NVIDIA GeForce GTX 1650 GDDR6 vs NVIDIA CMP 90HX
We compared two Desktop platform GPUs: 4GB VRAM GeForce GTX 1650 GDDR6 and 10GB VRAM CMP 90HX to see which GPU has better performance in key specifications, benchmark tests, power consumption, etc.
Main Differences
NVIDIA GeForce GTX 1650 GDDR6 's Advantages
Lower TDP (75W vs 320W)
NVIDIA CMP 90HX 's Advantages
Released 1 years and 3 months late
Boost Clock has increased by 8% (1710MHz vs 1590MHz)
More VRAM (10GB vs 4GB)
Larger VRAM bandwidth (760.3GB/s vs 192.0GB/s)
5504 additional rendering cores
Score
Benchmark
FP32 (float)
GeForce GTX 1650 GDDR6
2.849 TFLOPS
CMP 90HX
+668%
21.89 TFLOPS
Graphics Card
Apr 2020
Release Date
Jul 2021
GeForce 16
Generation
Mining GPUs
Desktop
Type
Desktop
PCIe 3.0 x16
Bus Interface
PCIe 1.0 x4
Clock Speeds
1410 MHz
Base Clock
1500 MHz
1590 MHz
Boost Clock
1710 MHz
1500 MHz
Memory Clock
1188 MHz
Memory
4GB
Memory Size
10GB
GDDR6
Memory Type
GDDR6X
128bit
Memory Bus
320bit
192.0GB/s
Bandwidth
760.3GB/s
Render Config
-
-
-
14
SM Count
50
896
Shading Units
6400
56
TMUs
200
32
ROPs
80
-
Tensor Cores
200
-
RT Cores
50
64 KB (per SM)
L1 Cache
128 KB (per SM)
1024 KB
L2 Cache
5 MB
-
-
-
Theoretical Performance
50.88 GPixel/s
Pixel Rate
136.8 GPixel/s
89.04 GTexel/s
Texture Rate
342.0 GTexel/s
5.699 TFLOPS
FP16 (half)
21.89 TFLOPS
2.849 TFLOPS
FP32 (float)
21.89 TFLOPS
89.04 GFLOPS
FP64 (double)
342.0 GFLOPS
Board Design
75W
TDP
320W
250 W
Suggested PSU
700 W
1x DVI
1x HDMI 2.0
1x DisplayPort 1.4a
Outputs
No outputs
None
Power Connectors
2x 8-pin
Graphics Processor
TU117
GPU Name
GA102
TU117-300-A1
GPU Variant
GA102-100-A1
Turing
Architecture
Ampere
TSMC
Foundry
Samsung
12 nm
Process Size
8 nm
4.7 billion
Transistors
28.3 billion
200 mm²
Die Size
628 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
7.5
CUDA
8.6
6.6
Shader Model
6.8
