GPU model | G98 | GA102 |
---|---|---|
Architecture | Tesla | Ampere |
Foundry | TSMC | TSMC |
Manufacturing process | 65 nm | 7 nm |
Number of transistors | 210 million | 40,000 million |
Die size | 86 mm² | 627 mm² |
Release date | Dec 4th, 2007 | Unknown |
---|---|---|
Generation | GeForce 8 | Quadro RTX |
Production | End-of-life | Active |
Interface | PCI | PCIe 4.0 x16 |
Reviews | 10 in our database | — |
Predecessor | GeForce 7 PCIe | — |
Successor | GeForce 9 | — |
GPU clock | 567 MHz | — |
---|---|---|
Shader clock | 1400 MHz | — |
Memory clock | 333 MHz 666 Mbps effective | 1750 MHz 14000 MHz effective |
Base clock | — | 1110 MHz |
Boost clock | — | 1500 MHz |
Memory size | 512 MB | 48 GB |
---|---|---|
Memory type | DDR2 | GDDR6 |
Memory bus width | 64 bit | 384 bit |
Bandwidth | 5.328 GB/s | 672.0 GB/s |
Shading units | 8 | 7552 |
---|---|---|
TMUs | 4 | 472 |
ROPs | 4 | 96 |
SM count | 1 | 118 |
Cache L2 | 16 KB | 6 MB |
Tensor cores | — | 472 |
RT cores | — | 118 |
Cache L1 | — | 64 KB (per SM) |
Pixel rate | 2.268 GPixel/s | 144.0 GPixel/s |
---|---|---|
Texture rate | 2.268 GTexel/s | 708.0 GTexel/s |
FP32 (float) performance | 22.40 GFLOPS | 22.66 TFLOPS |
FP16 (half) performance | — | 45.31 TFLOPS (2:1) |
FP64 (double) performance | — | 708.0 GFLOPS (1:32) |
Slot width | Single-slot | Dual-slot |
---|---|---|
Length | 170 mm 6.7 inches | 267 mm 10.5 inches |
TDP | 25 W | 260 W |
Suggested PSU | 200 W | 600 W |
Outputs | 1x DVI1x VGA1x S-Video | 4x DisplayPort1x USB Type-C |
Power connectors | None | 1x 6-pin + 1x 8-pin |
Board number | P561 | — |
DirectX | 11.1 (10_0) | 12 Ultimate (12_2) |
---|---|---|
OpenGL | 3.3 | 4.6 |
OpenCL | 1.1 | 2.0 |
Vulkan | — | 1.2.140 |
CUDA | 1.1 | 8.5 |
Shader model | 4.0 | 6.5 |