Home GPU Comparison AMD Radeon PRO W7900 vs NVIDIA H100 SXM5

AMD Radeon PRO W7900 vs NVIDIA H100 SXM5

AI GPU We compared a Professional market GPU: 48GB VRAM Radeon PRO W7900 and a GPU: 80GB VRAM H100 SXM5 to see which GPU has better performance in key specifications, benchmark tests, power consumption, etc.

Main Differences

AMD Radeon PRO W7900 's Advantages
Released 1 years and 1 months late
Boost Clock has increased by 26% (2495MHz vs 1980MHz)
Lower TDP (295W vs 700W)
NVIDIA H100 SXM5 's Advantages
More VRAM (80GB vs 48GB)
Larger VRAM bandwidth (3350GB/s vs 864.0GB/s)
10752 additional rendering cores

Score

Benchmark

FP32 (float)
Radeon PRO W7900
61.32 TFLOPS
H100 SXM5 +9%
66.91 TFLOPS
VS

Graphics Card

Apr 2023
Release Date
Mar 2022
Radeon Pro Navi
Generation
Tesla Hopper
Professional
Type
AI GPU
PCIe 4.0 x16
Bus Interface
PCIe 5.0 x16

Clock Speeds

1855 MHz
Base Clock
1590 MHz
2495 MHz
Boost Clock
1980 MHz
2250 MHz
Memory Clock
1313 MHz

Memory

48GB
Memory Size
80GB
GDDR6
Memory Type
HBM3
384bit
Memory Bus
5120bit
864.0GB/s
Bandwidth
3350GB/s

Render Config

-
SM Count
132
96
Compute Units
-
6144
Shading Units
16896
384
TMUs
528
192
ROPs
24
-
Tensor Cores
528
96
RT Cores
-
256 KB per Array
L1 Cache
256 KB (per SM)
6 MB
L2 Cache
50 MB

Theoretical Performance

479.0 GPixel/s
Pixel Rate
47.52 GPixel/s
958.1 GTexel/s
Texture Rate
1045 GTexel/s
122.6 TFLOPS
FP16 (half)
267.6 TFLOPS
61.32 TFLOPS
FP32 (float)
66.91 TFLOPS
1.916 TFLOPS
FP64 (double)
33.45 TFLOPS

Graphics Processor

Navi 31
GPU Name
GH100
Navi 31
GPU Variant
-
RDNA 3.0
Architecture
Hopper
TSMC
Foundry
TSMC
5 nm
Process Size
4 nm
57.7 billion
Transistors
80 billion
529 mm²
Die Size
814 mm²

Board Design

295W
TDP
700W
600 W
Suggested PSU
1100 W
3x DisplayPort 2.1 1x mini-DisplayPort 2.1
Outputs
No outputs
2x 8-pin
Power Connectors
8-pin EPS

Graphics Features

12 Ultimate (12_2)
DirectX
N/A
4.6
OpenGL
N/A
2.2
OpenCL
3.0
1.3
Vulkan
N/A
-
CUDA
9.0
6.7
Shader Model
N/A

Related GPU Comparisons

© 2024 - TopCPU.net   Contact Us Privacy Policy