Key Differences

Advantages of Radeon RX 7700 XT
Advantages of Apple M4 Max GPU (40-core)

Specifications

Performance

TGP 245 W 50 W

Storage Details

Bus 192-bit 512-bit

Ports

DisplayPort 2.1 -

Other

Forza Horizon 5 18311910072 -
The Witcher 3 18115711460 -
Counter-Strike 2 27419613473 -
Far Cry 6 14512710966 -
Hogwarts Legacy 107896739 -
CoD: Modern Warfare III 1511409963 -
Ghost of Tsushima 110917444 -
Cyberpunk 2077 1161026829 -
Shadow of the Tomb Raider 19017112765 -
1080p High 162 -
1080p Ultra 132 -
1440p Ultra 99 -
4K Ultra 57 -
Margin of Error Low High
Time Spy 17101 -
Solar Bay 64088 31561
Port Royal 9105 -
Fire Strike 42687 -
Wild Life Extreme 29765 20029
Night Raid 146439 -
Background Blur 285.8 img/sec 195 img/sec
Face Detection 181.6 img/sec 125.9 img/sec
Horizon Detection 4.9 Gpixels/sec 4.82 Gpixels/sec
Edge Detection 6.66 Gpixels/sec 7.47 Gpixels/sec
Gaussian Blur 5.68 Gpixels/sec 5.8 Gpixels/sec
Feature Matching 1.18 Gpixels/sec 1.06 Gpixels/sec
Stereo Matching 493 Gpixels/sec 417.4 Gpixels/sec
Particle Physics 19410.5 FPS 17079.9 FPS
API OpenGL Metal
G2D Mark 1162 -
DirectX 11 232 FPS -
DirectX 12 81 FPS -
GPU Compute 12954 Ops/s -
Aztec Ruins Normal Tier 1178.6 FPS 1316.3 FPS
Car Chase 1007.5 FPS 950.9 FPS
Manhattan 3.1.1 (1440p) 806.1 FPS 911 FPS
T-Rex 3466.6 FPS 3917.4 FPS
Tessellation 4717 FPS -
ALU 2 5953.9 FPS 1756 FPS
Driver Overhead 2 280.3 FPS 654.4 FPS
OS Windows OS X
Image Classification (SP) 11284 10302
Image Segmentation (HP) 14150 23763
Image Super Resolution (Q) 27972 29314
Face Detection (HP) 35311 42016
Pose Estimation (Q) 142235 105972
Text Classification (SP) 3836 2963
Machine Translation (HP) 4471 6594
Object Detection (SP) 12084 8935
Depth Estimation (Q) 49197 43143
Style Transfer (SP) 301429 221563
Framework ONNX Core ML
Backend DirectML GPU
Vendor Amd Apple
Build Discrete Integrated
Released September 6, 2023 October 30, 2024
Launch price (MSRP) $449 -
Case Desktop Laptop
Purpose Gaming Professional
Segment High-end Mid-range
Architecture RDNA 3.0 Apple M GPU
GPU Codename Navi 32 Custom
Rival Equivalent GeForce RTX 4060 Ti 8GB GeForce RTX 4070 Laptop
Base Clock 1435 MHz 500 MHz
Boost Clock 2544 MHz 1800 MHz
Shading Units 3456 5120
Texture Mapping Units (TMUs) 216 320
Render Output Units (ROPs) 96 160
Compute Units (Pipelines) - 640
Tensor Cores - -
Ray-tracing Cores 54 -
L1 Cache 128KB per cluster -
L2 Cache 2MB shared -
Instructions Per Cycle 4 IPC 2 IPC
Pixel Fill Rate 244 GPixel/s 288 GPixel/s
Texture Fill Rate 550 GTexel/s 576 GTexel/s
Interface PCIe 4.0 x16 Custom
Manufacturing TSMC TSMC
Fabrication Process 5 nm 3 nm
Die Size 346 mm² -
Transistor Count 28 billion -
Transistor Density 80.92 MTr/mm² -
Max. Temperature - 100°C
Memory Type GDDR6 System Shared
Memory Size 12 GB -
Memory Clock 2250 MHz 8533 MHz
Effective Memory Speed 18000 Mbps -
ECC No No
DirectX 12 -
Vulkan 1.3 -
OpenGL 4.6 -
OpenCL 2.2 -
CUDA - -
Ray Tracing Yes Yes
DLSS No No