AMD Radeon Pro Duo vs NVIDIA CMP 170HX

GPU Specs Compare

G

AMD Radeon Pro Duo

AMD 2016-04
G

NVIDIA CMP 170HX

NVIDIA 2021-09

Graphics Card

Release Date
Apr 2016
Sep 2021
Generation
Radeon Pro
Mining GPUs
Type
Desktop
Desktop
Bus Interface
PCIe 3.0 x16
PCIe 4.0 x4

Clock Speeds

Memory Clock
500 MHz
1458 MHz
Base Clock
-
1140 MHz
Boost Clock
-
1410 MHz

Memory

Memory Size
4GB
16GB
Memory Type
HBM
HBM2e
Memory Bus
4096bit
4096bit
Bandwidth
512.0GB/s
1493GB/s

Render Config

Compute Units
64
-
Shading Units
4096
4480
TMUs
256
280
ROPs
64
128
L1 Cache
16 KB (per CU)
192 KB (per SM)
L2 Cache
2 MB
8 MB
SM Count
-
70
Tensor Cores
-
280

Theoretical Performance

Pixel Rate
64.00 GPixel/s
180.5 GPixel/s
Texture Rate
256.0 GTexel/s
394.8 GTexel/s
FP16 (half)
8.192 TFLOPS
50.53 TFLOPS
FP32 (float)
8.192 TFLOPS
12.63 TFLOPS
FP64 (double)
512.0 GFLOPS
6.317 TFLOPS

Graphics Processor

GPU Name
Capsaicin
GA100
GPU Variant
Capsaicin XT
GA100-105F-A1
Architecture
GCN 3.0
Ampere
Foundry
TSMC
TSMC
Process Size
28 nm
7 nm
Transistors
8.9 billion
54.2 billion
Die Size
596 mm²
826 mm²

Board Design

TDP
350W
250W
Suggested PSU
750 W
600 W
Outputs
1x HDMI 1.4a3x DisplayPort 1.2
No outputs
Power Connectors
3x 8-pin
2x 8-pin

Graphics Features

DirectX
12 (12_0)
N/A
OpenGL
4.6
N/A
OpenCL
2.0
3.0
Vulkan
1.2
N/A
Shader Model
6.0
N/A
CUDA
-
8.0