Home / AMD / AMD Radeon RX 6900 XTX: Performance and Specs

AMD Radeon RX 6900 XTX

AMD Radeon RX 6900 XTX: Flagship for Gamers and Professionals

April 2025

Architecture and Key Features: RDNA 3 at Its Best

The AMD Radeon RX 6900 XTX graphics card is built on the RDNA 3 architecture, but with substantial improvements that elevate it beyond the original design from 2022. The chips are manufactured using 5nm TSMC technology, ensuring increased transistor density and energy efficiency. Key features include:

- Ray Accelerators 2.0: Enhanced ray tracing units that are 30% faster than those in the RX 6900 XT.

- FidelityFX Super Resolution 3.0: Upscaling technology with AI algorithms support, increasing FPS in games by up to 2x while maintaining detail.

- Hybrid Compute Units: A combination of graphics and compute cores for better resource optimization.

The card also supports AV1 decoding and HDMI 2.1a, making it ideal for 8K content and high refresh rate gaming.

Memory: GDDR6X with High Bandwidth

The RX 6900 XTX is equipped with 20GB of GDDR6X memory on a 320-bit bus and a bandwidth of 672GB/s (compared to 512GB/s in the RX 6900 XT). This allows for:

- Easy handling of textures in 4K and 8K.

- Reduced latency in open-world games (for example, Cyberpunk 2077: Phantom Liberty or Starfield: Shattered Space).

- Improved performance in professional applications that require large amounts of VRAM (such as Blender or DaVinci Resolve).

Gaming Performance: 4K without Compromises

In 2025 benchmarks, the RX 6900 XTX demonstrates the following results (Ultra settings, no FSR):

- Cyberpunk 2077: 78 FPS at 1440p, 54 FPS at 4K (with ray tracing — 48 FPS at 1440p, 32 FPS at 4K).

- Call of Duty: Black Ops 6: 144 FPS at 1440p, 98 FPS at 4K.

- Horizon Forbidden West PC Edition: 120 FPS at 1440p, 82 FPS at 4K.

With FSR 3.0 Quality Mode activated, performance increases by 40-60%. For example, in Alan Wake 3, at 4K, the average FPS rises from 45 to 68. However, ray tracing remains a weak point: NVIDIA's RTX 4080 Ti outperforms the RX 6900 XTX by 20-25% in comparable scenarios.

Professional Tasks: Power for Creativity

The card excels in:

- 3D Rendering: In Blender (Cycles engine), it is 15% faster than the RX 6900 XT thanks to optimized drivers and support for OpenCL 3.0.

- Video Editing: In DaVinci Resolve, rendering an 8K project takes 25% less time than with the RTX 4080.

- Scientific Calculations: Support for ROCm 6.0 allows the GPU to be used for machine learning and simulations, although NVIDIA CUDA still leads in this niche.

Power Consumption and Cooling: Requires Attention

- TDP: 330W (10% higher than RX 6900 XT).

- Recommendations:

- Power supply of at least 850W (preferably with an 80+ Gold certification).

- Case with good ventilation (minimum 3 fans) or liquid cooling for overclocking.

- Operating temperature: 70-75°C under load with stock cooling, 60-65°C with liquid cooling.

The card is noisy under load — sound levels reach 42 dB, which is higher than competing NVIDIA models.

Comparison with Competitors: Battle of the Flagships

- NVIDIA RTX 4080 Ti: Better in ray tracing (+25%) and DLSS 4.0, but more expensive ($1199 vs. $999 for AMD).

- AMD Radeon RX 7900 XTX: The younger sibling of the RX 6900 XTX lags behind in 4K by 15-20% but consumes less power (300W).

- Intel Arc Battlemage XT: A new model from Intel ($899) is strong in DX12 games but weak in professional tasks.

Practical Tips: How to Avoid Problems

1. Power Supply: Choose models with separate 12VHPWR cables (no adapters!).

2. Platform: Best compatibility with motherboards using AMD X670/B650 chipsets.

3. Drivers: Update Adrenalin Edition monthly — AMD actively optimizes support for new games.

4. Overclocking: Use the AMD Ryzen Master utility for fine-tuning — gains of up to 8% without risk.

Pros and Cons

Pros:

- Best price/performance ratio at 4K.

- Support for FSR 3.0 and AV1.

- 20GB of memory for future projects.

Cons:

- High power consumption.

- Weaker ray tracing performance compared to NVIDIA.

- Noisy cooling system.

Final Verdict: Who Is This Card For?

The RX 6900 XTX is ideal for:

- Gamers looking to play in 4K without compromises and reluctant to pay a premium for the RTX 4080 Ti.

- Video editors working with 8K materials.

