Home GPU Comparison NVIDIA H100 PCIe vs AMD Radeon Instinct MI300

NVIDIA H100 PCIe vs AMD Radeon Instinct MI300

AI GPU We compared a GPU: 80GB VRAM H100 PCIe and a Professional market GPU: 128GB VRAM Radeon Instinct MI300 to see which GPU has better performance in key specifications, benchmark tests, power consumption, etc.

Main Differences

NVIDIA H100 PCIe 's Advantages
Boost Clock has increased by 3% (1755MHz vs 1700MHz)
512 additional rendering cores
Lower TDP (350W vs 600W)
AMD Radeon Instinct MI300 's Advantages
Released 10 months late
More VRAM (128GB vs 80GB)
Larger VRAM bandwidth (3277GB/s vs 2039GB/s)

Score

Benchmark

FP32 (float)
H100 PCIe +6%
51.22 TFLOPS
Radeon Instinct MI300
47.87 TFLOPS
VS

Graphics Card

Mar 2022
Release Date
Jan 2023
Tesla Hopper
Generation
Radeon Instinct
AI GPU
Type
Professional
PCIe 5.0 x16
Bus Interface
PCIe 5.0 x16

Clock Speeds

1095 MHz
Base Clock
1000 MHz
1755 MHz
Boost Clock
1700 MHz
1593 MHz
Memory Clock
1600 MHz

Memory

80GB
Memory Size
128GB
HBM2e
Memory Type
HBM3
5120bit
Memory Bus
8192bit
2039GB/s
Bandwidth
3277GB/s

Render Config

114
SM Count
-
-
Compute Units
220
14592
Shading Units
14080
456
TMUs
880
24
ROPs
0
456
Tensor Cores
880
-
RT Cores
-
256 KB (per SM)
L1 Cache
16 KB (per CU)
50 MB
L2 Cache
16 MB

Theoretical Performance

42.12 GPixel/s
Pixel Rate
0 MPixel/s
800.3 GTexel/s
Texture Rate
1496 GTexel/s
204.9 TFLOPS
FP16 (half)
383.0 TFLOPS
51.22 TFLOPS
FP32 (float)
47.87 TFLOPS
25.61 TFLOPS
FP64 (double)
47.87 TFLOPS

Graphics Processor

GH100
GPU Name
Aqua Vanjaram
-
GPU Variant
-
Hopper
Architecture
CDNA 3.0
TSMC
Foundry
TSMC
4 nm
Process Size
5 nm
80 billion
Transistors
153 billion
814 mm²
Die Size
1017 mm²

Board Design

350W
TDP
600W
750 W
Suggested PSU
1000 W
No outputs
Outputs
No outputs
1x 16-pin
Power Connectors
2x 8-pin

Graphics Features

N/A
DirectX
N/A
N/A
OpenGL
N/A
3.0
OpenCL
3.0
N/A
Vulkan
N/A
9.0
CUDA
-
N/A
Shader Model
N/A

Related GPU Comparisons

© 2024 - TopCPU.net   Contact Us Privacy Policy