GeForce RTX 2080
Unlike the flagship RTX 2080 Ti accelerator, which is based on the TU102 chip, which is unprecedentedly large by consumer video card standards, the TU104 graphics processor that underlies the RTX 2080 has a close analogue among Pascal chips, namely the GP102.
The full TU104 chip contains six Graphics Processing Clusters (GPCs), each containing four Texture Processing Clusters (TPCs), consisting of one PolyMorph Engine and a pair of SM multiprocessors. Accordingly, each SM consists of: 64 CUDA cores, 256 KB of register memory and 96 KB of configurable L1 cache and shared memory, as well as four TMU texturing units.
For the needs of hardware ray tracing, each SM multiprocessor also has one RT core. In total, the full chip contains 48 SM multiprocessors, the same number of RT cores, 3072 CUDA cores and 384 tensor cores. The GeForce RTX 2080 is based on a stripped-down version of the chip with two SM units disabled by hardware. Accordingly, 2944 CUDA cores, 46 RT cores, 368 tensor cores and 184 TMU texturing units remained active in it.
Characteristics of GeForce GeForce RTX 2080
|
||||||
|
Chip
|
||||||
|
Frequencies
|
||||||
|
Memory
|
||||||
|
Interface and TDP
|
This video card has 8 gigabytes of 256-bit GDDR6 memory, and the frequency of the graphics chip ranges from 1515 to 1710 MHz.
