Home / AMD / AMD Radeon 820M: Performance and Specs

AMD Radeon 820M

AMD Radeon 820M
AMD Radeon 820M graphics card review

AMD Radeon 820M: Ryzen AI is present, gaming graphics are nearly absent

June 2026

AMD Radeon 820M is the most basic integrated graphics found in laptops with the Ryzen AI 5 330. Feature-wise, it is modern: it has AV1 support, external display support, and a platform with an NPU for Copilot+ PC. However, the graphical unit is significantly cut down - it has only 2 cores. Therefore, the Radeon 820M should be viewed not as an entry-level gaming iGPU, but as basic graphics for an inexpensive Ryzen AI laptop.

What is Radeon 820M

Radeon 820M is the basic integrated graphics in Ryzen AI 5 330. It belongs to the Radeon 800M family but occupies the junior position. For comparison: Radeon 840M has 4 graphical cores, Radeon 860M has 8, while Radeon 880M and 890M are found in more expensive and faster laptops.

In terms of capabilities, the Radeon 820M is not outdated. It includes DisplayPort 2.1, HDMI 2.1, FreeSync, AV1, and supports multiple displays. The limitation lies not in features, but in the number of compute units. Under 3D loads, the 2 graphical cores quickly become its main weak point.

The primary task of the Radeon 820M is to handle basic graphical tasks of a laptop without a discrete graphics card. This is sufficient for Windows, browsing, video, office applications, video calls, and older games. It is inadequate for modern gaming.

Why Radeon 820M is easy to overestimate

Radeon 820M is easy to overestimate due to the processor it is paired with. The Ryzen AI 5 330 features an NPU with up to 50 TOPS and meets the Copilot+ PC requirements. In specifications, such a laptop looks convincing: Ryzen AI, local AI features, fast memory, modern media block.

But the graphics remain entry-level. The Ryzen AI in the name does not mean that the laptop has a powerful iGPU. In this configuration, roles are strictly divided: the NPU handles AI functions, the CPU handles ordinary performance, and the Radeon 820M covers basic graphics and multimedia.

Performance: lower than Radeon 840M and 860M

According to tests, the Radeon 820M remains one of the weakest iGPUs in the Radeon 800M lineup. In 3DMark Time Spy Graphics, the result is around 786 points. This is acceptable for an office laptop, but insufficient for gaming.

The table shows the limits of its capabilities: light and older games still run reasonably well, while modern titles quickly hit the limitations of 2 graphical cores.

Game and Settings Approximate Result Radeon 820M
Dota 2 Reborn, 1080p Ultra around 40 FPS
GTA V, 1080p Low around 130 FPS
GTA V, 1080p High around 22 FPS
Cyberpunk 2077, 1080p Low around 13 FPS
Baldur’s Gate 3, 1080p Low around 13 FPS
Battlefield 6, 1080p Low around 11 FPS
Strange Brigade, 1080p Low around 76 FPS

Dota 2, older games, and light online projects are still feasible. Cyberpunk 2077, Baldur’s Gate 3, or Battlefield 6 are no longer suitable for comfortable gaming. Even on low settings and in Full HD, the Radeon 820M quickly reaches the limits of its basic graphical capabilities.

What Radeon 820M is suitable for

The Radeon 820M meets the needs for which it is used in inexpensive Ryzen AI laptops. It is a basic iGPU for everyday work rather than graphics with gaming potential.

It is suitable for such tasks:

  • office programs, documents, spreadsheets, and browsing;
  • video, streaming, and video conferencing;
  • connecting external monitors;
  • simple photo editing;
  • older games and light online projects;
  • laptops with Copilot+ PC features thanks to the NPU in Ryzen AI 5 330.

For such laptops, this is a rational configuration. The Radeon 820M does not increase the device's price, does not require complex cooling, and allows for a thin chassis. If the laptop is purchased for work, study, and everyday tasks, this level of graphics is adequate.

Main limitations of Radeon 820M

The main limitation of the Radeon 820M is its two graphical cores. Fast LPDDR5X memory helps the integrated graphics, but it does not change the device class. This iGPU has too few compute units for modern games and heavy 3D tasks.

Do not choose the Radeon 820M if you need:

  • modern gaming in Full HD;
  • stable FPS in AAA titles;
  • 3D rendering;
  • editing with active GPU acceleration;
  • work in Blender on the GPU;
  • a reserve of graphical power for the future.

Do not expect the Radeon 820M to perform at the level of the Radeon 680M or Radeon 780M. Those iGPUs had significantly more compute units and thus performed better in gaming. The Radeon 820M belongs to a different class: it is graphics for an affordable thin laptop.

Another important point is the dependence on the specific laptop model. Integrated graphics use system memory, and the result depends on its speed, operating mode, power limits, and cooling. But even in a favorable configuration, the Radeon 820M remains the lower-tier iGPU.

How to compare Radeon 820M

The most accurate comparison for the Radeon 820M is within AMD’s range:

Graphics Positioning
Radeon 820M basic graphics for office use, video, and older games
Radeon 840M superior due to 4 graphical cores
Radeon 860M reasonable minimum for light gaming
Radeon 880M / 890M significantly higher class of iGPU

With Intel, the situation depends on the specific processor and number of graphical units. Higher-end Intel Arc Graphics in Core Ultra may significantly outperform, but lower configurations are often limited as well. Therefore, it is necessary to compare not just series names, but specific laptops and tests.

