Home / AMD / AMD Radeon 660M: Performance and Specs

AMD Radeon 660M

AMD Radeon 660M: Compact Graphics for Everyday Tasks and Light Gaming

Analysis of Mobile GPU Capabilities in 2025

Introduction

The AMD Radeon 660M is an integrated graphics solution built into the Ryzen 7000/8000 series processors for laptops. Targeted at budget and mid-range devices, this GPU promises a balance between performance and energy efficiency. Let’s take a closer look at what it can do in gaming, work, and everyday tasks.

1. Architecture and Key Features

RDNA 3 Architecture: The Radeon 660M is based on the updated RDNA 3 microarchitecture, which brings energy consumption optimization and improved support for modern APIs (DirectX 12 Ultimate, Vulkan 1.3).

- Process Technology: 6nm TSMC - a choice that allows for reduced heat generation with minimal performance loss.

- Unique Features:

- FidelityFX Super Resolution 3.0: An upscaling technology that boosts FPS in games by 30-50% with minimal quality loss.

- Radeon Ray Tracing: Basic support for ray tracing, but with a caveat of low performance in AAA projects.

- Smart Access Memory: Optimization for CPU access to GPU memory in systems with Ryzen processors.

2. Memory: Type, Capacity, and Bandwidth

The Radeon 660M uses DDR5/LPDDR5 system memory, typical for integrated solutions.

- Dedicate Memory: Up to 4GB (dynamically adjusted based on workload).

- Bandwidth: Depends on the laptop's RAM speed. For example, with DDR5-5600, it can reach up to 89.6 GB/s.

- Impact on Gaming: In projects with high video memory requirements (e.g., Horizon Forbidden West), there may be stuttering due to limited bandwidth.

3. Gaming Performance

The Radeon 660M is positioned as a solution for 1080p gaming at low-medium settings.

- Cyber Games:

- Cyberpunk 2077 (without ray tracing): 28-35 FPS (Low), 22-27 FPS (Medium).

- Apex Legends: 60-70 FPS (Medium).

- Fortnite (FSR 3.0, Performance Mode): 90-100 FPS.

- Ray Tracing: Enabling RT reduces FPS by 40-60%. For example, Minecraft RTX achieves only 15-20 frames.

- 1440p and 4K: In less demanding games (CS:GO, Dota 2), 1440p is achievable at 60 FPS, but 4K is only feasible for indie projects.

4. Professional Tasks

The Radeon 660M is suitable for basic workloads but cannot replace discrete GPUs.

- Video Editing: In DaVinci Resolve, rendering a 1080p clip takes 20% longer than on an NVIDIA RTX 3050. Hardware acceleration via OpenCL speeds up export, but 4K effects will be processed slowly.

- 3D Modeling: In Blender, a scene with 500k polygons renders in 12-15 minutes (Cycles, GPU mode). For more complex tasks, cloud solutions are advisable.

- Scientific Calculations: OpenCL support allows for lightweight simulations, but more powerful graphics cards are needed for ML/AI tasks.

5. Power Consumption and Heat Generation

The integrated nature of the GPU minimizes power consumption:

- TDP: 15-28 Watts (within the overall TDP of the processor).

- Cooling: A passive heatsink or compact cooler is sufficient. In ultraportables, throttling may occur under prolonged loads.

- Case Recommendations: Laptops with vents on the bottom and aluminum chassis (e.g., Lenovo Yoga Slim 7) perform better than plastic counterparts.

6. Comparison with Competitors

AMD Radeon 660M vs NVIDIA GeForce MX570 vs Intel Arc A350M:

- Performance: The Radeon 660M is 10-15% faster than the MX570 in games thanks to FSR 3.0 but lags behind the Arc A350M in tasks supporting AV1 encoding.

- Price: Laptops with Radeon 660M range from $650 to $850, while those with MX570 start at $700 and with Arc A350M at $750.

- Energy Efficiency: AMD wins - battery life is 7-8 hours compared to 5-6 hours for the competitors.

7. Practical Tips

- Power Supply: A standard 65W adapter is sufficient for the laptop.

- Compatibility: Ensure the system uses DDR5 memory - this is critical for iGPU performance.

- Drivers: Regularly update the Adrenalin Edition - AMD actively optimizes support for new games.

- BIOS Settings: Allocate 3-4 GB of memory for the GPU if gaming is planned.

8. Pros and Cons

Pros:

- Energy efficiency.

- Support for FSR 3.0 and modern APIs.

- Affordable laptop prices.

Cons:

- Limited gaming performance.

- Dependency on RAM speed.

- Weak ray tracing capabilities.

Final Conclusion: Who is the Radeon 660M Suitable For?

This GPU is an ideal choice for:

- Students: Light video editing and work in CAD programs.

- Office Users: Multitasking and video conferencing.

- Casual Gamers: 1080p gaming at medium settings.

If you are looking for a laptop under $800 for work and light gaming, the Radeon 660M is a sensible compromise. However, for professional tasks or AAA games in 2025, consider models with discrete GPUs at the RTX 4050 level or higher.

Updated in April 2025. Prices are accurate as of the time of publication.

Basic

Label Name

AMD

Platform

Integrated

Launch Date

January 2022

Model Name

Radeon 660M

Generation

Rembrandt

Base Clock

1500MHz

Boost Clock

1900MHz

Bus Interface

PCIe 4.0 x8

Transistors

13,100 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.

Foundry

TSMC

Process Size

6 nm

Architecture

RDNA 2.0

Memory Specifications

Memory Size

System Shared

Memory Type

System Shared

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.

System Shared

Memory Clock

SystemShared

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.

System Dependent

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.

30.40 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.

45.60 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.

2.918 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.

91.20 GFLOPS

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.

1.43 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.

384

L1 Cache

128 KB per Array

L2 Cache

2MB

TDP

15W

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.

1.2

OpenCL Version

2.0

OpenGL

4.6

DirectX

12 Ultimate (12_2)

Power Connectors

None

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.