconds. It then automatically tunes detail levels to maintain a target frame rate, effortlessly demonstrating stable 120 FPS under heavy workloads.

Wind effects and environmental interactions update seamlessly, and the CPU's only job is to fill a small set of constants (camera matrices, timestamps, and so on) before dispatching a single work graph. There's no need for continuous host-device chatter or asset streaming, which simplifies integration into existing engines. Perhaps the most eye-opening result is how little memory the transient data consumes. A traditional buffer-heavy approach might need tens of GB, but AMD's demo holds onto just 51 KB of persistent state per frame—a mind-boggling 99.9999% reduction compared to conventional methods. A scratch buffer of up to 1.5 GB is allocated for work-graph execution, though actual usage varies by GPU driver and can be released or reused afterward. Static assets, such as meshes and textures, remain unaffected, leaving future opportunities for neural compression or procedural texturing to further enhance memory savings.

Colorful Technology Company Limited, a leading brand in gaming PC components, gaming laptops, and HiFi audio products, proudly introduces the SMART 900 AI Mini PC - a powerful, compact system designed to accelerate AI-driven content creation. Engineered for students and professionals, the SMART 900 combines cutting-edge performance with a sleek industrial design, pushing the boundaries of what a mini-PC can deliver.

At its core is the new Ryzen AI Max+ 395 processor, built on the advanced Zen 5 architecture. With 16 cores and 32 threads, a turbo frequency of up to 5.1 GHz, and 80 MB of cache, it delivers desktop-class performance in a 4-liter form factor. The integrated Radeon 8060S graphics, based on the RDNA 3.5 architecture with 40 compute units, offer GPU performance rivaling discrete mobile GPUs. For AI tasks, a dedicated 50 TOPS NPU provides outstanding acceleration for machine learning and generative workloads. The SMART 900 is built for creators on the move. It features a rich set of connectivity options including 3x USB 3.2 Gen 2, 2x USB 4.0, 2x USB 2.0, and a high-speed SD 4.0 card slot - making high-resolution image and video transfer seamless and efficient.

Just as the RDNA 4 rollout cycle is completing, the rumor mill for the next generation of Radeon GPUs is intensifying. According to @Kepler_L2 on X, a reliable leaker, AMD is equipping its next-generation UDNA-based GPUs with the HDMI 2.2 display protocol, capable of supporting 64 and 80 Gbit/s, meaning that the highest-end bandwidth of 96 Gbit/s (Ultra96) is not expected to arrive until the next generation. The HDMI Forum, the consortium behind the HDMI standard, officially released the HDMI 2.2 specifications in January during CES. The HDMI forum also foresees HDMI 2.2 cables hitting the shelves sometime in Q3 or Q4 of 2025, which means the rollout is near. However, AMD's next-generation GPUs will not utilize its entire feature spectrum in the coming GFX13 GPU architecture.

The HDMI 2.2 update introduces a future-proof Fixed Rate Link architecture that delivers up to 96 Gbps of bandwidth, enabling uncompressed 4K at 240 Hz or 8K at 60 Hz with full 4:4:4 chroma, and paving the way for even higher resolutions, such as 10K and 12K at 120 Hz. It accommodates both uncompressed and visually lossless compressed formats, as well as advanced chroma sampling options. This massive data throughput meets the growing demands of immersive audiovisual experiences from next-generation gaming and augmented and virtual reality to professional light-field displays, large-format digital signage, medical imaging, and machine-vision systems by doubling payload capacity every two to three years. The Ultra96 HDMI cable is designed to handle all HDMI 2.2 features. This cable is accompanied by the new Latency Indication Protocol, which ensures precise lip-sync and audio-video alignment across complex multi-hop setups involving receivers, soundbars, and other downstream devices.

Within the discrete graphics card sector, NVIDIA achieved a remarkable 92% share of the add-in board (AIB) GPU market in the first quarter of 2025, according to data released by Jon Peddie Research (JPR). This represents an 8.5% increase compared to NVIDIA's previous position. By contrast, AMD's share contracted to just 8%, down 7.3 points, while Intel's presence effectively disappeared, falling to 0% after losing 1.2 points. JPR reported that AIB shipments reached 9.2 million units during Q1 2025 despite desktop CPU shipments declining to 17.8 million units. The firm projects that the AIB market will face a compound annual decline of 10.3% from 2024 to 2028, although the installed base of discrete GPUs is expected to grow to 130 million units by the end of the forecast period. By 2028, an estimated 86% of desktop PCs are expected to feature a dedicated graphics card.

