Introduction

AMD today launched the Radeon RX 9060 XT graphics card, the third entry from their latest Radeon RX 9000-series cards powered by the RDNA 4 graphics architecture. We have with us the ASUS Radeon RX 9060 XT 16 GB Prime OC, a value oriented custom design rendition of this contender for the mainstream. As with the RTX 5060-series, the RX 9060 XT is being launched in two memory variants, 8 GB and 16 GB, both of which are available from tomorrow. The 8 GB variant starts at a MSRP of $300, with the 16 GB variant at $350. The RX 9060 XT is being served up to the sizable majority of gamers who still play at 1080p, but also offers some degree of future-proofing for 1440p, taking advantage of performance enhancing features such as the new FSR 4. With this generation, AMD has placed a big focus on improving the ray tracing and AI acceleration performance of its GPUs, given that ray tracing is now pervasive among AAA game titles, and AI acceleration not just enables new performance enhancements such improved image quality, but also paves the way for edge AI use-cases.



The Radeon RX 9060 XT debuts the new 4 nm Navi 44 silicon. This is the second chip based on the RDNA 4 architecture, and is built on the TSMC N4P foundry node. It logically succeeds the Navi 33 chip powering the Radeon RX 7600 from the previous generation, although AMD has made a conscious effort to make its mid-range 60-segment SKUs not just capable of maxed out 1080p pure raster performance, but also 1080p with ray tracing. The change in nomenclature with the RX 9000 series helps consumers more accurately compare the various SKUs to their NVIDIA competitors. The RX 9060 XT 8 GB at $300 would compete with the GeForce RTX 5060, while the RX 9060 XT 16 GB at $350 undercuts the RTX 5060 Ti 8 GB at $380, offering some future-proofing from its memory size.

The Navi 44 on paper is exactly half the chip when compared against the Navi 48 powering the RX 9070 series. It comes with 32 RDNA 4 compute units, which work out to 2,048 stream processors, 64 AI accelerators, 32 RT accelerators, 128 TMUs, and 64 ROPs. Its memory interface is 128-bit wide, driving 8 GB or 16 GB of 20 Gbps GDDR6 memory, for 320 GB/s of memory bandwidth, which is backed by 32 MB of on-die Infinity Cache. AMD has fallen behind NVIDIA on memory technology, which implements 28 Gbps GDDR7 on even its $300 RTX 5060, but the company is unfazed, and claims that the various memory management improvements introduced with RDNA 4 should cover for the modest growth in memory bandwidth over the previous generation.

The RDNA 4 graphics architecture is purpose built for the most performance per mm² die-area. The company claims a significant increase in performance-per-CU over the previous RDNA 3 architecture. It also claims a 100% increase in ray tracing performance over RDNA 3, which should reduce the performance cost of ray tracing. There is a similar leap in AI acceleration throughput, now close to 800 AI TOPS, paving the way for FSR 4, the biggest upgrade to the FSR suite of performance enhancements. FSR 4 uses a new AI ML-based upscaler that offers superior image quality at every performance preset.

FSR 4 is perhaps the biggest upgrade to AMD's in-house performance enhancement suite since its conception. It introduces a new AI ML-based upscaler that takes advantage of the generationally increased AI acceleration performance of RDNA 4 GPUs. The AI ML-based upscaler is significantly more accurate than the shader-based model driving previous generations of FSR, improving image-quality at every performance preset of FSR. FSR 4 is as easy to implement as the original FSR, which should mean practically every FSR-enabled title should get the technology soon.

The ASUS Radeon RX 9060 XT Prime OC is designed by the company to be offered close to the AMD baseline price of $350 for the 16 GB RX 9060 XT. It comes with a fairly large dual-slot, triple-fan cooling solution. The card is longer than the PCB underneath, which is barely two-thirds its length. This allows for much of the airflow from the third fan to flow through the heatsink, and out the back from a cutout in the backplate. Visually, the card covers just the basics, with a neutral, no-frills design. The card offers factory-overclocked speeds of 2741 MHz Game Clock, compared to 2530 MHz reference. This card is targeted at those who want an RX 9060 XT to install and forget about. ASUS is pricing the card slightly above the AMD baseline price, at $360.

