GPU model | GA102 | GT200B |
---|---|---|
GPU variant | GA102-890-A1 | — |
Architecture | Ampere | Tesla 2.0 |
Foundry | Samsung | TSMC |
Manufacturing process | 8 nm | 55 nm |
Number of transistors | 28,300 million | 1,400 million |
Die size | 628 mm² | 470 mm² |
Release date | Apr 12th, 2021 | Apr 9th, 2009 |
---|---|---|
gpu.details.availability | 2021 | — |
Generation | Tesla | Tesla |
Production | Active | End-of-life |
Interface | PCIe 4.0 x16 | PCIe 2.0 x16 |
Base clock | 885 MHz | — |
---|---|---|
Boost clock | 1695 MHz | — |
Memory clock | 1563 MHz 12.5 Gbps effective | 800 MHz 1600 Mbps effective |
GPU clock | — | 610 MHz |
Shader clock | — | 1296 MHz |
Memory size | 24 GB | 4 GB |
---|---|---|
Memory type | GDDR6 | GDDR3 |
Memory bus width | 384 bit | 512 bit |
Bandwidth | 600.2 GB/s | 102.4 GB/s |
Shading units | 9216 | 240 |
---|---|---|
TMUs | 288 | 80 |
ROPs | 96 | 32 |
SM count | 72 | 30 |
Tensor cores | 288 | — |
RT cores | 72 | — |
Cache L1 | 128 KB (per SM) | — |
Cache L2 | 6 MB | 256 KB |
Pixel rate | 162.7 GPixel/s | 19.52 GPixel/s |
---|---|---|
Texture rate | 488.2 GTexel/s | 48.80 GTexel/s |
FP16 (half) performance | 31.24 TFLOPS (1:1) | — |
FP32 (float) performance | 31.24 TFLOPS | 622.1 GFLOPS |
FP64 (double) performance | 976.3 GFLOPS (1:32) | 77.76 GFLOPS (1:8) |
Slot width | Single-slot | Dual-slot |
---|---|---|
Length | 267 mm 10.5 inches | 267 mm 10.5 inches |
Width | 112 mm 4.4 inches | — |
TDP | 150 W | 188 W |
Suggested PSU | 450 W | 450 W |
Outputs | No outputs | No outputs |
Power connectors | 8-pin EPS | — |
Board number | PG133 SKU 215 | — |
DirectX | 12 Ultimate (12_2) | 11.1 (10_0) |
---|---|---|
OpenGL | 4.6 | 3.3 |
OpenCL | 3.0 | 1.1 |
Vulkan | 1.2 | — |
CUDA | 8.6 | 1.3 |
Shader model | 6.6 | 4.0 |