NVIDIA H100 SXM5 vs AMD Radeon Pro W6400. Specifications, performance, tests

Comparison of basic technical data of graphics cards NVIDIA H100 SXM5 and AMD Radeon Pro W6400, chip, information processing units.

• Architecture
  Hopper -
• Execution Units
  8448 12
• GPU Chip
  GH100 Navi 24 XL-W
• L1 Cache
  192 KB -
• L2 Cache
  50 MB -
• Render Output Units
  24 32
• Texture Units
  528 48
• Based on
  - AMD Radeon PRO
• Shader
  - 768

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

• Memory Bandwidth
  1,920 GB/s 112 GB/s
• Memory Bus
  5120 bit 64 bit
• Memory Clock
  1500 MHz -
• Memory Data Rate
  3 Gbps -
• Memory Size
  80 GB 4 GB
• Memory Type
  HBM3 GDDR6
• Memory Speed (legacy frequency label)
  - 1750 MHz

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

• Base Clock
  1065 MHz 2330 MHz
• Boost Clock
  1780 MHz 2330 MHz
• Overclocking
  - No

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

• Board Number
  PG520 -
• System Power Supply
  1100 W (PSU recommendation) -
• TDP
  700 W 50 W

Connectivity and connections.

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

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

• CUDA
  9.0 -
• DirectX
  No 12_2
• Max. resolution
  3840x2160@60Hz 7680 × 4320
• OpenCL
  3.0 2.2
• OpenGL
  No 4.6
• Shader Model
  No 6.6
• Vulkan
  No 1.3

Comparison of interfaces and release dates for GPU AMD Radeon Pro W6400 and NVIDIA H100 SXM5.

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

• FP16 (theoretical)
  120.3 TFLOPS -
• FP32 (theoretical)
  30.07 TFLOPS -
• FP64 (theoretical)
  15.04 TFLOPS -
• Pixel Rate
  42.72 GPixel/s -
• Texture Rate
  939.8 GTexel/s -