NVIDIA's success this quarter can be attributed to its launch of the RTX 50 series GPUs. In contrast, AMD's RDNA 4 GPUs were released significantly later in Q1. Additionally, Intel's Battlemage Arc GPUs, which were launched in Q4 2024, have struggled to gain traction, likely due to limited availability and low demand in the mainstream market. The broader PC GPU market, which includes integrated solutions, contracted by 12% from the previous quarter, with a total of 68.8 million units shipped. Desktop graphics unit sales declined by 16%, while notebook GPUs decreased by 10%. Overall, NVIDIA's total GPU share rose by 3.6 points, AMD's dipped by 1.6 points, and Intel's declined by 2.1 points. Meanwhile, data center GPUs bucked the overall downward trend, rising by 9.6% as enterprises continue to invest in artificial intelligence applications. On the CPU side, notebook processors accounted for 71% of shipments, with desktop CPUs comprising the remaining 29%.

ASRock, the leading global motherboard, graphics card and mini PC manufacturer, today launched the new Steel Legend and Challenger series graphics cards based on the latest AMD Radeon RX 9060 XT GPU. As the latest member of the AMD Radeon RX 9000 series GPUs, AMD Radeon RX 9060 XT also uses the AMD RDNA 4 architecture, which is equipped with the 3rd Gen ray tracing and 2nd Gen AI accelerators, provides up to 32 AMD RDNA 4 compute units, and supports AMD HYPR-RX, AMD FidelityFX Super Resolution (FSR) 4 and advanced Frame Generation..., and other AMD's latest exclusive technologies, plus the latest specifications such as DisplayPort 2.1a, PCI Express 5.0, and also a large-capacity up to 16 GB GDDR6 memory. These rich features and specifications provide end users with an excellent AAA gaming experience.

Otherwise, ASRock AMD Radeon RX 9060 XT Series graphics cards also support various ASRock exclusive features, including the Striped Ring/Axial Fan, Air Deflecting Fin, Ultra-fit Heatpipe, Metal Backplate, and Polychrome SYNC technology to provide great cooling efficiency, solid construction and fancy ARGB lighting effects. With these exclusive features, ASRock AMD Radeon RX 9060 XT Series graphics cards are premium choices for gamers and system integrators.

It seems as though an official international launch for the elusive AMD Ryzen 9000G APUs might still be on the cards for later this year, after all. While an announcement was expected at Computex 2025, along with a full-scale retail launch later this year, AMD was suspiciously quiet about its CPUs at the electronics trade show. Now, it looks as though Gigabyte has seemingly started preparing its B650 motherboard line-up for the arrival of AMD's Ryzen 9000G APUs.

On Gigabyte's B650 Gaming motherboard memory compatibility page, the hardware maker seems to have replaced any mention of Ryzen 9000 series CPUs with "Ryzen 9000 G-series CPUs," altogether. It's unclear whether this change means the same memory compatibility table will apply for both AMD Ryzen 9000 and 9000G CPUs, but that seems to be the case. At any rate, this leak suggests that previous rumors claiming a Q4 launch may have been correct. Those same leaks suggested that the Ryzen 9000G APUs would feature AMD Zen 5 CPU cores paired with AMD RDNA 3.5 iGPUs, with the top-spec iGPU likely being capable of running at least some lighter-weight games and e-sports titles. As it stands, the iGPUs found in recent AMD Ryzen CPUs are merely there for basic video output and diagnostic purposes, while the G-series APUs are generally intended for use without a dedicated GPU. It seems reasonable to expect AMD to launch at least one variant in the AMD Ryzen 9000G APU series equipped with the Radeon 890M, which also powers some of the most popular Windows gaming handhelds on the market.

