Intel Arc B580 vs Arc B390

We performed a head-to-head comparison of the Intel Arc B580 12 GB with 20 pipelines and 2560 shaders against the 1 year and 2 months newer Arc B390 that utilizes 12 pipelines and 1536 shaders. Here you will find complete details about specs, efficiency, performance tests, and more.

Review

General comparison of performance in games, applications, power efficiency, and other metrics
Gaming
Performance in DirectX, OpenCL, and Vulkan games
Workstation
Perf. in 3D modeling, video editing and rendering apps
AI/ML
Capabilities for machine learning and AI-related tasks
Energy Efficiency
Power consumption efficiency in different scenarios
NanoReview Final Score
Overall video card score
The "Energy Efficiency" metric has less impact on the NanoReview Score for desktop GPU.

Key differences

Key distinctions and advantages of Arc B390 over Arc B580
Reasons to consider the Intel Arc B580
  • Performs significantly better (up to 2x) in 3DMark Steel Nomad Lite
  • Includes 12 GB of dedicated GDDR6 memory
  • Shows 2x higher average frame rate in modern games at QHD resolution – 67 vs 33 FPS
  • 78% higher maximum theoretical performance (13.7 vs 7.7 TFLOPS)
  • Achieves 2x more points in the GeekBench 6 Compute test (111K vs 56K)
  • Features 64 more tensor cores for effective ML and AI workloads
  • Has 67% more shading units (2560 vs 1536)
Reasons to consider the Intel Arc B390
  • Manufactured using a more efficient 3 nm process technology

Gaming Performance

Frame rate comparison across popular AAA titles at different resolutions

Games

FPS Table
Forza Horizon 5
123
78
67
51
80
41
36
20
The Witcher 3
147
128
91
46
76
63
44
19
Counter-Strike 2
162
116
83
40
96
70
46
23
Far Cry 6
108
98
76
43
63
56
38
18
Hogwarts Legacy
71
57
44
26
33
24
18
-
CoD: Modern Warfare III
82
85
61
38
37
35
28
13
Ghost of Tsushima
72
61
42
25
37
31
24
12
Cyberpunk 2077
90
82
55
26
43
38
21
9
Shadow of the Tomb Raider
124
107
83
43
65
57
39
23
Average FPS by Resolution
1080p High 109 59
1080p Ultra 90 46
1440p Ultra 67 33
4K Ultra 38 16
Margin of Error Low Low

Benchmarks

Graphics cards’ performance in recent benchmarking apps

3D Mark

Multiplatform graphics benchmark suite that directly correlates with performance in modern games
Steel Nomad Lite Score
Arc B580 +105%
11186
5461
Time Spy 14685 6880
Solar Bay 55846 26783
Port Royal 7939 3880
Fire Strike 36801 16845
Wild Life Extreme 28331 11778
Night Raid 125066 72249
Sources: 3DMark [1], [2]

GeekBench 6 OpenCL

GPU test for computational tasks (image processing, photography, computer vision, and ML)
GB6 Compute Score
Arc B580 +98%
111561
56204
Background Blur 284.6 img/sec 87.6 img/sec
Face Detection 150 img/sec 58.5 img/sec
Horizon Detection 2.67 Gpixels/sec 1.9 Gpixels/sec
Edge Detection 5.22 Gpixels/sec 2.68 Gpixels/sec
Gaussian Blur 6.35 Gpixels/sec 4.74 Gpixels/sec
Feature Matching 1.18 Gpixels/sec 1.01 Gpixels/sec
Stereo Matching 397 Gpixels/sec 249.7 Gpixels/sec
Particle Physics 15177.3 FPS 10821.1 FPS
API OpenCL OpenCL
Sources: Geekbench [3], [4]

Passmark Graphics

Videocard test that focuses on compute shaders, multi-texturing, tessellation, and other features
G3D Mark Score
Arc B580 +80%
16050
8926
G2D Mark 702 812
DirectX 11 126 FPS 70 FPS
DirectX 12 78 FPS 48 FPS
GPU Compute 8003 Ops/s 3669 Ops/s
Sources: PassMark [5], [6]3665 & 425 samples

