NVIDIA GeForce GTX 1070 vs NVIDIA H100 PCIe. Specifications, performance, tests

Comparison of basic technical data of graphics cards NVIDIA GeForce GTX 1070 and NVIDIA H100 PCIe, chip, information processing units.

• GPU Chip
  GP104 GH100
• Execution units
  1920 7296
• Render Output Units
  64 24
• Texture Units
  120 456

Comparison of the amount of memory on board graphics cards. The more the better.

• Memory Size
  8 GB 80 GB
• Memory Type
  GDDR5 HBM2e
• Memory Clock
  2 GHz 1 GHz
• Memory Speed
  8 Gbps 2 Gbps
• Memory Bandwith
  256.3 GB/s 1,280 GB/s
• Memory Interface
  256 bit 5120 bit

Let's compare the memory frequency of graphics cards NVIDIA GeForce GTX 1070 and NVIDIA H100 PCIe. The higher the better.

• Base Clock
  1.506 GHz 1.065 GHz
• Boost Clock
  1.683 GHz 1.65 GHz

Connectors, the number of thermal watts emitted in normal mode and at overclocking.

• TDP
  150 W 350 W
• Power Connectors
  1x 8-pin 8-pin EPS
• Suggested PSU
  450 W 750 W

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

• DirectX
  12.1 No
• OpenGL
  4.6 No
• OpenCL
  3.0 3.0
• Vulkan
  1.3 No
• CUDA
  6.1 9.0
• Shader Model
  6.7 (6.4) No

Comparison of interfaces and release dates for GPU NVIDIA H100 PCIe and NVIDIA GeForce GTX 1070.

• Architecture
  Pascal Hopper
• GPU Interface
  PCIe 3.0 x16 PCIe 5.0 x16
• Structure size
  16 nm 4 nm
• Release date
  Q3/2016 Q2/2022

• Pixel Rate
  107.7 GPixel/s 39.60 GPixel/s
• Texture Rate
  202.0 GTexel/s 752.4 GTexel/s
• FP16 (half)
  101.0 GFLOPS (1:64) 96.31 TFLOPS
• FP32 (float)
  6.463 TFLOPS 24.08 TFLOPS
• FP64 (double)
  202.0 GFLOPS (1:32) 12.04 TFLOPS