NVIDIA Tesla K80 vs NVIDIA T400 4GB

Comparative analysis of NVIDIA Tesla K80 and NVIDIA T400 4GB 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: PassMark - G2D Mark, PassMark - G3D Mark, Geekbench - OpenCL.

 

Differences

Reasons to consider the NVIDIA Tesla K80

  • Around 34% higher core clock speed: 562 MHz vs 420 MHz
  • 5x more texture fill rate: 171.4 GTexel/s vs 34.20 GTexel/s
  • 6.5x more pipelines: 2496 vs 384
  • 3x more maximum memory size: 12 GB vs 4 GB
  • Around 56% better performance in PassMark - G3D Mark: 5852 vs 3760
Specifications (specs)
Core clock speed 562 MHz vs 420 MHz
Texture fill rate 171.4 GTexel/s vs 34.20 GTexel/s
Pipelines 2496 vs 384
Maximum memory size 12 GB vs 4 GB
Memory clock speed 1253 MHz, 5 Gbps effective vs 1250 MHz, 10 Gbps effective
Benchmarks
PassMark - G3D Mark 5852 vs 3760

Reasons to consider the NVIDIA T400 4GB

  • Videocard is newer: launch date 6 year(s) 5 month(s) later
  • Around 73% higher boost clock speed: 1425 MHz vs 824 MHz
  • A newer manufacturing process allows for a more powerful, yet cooler running videocard: 12 nm vs 28 nm
  • 10x lower typical power consumption: 30 Watt vs 300 Watt
  • Around 75% better performance in PassMark - G2D Mark: 653 vs 374
Specifications (specs)
Launch date 6 May 2021 vs 17 Nov 2014
Boost clock speed 1425 MHz vs 824 MHz
Manufacturing process technology 12 nm vs 28 nm
Thermal Design Power (TDP) 30 Watt vs 300 Watt
Benchmarks
PassMark - G2D Mark 653 vs 374

Compare benchmarks

GPU 1: NVIDIA Tesla K80
GPU 2: NVIDIA T400 4GB

PassMark - G2D Mark
GPU 1
GPU 2
374
653
PassMark - G3D Mark
GPU 1
GPU 2
5852
3760
Name NVIDIA Tesla K80 NVIDIA T400 4GB
PassMark - G2D Mark 374 653
PassMark - G3D Mark 5852 3760
Geekbench - OpenCL 17217

Compare specifications (specs)

NVIDIA Tesla K80 NVIDIA T400 4GB

Essentials

Architecture Kepler 2.0 Turing
Code name GK210 TU117
Launch date 17 Nov 2014 6 May 2021
Place in performance rating 256 259

Technical info

Boost clock speed 824 MHz 1425 MHz
Core clock speed 562 MHz 420 MHz
Manufacturing process technology 28 nm 12 nm
Peak Double Precision (FP64) Performance 1,371 GFLOPS (1:3) 34.20 GFLOPS (1:32)
Peak Single Precision (FP32) Performance 4.113 TFLOPS 1,094 GFLOPS
Pipelines 2496 384
Pixel fill rate 42.85 GPixel/s 22.80 GPixel/s
Texture fill rate 171.4 GTexel/s 34.20 GTexel/s
Thermal Design Power (TDP) 300 Watt 30 Watt
Transistor count 7100 million 4700 million
Peak Half Precision (FP16) Performance 2.189 TFLOPS (2:1)

Video outputs and ports

Display Connectors No outputs 3x mini-DisplayPort 1.4a

Compatibility, dimensions and requirements

Form factor Dual-slot Single-slot
Interface PCIe 3.0 x16 PCIe 3.0 x16
Length 267 mm, 10.5 inches
Recommended system power (PSU) 700 Watt 200 Watt
Supplementary power connectors 1x 8-pin None

API support

DirectX 12 (11_1) 12 (12_1)
OpenCL 3.0 3.0
OpenGL 4.6 4.6
Shader Model 6.5 (5.1) 6.7 (6.4)
Vulkan

Memory

Maximum RAM amount 12 GB 4 GB
Memory bandwidth 240.6 GB/s 80.00 GB/s
Memory bus width 384 bit 64 bit
Memory clock speed 1253 MHz, 5 Gbps effective 1250 MHz, 10 Gbps effective
Memory type GDDR5 GDDR6