NVIDIA Tesla M10 vs NVIDIA Tesla S870
Graphics Processor
| GPU model | GM107 | G80 |
|---|---|---|
| GPU variant | GM107-570-A2 | — |
| Architecture | Maxwell | Tesla |
| Foundry | TSMC | TSMC |
| Manufacturing process | 28 nm | 90 nm |
| Number of transistors | 1,870 million | 681 million |
| Die size | 148 mm² | 484 mm² |
Graphics Card
| Release date | May 18th, 2016 | May 2nd, 2007 |
|---|---|---|
| Generation | Tesla | Tesla |
| Production | End-of-life | End-of-life |
| Interface | PCIe 3.0 x16 | PCIe 1.0 x16 |
| Launch price | — | 11,999 USD |
Clocks
| Base clock | 1033 MHz | — |
|---|---|---|
| Boost clock | 1306 MHz | — |
| Memory clock | 1300 MHz 5.2 Gbps effective | 800 MHz 1600 Mbps effective |
| GPU clock | — | 600 MHz |
| Shader clock | — | 1350 MHz |
Memory Configuration
| Memory size | 8 GB | 1536 MB |
|---|---|---|
| Memory type | GDDR5 | GDDR3 |
| Memory bus width | 128 bit | 384 bit |
| Bandwidth | 83.20 GB/s | 76.80 GB/s |
Render Configuration
| Shading units | 640 | 128 |
|---|---|---|
| TMUs | 40 | 32 |
| ROPs | 16 | 24 |
| SMM count | 5 | — |
| Cache L1 | 64 KB (per SMM) | — |
| Cache L2 | 2 MB | 96 KB |
| SM count | — | 16 |
Performance
| Pixel rate | 20.90 GPixel/s | 14.40 GPixel/s |
|---|---|---|
| Texture rate | 52.24 GTexel/s | 38.40 GTexel/s |
| FP32 (float) performance | 1.672 TFLOPS | 345.6 GFLOPS |
| FP64 (double) performance | 52.24 GFLOPS (1:32) | — |
Dimensions & Outputs
| Slot width | Dual-slot | Dual-slot |
|---|---|---|
| Length | 267 mm 10.5 inches | — |
| TDP | 225 W | 800 W |
| Suggested PSU | 550 W | 1200 W |
| Outputs | No outputs | No outputs |
| Power connectors | 1x 8-pin | — |
| Board number | P2405 SKU 70 | — |
API Support & Features
| DirectX | 12 (11_0) | 11.1 (10_0) |
|---|---|---|
| OpenGL | 4.6 | 3.3 |
| OpenCL | 3.0 | 1.1 (1.0) |
| Vulkan | 1.1 | — |
| CUDA | 5.0 | 1.0 |
| Shader model | 5.1 | 4.0 |