A practical guideline is this: if gaming is not important, the Radeon 820M is sufficient. If you occasionally wish to run modern projects, it’s better to consider at least the Radeon 860M. If stable gaming performance is needed, a discrete graphics card is required.

Conclusion

AMD Radeon 820M is functionally modern but very limited in 3D performance integrated graphics. Its main value is not FPS, but its role in an affordable platform with Ryzen AI featuring a full NPU.

Purchasing a laptop with the Radeon 820M for gaming is not worthwhile. It is a lower configuration with 2 graphical cores: it provides modern media capabilities, but not a reserve for 3D loads.

If an inexpensive Copilot+ PC for work and study is needed, the Radeon 820M makes sense. If gaming, 3D work, or heavy GPU acceleration is important, it’s better to look at models with Radeon 860M, 880M, 890M, or a laptop with a discrete graphics card.

Basic

Label Name
Intel
Platform
Integrated
Launch Date
July 2025
Former Codename
Krackan Point
GPU Lithography
4 nm
Model Name
AMD Radeon 820M
Generation
Radeon 800M Series
Boost Clock
2800 MHz
Bus Interface
Integrated
RT Cores
2
Compute Units
2
Tensor Cores
?
Tensor Cores are specialized processing units designed specifically for deep learning, providing higher training and inference performance compared to FP32 training. They enable rapid computations in areas such as computer vision, natural language processing, speech recognition, text-to-speech conversion, and personalized recommendations. The two most notable applications of Tensor Cores are DLSS (Deep Learning Super Sampling) and AI Denoiser for noise reduction.
No
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.
8
Foundry
TSMC
Process Size
4 nm
Architecture
RDNA 3.5

Memory Specifications

Memory Size
Shared system memory
Memory Type
DDR5 / LPDDR5X shared system memory
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.
Dual-channel system memory, platform dependent
Memory Clock
DDR5-5600 / LPDDR5X-8000, platform dependent
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 memory dependent

Display and Media

AMD FreeSync
Yes
AV1 Encode/Decode
Encode/Decode
DisplayPort Extensions
Adaptive-Sync, HBR3, UHBR10
H.264 Hardware Encode/Decode
Encode/Decode
H.265 HEVC Hardware Encode/Decode
Encode/Decode
H.266 VVC Hardware Encode/Decode
No hardware support
HDCP Version
2.3
HDMI Version
2.1
Intel Quick Sync Video
No
Max Resolution DP
7680x4320 @ 60Hz
Max Resolution HDMI
7680x4320 @ 60Hz
Max Video Decode Bandwidth
1080p60 8bpc MPEG2, 1080p60 8bpc VC1, 1080p786 8/10bpc VP9, 2160p196 8/10bpc VP9, 4320p49 8/10bpc VP9, 1080p1200 8bpc H.264, 2160p300 8bpc H.264, 4320p75 8bpc H.264, 1080p786 8/10bpc H.265, 2160p196 8/10bpc H.265, 4320p49 8/10bpc H.265, 1080p960 8/10bpc AV1, 2160p240 8/10bpc AV1, 4320p60 8/10bpc AV1
Max Video Encode Bandwidth
1080p630 8bpc H.264, 1440p373 8bpc H.264, 2160p175 8bpc H.264, 1080p630 8bpc H.265, 1440p373 8bpc H.265, 2160p175 8bpc H.265, 4320p43 8bpc H.265, 1080p864 8/10bpc AV1, 1440p513 8/10bpc AV1, 2160p240 8/10bpc AV1, 4320p60 8/10bpc AV1
Number of Displays Supported
4
Outputs
HDMI 2.1, DisplayPort 2.1, USB-C DisplayPort Alt Mode; device dependent
USB Type-C DisplayPort Alternate Mode
Yes
Wireless Display
Miracast

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.
11.2 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.
22.4 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.
1.43 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.
44.8 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.
0.72 TFLOPS

AI Features

Intel Deep Learning Boost on GPU
No
NPU TOPS
Up to 50 TOPS
Processor Overall TOPS
Up to 56 TOPS

Miscellaneous

Native PCIe Lanes
14 total / 14 usable
PCI Express Version
PCIe 4.0
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.
128
TDP
Shared with processor; 15-28 W cTDP
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.
1.4
OpenCL Version
2.1
OpenGL
4.6
CUDA
No
DirectX
12 Ultimate (12_2)
Power Connectors
None
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.
4
Shader Model
6.7

Benchmarks

FP32 (float)
Score
0.72 TFLOPS
3DMark Time Spy
Score
786
Vulkan
Score
10321
OpenCL
Score
6593

Compared to Other GPU

FP32 (float) / TFLOPS
T400 4 GB
1.072 +48.9%
GeForce GTX 760A
1.029 +42.9%
Tesla C2050
1.007 +39.9%
Quadro M1000M
0.997 +38.5%
Radeon 820M
0.72
3DMark Time Spy
RTX A1000 Mobile 6 GB
4844 +516.3%
GeForce GTX 1060 Mobile
3723 +373.7%
Radeon 680M
2399 +205.2%
GeForce GTX 760
1705 +116.9%
Radeon 820M
786
Vulkan
GeForce RTX 3060
84816 +721.8%
Radeon RX 5600 XT
60350 +484.7%
Arc 140T
37148 +259.9%
Radeon R9 M380
18210 +76.4%
Radeon 820M
10321
OpenCL
CMP 30HX
57474 +771.7%
Radeon Pro W6400
35443 +437.6%
P106 090
20338 +208.5%
Quadro P600
11181 +69.6%
Radeon 820M
6593