Splošni opis:
ModelNVIDIA GeForce RTX 5080
memory16GB GDDR7, 256bit, 30Gbps, 1875MHz, 960GB/s
Clock speed2295MHz (Standard Profile)
Clock speed - boost2655MHz (standard profile), 2685MHz (OC profile)
Overclocking+68MHz Boost (OC profile), +38MHz Boost (Standard profile)
cooling3x axial fans (90mm) with 0dB zero-fan mode (up to 55°C)
TDP/TGP360W (NVIDIA)
Power connections1x 12V-2x6 (via adapter: 3x 8-pin PCIe)
Connections1x HDMI 2.1b, 3x DisplayPort 2.1b UHBR20
ConnectionPCIe 5.0 x16
Special featuresBackplate, Dual BIOS Switch, AI Acceleration, Clock Profiles (Default Mode/OC Mode)
Total heightTriple Slot (2.5 Slots)
Dimensions306x126x50mm (LxWxH)
DesignPCIe card (full height)
Slot coverfull height
AI computing power (dense)900.4 TOPS (FP4), 450.2 TOPS (FP8), 225.1 TOPS (FP16)
AI computing power (sparse)1800.8 TOPS (FP4), 900.4 TOPS (FP8), 450.2 TOPS (FP16)
GPU computing power57.74 TFLOPS (FP16), 57.74 TFLOPS (FP32), 0.90 TFLOPS (FP64)
Chip data
Chip architectureBlackwell (from 2025/Q1)
Chip designationGB203 (GB203-400-A1)
Chip expansionFully expanded (84/84 SM)
Chip configuration84SM (10752SP/336TMU/112ROP), 84 RT cores, 336 Tensor cores, 64MiB L2 cache
Chip manufacturingTSMC N5P [NVIDIA 4N]
Chip size378 mm², monolithic, 45.6 billion transistors
Chip functionsRay tracing (4th Gen NVIDIA RTX, 171 TFLOPS), NVIDIA Tensor (5th Gen, 225.1 TOPS), HDCP 2.3
API supportDirectX 12 Ultimate (12_2) / CUDA 12.8 / Vulkan 1.3 / OpenCL 3.0 / OpenGL 4.6 / Shader Model 6.8
Transcoding engine2x NVENC (9th Gen, max. Sessions: 8), 2x NVDEC (6th Gen)
EncodingAV1 (YUV 4:2:0), H.265 (4K YUV 4:2:0 / 4K YUV 4:2:2 / 4K YUV 4:4:4 / 4K Lossless / 8K / HEVC 10bit support / HEVC B Frame support), H.264 (YUV 4:2:0 / YUV 4:2:2 / YUV 4:4:4 / Lossless)
DecodingAV1 (8bit / 10bit), H.265 (8bit 4:2:0 / 10bit 4:2:0 / 12bit 4:2:0 / 8bit 4:2:2 / 10bit 4:2:2 / 8bit 4:4:4 / 10bit 4:4:4 / 12bit 4:4:4), H.264 (8bit 4:2:0, 10bit 4:2:0, 8bit 4:2:2, 10bit 4:2:2), VP9 (8bit / 10bit / 12bit), VP8, VC-1, MPEG-2, MPEG-1
