AMD Radeon Pro W6900X vs NVIDIA A800 SXM4 80 GB

Comparison of AMD Radeon Pro W6900X with 32 GB GDDR6 and 5,120 cores vs NVIDIA A800 SXM4 80 GB with 80 GB HBM2e and 6,912 cores.

Loading...

Performance Rating

NVIDIA A800 SXM4 80 GB outperforms AMD Radeon Pro W6900X by 89.81% in the overall GPU ARK performance rating

A100 A100
H200 H200
MI325X MI325X

AMD Radeon Pro W6900X

13.6

AMD Radeon Pro W6900X

13.6
RX 7900 XTX RX 7900 XTX
MI250 MI250
Instinct MI300X Instinct MI300X

NVIDIA A800 SXM4 80 GB

25.9

NVIDIA A800 SXM4 80 GB

25.9

Contents:

Memory ML Performance Compute Power Architecture & Compatibility ML Software Support Clocks & Performance Power Consumption Rendering Additional

Memory

Memory Size

32 ГБ
🔥 +150% 80 ГБ

Memory Type

GDDR6 HBM2e

Memory Bandwidth

512.0 GB/s
🔥 2.04 TB/s

Memory Bus Width

256 бит 5,120 бит

ML Performance

FP16 (Half Precision)

44.46 TFLOPS
🔥 +75% 77.97 TFLOPS

BF16 (Brain Float)

No
🔥 311.84 TFLOPS

TF32 (TensorFloat)

No
🔥 155.92

Compute Power

FP32 (Single Precision)

🔥 +14% 22.23 TFLOPS
19.49 TFLOPS

FP64 (Double Precision)

No
🔥 9.746 TFLOPS

CUDA Cores

5,120
🔥 +35% 6,912

RT Cores

🔥 80
No

Architecture & Compatibility

GPU Architecture

RDNA 2.0 Ampere

SM (Streaming Multiprocessor)

No
🔥 108

PCIe Version

Apple MPX PCIe 4.0 x16

ML Software Support

CUDA Version

No 8.0

Clocks & Performance

Base Clock

🔥 +58% 1,825
1,155

Boost Clock

🔥 +54% 2,171
1,410

Memory Clock

🔥 +26% 2,000
1,593

Power Consumption

TDP/TGP

🔥 -25% 300 W
400 W

Recommended PSU

🔥 -12% 700 W
800 W

Power Connector

No None

Rendering

Texture Units (TMU)

320
🔥 +35% 432

ROP

🔥 80
No

L2 Cache

4 MB
🔥 +900% 40 MB

Additional

Slots

No
🔥 SXM Module

Release Date

Aug. 3, 2021 Aug. 11, 2022

Display Outputs

1x HDMI 2.1
4x Thunderbolt
No outputs

Renting is cheaper than buying