AMD has officially announced the AMD Radeon RX 9070 GRE, the China-exclusive high-end RDNA 4 gaming GPU, along with a May 8 pre-order date. As with previous iterations of AMD's GRE GPUs, the RX 9070 GRE is a trimmed-back version of the higher-tier models, with both less VRAM and fewer compute units than its other RDNA 4 counterparts. Along with the official announcement and the May 8 release date, at least one pre-order retail listing has been spotted online, with retail prices coming in at 4499 CNY ($617 converted). The retail unit in question is an ASUS tri-fan card, so it comes in slightly above the claimed 4199 CNY MSRP.

According to the specifications released by AMD, the RX 9070 GRE features 12 GB of GDDR6 on a 192-bit bus with up to 432 GB/s bandwidth. Instead of the 56 compute units found in the Radeon RX 9070, the 9070 GRE has 48 CUs, which should make it a fair bit slower than the 9070 and 9070 XT. However, the GPU boost frequency has been bumped up to 2,790 MHz on the 9070 GRE, from 2,520 MHz on the RX 9070. AMD claims the new GPU is 6% faster than the RX 7900 GRE in "more than 30 games." The RX 9070 GRE also features a standard 2×8-pin power connector, so no need to worry about melting 12VHPWR cables. Much like the other RDNA 4 GPUs, the RX 9070 GRE has launched without a first-party reference design, meaning it may become difficult to find MSRP variants of the 9070 GRE.

AMD officially released ROCm 6.4, its latest open‑source GPU compute stack, bringing several under‑the‑hood improvements while still lacking official RDNA 4 support. The update improves compatibility between ROCm's user‑space libraries and the AMDKFD kernel driver, making it easier to run across a wider range of Linux kernels. AMD has also expanded its internal testing to cover more combinations of user and kernel versions, which should reduce integration headaches for HPC and AI workloads. On the framework side, ROCm 6.4 now supports PyTorch 2.5 and 2.6 out of the box, so developers can use the latest deep‑learning features without building from source. The Megatron‑LM integration adds three new fused kernels, Attention (QKV), Layer Norm, and ROPE, to speed up transformer model training by combining multiple operations into single GPU passes. Video decoding gets a boost, too, with VP9 support in both rocDecode and rocPyDecode, plus a new bitstream reader module to streamline media pipelines.

Oracle Linux 9 is now officially supported, and the Radeon PRO W7800 48 GB workstation card has been validated under ROCm. AMD also enabled CPX mode with NPS4 memory configurations, catering to advanced memory bandwidth scenarios on MI Instinct accelerators. Despite these updates, ROCm 6.4 still does not officially support RDNA 4 GPUs, such as the RX 9070 series. While community members report that the new release can run on those cards unofficially, the lack of formal enablement means RDNA 4's doubled FP16 throughput, eight times INT4 sparsity acceleration, and FP8 capabilities remain largely untapped in ROCm workflows. On Linux, consumer Radeon support is limited to just a few models, even though Windows coverage for RDNA 2 and 3 families has expanded since 2022. With AMD's "Advancing AI" event coming in June, many developers are hoping for an announcement about RDNA 4 integration. Until then, those who need guaranteed, day‑one GPU support may continue to look at alternative ecosystems.

Surface defects on the GPU die of a PowerColor Radeon RX 9070 XT Hellhound have been linked to excessive thermal hotspots, according to analysis following Igor's Lab's investigation. Microscopic examination revealed 1,934 pits across the silicon die surface despite proper thermal interface material application. The defects affect over one percent of the chip's surface area, with the largest measuring 12.59 µm in depth and 212.36 µm in diameter—significantly exceeding industry tolerances. These imperfections impede heat transfer, causing localized temperatures to reach 113°C, exceeding the 110°C threshold for RDNA GPUs and triggering thermal throttling. The defects appear to stem from flaws in the backgrinding process that thins silicon wafers for packaging. Improper process control creates surface irregularities that compromise thermal conductivity and structural integrity.

