NVIDIA H100 PCIe 96 GB vs NVIDIA Tesla V100 DGXS

Comparative analysis of NVIDIA H100 PCIe 96 GB and NVIDIA Tesla V100 DGXS 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.

NVIDIA H100 PCIe 96 GB

NVIDIA Tesla V100 DGXS

Differences

Reasons to consider the NVIDIA H100 PCIe 96 GB

Videocard is newer: launch date 3 year(s) 11 month(s) later
2x more core clock speed: 1665 MHz vs 823 MHz
2x more boost clock speed: 1837 MHz vs 918 MHz
2094.8x more texture fill rate: 969.9 GTexel/s vs 463.0 GTexel / s
3.3x more pipelines: 16896 vs 5120
A newer manufacturing process allows for a more powerful, yet cooler running videocard: 4 nm vs 12 nm
3x more maximum memory size: 96 GB vs 32 GB

Launch date	22 Mar 2022 vs 27 March 2018
Core clock speed	1665 MHz vs 823 MHz
Boost clock speed	1837 MHz vs 918 MHz
Texture fill rate	969.9 GTexel/s vs 463.0 GTexel / s
Pipelines	16896 vs 5120
Manufacturing process technology	4 nm vs 12 nm
Maximum memory size	96 GB vs 32 GB

Reasons to consider the NVIDIA Tesla V100 DGXS

2.8x lower typical power consumption: 250 Watt vs 700 Watt
Around 33% higher memory clock speed: 1752 MHz vs 1313 MHz, 2.6 Gbps effective

Thermal Design Power (TDP)	250 Watt vs 700 Watt
Memory clock speed	1752 MHz vs 1313 MHz, 2.6 Gbps effective

Compare benchmarks

GPU 1: NVIDIA H100 PCIe 96 GB
GPU 2: NVIDIA Tesla V100 DGXS

Name	NVIDIA H100 PCIe 96 GB	NVIDIA Tesla V100 DGXS
Geekbench - OpenCL		161295

Compare specifications (specs)

	NVIDIA H100 PCIe 96 GB	NVIDIA Tesla V100 DGXS
Essentials
Architecture	Hopper	Volta
Code name	GH100	GV100
Launch date	22 Mar 2022	27 March 2018
Place in performance rating	not rated	57
Type		Workstation
Technical info
Boost clock speed	1837 MHz	918 MHz
Core clock speed	1665 MHz	823 MHz
Manufacturing process technology	4 nm	12 nm
Pipelines	16896	5120
Pixel fill rate	44.09 GPixel/s
Texture fill rate	969.9 GTexel/s	463.0 GTexel / s
Thermal Design Power (TDP)	700 Watt	250 Watt
Transistor count	80000 million	21,100 million
Floating-point performance		14,817 gflops
Video outputs and ports
Display Connectors	No outputs	No outputs
Compatibility, dimensions and requirements
Form factor	SXM Module
Interface	PCIe 5.0 x16	PCIe 3.0 x16
Recommended system power (PSU)	1100 Watt
Supplementary power connectors	8-pin EPS	None
API support
OpenCL	3.0
DirectX		12.0 (12_1)
OpenGL		4.6
Memory
Maximum RAM amount	96 GB	32 GB
Memory bandwidth	1,681 GB/s	897.0 GB / s
Memory bus width	5120 bit	4096 Bit
Memory clock speed	1313 MHz, 2.6 Gbps effective	1752 MHz
Memory type	HBM3	HBM2