NVIDIA CMP 90HX vs NVIDIA CMP 70HX
We compared two Desktop platform GPUs: 10GB VRAM CMP 90HX and 8GB VRAM CMP 70HX to see which GPU has better performance in key specifications, benchmark tests, power consumption, etc.
Main Differences
NVIDIA CMP 90HX 's Advantages
Released 6 months late
Boost Clock has increased by 23% (1710MHz vs 1395MHz)
More VRAM (10GB vs 8GB)
Larger VRAM bandwidth (760.3GB/s vs 608.3GB/s)
2560 additional rendering cores
Score
Benchmark
FP32 (float)
CMP 90HX
+104%
21.89 TFLOPS
CMP 70HX
10.71 TFLOPS
Graphics Card
Jul 2021
Release Date
Jan 2021
Mining GPUs
Generation
Mining GPUs
Desktop
Type
Desktop
PCIe 1.0 x4
Bus Interface
PCIe 1.0 x4
Clock Speeds
1500 MHz
Base Clock
1365 MHz
1710 MHz
Boost Clock
1395 MHz
1188 MHz
Memory Clock
1188 MHz
Memory
10GB
Memory Size
8GB
GDDR6X
Memory Type
GDDR6X
320bit
Memory Bus
256bit
760.3GB/s
Bandwidth
608.3GB/s
Render Config
-
-
-
50
SM Count
30
6400
Shading Units
3840
200
TMUs
120
80
ROPs
64
200
Tensor Cores
120
50
RT Cores
30
128 KB (per SM)
L1 Cache
128 KB (per SM)
5 MB
L2 Cache
4 MB
-
-
-
Theoretical Performance
136.8 GPixel/s
Pixel Rate
89.28 GPixel/s
342.0 GTexel/s
Texture Rate
167.4 GTexel/s
21.89 TFLOPS
FP16 (half)
10.71 TFLOPS
21.89 TFLOPS
FP32 (float)
10.71 TFLOPS
342.0 GFLOPS
FP64 (double)
167.4 GFLOPS
Board Design
320W
TDP
Unknown
700 W
Suggested PSU
200 W
No outputs
Outputs
No outputs
2x 8-pin
Power Connectors
1x 12-pin
Graphics Processor
GA102
GPU Name
GA104
GA102-100-A1
GPU Variant
GA104-100-A1
Ampere
Architecture
Ampere
Samsung
Foundry
Samsung
8 nm
Process Size
8 nm
28.3 billion
Transistors
17.4 billion
628 mm²
Die Size
392 mm²
Graphics Features
12 Ultimate (12_2)
DirectX
12 Ultimate (12_2)
4.6
OpenGL
4.6
3.0
OpenCL
3.0
1.3
Vulkan
1.3
8.6
CUDA
8.6
6.8
Shader Model
6.8
