NVIDIA A100 SXM4 40 GB vs AMD Radeon Pro W6800

Comparative analysis of NVIDIA A100 SXM4 40 GB and AMD Radeon Pro W6800 videocards for all known characteristics in the following categories: Essentials, Technical info, Video outputs and ports, Compatibility, dimensions and requirements, API support, Memory. Benchmark videocards performance analysis: Geekbench - OpenCL, PassMark - G2D Mark, PassMark - G3D Mark.

 

Differences

Reasons to consider the NVIDIA A100 SXM4 40 GB

  • Around 1% higher texture fill rate: 609.1 GTexel/s vs 601.7 GTexel/s
  • Around 80% higher pipelines: 6912 vs 3840
  • Around 13% lower typical power consumption: 400 Watt vs 450 Watt
  • Around 25% higher maximum memory size: 40 GB vs 32 GB
Texture fill rate 609.1 GTexel/s vs 601.7 GTexel/s
Pipelines 6912 vs 3840
Thermal Design Power (TDP) 400 Watt vs 450 Watt
Maximum memory size 40 GB vs 32 GB

Reasons to consider the AMD Radeon Pro W6800

  • Videocard is newer: launch date 1 year(s) 0 month(s) later
  • Around 89% higher core clock speed: 2075 MHz vs 1095 MHz
  • Around 78% higher boost clock speed: 2507 MHz vs 1410 MHz
  • Around 65% higher memory clock speed: 2000 MHz (16 Gbps effective) vs 1215 MHz (2.4 Gbps effective)
Launch date 8 Jun 2021 vs 14 May 2020
Core clock speed 2075 MHz vs 1095 MHz
Boost clock speed 2507 MHz vs 1410 MHz
Memory clock speed 2000 MHz (16 Gbps effective) vs 1215 MHz (2.4 Gbps effective)

Compare benchmarks

GPU 1: NVIDIA A100 SXM4 40 GB
GPU 2: AMD Radeon Pro W6800

Name NVIDIA A100 SXM4 40 GB AMD Radeon Pro W6800
Geekbench - OpenCL 191749
PassMark - G2D Mark 929
PassMark - G3D Mark 17906

Compare specifications (specs)

NVIDIA A100 SXM4 40 GB AMD Radeon Pro W6800

Essentials

Architecture Ampere RDNA 2.0
Code name GA100 Navi 21
Launch date 14 May 2020 8 Jun 2021
Place in performance rating 3 6
Launch price (MSRP) $2249

Technical info

Boost clock speed 1410 MHz 2507 MHz
Core clock speed 1095 MHz 2075 MHz
Manufacturing process technology 7 nm 7 nm
Peak Double Precision (FP64) Performance 9.746 TFLOPS (1:2) 1203 GFLOPS (1:16)
Peak Half Precision (FP16) Performance 77.97 TFLOPS (4:1) 38.51 TFLOPS (2:1)
Peak Single Precision (FP32) Performance 19.49 TFLOPS 19.25 TFLOPS
Pipelines 6912 3840
Pixel fill rate 225.6 GPixel/s 240.7 GPixel/s
Texture fill rate 609.1 GTexel/s 601.7 GTexel/s
Thermal Design Power (TDP) 400 Watt 450 Watt
Transistor count 54200 million 26800 million
Compute units 60

Video outputs and ports

Display Connectors No outputs 6x mini-DisplayPort

Compatibility, dimensions and requirements

Form factor IGP Dual-slot
Interface PCIe 4.0 x16 PCIe 4.0 x16
Recommended system power (PSU) 800 Watt 850 Watt
Supplementary power connectors None 2x 8-pin
Height 50 mm (2 inches)
Length 267 mm (10.5 inches)
Width 120 mm (4.7 inches)

API support

OpenCL 3.0 2.1
DirectX 12.2
OpenGL 4.6
Shader Model 6.5
Vulkan

Memory

High bandwidth memory (HBM)
Maximum RAM amount 40 GB 32 GB
Memory bandwidth 1555 GB/s 512 GB/s
Memory bus width 5120 bit 256 bit
Memory clock speed 1215 MHz (2.4 Gbps effective) 2000 MHz (16 Gbps effective)
Memory type HBM2e GDDR6