NVIDIA CMP 170HX vs AMD Radeon E8950

GPU Specs Compare

G

NVIDIA CMP 170HX

NVIDIA 2021-09
G

AMD Radeon E8950

AMD 2015-09

Graphics Card

Release Date
Sep 2021
Sep 2015
Generation
Mining GPUs
Embedded
Type
Desktop
Mobile
Bus Interface
PCIe 4.0 x4
MXM-B (3.0)

Clock Speeds

Base Clock
1140 MHz
735 MHz
Boost Clock
1410 MHz
1000 MHz
Memory Clock
1458 MHz
1500 MHz

Memory

Memory Size
16GB
8GB
Memory Type
HBM2e
GDDR5
Memory Bus
4096bit
256bit
Bandwidth
1493GB/s
192.0GB/s

Render Config

SM Count
70
-
Shading Units
4480
2048
TMUs
280
128
ROPs
128
32
Tensor Cores
280
-
L1 Cache
192 KB (per SM)
16 KB (per CU)
L2 Cache
8 MB
512 KB
Compute Units
-
32

Theoretical Performance

Pixel Rate
180.5 GPixel/s
32.00 GPixel/s
Texture Rate
394.8 GTexel/s
128.0 GTexel/s
FP16 (half)
50.53 TFLOPS
4.096 TFLOPS
FP32 (float)
12.63 TFLOPS
4.096 TFLOPS
FP64 (double)
6.317 TFLOPS
256.0 GFLOPS

Graphics Processor

GPU Name
GA100
Amethyst
GPU Variant
GA100-105F-A1
E8950
Architecture
Ampere
GCN 3.0
Foundry
TSMC
TSMC
Process Size
7 nm
28 nm
Transistors
54.2 billion
5 billion
Die Size
826 mm²
366 mm²

Board Design

TDP
250W
95W
Suggested PSU
600 W
-
Outputs
No outputs
No outputs
Power Connectors
2x 8-pin
None

Graphics Features

DirectX
N/A
12 (12_0)
OpenGL
N/A
4.6
OpenCL
3.0
2.0
Vulkan
N/A
1.2
CUDA
8.0
-
Shader Model
N/A
6.3