AMD Radeon Pro W6400 vs NVIDIA H100 PCIe. Specifications, performance, tests

Comparison of basic technical data of graphics cards AMD Radeon Pro W6400 and NVIDIA H100 PCIe, chip, information processing units.

• Based on
  AMD Radeon PRO -
• Execution Units
  12 7296
• GPU Chip
  Navi 24 XL-W GH100
• Render Output Units
  32 24
• Shader
  768 -
• Texture Units
  48 456
• Architecture
  - Hopper

Video memory capacity, type, bus width and bandwidth describe different properties. These values alone do not determine graphics performance.

• Memory Bandwidth
  112 GB/s 1,280 GB/s
• Memory Bus
  64 bit 5120 bit
• Memory Size
  4 GB 80 GB
• Memory Speed (legacy frequency label)
  1750 MHz -
• Memory Type
  GDDR6 HBM2e
• Memory Clock
  - 1000 MHz
• Memory Data Rate
  - 2 Gbps

Published GPU clock frequencies describe different operating modes. Clock speed alone does not determine graphics performance.

• Base Clock
  2330 MHz 1065 MHz
• Boost Clock
  2330 MHz 1650 MHz
• Overclocking
  No -

Graphics card power limits and recommended system power supply describe different specifications.

• TDP
  50 W 350 W
• System Power Supply
  - 750 W (PSU recommendation)

Connectivity and connections.

• DisplayPort Ports
  2 × DisplayPort 1.4a -
• Power Connectors
  No 8-pin EPS
• Max. Displays
  - 1
• PCIe Interface
  - PCIe 5.0 x16

Technical data that is used to its full potential in computer games.

• DirectX
  12_2 No
• Max. resolution
  7680 × 4320 -
• OpenCL
  2.2 3.0
• OpenGL
  4.6 No
• Shader Model
  6.6 No
• Vulkan
  1.3 No
• CUDA
  - 9.0

Comparison of interfaces and release dates for GPU NVIDIA H100 PCIe and AMD Radeon Pro W6400.

• Process Technology
  - 4 nm
• Release date
  - Q2/2022

• FP16 (theoretical)
  - 96.31 TFLOPS
• FP32 (theoretical)
  - 24.08 TFLOPS
• FP64 (theoretical)
  - 12.04 TFLOPS
• Pixel Rate
  - 39.60 GPixel/s
• Texture Rate
  - 752.4 GTexel/s