Mercury GSC6201 6U OpenVPX vs NVIDIA A800 SXM4 80 GB

Comparison of Mercury GSC6201 6U OpenVPX with 2 GB GDDR5 and 240 cores vs NVIDIA A800 SXM4 80 GB with 80 GB HBM2e and 6,912 cores.

Loading...

Performance Rating

H200 H200
MI325X MI325X
A100 A100

Mercury GSC6201 6U OpenVPX

Mercury GSC6201 6U OpenVPX

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

NVIDIA A800 SXM4 80 GB

NVIDIA A800 SXM4 80 GB

Contents:

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

Memory

Memory Size

2 GB ×2 (4 ГБ) 80 GB

Memory Type

GDDR5 HBM2e

Memory Bandwidth

80.00 GB/s ×2 (160 GB/s) 2.04 TB/s

Memory Bus Width

256 бит ×2 (512 бит) 5,120 бит

ML Performance

FP16 (Half Precision)

No 77.97 TFLOPS

BF16 (Brain Float)

No 311.84 TFLOPS

TF32 (TensorFloat)

No 155.92

Compute Power

FP32 (Single Precision)

0.432 TFLOPS 19.49 TFLOPS

FP64 (Double Precision)

0.036 TFLOPS 9.746 TFLOPS

CUDA Cores

240 ×2 (480) 6,912

RT Cores

No No

Architecture & Compatibility

GPU Architecture

Fermi Ampere

SM (Streaming Multiprocessor)

5 108

PCIe Version

MXM-B (3.0) PCIe 4.0 x16

ML Software Support

CUDA Version

2.1
🔥 8.0

Clocks & Performance

Base Clock

No 1,155

Boost Clock

No 1,410

Memory Clock

625 1,593

Power Consumption

TDP/TGP

🔥 -62% 150 W
400 W

Recommended PSU

No 800 W

Power Connector

None None

Rendering

Texture Units (TMU)

40 ×2 (80) 432

ROP

No No

L2 Cache

512 KB 40 MB

Additional

Slots

MXM Module SXM Module

Release Date

Feb. 22, 2011 Aug. 11, 2022

Display Outputs

Portable Device Dependent
No outputs

Renting is cheaper than buying