The discovery raises questions about the effectiveness of automated optical inspection systems in detecting subtle surface abnormalities. TSMC, the maker of these chips, uses optical recognition tools to see if any defects occurred. AMD stated: "We are aware of the reported issue and believe this to be an isolated incident. We are working with our partners and internal teams to understand the issue." The company maintains this is an isolated case, though the findings suggest potential gaps in quality control protocols. Users experiencing thermal issues with affected RX 9070 XT units should initiate the RMA process for replacement while AMD and PowerColor investigate whether these defects exist beyond the identified sample. We hope no further issues are present, and so far, PowerColor hasn't received any customer complaints. We are on the lookout for further situation development. We reviewed PowerColor's Radeon RX 9070 Hellhound, and found no issues on our unit, so this indeed remains an isolated case so far.

At the tail end of 2024, AMD technical staffers added the "GFX1153" target to their first-party GPU supported chip list. Almost three months later, PC hardware news outlets and online enthusiasts have just picked up on this development. "GFX1150" family IPs were previously linked to Team Red's RDNA 3.5 architecture. This graphics technology debuted with the launch of Ryzen AI "Strix Halo," "Strix Point" and "Krackan Point" mobile processors. Recent leaks have suggested that Team Red is satisfied with the performance of RDNA 3.5-based Radeon iGPUs; warranting a rumored repeat rollout with next-gen "Medusa Point" APU designs.

Both "Medusa Point" and "Gorgon Point" mobile CPU families are expected to launch next year, with leaks pointing to the utilization of "Zen 6" and "Zen 5" processor cores (respectively) and RDNA 3.5 graphics architecture. RDNA 4 seems to be a strictly desktop-oriented generation. AMD could be reserving the "further out" UDNA tech for truly next-generation integrated graphics solutions. In the interim, Team Red's "GFX1153" IP will likely serve as "Medusa Point's" onboard GPU, according to the latest logical theories. Last year, the "GFX1152" target was associated with Ryzen AI 7 300-series "Krackan Point" APUs.

Last week, we got a leaked picture of AMD's reference/OEM/Made by AMD (MBA) design of the Radeon RX 9070 XT in China. However, we have a non-XT card on the menu today, with the reference design also pictured in China. We already know that AMD is not releasing any MBA reference designs of its latest Radeon RX 9000 series GPUs based on the RDNA 4. However, the Chinese black market is the place to be when it comes to finding these MBA cards, probably manufactured in high double-digit or low tripe-digits of units. AMD usually has one of its AIB partners manufacture these for OEM integrated system distribution or partner testing/software optimization.

Pictured below is the Radeon RX 9070 non-XT GPU with an all-black shroud. Unlike official renders, this design is much more toned down. The official render shows a black/gray color combination with an LED-illuminated Radeon logo. However, the official version appears with only a simple aesthetic, without a hint of LED illumination. The card wasn't pictured running, but we assume that there is no LED illumination. Below, you can compare the pictured/leaked Radeon RX 9070 GPU with the official render.

Acer today unveiled its latest Predator BiFrost and Acer Nitro graphics cards, powered by the next-generation AMD Radeon RX 9000 Series GPUs. This cutting-edge line-up includes six models designed to elevate gaming and creative experiences with next-level performance and visual fidelity. The models include the Predator BiFrost Radeon RX 9070 XT OC 16GB, Predator BiFrost Radeon RX 9070 OC 16GB, and four Nitro graphics cards: Acer Nitro Radeon RX 9070 XT OC 16GB, Acer Nitro Radeon RX 9070 OC 16GB, and standard versions of the Acer Nitro Radeon RX 9070 XT 16GB and Acer Nitro Radeon RX 9070 16GB.

The new GPUs are engineered for ultra-fast gaming and demanding tasks, leveraging AMD RDNA 4 compute units and up to 16 GB of memory to deliver immersive 8K visuals. The advanced cooling technology, featuring three FrostBlade 4.0 fans, and reinforced designs ensures systems can be pushed to their limits while maintaining peak performance. DIY gamers will appreciate new software features such as Acer Intelligence Space, which offers intuitive AI system detection, and Acer Game Assistance's adaptive aim system to give players a competitive edge.

