NVIDIA H100 SXM5 contre NVIDIA GeForce RTX 3090 Ti. Spécifications, performances, tests

Comparaison des données techniques de base des cartes graphiques NVIDIA H100 SXM5 et NVIDIA GeForce RTX 3090 Ti, puce, unités de traitement de l'information.

• Architecture
  Hopper Ampere
• Execution Units
  8448 -
• Puce GPU
  GH100 GA102-350-A1
• L1 Cache
  192 KB 128 KB
• L2 Cache
  50 MB 6 MB
• Unités de sortie de rendu
  24 112
• Unités de texture
  528 336
• Sur la base de
  - GA102
• CUDA Cores
  - 10752
• Shader
  - 10752

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 1,008 GB/s
• Memory Bus
  5120 bit 384 bit
• Memory Clock
  1500 MHz -
• Memory Data Rate
  3 Gbps -
• Taille de la mémoire
  80 GB 24 GB
• Type de mémoire
  HBM3 GDDR6X
• Memory Speed (legacy frequency label)
  - 1300 MHz

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

• Fréquence de base
  1065 MHz 1560 MHz
• Fréquence maximale
  1780 MHz 1860 MHz
• Overclocking
  - Non

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

• Board Number
  PG520 -
• System Power Supply
  1100 W (PSU recommendation) 850 W (System power requirement)
• TDP
  700 W 450 W
• Graphics Card Power
  - 450 W
• Maximum GPU Temperature
  - 92 °C

Connectivité et connexions.

• Max. Affichages
  1 4
• PCIe Interface
  PCIe 5.0 x16 PCIe 4.0 x16
• Power Connectors
  8-pin EPS 3x PCIe 8-pin cables (adapter in box) OR 450W or greater PCIe Gen 5 cable
• DisplayPort Ports
  - 3 × DisplayPort 1.4a
• HDCP Version
  - 2.3
• Ports HDMI
  - 1 × HDMI 2.1
• Other Display Outputs (unparsed)
  - HDMI

Des données techniques utilisées à leur plein potentiel dans les jeux informatiques.

• CUDA
  9.0 -
• DirectX
  Non 12_2
• Résolution maximale
  3840x2160@60Hz 7680 × 4320
• OpenCL
  3.0 3.0
• OpenGL
  Non 4.6
• Shader Model
  Non 6.6
• Vulkan
  Non 1.3

Comparaison des interfaces et des dates de sortie des GPU NVIDIA GeForce RTX 3090 Ti et NVIDIA H100 SXM5.

• Process Technology
  4 nm 8 nm
• Date de sortie
  Q2/2022 -
• Manufacturer specification scope
  - NVIDIA Founders Edition/reference design. Clock specifications apply to gaming use. AIB shipping specifications and complete-system power needs may differ.

• FP16 (theoretical)
  120.3 TFLOPS 40 TFLOPS
• FP32 (theoretical)
  30.07 TFLOPS 40 TFLOPS
• FP64 (theoretical)
  15.04 TFLOPS 0.625 TFLOPS
• Pixel Rate
  42.72 GPixel/s 208.3 GPixel/s
• Texture Rate
  939.8 GTexel/s 625.0 GTexel/s