GPU Comparison

NVIDIA
GEFORCE

NVIDIA CMP 90HX

CORE STATE GA102
VRAM 10 GB
CLOCK SPEED 1710 MHz
TDP 320 W
BUS WIDTH 320 bit
ARCHITECTURE Ampere
nm
PROCESS 8 nm
LAUNCH DATE 2021
VS
NVIDIA
GEFORCE

H200 NVL

CORE STATE GH100
VRAM 141 GB
CLOCK SPEED 1785 MHz
TDP 600 W
BUS WIDTH 6144 bit
ARCHITECTURE Hopper
nm
PROCESS 5 nm
LAUNCH DATE 2024

PERFORMANCE BENCHMARKS

geekbench_opencl
69,000
334,891

Analysis: NVIDIA CMP 90HX vs NVIDIA H200 NVL

This graphics card evaluation between GeForce NVIDIA H200 NVL and Radeon NVIDIA CMP 90HX shows surprisingly close performance. Each cards trade wins in different games, with not one claiming a definitive advantage. The score numbers demonstrates this assessment. Not one option leads steadily. Those encounter a genuine choice with these options.

Considering a specification standpoint, GeForce NVIDIA H200 NVL and Radeon NVIDIA CMP 90HX have matching specs. Video memory capacity, CUDA/stream count, and bus speed are nearly aligned. This spec balance explains the similar benchmark data. Neither holds a significant specification edge. The competition is level.

Ultimately, consumers might select based on availability instead of absolute frame rates only. Both GeForce NVIDIA H200 NVL and Radeon NVIDIA CMP 90HX provide strong graphics performance for the respective segments. The choice comes down to individual needs. Each option is solid. You won't go badly here.

DETAILED SPECIFICATIONS

SPECIFICATION
CMP 90HX
H200 NVL
Core Specs
Shading Units
6,400
16,896 +164.0%
Shaders
6,400
16,896 +164.0%
TMUs
200
528 +164.0%
ROPs
80
24 -70.0%
SM Count
50
132 +164.0%
Clocks
Base Clock
1500 MHz
1365 MHz
Boost Clock
1710 MHz
1785 MHz
Memory Clock
1188 MHz 19 Gbps effective
1593 MHz 6.4 Gbps effective
Memory
Memory Size
10 GB
141 GB
VRAM (MB)
10,240
144,384 +1310.0%
Memory Type
GDDR6X
HBM3e
Memory Bus
320 bit
6144 bit
Bandwidth
760.3 GB/s
4.89 TB/s
Cache
L1 Cache
128 KB (per SM)
256 KB (per SM)
L2 Cache
5 MB
50 MB
Performance
Pixel Rate
136.8 GPixel/s
42.84 GPixel/s
Texture Rate
342.0 GTexel/s
942.5 GTexel/s
FP32 (TFLOPS)
21.89 TFLOPS
60.32 TFLOPS
FP64 (TFLOPS)
342.0 GFLOPS (1:64)
30.16 TFLOPS (1:2)
FP16 (TFLOPS)
21.89 TFLOPS (1:1)
241.3 TFLOPS (4:1)
AI/RT
RT Cores
50
Tensor Cores
200
528 +164.0%
Power
TDP
320 W
600 W
TDP (W)
320
600 +87.5%
Suggested PSU
700 W
1000 W
Power Connectors
2x 8-pin
8-pin EPS
Architecture
Architecture
Ampere
Hopper
GPU Name
GA102
GH100
Generation
Mining GPUs
Server Hopper (Hxx)
Process Size
8 nm
5 nm
Transistors
28,300 million
80,000 million
Die Size
628 mm²
814 mm²
Foundry
Samsung
TSMC
Density
45.1M / mm²
98.3M / mm²
API Support
DirectX
12 Ultimate (12_2)
OpenGL
4.6
Vulkan
1.4
OpenCL
3.0
3.0
CUDA
8.6
9.0
Shader Model
6.8
Physical
Slot Width
Dual-slot
Dual-slot
Length
285 mm 11.2 inches
267 mm 10.5 inches
Height
112 mm 4.4 inches
111 mm 4.4 inches
Outputs
No outputs
No outputs
Bus Interface
PCIe 1.0 x4
PCIe 5.0 x16
Other
Production
End-of-life
Active
Predecessor
Server Ada
Successor
Server Blackwell
View CMP 90HX Details View H200 NVL Details