Key Differences

Advantages of GeForce RTX 3070
Advantages of Radeon 840M

Specifications

Performance

TGP 220 W 15-54 W (configurable)

Storage Details

Bus 256-bit -

Ports

DisplayPort 1.4a 2.1

Other

Forza Horizon 5 145958562 301412 -
The Witcher 3 17014110356 1012 - -
Counter-Strike 2 25119213676 252013 -
Far Cry 6 1271189557 241910 -
Hogwarts Legacy 97806233 12 - - -
CoD: Modern Warfare III 1251138452 2218139
Ghost of Tsushima 87755630 139 - -
Cyberpunk 2077 108925830 1110 - -
Shadow of the Tomb Raider 16715211462 181410 -
1080p High 142 18
1080p Ultra 118 14
1440p Ultra 88 9
4K Ultra 51 -
Margin of Error Low High
Time Spy 13642 1673
Solar Bay 64327 7110
Port Royal 8311 611
Fire Strike 34046 5293
Wild Life Extreme 28473 -
Night Raid 121837 22773
Background Blur 161.6 img/sec 31.8 img/sec
Face Detection 117.2 img/sec 16.8 img/sec
Horizon Detection 5.89 Gpixels/sec 0.46 Gpixels/sec
Edge Detection 8.05 Gpixels/sec 0.58 Gpixels/sec
Gaussian Blur 6.65 Gpixels/sec 0.42 Gpixels/sec
Feature Matching 1.25 Gpixels/sec 0.18 Gpixels/sec
Stereo Matching 550.6 Gpixels/sec 52.5 Gpixels/sec
Particle Physics 17613.1 FPS 2048.6 FPS
API OpenGL OpenGL
G2D Mark 1003 -
DirectX 11 182 FPS -
DirectX 12 85 FPS -
GPU Compute 11130 Ops/s -
Aztec Ruins Normal Tier 856 FPS 139.4 FPS
Car Chase 668 FPS 124 FPS
Manhattan 3.1.1 (1440p) 592.9 FPS 98.1 FPS
T-Rex 3078.4 FPS 441.1 FPS
Tessellation 4563.3 FPS 605.3 FPS
ALU 2 4531.5 FPS 635.4 FPS
Driver Overhead 2 283 FPS 286.6 FPS
OS Windows Windows
Image Classification (SP) 7624 1988
Image Segmentation (HP) 23420 2184
Image Super Resolution (Q) 17898 4788
Face Detection (HP) 40535 6857
Pose Estimation (Q) 83624 17396
Text Classification (SP) 2724 1628
Machine Translation (HP) 4481 2053
Object Detection (SP) 10304 1825
Depth Estimation (Q) 30418 8471
Style Transfer (SP) 180089 31032
Framework ONNX ONNX
Backend DirectML DirectML
Vendor Nvidia Amd
Build Discrete Integrated
Released October 22, 2020 January 7, 2025
Launch price (MSRP) $499 -
Case Desktop Laptop
Purpose Gaming Generic
Segment Mid-range Entry-level
Architecture Ampere RDNA 3.5
GPU Codename GA104 Krackan Point
Rival Equivalent Radeon RX 7800 XT Intel Arc 130T
Base Clock 1500 MHz -
Boost Clock 1725 MHz 2900 MHz
Shading Units 5888 256
Texture Mapping Units (TMUs) 184 16
Render Output Units (ROPs) 96 8
Compute Units (Pipelines) 46 4
Tensor Cores 184 -
Ray-tracing Cores 46 4
L1 Cache 128KB per cluster 128KB per cluster
L2 Cache 4MB shared 2MB shared
Instructions Per Cycle 2 IPC 4 IPC
Pixel Fill Rate 166 GPixel/s 23 GPixel/s
Texture Fill Rate 317 GTexel/s 46 GTexel/s
Interface PCIe 4.0 x16 PCIe 4.0 x16
Manufacturing Samsung TSMC
Fabrication Process 8 nm 4 nm
Die Size 392 mm² -
Transistor Count 17 billion -
Transistor Density 43.37 MTr/mm² -
Max. Temperature - 100°C
Memory Type GDDR6 System Shared
Memory Size 8 GB -
Memory Clock 1750 MHz -
Effective Memory Speed 14000 Mbps -
ECC No -
DirectX 12 12
Vulkan 1.3 1.4
OpenGL 4.6 4.6
OpenCL 3.0 2.1
CUDA 8.6 No
Ray Tracing Yes Yes
DLSS DLSS 2 No