NVIDIA GeForce RTX 4070 GDDR6 vs AMD Radeon Pro W6800X
We compared two Desktop platform GPUs: 12GB VRAM GeForce RTX 4070 GDDR6 and 32GB VRAM Radeon Pro W6800X to see which GPU has better performance in key specifications, benchmark tests, power consumption, etc.
Main Differences
NVIDIA GeForce RTX 4070 GDDR6 's Advantages
Released 3 years late
Boost Clock has increased by 19% (2475MHz vs 2087MHz)
2048 additional rendering cores
AMD Radeon Pro W6800X 's Advantages
More VRAM (32GB vs 12GB)
Larger VRAM bandwidth (512.0GB/s vs 480.0GB/s)
Score
Benchmark
FP32 (float)
GeForce RTX 4070 GDDR6
+81%
29.15 TFLOPS
Radeon Pro W6800X
16.03 TFLOPS
Graphics Card
Aug 2024
Release Date
Aug 2021
GeForce 40
Generation
Radeon Pro Mac
Desktop
Type
Desktop
PCIe 4.0 x16
Bus Interface
Apple MPX
Clock Speeds
1920 MHz
Base Clock
1800 MHz
2475 MHz
Boost Clock
2087 MHz
2500 MHz
Memory Clock
2000 MHz
Memory
12GB
Memory Size
32GB
GDDR6
Memory Type
GDDR6
192bit
Memory Bus
256bit
480.0GB/s
Bandwidth
512.0GB/s
Render Config
-
Compute Units
60
46
SM Count
-
5888
Shading Units
3840
184
TMUs
240
64
ROPs
96
184
Tensor Cores
-
46
RT Cores
60
128 KB (per SM)
L1 Cache
128 KB per Array
36 MB
L2 Cache
4 MB
-
L3 Cache
128 MB
Theoretical Performance
158.4 GPixel/s
Pixel Rate
200.4 GPixel/s
455.4 GTexel/s
Texture Rate
500.9 GTexel/s
29.15 TFLOPS
FP16 (half)
32.06 TFLOPS
29.15 TFLOPS
FP32 (float)
16.03 TFLOPS
455.4 GFLOPS
FP64 (double)
1002 GFLOPS
Board Design
200W
TDP
200W
550 W
Suggested PSU
550 W
1x HDMI 2.1
3x DisplayPort 1.4a
Outputs
1x HDMI 2.1
4x Thunderbolt
1x 16-pin
Power Connectors
Apple MPX
Graphics Processor
AD104
GPU Name
Navi 21
AD104-2XX-A1
GPU Variant
Navi 21 Pro-XLA
Ada Lovelace
Architecture
RDNA 2.0
TSMC
Foundry
TSMC
5 nm
Process Size
7 nm
35.8 billion
Transistors
26.8 billion
294 mm²
Die Size
520 mm²
Graphics Features
12 Ultimate (12_2)
DirectX
12 Ultimate (12_2)
4.6
OpenGL
4.6
3.0
OpenCL
2.1
1.3
Vulkan
1.3
8.9
CUDA
-
6.8
Shader Model
6.7
