NVIDIA partners have released the entry-level T1000 professional graphics accelerator for Computer Aided Design and Manufacturing (CAD/CAM) applications.
The solution is based on the TU117 chip with the Turing architecture. The configuration includes 896 CUDA cores, and the amount of GDDR6 memory with a 128-bit bus can be 4 or 8 GB.
Delivers up to 2.5 teraflops of performance in single-precision operations. The memory bandwidth reaches 160 GB/s. The graphics accelerator uses the PCI Express 3.0 x16 interface.
Power consumption is about 50W; additional power connection is not required. The card has a single-slot design and a low-profile form factor. The active cooling system includes a fan.
The NVIDIA T1000 accelerator is equipped with four Mini DisplayPort interfaces. You can simultaneously output images to four 4K monitors or two 8K monitors with a refresh rate of 60 Hz.
The price of the accelerator varies from $400 to $500 depending on the amount of memory and equipment, which can include two or four Mini DisplayPort to HDMI adapters.