NVIDIA T400 videocard review

NVIDIA T400

T400 videocard released by NVIDIA; release date: 6 May 2021. The videocard is based on Turing microarchitecture codenamed TU117.

Core clock speed - 420 MHz. Boost clock speed - 1425 MHz. Texture fill rate - 34.20 GTexel/s. Pipelines - 384. Manufacturing process technology - 12 nm. Transistors count - 4700 million. Power consumption (TDP) - 30 Watt.

Memory type: GDDR6. Maximum RAM amount - 2 GB. Memory bus width - 64 bit. Memory clock speed - 1250 MHz, 10 Gbps effective. Memory bandwidth - 80.00 GB/s.

Benchmarks

Geekbench
OpenCL
Top 1 GPU
This GPU
345150
16965
CompuBench 1.5 Desktop
Face Detection
Top 1 GPU
This GPU
939.363 mPixels/s
58.805 mPixels/s
CompuBench 1.5 Desktop
Ocean Surface Simulation
Top 1 GPU
This GPU
14101.507 Frames/s
785.059 Frames/s
CompuBench 1.5 Desktop
T-Rex
Top 1 GPU
This GPU
95.851 Frames/s
3.961 Frames/s
CompuBench 1.5 Desktop
Video Composition
Top 1 GPU
This GPU
530.303 Frames/s
53.111 Frames/s
CompuBench 1.5 Desktop
Bitcoin Mining
Top 1 GPU
This GPU
4429.590 mHash/s
222.150 mHash/s
GFXBench 4.0
Car Chase Offscreen
Top 1 GPU
This GPU
34770 Frames
5088 Frames
GFXBench 4.0
Manhattan
Top 1 GPU
This GPU
27823 Frames
2173 Frames
GFXBench 4.0
T-Rex
Top 1 GPU
This GPU
69225 Frames
3346 Frames
GFXBench 4.0
Car Chase Offscreen
Top 1 GPU
This GPU
34770.000 Fps
5088.000 Fps
GFXBench 4.0
Manhattan
Top 1 GPU
This GPU
27823.000 Fps
2173.000 Fps
GFXBench 4.0
T-Rex
Top 1 GPU
This GPU
69225.000 Fps
3346.000 Fps
Name Value
Geekbench - OpenCL 16965
CompuBench 1.5 Desktop - Face Detection 58.805 mPixels/s
CompuBench 1.5 Desktop - Ocean Surface Simulation 785.059 Frames/s
CompuBench 1.5 Desktop - T-Rex 3.961 Frames/s
CompuBench 1.5 Desktop - Video Composition 53.111 Frames/s
CompuBench 1.5 Desktop - Bitcoin Mining 222.150 mHash/s
GFXBench 4.0 - Car Chase Offscreen 5088 Frames
GFXBench 4.0 - Manhattan 2173 Frames
GFXBench 4.0 - T-Rex 3346 Frames
GFXBench 4.0 - Car Chase Offscreen 5088.000 Fps
GFXBench 4.0 - Manhattan 2173.000 Fps
GFXBench 4.0 - T-Rex 3346.000 Fps

Specifications (specs)

Essentials

Architecture Turing
Code name TU117
Launch date 6 May 2021
Place in performance rating 901

Technical info

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

Video outputs and ports

Display Connectors 3x mini-DisplayPort 1.4a

Compatibility, dimensions and requirements

Form factor Single-slot
Interface PCIe 3.0 x16
Recommended system power (PSU) 200 Watt
Supplementary power connectors None

API support

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

Memory

Maximum RAM amount 2 GB
Memory bandwidth 80.00 GB/s
Memory bus width 64 bit
Memory clock speed 1250 MHz, 10 Gbps effective
Memory type GDDR6