CMP 170HX vs A40 PCIe
We compared two Desktop platform GPUs: 16GB VRAM CMP 170HX and 48GB VRAM A40 PCIe to see which GPU has better performance in key specifications, benchmark tests, power consumption, etc.
Main Differences
CMP 170HX 's Advantages
Released 11 months late
Larger VRAM bandwidth (1493GB/s vs 695.8GB/s)
Lower TDP (250W vs 300W)
A40 PCIe 's Advantages
Boost Clock has increased by 23% (1740MHz vs 1410MHz)
More VRAM (48GB vs 16GB)
6272 additional rendering cores
Score
Benchmark
FP32 (float)
CMP 170HX
12630
A40 PCIe
+196%
37420
Graphics Card
Sep 2021
Release Date
Oct 2020
Mining GPUs
Generation
Tesla
Desktop
Type
Desktop
PCIe 4.0 x4
Bus Interface
PCIe 4.0 x16
Clock Speeds
1140 MHz
Base Clock
1305 MHz
1410 MHz
Boost Clock
1740 MHz
1458 MHz
Memory Clock
1812 MHz
Memory
16GB
Memory Size
48GB
HBM2e
Memory Type
GDDR6
4096bit
Memory Bus
384bit
1493GB/s
Bandwidth
695.8GB/s
Render Config
70
SM Count
84
4480
Shading Units
10752
280
TMUs
336
128
ROPs
112
280
Tensor Cores
336
-
RT Cores
84
192 KB (per SM)
L1 Cache
128 KB (per SM)
8 MB
L2 Cache
6 MB
Theoretical Performance
180.5 GPixel/s
Pixel Rate
194.9 GPixel/s
394.8 GTexel/s
Texture Rate
584.6 GTexel/s
50.53 TFLOPS
FP16 (half)
37.42 TFLOPS
12.63 TFLOPS
FP32 (float)
37.42 TFLOPS
6.317 TFLOPS
FP64 (double)
584.6 GFLOPS
Board Design
250W
TDP
300W
600 W
Suggested PSU
700 W
No outputs
Outputs
3x DisplayPort 1.4a
2x 8-pin
Power Connectors
8-pin EPS
Graphics Processor
GA100
GPU Name
GA102
GA100-105F-A1
GPU Variant
-
Ampere
Architecture
Ampere
TSMC
Foundry
Samsung
7 nm
Process Size
8 nm
54.2 billion
Transistors
28.3 billion
826 mm²
Die Size
628 mm²
Graphics Features
N/A
DirectX
12 Ultimate (12_2)
N/A
OpenGL
4.6
3.0
OpenCL
3.0
N/A
Vulkan
1.3
8.0
CUDA
8.6
N/A
Shader Model
6.6
