|
gptkbp:instanceOf
|
gptkb:graphics_card
|
|
gptkbp:announced
|
2020-05-14
|
|
gptkbp:architecture
|
parallel computing
|
|
gptkbp:cudaComputeCapability
|
8.0
|
|
gptkbp:developedBy
|
gptkb:Nvidia
|
|
gptkbp:fabricationProcess
|
7nm
|
|
https://www.w3.org/2000/01/rdf-schema#label
|
Nvidia Ampere GPU
|
|
gptkbp:market
|
gptkb:cloud_service
professional
consumer
|
|
gptkbp:memoryType
|
gptkb:GDDR6X
gptkb:HBM2e
|
|
gptkbp:notableFeature
|
improved energy efficiency
PCIe Gen 4 support
second-generation Ray Tracing Cores
third-generation Tensor Cores
|
|
gptkbp:notableProduct
|
gptkb:Nvidia_A100
gptkb:Nvidia_RTX_3090
Nvidia RTX 3060
Nvidia RTX 3070
Nvidia RTX 3080
|
|
gptkbp:predecessor
|
Nvidia Turing GPU
|
|
gptkbp:processNode
|
Samsung 8nm (GA102, GA104, GA106, GA107)
TSMC 7nm (GA100)
|
|
gptkbp:product
|
gptkb:Nvidia_A100
|
|
gptkbp:releaseYear
|
2020
|
|
gptkbp:successor
|
Nvidia Ada Lovelace GPU
|
|
gptkbp:supports
|
gptkb:HDMI_2.1
gptkb:OpenGL_4.6
gptkb:Vulkan_1.2
gptkb:NVLink
gptkb:PCIe_4.0
gptkb:DLSS
gptkb:DirectX_12_Ultimate
gptkb:OpenCL_3.0
gptkb:Tensor_Cores
gptkb:CUDA_11
GDDR6X memory
Ray tracing
|
|
gptkbp:transistorCount
|
28.3 billion (GA102)
54.2 billion (GA100)
|
|
gptkbp:usedIn
|
gptkb:Nvidia_A100
gptkb:Nvidia_RTX_3090
gptkb:Nvidia_GeForce_RTX_30_Series
gptkb:Nvidia_Quadro_RTX_A6000
Nvidia RTX 3060
Nvidia RTX 3070
Nvidia RTX 3080
Nvidia RTX A4000
Nvidia RTX A5000
|
|
gptkbp:bfsParent
|
gptkb:Nvidia_Drive_AGX
|
|
gptkbp:bfsLayer
|
7
|