MSI has officially confirmed that it will not manufacture graphics cards based on AMD's latest RDNA 4 architecture, effectively bypassing the entire Radeon RX 9000 series lineup. In a statement to Tom's Hardware, an MSI representative briefly noted the company "is not manufacturing AMD GPUs this generation," pausing its AMD partnership while leaving the door open for future collaborations. Data compiled by Tom's Hardware shows MSI produced 45 distinct models during the RDNA 2 generation (RX 6000-series), but dramatically scaled back to just four custom designs for RDNA 3 (RX 7000-series)—representing a 91% reduction in AMD SKU diversity. Those limited RDNA 3 offerings, including the flagship RX 7900 XTX, notably reused cooling solutions from previous-generation AMD cards, indicating reduced R&D allocation compared to the company's NVIDIA lineup.

MSI's withdrawal from offering AMD-based solutions is due to several factors. NVIDIA's dominance in market share (83%, according to Steam hardware surveys) provides partners with stronger return-on-investment potential, while EVGA's 2022 exit from the GPU market created an opportunity for MSI to strengthen its position as a premier NVIDIA partner. There were reportedly delays in finalizing the MSRP for the RX 9000 series, which complicated manufacturers' calculations, production planning, and shipments. The vacuum left by MSI creates potential opportunities for other emerging players like Acer, which has been slowly expanding its AMD graphics card presence. However, MSI's decision—likely finalized during earlier strategic planning cycles—could potentially push away AMD enthusiasts despite strong projected demand for the Radeon RX 9070 series. Whether this represents a permanent strategic realignment or a temporary market response, we are yet to find out. AMD plans to return to the high-end GPU segment with its UDNA generation, so we have to wait and see if MSI makes a comeback here. For AIBs, partnerships with GPU makers are essential to get R&D resources behind a new product. Once partnerships pause, it is hard to get that wheel going again.

AMD is not releasing any "Made by AMD" (MBA) reference designs of its latest Radeon RX 9070 XT GPUs based on the RDNA 4 IP. However, leakers in China managed to get ahold of what appears to be an MBA Radeon RX 9070 XT design, assumingly being used as a prototype. While there are custom designs by AMD's AIB partners, AMD itself hasn't released the reference design to the public. The latest leak from Chinese forums confirms that this GPU actually exists beyond the standard press renders/mockups, meaning that someone can get their hands on it. The seller is offering a brand-new reference edition of the AMD Radeon RX 9070 XT with a warranty for one year, with the second year requiring self-paid repairs. They specify no disassembly and no damage to the unit.

The standard price is set at 5000 RMB, with a preference for local pickup. For buyers outside the city, shipping via SF Express is available with insurance, requiring a payment of at least 5800 RMB. The physical card matches the previously leaked render with its three-fan design, though with a notable color difference. Instead of the expected gray finish, the actual unit features a complete black design. Currently sealed in an antistatic bag, the card appears unused. The asking price of approximately $800 initially seems high but aligns with current market rates for custom versions. The listing has already disappeared, suggesting someone has already snagged this rare prototype. With serial codes visible, AMD can potentially trace the person who put it up for sales, so its not a good outlook for anyone that wanted to sell it.

GIGABYTE TECHNOLOGY Co. Ltd, a leading manufacturer of premium gaming hardware, today launched the GIGABYTE Radeon RX 9000 series graphics cards powered by unified AMD RDNA 4 compute units with enhanced ray tracing capabilities for an ultra-immersive experience across all key gaming resolutions. To satisfy gamers' pursuit of speed, performance, and visuals for a smooth and immersive experience, GIGABYTE offers the AORUS Radeon RX 9070 XT ELITE 16G, Radeon RX 9070 XT GAMING OC 16G, and Radeon RX 9070 GAMING OC 16G, catering to various gaming needs.

GIGABYTE has upgraded the WINDFORCE cooling system, balancing performance and thermal efficiency. The new Hawk Fan design minimizes turbulence and noise, resulting in up to a 53.6% increase in air pressure and a 12.5% increase in air volume without compromising acoustics. Server-grade thermal conductive gel is used to cool critical components such as VRAM and MOSFETs. This highly deformable, non-fluid gel provides optimal contact even on uneven surfaces and effectively resists deformation from transport or long-term use. Combined with advanced thermal technologies like a larger heatsink, vapor chamber and copper plate, these cards ensure optimal performance and quiet operation under demanding workloads.⁠

AMD has announced that beginning with its RDNA 4 architecture, only Unified Extensible Firmware Interface (UEFI) Mode will be officially supported for the Radeon RX 9000 Series graphics cards and later models. This decision is intended to harness the full capabilities of the advanced UEFI, ensuring improved performance, enhanced security, and a smoother user experience. UEFI, a modern replacement for the traditional BIOS, functions as a compact operating system residing on the motherboard. It plays a vital role in initializing processors, configuring memory, and managing essential hardware components. Its features include faster boot times, support for storage devices larger than 2.2 TB, and robust protection mechanisms such as Windows Secure Boot. These capabilities represent a significant leap forward compared to legacy BIOS systems.