AMD RDNA 4 Architecture

AMD didn't provide a block diagram for Navi 44, so we puzzled one together

The Radeon RX 9060 XT is powered by the new Navi 44, AMD's second gaming GPU silicon based on the RDNA 4 graphics architecture. This is a monolithic chip built on the 4 nm TSMC N4P foundry node, and packs 29.7 billion transistors across a die-area of 199 mm². On paper, this is half the chip as the Navi 48 powering the Radeon RX 9070 series, but with one key change that sets it apart from its predecessors, Navi 33 and Navi 23—the chip comes with a full PCI-Express 5.0 x16 host interface, just like Navi 48, and not a truncated PCI-Express 5.0 x8. This is also unlike the NVIDIA GB206 chip powering the RTX 5060 Ti, which still uses x8, like it's smaller brother the RTX 5060. This full x16-wide PCIe Gen 5 interface may help in certain cases, such as trying to use this GPU on older systems with Gen 3 x16 host interfaces. Given that AMD is still launching processors with PCIe Gen 3, such as the Ryzen 7 5700GT "Cezanne," the company might be onto something.

Navi 44 comes with 32 compute units across 2 shader engines, which each have 16 RDNA 4 Dual Compute Units. Each CU contains 64 stream processors, and hence the Navi 44 has 2,048 of them; 64 AI accelerators, 32 RT accelerators, 128 TMUs, and 64 ROPs. The chip also has 32 MB of Infinity Cache memory, all of which is enabled on the RX 9060 XT. The memory interface is 128-bit GDDR6, which drives either 16 GB or 8 GB of 20 Gbps GDDR6 memory for 320 GB/s of memory bandwidth. This might be low compared to the RTX 5060 series with its 28 Gbps GDDR7 yielding 448 GB/s of bandwidth, but AMD claims that it has made several memory management advances with RDNA 4, which should generationally improve memory performance akin to a significant bandwidth increase. Both the 16 GB and 8 GB variants of the RX 9060 XT max out the Navi 44 silicon. The 16 GB variant comes with a total board power value of 160 W.


At the heart of the RDNA 4 graphics architecture is the new dual compute unit, with a vastly improved memory sub-system, improvements made to the scalar units, a new technology called dynamic register allocation, and improvements to CU efficiency and engine clocks. Each CU has two scheduler blocks, driving a 192 KB general purpose register (GPR), an 8 KB scalar GPR, 32 FMA ALUs, and 32 FMA+INT ALUs. There are also 8 transcendental logic units. RDNA 4 introduces the concept of dual SIMD32 vector units, for even more parallelism. The Scalar Unit comes with support for newer Float32 ops. Schedulers are updated with accelerated spill/fill operations. Instruction prefetching is improved. The new generation AI Accelerator comes with two 16-bit and four 8-bit/4-bit dense matrix compute rates, support for 4:2 structured sparsity for doubling throughput, and matrix loads with transpose. AMD has incorporated many technologies from its CDNA 3 Instinct AI ML accelerators on the AI Accelerators of RDNA 4, including enhanced and power-optimized WMMA, improvements to the ops per CU, support for FP8, E4M3 and E5M2 formats, and 4:2 structured sparsity.


The new generation AI Accelerator comes with two 16-bit and four 8-bit/4-bit dense matrix compute rates, support for 4:2 structured sparsity for doubling throughput, and matrix loads with transpose. AMD has incorporated many technologies from its CDNA 3 Radeon Instinct AI ML accelerators on the AI Accelerators of RDNA 4, including enhanced and power-optimized WMMA, improvements to the ops per CU, support for FP8, E4M3 and E5M2 formats, and 4:2 structured sparsity.


The new generation Ray Accelerator comes with double the box and triangle intersection resources as RDNA 3 RT accelerator, support for hardware instance transforms, improvements to the RT stack management, BVH8 node compression, and a revolutionary feature called oriented bounding boxes. To contain the number of rays really needed to be tested against an object, modern ray tracing technologies use something called a bounding box, which defines a region in which a geometry has to be tested against rays. Most of the time, the geometry is of a vastly different shape and smaller than the shape of a bounding box, which introduces false intersections, and wastes ray testing resources. AMD innovated a way to turn this bounding box into a 3D shape by giving it a Z-axis component, so the bounding box is oriented closer to the shape of the object to be tested, reducing the number of rays needed to be tested against it.


