NVIDIA P106 090 vs NVIDIA CMP 70HX
We compared two Desktop platform GPUs: 3GB VRAM P106 090 and 8GB VRAM CMP 70HX to see which GPU has better performance in key specifications, benchmark tests, power consumption, etc.
Main Differences
NVIDIA P106 090 's Advantages
Boost Clock has increased by 10% (1531MHz vs 1395MHz)
NVIDIA CMP 70HX 's Advantages
Released 3 years and 6 months late
More VRAM (8GB vs 3GB)
Larger VRAM bandwidth (608.3GB/s vs 192.2GB/s)
3072 additional rendering cores
Score
Benchmark
FP32 (float)
P106 090
2.352 TFLOPS
CMP 70HX
+355%
10.71 TFLOPS
Graphics Card
Jul 2017
Release Date
Jan 2021
Mining GPUs
Generation
Mining GPUs
Desktop
Type
Desktop
PCIe 1.0 x1
Bus Interface
PCIe 1.0 x4
Clock Speeds
1354 MHz
Base Clock
1365 MHz
1531 MHz
Boost Clock
1395 MHz
2002 MHz
Memory Clock
1188 MHz
Memory
3GB
Memory Size
8GB
GDDR5
Memory Type
GDDR6X
192bit
Memory Bus
256bit
192.2GB/s
Bandwidth
608.3GB/s
Render Config
-
-
-
6
SM Count
30
768
Shading Units
3840
48
TMUs
120
48
ROPs
64
-
Tensor Cores
120
-
RT Cores
30
48 KB (per SM)
L1 Cache
128 KB (per SM)
1536 KB
L2 Cache
4 MB
-
-
-
Theoretical Performance
73.49 GPixel/s
Pixel Rate
89.28 GPixel/s
73.49 GTexel/s
Texture Rate
167.4 GTexel/s
36.74 GFLOPS
FP16 (half)
10.71 TFLOPS
2.352 TFLOPS
FP32 (float)
10.71 TFLOPS
73.49 GFLOPS
FP64 (double)
167.4 GFLOPS
Board Design
75W
TDP
Unknown
250 W
Suggested PSU
200 W
No outputs
Outputs
No outputs
1x 6-pin
Power Connectors
1x 12-pin
Graphics Processor
GP106
GPU Name
GA104
GP106-090-A1
GPU Variant
GA104-100-A1
Pascal
Architecture
Ampere
TSMC
Foundry
Samsung
16 nm
Process Size
8 nm
4.4 billion
Transistors
17.4 billion
200 mm²
Die Size
392 mm²
Graphics Features
12 (12_1)
DirectX
12 Ultimate (12_2)
4.6
OpenGL
4.6
3.0
OpenCL
3.0
1.3
Vulkan
1.3
6.1
CUDA
8.6
6.8
Shader Model
6.8
