Intel Arc B390
vs
AMD Radeon 890M

vs
Intel Arc B390 vs AMD Radeon 890M graphics card comparison

GPU Comparison Result

Intel Arc B390 vs AMD Radeon 890M: How Much Faster is Intel's New GPU?

The AMD Radeon 890M has long remained one of the most powerful integrated GPUs for Windows laptops. It allowed modern games to be played at 1080p without a discrete graphics card and became a significant advantage for Ryzen AI 300 processors. The Intel Arc B390 offers a higher level of performance: the Xe3 architecture, 12 Xe-cores, and XMX matrix engines provide a noticeable edge in gaming and video processing tasks.

However, comparing two iGPUs solely based on their names or maximum frequencies is not feasible. They utilize system RAM and share the overall power limit with the CPU cores. Thus, the outcome depends not only on the GPU itself but also on the LPDDR5X frequency, cooling, and specific laptop settings.

Key Differences

Feature Intel Arc B390 AMD Radeon 890M
Architecture Xe3 RDNA 3.5
Compute Configuration 12 Xe-cores 16 Compute Units
Max Frequency Up to 2.5 GHz Up to 2.9 GHz
Matrix Accelerators XMX No direct analogue
Scaling Technology XeSS with XMX acceleration AMD FSR
Ray Tracing Hardware Hardware
Video Processing AV1 encode/decode, VVC decode, Quick Sync AV1 encode/decode
Main Advantage Higher gaming performance Availability and variety in laptops

The number of compute units and frequency cannot be directly compared: Intel's Xe-core and AMD's Compute Unit perform tasks differently. The Radeon 890M formally has more units and a higher frequency, but the Arc B390 uses a newer organization of its graphics component and specialized accelerators for neural operations.

Why Arc B390 Pulls Ahead

The Arc B390's strength lies in its combination of computational resources, XMX blocks, and high memory bandwidth. For integrated graphics, the latter is particularly important: it has no dedicated video memory, so textures and other data are stored in shared RAM.

In laptops with fast LPDDR5X memory, Intel graphics are less likely to hit bandwidth limits at a resolution of 1920 × 1080. This helps maintain a more stable frame rate during demanding scenes and reduces texture quality drops.

The XMX blocks accelerate the performance of XeSS and other matrix algorithms. The Radeon 890M uses FSR, which does not require specialized hardware and works on a wide range of graphics cards. AMD's approach is more universal, but in compatible games, the Arc B390 can perform neural scaling more effectively.

Gaming Performance

Independent tests show the Arc B390's advantage in most modern games, though its magnitude depends on the specific platform. In some projects, the Radeon 890M lags moderately, while in others, the difference becomes more noticeable, especially under heavy graphical loads.

Comparing results from different reviews should be done cautiously. Laptops may differ in:

  • Sustained power limit;
  • RAM speed;
  • Cooling system;
  • Scaling settings;
  • BIOS and graphics driver versions.

With sufficient power consumption, the Arc B390 handles 1080p gaming more confidently. It more often allows for medium settings instead of low and uses a less aggressive scaling mode. In the most demanding projects, lowering some parameters is still necessary, but Intel's headroom is greater.

The Radeon 890M holds its ground well in esports titles. For Counter-Strike 2, Valorant, Dota 2, and other comparatively less demanding projects, its performance is adequate. The B390's advantage is felt more strongly in modern AAA games, where the Radeon is already operating near its limits.

Neither of these iGPUs replaces a powerful discrete graphics card. Their function is to provide acceptable gaming speeds in a thin laptop without a dedicated GPU, rather than to run new games at maximum settings.

XeSS and FSR

XeSS restores images from a lower internal resolution and reduces the load on the GPU. On the Arc B390, the technology utilizes XMX blocks, giving Intel additional resources for neural processing.

FSR is not tied to a specific architecture and works on a wider range of GPUs. This is its main advantage. The image quality and performance gains of both technologies depend on the version of the algorithm and its implementation in specific games.

Frame generation can also significantly increase displayed FPS but does not fix excessively low base performance. If a game runs slowly from the outset, input lag will remain noticeable even after adding intermediate frames.

Ray Tracing

Both graphics solutions support hardware ray tracing, but for integrated GPUs, it remains a heavy load. Even the Arc B390 will require scaling and quality reductions in demanding games.

Intel's advantage here is determined primarily by its higher overall performance. The B390 should not be seen as a solution for maximum ray tracing, but moderate RT effects are available more often than with the Radeon 890M.