To accommodate modern hardware innovations like AMD's next-generation graphics, motherboards typically offer UEFI Mode and Compatibility Support Module (CSM) Mode. CSM emulates older BIOS functions but restricts access to many of UEFI's advanced features, such as AMD Smart Access Memory. By mandating UEFI Mode, AMD ensures that users of its latest graphics cards enjoy optimal compatibility, performance, and security benefits. This requirement is a natural evolution in PC technology, encouraging users to upgrade their system configurations to meet modern standards. As consumers prepare for the next wave of AMD GPUs, verifying that both hardware and operating systems are properly set up for UEFI Mode is a necessary step to fully leverage the advantages offered by the RDNA 4 graphics cards.

ASRock, the leading global motherboard, graphics card and mini PC manufacturer, today launched the new Taichi, Steel Legend and Challenger series graphics cards based on AMD Radeon RX 9070 XT and Radeon RX 9070 GPUs. The new graphics cards are built-in the ultra-fast AMD RDNA 4 architecture graphics which features 3rd Gen ray tracing and 2nd Gen AI accelerators to deliver the performance you need to max out your gaming. Powered by AMD RDNA 4 architecture, the AMD Radeon RX 9070 XT GPU features 64 compute units (3rd Gen ray tracing and 2nd Gen AI accelerators), AMD HYPR-RX, AMD FidelityFX Super Resolution, DisplayPort 2.1, and PCI Express 5.0 ready. And the AMD Radeon RX 9070 features 56 compute units (3rd Gen ray tracing and 2nd Gen AI accelerators), AMD HYPR-RX, AMD FidelityFX Super Resolution, DisplayPort 2.1, and PCI Express 5.0 ready.

The AMD RDNA 4 architecture also delivers various features such as Radiance Display Engine, Fluid Motion Frames 2 technology, Radeon Anti-Lag 2 technology, AV1 Encode/Decode, AMD FidelityFX technologies and AMD Noise Suppression etc. These rich features enhance the gaming comprehensive performance of RDNA 4 and bring much powerful gaming platform to gaming users.

VideoCardz obtained AMD Radeon RX 9070 series specifications, which appear to be the official final configurations of the upcoming RDNA 4 GPUs. As we previously expected, the lineup consists of two models based on the Navi 48 GPU, which integrates 53.9 billion transistors on a 357 mm² die using a 4 nm (N5) process from TSMC. Both the RX 9070 XT and RX 9070 utilize identical memory configurations: 16 GB of GDDR6 memory running at 20 Gbps across a 256-bit bus, delivering 640 GB/s bandwidth. Each card implements 64 MB of 3rd Generation Infinity Cache and supports PCIe 5.0 x16 interface standards. The RX 9070 XT features 64 RDNA 4 Compute Units, equating to 4096 Stream Processors, 64 Ray Accelerators, and 128 AI Accelerators. It operates at a 2400 MHz game clock and 2970 MHz boost clock, providing 48.7 TFLOPS of single-precision FP32 compute performance.

Power requirements include a 304 W TBP and a recommended 750 W power supply. The standard RX 9070 reduces specifications to 56 Compute Units (3584 Stream Processors), 56 Ray Accelerators, and 112 AI Accelerators. Clock speeds decrease to 2070 MHz game clock and 2540 MHz boost clock, with correspondingly lower power requirements of 220 W TBP and a recommended 650 W power supply. Since both SKUs use the same Navi 48 die, the separation between them is likely better binning for the XT version, and lower bins end up for the non-XT version. Both models support HDMI 2.1b and DisplayPort 2.1a UHBR13.5 outputs. AMD has confirmed the cards will launch exclusively through board partners with no reference designs planned and that the official unveiling will be in March. Earlier rumors have suggested a $699 price tag for the Radeon RX 9070 XT SKU, putting its expected price/performance near NVIDIA's GeForce RTX 5070 Ti. AMD notes that 85% of gamers buy cards below $700, which the RDNA 4 series will focus on.

