NVIDIA GeForce RTX 4080 12 GB vs NVIDIA CMP 90HX
We compared two Desktop platform GPUs: 12GB VRAM GeForce RTX 4080 12 GB and 10GB VRAM CMP 90HX to see which GPU has better performance in key specifications, benchmark tests, power consumption, etc.
Main Differences
NVIDIA GeForce RTX 4080 12 GB 's Advantages
Boost Clock has increased by 53% (2610MHz vs 1710MHz)
More VRAM (12GB vs 10GB)
1280 additional rendering cores
Lower TDP (285W vs 320W)
NVIDIA CMP 90HX 's Advantages
Larger VRAM bandwidth (760.3GB/s vs 504.2GB/s)
Score
Benchmark
FP32 (float)
GeForce RTX 4080 12 GB
+83%
40.09 TFLOPS
CMP 90HX
21.89 TFLOPS
Graphics Card
Unknown
Release Date
Jul 2021
GeForce 40
Generation
Mining GPUs
Desktop
Type
Desktop
PCIe 4.0 x16
Bus Interface
PCIe 1.0 x4
Clock Speeds
2310 MHz
Base Clock
1500 MHz
2610 MHz
Boost Clock
1710 MHz
1313 MHz
Memory Clock
1188 MHz
Memory
12GB
Memory Size
10GB
GDDR6X
Memory Type
GDDR6X
192bit
Memory Bus
320bit
504.2GB/s
Bandwidth
760.3GB/s
Render Config
-
-
-
60
SM Count
50
7680
Shading Units
6400
240
TMUs
200
80
ROPs
80
240
Tensor Cores
200
60
RT Cores
50
128 KB (per SM)
L1 Cache
128 KB (per SM)
48 MB
L2 Cache
5 MB
-
-
-
Theoretical Performance
208.8 GPixel/s
Pixel Rate
136.8 GPixel/s
626.4 GTexel/s
Texture Rate
342.0 GTexel/s
40.09 TFLOPS
FP16 (half)
21.89 TFLOPS
40.09 TFLOPS
FP32 (float)
21.89 TFLOPS
626.4 GFLOPS
FP64 (double)
342.0 GFLOPS
Board Design
285W
TDP
320W
600 W
Suggested PSU
700 W
1x HDMI 2.1
3x DisplayPort 1.4a
Outputs
No outputs
1x 16-pin
Power Connectors
2x 8-pin
Graphics Processor
AD104
GPU Name
GA102
AD104-400-A1
GPU Variant
GA102-100-A1
Ada Lovelace
Architecture
Ampere
TSMC
Foundry
Samsung
4 nm
Process Size
8 nm
35.8 billion
Transistors
28.3 billion
295 mm²
Die Size
628 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
8.9
CUDA
8.6
6.6
Shader Model
6.8
