GPUNVIDIA GeForce RTX 4070
Memory12GB GDDR6X, 192bit, 21Gbps, 1313MHz, 504GB/s
overclock-
clock Boost2475MHz
clock Base1920MHz
Cooling2x Axial-fan (100mm)
TDP/TGP200W (NVIDIA), max. 200W (Manli)
Power connections1x 8-Pin PCIe
Interfaces1x HDMI 2.1a, 3x DisplayPort 1.4a
InterfacePCIe 4.0 x16
Special featuresBackplate
total heightTriple Slot
Dimensions254x128x53mm (LxWxH)
TypePCIe card (full height)
PCI slot bracketfull height
Computing power29.15 TFLOPS (FP16), 29.15 TFLOPS (FP32), 0.46 TFLOPS (FP64)
Chip-data
Chip-ArchitectureAda Lovelace (from 2022/Q3)
Chip-titleAD104-250-A1 / AD104-251-A1
Chip-Configuration46SM (5888 ALUs / 184 TMUs / 64 ROPs), 46 RT Cores, 184 Tensor Cores, 36MB L2 cache
Chip-Manufacturing ProcessTSMC 5nm "NVIDIA 4N"
Chip size295mm², monolithisch
Chip-functionsRaytracing (3rd Gen NVIDIA RTX), NVIDIA Tensor (4th Gen), HDCP 2.3
API-supportDirectX 12 Ultimate (12_2) / CUDA 8.9 / Vulkan 1.3 / OpenCL 3.0 / OpenGL 4.6 / Shader Model 6.7
Transcoding-Engine1x NVENC (8th Gen, max. Sessions: 5), 1x NVDEC (5th Gen), Info
encodingAV1, H.265 (4K YUV 4:2:0 / 4K YUV 4:4:4 / 4K Lossless / 8K / HEVC 10bit support / HEVC B Frame support), H.264 (YUV 4:2:0 / YUV 4:4:4 / Lossless)
decodingAV1 (8bit / 10bit), H.265 (8bit 4:2:0 / 10bit 4:2:0 / 12bit 4:2:0 / 8bit 4:4:4 / 10bit 4:4:4 / 12bit 4:4:4), H.264, VP9 (8bit / 10bit / 12bit), VP8, VC-1, MPEG-2, MPEG-1