- AMD enthusiasts who value open-source software (ROCm, OpenCL) and frequent driver updates.

Priced at $999, it makes for a compelling alternative to NVIDIA, especially if ray tracing is not a top priority.

Basic

Label Name

AMD

Platform

Desktop

Model Name

Radeon RX 6900 XTX

Generation

Navi II

Base Clock

2075MHz

Boost Clock

2435MHz

Bus Interface

PCIe 4.0 x16

Transistors

26,800 million

RT Cores

Compute Units

TMUs

Texture Mapping Units (TMUs) serve as components of the GPU, which are capable of rotating, scaling, and distorting binary images, and then placing them as textures onto any plane of a given 3D model. This process is called texture mapping.

320

Foundry

TSMC

Process Size

7 nm

Architecture

RDNA 2.0

Memory Specifications

Memory Size

16GB

Memory Type

GDDR6

Memory Bus

The memory bus width refers to the number of bits of data that the video memory can transfer within a single clock cycle. The larger the bus width, the greater the amount of data that can be transmitted instantaneously, making it one of the crucial parameters of video memory. The memory bandwidth is calculated as: Memory Bandwidth = Memory Frequency x Memory Bus Width / 8. Therefore, when the memory frequencies are similar, the memory bus width will determine the size of the memory bandwidth.

256bit

Memory Clock

2250MHz

Bandwidth

Memory bandwidth refers to the data transfer rate between the graphics chip and the video memory. It is measured in bytes per second, and the formula to calculate it is: memory bandwidth = working frequency × memory bus width / 8 bits.

576.0 GB/s

Theoretical Performance

Pixel Rate

Pixel fill rate refers to the number of pixels a graphics processing unit (GPU) can render per second, measured in MPixels/s (million pixels per second) or GPixels/s (billion pixels per second). It is the most commonly used metric to evaluate the pixel processing performance of a graphics card.

311.7 GPixel/s

Texture Rate

Texture fill rate refers to the number of texture map elements (texels) that a GPU can map to pixels in a single second.

779.2 GTexel/s

FP16 (half)

An important metric for measuring GPU performance is floating-point computing capability. Half-precision floating-point numbers (16-bit) are used for applications like machine learning, where lower precision is acceptable. Single-precision floating-point numbers (32-bit) are used for common multimedia and graphics processing tasks, while double-precision floating-point numbers (64-bit) are required for scientific computing that demands a wide numeric range and high accuracy.

49.87 TFLOPS

FP64 (double)

An important metric for measuring GPU performance is floating-point computing capability. Double-precision floating-point numbers (64-bit) are required for scientific computing that demands a wide numeric range and high accuracy, while single-precision floating-point numbers (32-bit) are used for common multimedia and graphics processing tasks. Half-precision floating-point numbers (16-bit) are used for applications like machine learning, where lower precision is acceptable.

1.558 TFLOPS

FP32 (float)

An important metric for measuring GPU performance is floating-point computing capability. Single-precision floating-point numbers (32-bit) are used for common multimedia and graphics processing tasks, while double-precision floating-point numbers (64-bit) are required for scientific computing that demands a wide numeric range and high accuracy. Half-precision floating-point numbers (16-bit) are used for applications like machine learning, where lower precision is acceptable.

24.431 TFLOPS

Miscellaneous

Shading Units

The most fundamental processing unit is the Streaming Processor (SP), where specific instructions and tasks are executed. GPUs perform parallel computing, which means multiple SPs work simultaneously to process tasks.

5120

L1 Cache

128 KB per Array

L2 Cache

4MB

TDP

330W

Vulkan Version

Vulkan is a cross-platform graphics and compute API by Khronos Group, offering high performance and low CPU overhead. It lets developers control the GPU directly, reduces rendering overhead, and supports multi-threading and multi-core processors.

1.3

OpenCL Version

2.1

OpenGL

4.6

DirectX

12 Ultimate (12_2)

Power Connectors

2x 8-pin

Shader Model

6.5

ROPs

The Raster Operations Pipeline (ROPs) is primarily responsible for handling lighting and reflection calculations in games, as well as managing effects like anti-aliasing (AA), high resolution, smoke, and fire. The more demanding the anti-aliasing and lighting effects in a game, the higher the performance requirements for the ROPs; otherwise, it may result in a sharp drop in frame rate.

128

Suggested PSU

700W

Benchmarks

FP32 (float)

Score

24.431 TFLOPS

Compared to Other GPU

FP32 (float) / TFLOPS

A10G

32.15 +31.6%

GeForce RTX 4070 GDDR6

29.733 +21.7%

Radeon RX 6900 XTX

24.431

Instinct MI210

23.083 -5.5%

GeForce RTX 4060 Ti 8 GB

22.501 -7.9%