Video Processing

The Arc B390 supports hardware AV1 encoding and decoding, VVC decoding, and Intel Quick Sync. This suite is useful for video editing, converting clips, screen recording, and streaming.

The Radeon 890M also handles AV1 and is suitable for most multimedia tasks. However, Quick Sync has long been supported by video editing and encoding software, giving Intel an additional advantage in compatible applications.

The outcome depends on the specific software. Some editors make better use of Intel's media engine, while others work more efficiently with AMD's compute units. For professional use, it’s worth looking at benchmarks of the needed application rather than just gaming results.

What to Check Before Buying

The name of the integrated GPU does not guarantee identical performance across all laptops. A compact model with a low power limit may lag considerably behind a larger device with the same processor.

Before making a purchase, three parameters are particularly important:

  • Speed and capacity of RAM;
  • Performance after sustained load;
  • Power limits in gaming mode.

It is also advisable to consider noise, chassis temperature, battery life, and screen quality. A slight gain in FPS does not always justify a higher price or a less convenient laptop.

Conclusion

The Intel Arc B390 sets a new benchmark among high-performance integrated GPUs. It outperforms the Radeon 890M in most gaming scenarios, is equipped with XMX blocks, and offers more advanced video processing capabilities. For gaming without a discrete graphics card, this is the preferred choice under comparable conditions.

At the same time, the Radeon 890M does not become a weak option. It remains suitable for esports projects, older AAA games, and new releases at moderate settings. Its main advantage continues to be the price of the ready-made laptop.

If both models are close in cost, memory, cooling, and screen quality, the Arc B390 appears more compelling. The Radeon 890M makes sense to choose when an AMD laptop is considerably cheaper, has longer battery life, or excels in other important characteristics.

Advantages

  • More Shading Units: 1536 (1536 vs 1024)
  • Newer Launch Date: January 2026 (January 2026 vs July 2024)
  • Higher Boost Clock: 2900 MHz (2.5 GHz vs 2900 MHz)

Basic

Intel
Label Name
AMD
January 2026
Launch Date
July 2024
Integrated
Platform
Integrated
TSMC N3E
GPU Lithography
-
Intel Arc B390 GPU
Model Name
Radeon 890M
Arc B-Series
Generation
Navi III IGP
300 MHz
Base Clock
400 MHz
2.5 GHz
Boost Clock
2900 MHz
-
Bus Interface
PCIe 4.0 x8
-
Transistors
25.39 billion
12
RT Cores
16
12 Xe-cores
Compute Units
16
48
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.
64
TSMC
Foundry
TSMC
3 nm
Process Size
4 nm
Xe3
Architecture
RDNA 3.0

Memory Specifications

-
Memory Size
System Shared
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
System Shared
-
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

Display and Media

Yes
AV1 Encode/Decode
-
Yes
H.264 Hardware Encode/Decode
-
Yes
H.265 HEVC Hardware Encode/Decode
-
Decode Only
H.266 VVC Hardware Encode/Decode
-
Yes
Intel Quick Sync Video
-
7680 x 4320 @ 60Hz
Max Resolution DP
-
3840 x 2400 @ 120Hz
Max Resolution eDP
-
4
Number of Displays Supported
-
eDP 1.5, DisplayPort 2.1 UHBR20, HDMI 2.1 FRL
Outputs
Portable Device Dependent

Theoretical Performance

60 GPixel/s
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.
92.80 GPixel/s
120 GTexel/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.
185.6 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.
23.76 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.
742.4 GFLOPS
7.7 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.
11.642 TFLOPS

AI Features

OpenVINO, WindowsML, DirectML, ONNX RT, WebGPU, WebNN
AI Software Frameworks Supported by GPU
-
122
GPU Peak TOPS (Int8)
-
Yes
Intel Deep Learning Boost on GPU
-

Miscellaneous

1536
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.
1024
768 KB
L1 Cache
128 KB per Array
16 MB
L2 Cache
2 MB
-
TDP
15W
1.4
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
3.0
OpenCL Version
2.1
4.6
OpenGL
4.6
DirectX 12 Ultimate
DirectX
12 Ultimate (12_2)
-
Power Connectors
None
24
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.
32
-
Shader Model
6.7

Benchmarks

FP32 (float) / TFLOPS
Arc B390
7.7
Radeon 890M
11.642 +51%
3DMark Steel Nomad
Arc B390
1667 +200%
Radeon 890M
555
Blender
Arc B390
1281.07 +244%
Radeon 890M
372.13