NVIDIA Rubin GPU vs NVIDIA CMP 170HX

GPU Specs Compare

G

NVIDIA Rubin GPU

NVIDIA 2026-01
G

NVIDIA CMP 170HX

NVIDIA 2021-09

Graphics Card

Release Date
Jan 2026
Sep 2021
Generation
Server Rubin
Mining GPUs
Type
Desktop
Desktop
Bus Interface
PCIe 6.0 x16
PCIe 4.0 x4

Clock Speeds

Base Clock
700 MHz
1140 MHz
Boost Clock
2267 MHz
1410 MHz
Memory Clock
2695 MHz
1458 MHz

Memory

Memory Size
288GB
16GB
Memory Type
HBM4
HBM2e
Memory Bus
16384bit
4096bit
Bandwidth
22.1TB/s
1493GB/s

Render Config

SM Count
224
70
Shading Units
28672
4480
TMUs
896
280
ROPs
24
128
Tensor Cores
896
280
L1 Cache
256 KB (per SM)
192 KB (per SM)
L2 Cache
128 MB
8 MB

Theoretical Performance

Pixel Rate
54.41 GPixel/s
180.5 GPixel/s
Texture Rate
2031.2 GTexel/s
394.8 GTexel/s
FP16 (half)
260.0 TFLOPS
50.53 TFLOPS
FP32 (float)
130.0 TFLOPS
12.63 TFLOPS
FP64 (double)
32.50 TFLOPS
6.317 TFLOPS

Graphics Processor

GPU Name
GR100
GA100
Architecture
Rubin
Ampere
Foundry
TSMC
TSMC
Process Size
3 nm
7 nm
Transistors
336 billion
54.2 billion
Die Size
1456 mm²
826 mm²
GPU Variant
-
GA100-105F-A1

Board Design

TDP
2300W
250W
Suggested PSU
2700 W
600 W
Outputs
No outputs
No outputs
Power Connectors
-
2x 8-pin

Graphics Features

DirectX
N/A
N/A
OpenGL
N/A
N/A
OpenCL
3.0
3.0
Vulkan
N/A
N/A
CUDA
10.7
8.0
Shader Model
N/A
N/A