AMD Radeon R9 M470 vs NVIDIA GeForce GTX 485M
Graphics Processor
| GPU model | Emerald | GF104 |
|---|---|---|
| GPU variant | Emerald PRO | N12E-GTX-A1 |
| Architecture | GCN 2.0 | Fermi |
| Foundry | TSMC | TSMC |
| Manufacturing process | 28 nm | 40 nm |
| Number of transistors | 2,080 million | 1,950 million |
| Die size | 160 mm² | 332 mm² |
Mobile graphics
| Release date | May 15th, 2016 | Jan 5th, 2011 |
|---|---|---|
| Generation | Crystal System (R9 M400) | GeForce 400M |
| Production | End-of-life | End-of-life |
| Interface | PCIe 3.0 x16 | MXM-B (3.0) |
| Predecessor | Solar System | GeForce 300M |
| Successor | Mobility Radeon | GeForce 500M |
Clocks
| Base clock | 900 MHz | — |
|---|---|---|
| Boost clock | 1000 MHz | — |
| Memory clock | 1500 MHz 6 Gbps effective | 750 MHz 3 Gbps effective |
| GPU clock | — | 575 MHz |
| Shader clock | — | 1150 MHz |
Memory Configuration
| Memory size | 4 GB | 2 GB |
|---|---|---|
| Memory type | GDDR5 | GDDR5 |
| Memory bus width | 128 bit | 256 bit |
| Bandwidth | 96.00 GB/s | 96.00 GB/s |
Render Configuration
| Shading units | 768 | 384 |
|---|---|---|
| TMUs | 48 | 64 |
| ROPs | 16 | 32 |
| Compute units | 12 | — |
| Cache L1 | 16 KB (per CU) | 64 KB (per SM) |
| Cache L2 | 256 KB | 512 KB |
| SM count | — | 8 |
Performance
| Pixel rate | 16.00 GPixel/s | 9.200 GPixel/s |
|---|---|---|
| Texture rate | 48.00 GTexel/s | 36.80 GTexel/s |
| FP32 (float) performance | 1.536 TFLOPS | 883.2 GFLOPS |
| FP64 (double) performance | 96.00 GFLOPS (1:16) | 73.60 GFLOPS (1:12) |
Dimensions & Outputs
| TDP | unknown | 100 W |
|---|---|---|
| Outputs | No outputs | No outputs |
| Slot width | — | MXM Module |
| Power connectors | — | None |
| Board number | — | P1044 |
API Support & Features
| DirectX | 12 (12_0) | 12 (11_0) |
|---|---|---|
| OpenGL | 4.6 | 4.6 |
| OpenCL | 2.0 | 1.1 |
| Vulkan | 1.2 | — |
| Shader model | 6.3 | 5.1 |
| CUDA | — | 2.1 |
