GPU model | G80 | TU106 |
---|---|---|
GPU variant | G80-100-K0-A2 | TU106-200A-KA-A1 |
Architecture | Tesla | Turing |
Foundry | TSMC | TSMC |
Manufacturing process | 90 nm | 12 nm |
Number of transistors | 681 million | 10,800 million |
Die size | 484 mm² | 445 mm² |
Release date | Nov 8th, 2006 | Jan 7th, 2019 |
---|---|---|
Generation | GeForce 8 | GeForce 20 |
Production | End-of-life | Active |
Launch price | 449 USD | 349 USD |
Interface | PCIe 1.0 x16 | PCIe 3.0 x16 |
Reviews | 403 in our database | 85 in our database |
Predecessor | GeForce 7 PCIe | GeForce 10 |
Successor | GeForce 9 | GeForce 30 |
GPU clock | 513 MHz | — |
---|---|---|
Shader clock | 1188 MHz | — |
Memory clock | 792 MHz 1584 Mbps effective | 1750 MHz 14 Gbps effective |
Base clock | — | 1365 MHz |
Boost clock | — | 1680 MHz |
Memory size | 640 MB | 6 GB |
---|---|---|
Memory type | GDDR3 | GDDR6 |
Memory bus width | 320 bit | 192 bit |
Bandwidth | 63.36 GB/s | 336.0 GB/s |
Shading units | 96 | 1920 |
---|---|---|
TMUs | 24 | 120 |
ROPs | 20 | 48 |
SM count | 12 | 30 |
Cache L2 | 80 KB | 3 MB |
Tensor cores | — | 240 |
RT cores | — | 30 |
Cache L1 | — | 64 KB (per SM) |
Pixel rate | 10.26 GPixel/s | 80.64 GPixel/s |
---|---|---|
Texture rate | 24.62 GTexel/s | 201.6 GTexel/s |
FP32 (float) performance | 228.1 GFLOPS | 6.451 TFLOPS |
FP16 (half) performance | — | 12.90 TFLOPS (2:1) |
FP64 (double) performance | — | 201.6 GFLOPS (1:32) |
Slot width | Dual-slot | Dual-slot |
---|---|---|
Length | 267 mm 10.5 inches | 229 mm 9 inches |
TDP | 143 W | 160 W |
Suggested PSU | 300 W | 450 W |
Outputs | 2x DVI1x S-Video | 1x DVI1x HDMI2x DisplayPort1x USB Type-C |
Power connectors | 1x 6-pin | 1x 8-pin |
Board number | P356 | PG160 SKU 42 |
Width | — | 113 mm 4.4 inches |
Height | — | 35 mm 1.4 inches |
DirectX | 11.1 (10_0) | 12 Ultimate (12_2) |
---|---|---|
OpenGL | 3.3 | 4.6 |
OpenCL | 1.1 (1.0) | 3.0 |
Vulkan | — | 1.2 |
CUDA | 1.0 | 7.5 |
Shader model | 4.0 | 6.6 |