GPU Comparison
GEFORCE
NVIDIA CMP 30HX
CORE STATE
TU116
VRAM
6 GB
CLOCK SPEED
1785 MHz
TDP
125 W
BUS WIDTH
192 bit
ARCHITECTURE
Turing
PROCESS
12 nm
LAUNCH DATE
2021
VS
GEFORCE
GB10
CORE STATE
GB20B
VRAM
128 GB
CLOCK SPEED
2418 MHz
TDP
140 W
BUS WIDTH
256 bit
ARCHITECTURE
Blackwell 2.0
PROCESS
5 nm
LAUNCH DATE
2025
PERFORMANCE BENCHMARKS
geekbench_opencl
geekbench_vulkan
Analysis: NVIDIA CMP 30HX vs NVIDIA GB10
Looking at the test results for NVIDIA NVIDIA CMP 30HX GPU and the NVIDIA GB10 card from NVIDIA, the picture of close performance unfolds. Not one graphics card leads repeatedly across the benchmarks. This swapping of leads suggests inherent equivalence. Buyers face a genuine choice between them. Either options achieve competent visual throughput.
All things considered, the analysis between NVIDIA NVIDIA CMP 30HX GPU and the NVIDIA GB10 card from NVIDIA ends in a practical stalemate. Gaming variations are marginal and questionable to affect practical sessions. Direct your selection on availability, driver options, and individual extra preferences. Both are capable options. The draw is genuine.
DETAILED SPECIFICATIONS
SPECIFICATION
CMP 30HX
GB10
Core Specs
Shading Units
1,408
6,144
+336.4%
Shaders
1,408
6,144
+336.4%
TMUs
88
384
+336.4%
ROPs
48
48
0.0%
SM Count
22
48
+118.2%
Clocks
Base Clock
1530 MHz
1665 MHz
Boost Clock
1785 MHz
2418 MHz
Memory Clock
1750 MHz
14 Gbps effective
1067 MHz
8.5 Gbps effective
Memory
Memory Size
6 GB
128 GB
VRAM (MB)
6,144
131,072
+2033.3%
Memory Type
GDDR6
LPDDR5X
Memory Bus
192 bit
256 bit
Bandwidth
336.0 GB/s
273.2 GB/s
Cache
L1 Cache
64 KB (per SM)
256 KB (per SM)
L2 Cache
1536 KB
50 MB
Performance
Pixel Rate
85.68 GPixel/s
116.1 GPixel/s
Texture Rate
157.1 GTexel/s
928.5 GTexel/s
FP32 (TFLOPS)
5.027 TFLOPS
29.71 TFLOPS
FP64 (TFLOPS)
157.1 GFLOPS (1:32)
464.3 GFLOPS (1:64)
FP16 (TFLOPS)
10.05 TFLOPS (2:1)
29.71 TFLOPS (1:1)
AI/RT
RT Cores
—
48
Tensor Cores
—
384
Power
TDP
125 W
140 W
TDP (W)
125
140
+12.0%
Suggested PSU
300 W
300 W
Power Connectors
1x 8-pin
None
Architecture
Architecture
Turing
Blackwell 2.0
GPU Name
TU116
GB20B
Generation
Mining GPUs
Server Blackwell
(Bxx)
Process Size
12 nm
5 nm
Transistors
6,600 million
—
Die Size
284 mm²
—
Foundry
TSMC
TSMC
Density
23.2M / mm²
—
API Support
DirectX
12 (12_1)
—
OpenGL
4.6
—
Vulkan
1.4
—
OpenCL
3.0
3.0
CUDA
7.5
12.1
Shader Model
6.8
—
Physical
Slot Width
Dual-slot
IGP
Length
229 mm
9 inches
150 mm
5.9 inches
Height
111 mm
4.4 inches
51 mm
2 inches
Outputs
No outputs
1x HDMI
Bus Interface
PCIe 1.0 x4
PCIe 5.0 x16
Other
Launch Price
799 USD
3,999 USD
Production
End-of-life
Active
Predecessor
—
Server Hopper
Successor
—
Server Rubin