This graph highlights the contribution of various components toward the 100% generational ray traversal performance gain, allowing AMD to make do with a CU count of 64, with RDNA 3 being the baseline.


Both ray tracing and ML acceleration are memory sensitive applications, so AMD innovated a revolutionary change to its memory management system with the introduction of new out-of-order memory. All math is executed in waves on an RDNA GPU, and mutual dependencies between waves can cause memory request stream misses, as one wave's memory request queue waits for the other wave to complete its task. This is solved with a new out-of-order (relaxed ordering) memory management. This graph highlights the contribution of various components toward the 100% generational ray traversal performance gain, allowing AMD to make do with a CU count of 64, with RDNA 3 being the baseline.


On AMD, a fairly big chunk of the ray tracing stack continues to be executed on shaders, but the company has made advances to ensure the cost of ray tracing on the shader resources of the GPU is minimal, with the introduction of Dynamic Registers to improve parallelism.


The new Radiance 2 Display Engine comes with major hardware updates that reduce GPU idle power draw in multi-monitor setups. The engine also comes with hardware flip-metering support (something NVIDIA also introduced with Blackwell, and which enables Multi-Frame Gen on the RTX 50-series). Flip-metering improves video frame pacing to the GPU and reduces CPU overhead for video playback. There is also a display engine level hardware image sharpening component that drives Radeon Image Sharpening. As for I/O, you get contemporary DisplayPort 2.1a and HDMI 2.1b, the maximum bitrate is UHBR 13.5.


Navi 44 comes with an updated media engine which can perform concurrent encoding and decoding. The new generation media engine offers a 25% increase in H.264 low-bitrate encode image quality, and an 11% improvement in HEVC encode quality. AV1 encode and decode get B-frames support, vastly improving bitrates. The media engine posts a 50% generational performance uplift (measured in encoder/decoder frame rates), with reductions in memory overhead.

AMD FSR 4 and New AMD Software Features

With the Radeon RX 9000-series, AMD is debuting FSR 4, the latest version of its performance enhancement suite. FSR 4 introduces the biggest tech update since FSR 2, which also makes it exclusive to the RX 9000-series (and presumably next-gen cards yet to come), breaking the FSR tradition of being able to be used with almost any modern graphics card, regardless of vendor. With the RDNA 4 generation of GPUs offering significantly higher ML compute throughput, AMD designed a new ML-based upscaler to reconstruct details of the upscaled frame. In that sense, FSR 4 is technologically close to DLSS. RDNA 4 GPUs also leverage AI DNNs to de-noise ray tracing and path tracing, unburdening the shaders.

An AI ML based upscaler requires ground truth data, and just as NVIDIA goes through the painstaking task of training game-specific ML models that can later be shipped to end-users through game developers and as driver updates; AMD has set up a large FSR 4 data-center of its own, powered by Instinct MI300-series AI GPUs. On the client side, the ML-based upscaler needs the dedicated AI acceleration hardware from the RDNA 4 family. The frame generation technology appears to be carried over from FSR 3 Frame Generation, and is essentially a smart frame interpolation technology.


AMD in its presentation provided examples of image quality for FSR 4, claiming that in some cases, FSR 4 presents geometry more accurately than even native resolution. While AMD didn't detail it, FSR 4 comes with preparation for neural rendering. Given that Microsoft has now standardized neural shaders, and the ability for applications to directly address AI acceleration hardware on the GPU via the DirectX API, AMD could in the near future implement neural rendering, however, the company hasn't detailed a timeline. FSR 4 is ready for neural rendered objects.


Fluid Motion Frames gets an update with FMF 2.1, with improved image quality, reduced ghosting, and improved temporal tracking. The company also updated its API-agnostic image sharpening tool, with the introduction of Radeon Image Sharpening 2 (RIS 2).


AMD provides a slick driver frontend application called simply "AMD Software." AMD improved the way in which users can report bugs, artifacts, or corruption in games, through a voluntary (opt-in) feature called AMD Image Inspector.


As you submit a screenshot or video sequence to AMD, the company uses an AI model to identify rendering bugs, or corruption in the display. The entire driver-level game optimization, testing, and bug fixing process has been revamped with AI doing the heavy lifting.