ASUS's Republic of Gamers China account on Weibo has teased the ROG Magic X mobile device that combines laptop/tablet form with a detachable keyboard. Inside, the device is powered by AMD Ryzen AI MAX+ SoC, which ASUS called a "three-in-one" chip, mainly due to its CPU, NPU, and iGPU combination. One possible SKU for ROG Magic X is AMD's top-end Ryzen AI MAX+ 395 processor. Carrying 16 "Zen 5" cores and 32 threads, the chip is designed for AI-enhanced laptops with 126 combined TOPS of AI processing power. The Ryzen AI MAX+ 395 features 64 MB of L3 cache and operates at a base clock of 3 GHz, with boost capabilities up to 5.1 GHz, depending on workload conditions. Manufactured on TSMC's 4 nm process node, the processor maintains a modest 55 W TDP, suitable for high-performance mobile systems. The chip includes support for ECC memory and PCIe Gen 5. It integrates the Radeon 8060S solution based on RDNA 3.5 architecture for graphics.

We are yet to see more details about the ROG Magic X, but with the arrival of AMD Ryzen AI MAX+, we assume this machine will result in a powerful gaming device for users on the go. More details are expected on February 25, when ASUS plans to showcase it. Pricing and availability are also expected to follow soon after.

David McAfee—AMD's corporate vice president and general manager of client channel business—has highlighted February 28 as a highly important date for next-gen graphics technology. The much-anticipated (and teased) Radeon RX 9000 series unveiling event has a confirmed time slot on that day: 8 AM EST. The overseer of Ryzen CPU and Radeon GPUs has warded off curious queries from journalists and members of the PC hardware for several weeks, since the conclusion of CES 2025. A confusing early January presentation did not include a segment dedicated to upcoming RDNA 4 products. Online conjecture pointed to Team Red delaying and restrategizing the launch of Radeon RX 9070 XT and RX 9070 models.

AMD enthusiasts will breathe a collective sigh of relief, after reading McAfee's announcement: "the wait is almost over. Join us on February 28 at 8 AM EST for the reveal of the next-gen AMD Radeon RX 9000 Series. Get ready to make it yours when it hits shelves in early March. RSVP by subscribing to the AMD YouTube channel." Insider sources reckon that retailers will have stock on shelves by the rumored March 6 launch day. Mid-to-late January leaks suggested a fairly comprehensive distribution of board partner custom cards across European retail channels. Alleged specifications and performance results have leaked out over the past month and a half—will AMD (and AIBs) have any surprises lined up for the February 28 event?

According to ITHome, AMD is preparing to disrupt its competition with aggressive pricing for its upcoming RX 9000 series. The RX 9070 XT, built on the RDNA 4 architecture, is expected to launch at $599, positioning it directly against NVIDIA's RTX 5070 Ti, which carries a $749 price tag. With this competitive pricing, AMD aims to revitalize its market position following lower-than-expected sales of the RX 7000 series, causing it to lose some market share. The upcoming RX 9070 XT features the Navi 48 core running at 2.97 GHz, complemented by 16 GB of GDDR6 memory across a 256-bit bus. Architecture's enhanced AI upscaling capabilities, already demonstrated in the PlayStation 5 Pro, could offer compelling performance advantages over current-generation cards. The base RX 9070 model is anticipated to debut at $499, creating a focused attack on multiple market segments, including NVIDIA's RTX 5070, priced at $549.

AMD reportedly plans to accelerate the end-of-life timeline for its RX 7800 XT, currently priced at $479. Sources from IT Home suggest production ceased as early as January, months ahead of the planned initial third-quarter 2025 termination. This accelerated timeline suggests AMD's confidence in the RX 9000 series' ability to deliver superior price-to-performance metrics. The March 2025 launch window for the RX 9000 series arrives at a critical point in the GPU market, as NVIDIA rolls out its Blackwell-based RTX 50 series. AMD's aggressive pricing strategy and the architectural improvements in RDNA 4 positions the company to challenge NVIDIA's market dominance, at least in the $500-$600 price range. This competitive positioning could trigger NVIDIA price adjustments, potentially benefiting consumers who have faced consistently high GPU prices in recent years.

