NVIDIA A10 videocard review

A10 videocard released by NVIDIA; release date: 12 Apr 2021. The videocard is based on Ampere microarchitecture codenamed GA102.

Core clock speed - 885 MHz. Boost clock speed - 1695 MHz. Texture fill rate - 488.2 GTexel/s. Pipelines - 9216. Manufacturing process technology - 8 nm. Transistors count - 28300 million. Power consumption (TDP) - 150 Watt.

Memory type: GDDR6. Maximum RAM amount - 24 GB. Memory bus width - 384 bit. Memory clock speed - 1563 MHz, 12.5 Gbps effective. Memory bandwidth - 600.2 GB/s.

Benchmarks

PassMark
G2D Mark

Top 1 GPU

This GPU

1295

1007

PassMark
G3D Mark

Top 1 GPU

This GPU

38588

22064

Geekbench
OpenCL

Top 1 GPU

This GPU

345150

158472

Name	Value
PassMark - G2D Mark	1007
PassMark - G3D Mark	22064
Geekbench - OpenCL	158472

Specifications (specs)

Essentials
Architecture	Ampere
Code name	GA102
Launch date	12 Apr 2021
Place in performance rating	29
Technical info
Boost clock speed	1695 MHz
Core clock speed	885 MHz
Manufacturing process technology	8 nm
Peak Double Precision (FP64) Performance	976.3 GFLOPS (1:32)
Peak Half Precision (FP16) Performance	31.24 TFLOPS (1:1)
Peak Single Precision (FP32) Performance	31.24 TFLOPS
Pipelines	9216
Pixel fill rate	162.7 GPixel/s
Texture fill rate	488.2 GTexel/s
Thermal Design Power (TDP)	150 Watt
Transistor count	28300 million

Video outputs and ports
Display Connectors	No outputs
Compatibility, dimensions and requirements
Form factor	Single-slot
Interface	PCIe 4.0 x16
Length	267 mm, 10.5 inches
Recommended system power (PSU)	450 Watt
Supplementary power connectors	1x 8-pin
Width	112 mm, 4.4 inches
API support
DirectX	12 Ultimate (12_2)
OpenCL	3.0
OpenGL	4.6
Shader Model	6.7
Vulkan
Memory
Maximum RAM amount	24 GB
Memory bandwidth	600.2 GB/s
Memory bus width	384 bit
Memory clock speed	1563 MHz, 12.5 Gbps effective
Memory type	GDDR6