The company also introduced AMD Chat, which is an AI chatbot that runs locally, and is accelerated by the GPU. You can make plain language queries on configuring AMD Software features and settings; and also chat with it on AMD technologies. This is essentially AMD's answer to NVIDIA's ChatRTX, but we don't know if the utility can be trained with custom datasets.


For streamers, AMD Software simplifies connecting their gameplay with their social media and streaming accounts. You can then take advantage of the new dual media engine of RDNA 4 that improves H.264 and HEVC image quality, and use other utilities such as AMD Noise Suppression, which is a 2-way AI-based audio filter.

AMD AI Apps Manager consolidates all your AI applications into a simple launcher-like interface resembling a game launcher.

AMD FSR "Project Redstone"

In the second half of 2025, AMD is planning to introduce a major update to FSR 4, which is designed to bring it technologically up to par with NVIDIA. AMD has given this update the working title "Project Redstone." The FSR 4 implementation that launched with the Radeon RX 9070-series comes with an AI ML-based upscaler, which only covers the super resolution-based performance enhancement of the FSR suite. As of June 5, 2025, AMD claims that over 60 titles will have implemented FSR 4 super resolution. "Project Redstone" is a significant step up from this.


FSR "Project Redstone" introduces four new components besides the AI ML-based upscaler for super resolution AMD introduced earlier this year. The first component is an AI-ML model for Neural Radiance Caching, a method with which the GPU continuously learns how light interacts with ray traced objects in a scene, so it could store indirect lighting data to conserve performance cost of ray tracing. The second key feature is an AI ML-based model for ray regeneration. This should enhance image quality for high sparsity (noisy) low-sampled renders. It predicts and filters grainy noise in real time, and contributes to reduction of the cost of ray tracing.

The third of course is AI ML-based super resolution, which has been rolled out since the RX 9070 series, and significantly uplifts image quality at all performance presets of FSR compared to the previous FSR 2.2. Lastly, AI ML-based Frame Generation. This is a classic single frame generation akin to FSR 3, but with a new ML model that more accurately determines what an intermediate frame to two conventionally-rendered ones could look like, for more visual accuracy and reduced ghosting associated with frame generation tech.

AMD's main aim with FSR 4 and "Redstone" isn't just to improve performance across the board, but make Radeon RX 9000 series GPUs capable of games with path tracing within their recommended resolution and use-cases.

Packaging

The Card

I really like the ASUS Prime aesthetic for the RX 9000-series. It's a clean design with highlights that results in a clean look that doesn't appear too plasticky. On the back you get a high quality metal backplate with a large cutout for air to flow through, the front cooler shroud is made from plastic.


Dimensions of the card are 30.5 x 12.5 cm, and it weighs 866 g.


Installation requires three slots in your system. We measured the card's width to be 50 mm.


Display connectivity includes two standard DisplayPort 2.1b and a HDMI 2.1b.

AMD put in some extra work for creative professionals, particularly those into video editing, by giving Navi 44 its latest video engine. The video engine supports concurrent hardware-accelerated encode and decode for AV1, HEVC, and H.264 video codecs. With RDNA 4, AMD has improved the image quality of H.264 encoding at lower bitrates, which should make AMD appeal to content creators. The AV1 encoder and decoder support B-frames, a technology with which there can be a significant reduction in bitrates without a loss in image quality. While the RX 9070 Series has a dual encoder setup, the GPU of the RX 9060 XT comes with just a single encoder.


The card uses a single 8-pin connector, which allows a maximum power draw of 150 W, plus 75 W from the slot, but the actual board power limit is set much lower of course.


There is no lighting on the ASUS RX 9060 XT Prime OC.


This BIOS switch lets you toggle between the default Performance BIOS and an optional "Quiet" BIOS.

Teardown

The fan assembly can be removed easily, which makes it easy to replace a broken fan in a couple of years—no need to touch the thermal paste of the card.


ASUS has installed two heatpipes on their thermal solution, the main heatsink also provides cooling for the memory chips and VRM circuitry.


The backplate protects the card against damage during installation and handling. Note the thermal pads, which provide cooling for the memory chips on the back side of the PCB.