When the next-generation AMD Radeon RX 9000 series of GPUs, headed by RX 9070 XT and RX 9070, are surrounded by rumors, AMD's officials are coming to the rescue. According to the Bulgarian retailer's disclosure, AMD's initial pricing strategy for the new cards caused concerns, given their reported performance levels. The RX 9070 XT was reportedly positioned at around $899, matching the price point of the RX 7900 XT. The standard RX 9070 was said to carry a $749 price tag. To clarify the situation, AMD's Frank Azor jumped on social media platform X and explained, "While we aren't going to comment on all the price rumors, I can say that an $899 USD starting price point was never part of the plan."

Earlier reports indicate AMD has distributed its first wave of RDNA 4 graphics cards to various partners and retailers globally. However, these companies are currently unable to sell the new GPUs, as AMD has apparently set a March timeline for their release. This information gained additional credibility when a retailer in Bulgaria provided insights into AMD's preliminary launch strategy for the RX 9000 series. The retailer demonstrated the PowerColor Red Devil RX 9070 XT, one of three RX 9070 XT models that PowerColor unveiled during CES. While several AMD board partners have completed their RX 9070 XT designs, they have not yet disclosed official specifications or retail prices. Until March, we have limited information on pricing strategy.

When AMD announced its upcoming Radeon RX 9000 series of GPUs based on RDNA 4 IP, we expected the general availability to follow soon after the CES announcement. However, it turns out that AMD has scheduled its Radeon RX 9000 series availability for March, as the company is allegedly optimizing the software stack and its FidelityFX Super Resolution 4 (FSR 4) for a butter smooth user experience. In a response on X to Hardware Unboxed, AMD's David McAfee shared, "I really appreciate the excitement for RDNA 4. We are focused on ensuring we deliver a great set of products with Radeon 9000 series. We are taking a little extra time to optimize the software stack for maximum performance and enable more FSR 4 titles. We also have a wide range of partners launching Radeon 9000 series cards, and while some have started building initial inventory at retailers, you should expect many more partner cards available at launch."

AMD is taking its RDNA 4 launch more cautiously than before, as it now faces a significant problem with NVIDIA and its waste portfolio of software optimization and AI-enhanced visualization tools. The FSR 4 introduces a new machine learning (ML) based upscaling component to handle Super Resolution. This will be paired with Frame Generation and an updated Anti-Lag 2 to make up the FSR 4 feature set. Optimizing this is the number one priority, and AMD plans to get more games on FSR 4 so gamers experience out-of-the-box support.

AMD's upcoming Radeon RX 9070 series graphics cards have hit an unexpected roadblock, according to recent reports from PC Games Hardware. Despite physical units already reaching select retailers, the launch appears to be delayed due to ongoing pricing negotiations. Industry insider and forum moderator "pokerclock," known for accurate predictions about NVIDIA's GeForce RTX 50 series, reveals that AMD's initial pricing strategy has created tension with retail partners. While boxes bearing the RX 9070 branding have been spotted in retail channels, disagreements over costs have prevented an official release. The core issue stems from AMD's aggressive pricing approach for both the RX 9070 and RX 9070 XT models. Retailers have pushed back against what they consider excessive wholesale costs, forcing AMD to reconsider its strategy.

The company now faces the complex task of potentially reducing prices while compensating retailers who have already purchased inventory at higher rates. Sources suggest AMD may offer marketing funds or cashback incentives to bridge the price gap, though negotiations have reportedly stalled. For example, we recently reported on the AMD Radeon RX 9070 XT AIB model costing around $549. However, NVIDIA has announced its GeForce RTX 5070 at the same $549 price point, with potentially equal or higher raster, ray tracing, and AI capabilities across the board. For AMD to make the value case, the company would need to undercut NVIDIA's pricing. Until that is resolved, retailers aren't allowed to place RDNA 4 GPUs in general sale yet.