Blender

Rendering performance test for 3D modeling
Blender GPU
Arc B580 +71%
2175.32
1269.36
Sources: Blender [9], [10]357 & 95 samples

Recent User Tests

The latest benchmark tests that have been submitted by users
Intel Arc B580
DateBenchmarkResult
📘 2026-07-07 (Mike)GeekBench113977
📘 2026-07-07 (Mike)Steel Nomad Light11944
📘 2026-06-17 (Mike)GeekBench103868
Intel Arc B390
DateBenchmarkResult
📘 2026-09-05 -> asterismxGeekBench57164

Artificial Intelligence Tests

Performance in machine learning and artificial intelligence tasks

GeekBench 6 ML

Tests throughput of AI operations in single, half, and quantized precision
GB6 ML Single Precision
Arc B580 +68%
21661
12930
GB6 ML Half Precision
Arc B580 +45%
34463
23819
GB6 ML Quantized
Arc B580 +70%
14366
8437
Image Classification (SP) 10654 5487
Image Segmentation (HP) 18608 16864
Image Super Resolution (Q) 16715 8788
Face Detection (HP) 40348 25029
Pose Estimation (Q) 64494 18852
Text Classification (SP) 3091 2115
Machine Translation (HP) 3530 4266
Object Detection (SP) 13690 7449
Depth Estimation (Q) 36797 22207
Style Transfer (SP) 197057 95039
Framework ONNX ONNX
Backend DirectML DirectML
Sources: Geekbench [9], [10]

Specifications

Technical specifications of Intel Arc B580 and Arc B390

General

Vendor Intel Intel
Build Discrete Integrated
Released December 3, 2024 January 5, 2026
Launch price (MSRP) $249 -
Case Desktop Laptop
Purpose Gaming Gaming
Segment Mid-range Mid-range
Architecture Generation 12.7 Celestial
GPU Codename BMG-G21 Panther Lake
Rival Equivalent - GeForce RTX 4060 -
Recommended CPU - Intel Core Ultra 5 245K or above - Intel Core Ultra X9 388H or above
Used in CPUs - - Intel Core Ultra X9 378H
- Intel Core Ultra X7 368H
- Intel Core Ultra X9 388H
- Intel Core Ultra X7 358H
Desktop GPU rating (#49th place)
Laptop GPU ranking (#62nd place)

Graphics Processing Unit

Base Clock 1700 MHz 300 MHz
Boost Clock 2670 MHz 2500 MHz
Shading Units 2560 1536
Texture Mapping Units (TMUs) 160 48
Render Output Units (ROPs) 80 24
Compute Units (Pipelines) 20 12
Tensor Cores 160 96
Ray-tracing Cores 20 12
L1 Cache 256KB per cluster 64KB per cluster
L2 Cache 12MB shared 16MB shared
Instructions Per Cycle 2 IPC 2 IPC

Raw Performance

Pixel Fill Rate 214 GPixel/s 60 GPixel/s
Texture Fill Rate 427 GTexel/s 120 GTexel/s
FLOPS (FP32)
Arc B580 +78%
13.7 TFLOPS
7.7 TFLOPS

Physical

Interface PCIe 4.0 x8 IGP
TGP 190 W 30 W
Manufacturing TSMC TSMC
Fabrication Process 5 nm 3 nm
Die Size 272 mm² -
Transistor Count 19.6 billion -
Transistor Density 72.06 MTr/mm² -
Max. Temperature - 100°C

Memory

Memory Type GDDR6 System Shared
Memory Size 12 GB -
Memory Clock 2375 MHz -
Effective Memory Speed 19000 Mbps -
Bus 192-bit -
ECC No No
Memory Bandwidth
456 GB/s
n/a

API

DirectX 12 12.1
Vulkan 1.3 1.4
OpenGL 4.6 4.6
OpenCL 3.0 3.0
CUDA No No
Ray Tracing Yes Yes
DLSS No No
DisplayPort 2.1 2.1

Cast your vote

Choose between two graphics cards
5 (62.5%)
3 (37.5%)
Total votes: 8

User opinions

You can share your opinion or ask a question in the comments below
🌐 Register your profile and become part of NanoReview community!