AMD Radeon R7 340 OEM vs NVIDIA Switch GPU 20nm
Graphics Processor
| GPU model | Oland | GM20B |
|---|---|---|
| Architecture | GCN 1.0 | Maxwell 2.0 |
| Foundry | TSMC | TSMC |
| Manufacturing process | 28 nm | 20 nm |
| Number of transistors | 950 million | 2,000 million |
| Die size | 77 mm² | 118 mm² |
| GPU variant | — | ODNX02-A2 |
Graphics Card
| Release date | May 5th, 2015 | Mar 17th, 2017 |
|---|---|---|
| Generation | Pirate Islands | Console GPU |
| Production | End-of-life | End-of-life |
| Interface | PCIe 3.0 x8 | IGP |
| Predecessor | Volcanic Islands | — |
| Successor | Arctic Islands | — |
| Launch price | — | 299 USD |
Clocks
| Base clock | 730 MHz | 384 MHz |
|---|---|---|
| Boost clock | 780 MHz | 768 MHz |
| Memory clock | 1000 MHz 4 Gbps effective | 1600 MHz 3.2 Gbps effective |
Memory Configuration
| Memory size | 2 GB | 4 GB |
|---|---|---|
| Memory type | GDDR5 | DDR4 |
| Memory bus width | 128 bit | 64 bit |
| Bandwidth | 64.00 GB/s | 25.60 GB/s |
Render Configuration
| Shading units | 384 | 256 |
|---|---|---|
| TMUs | 24 | 16 |
| ROPs | 8 | 16 |
| Compute units | 6 | — |
| Cache L1 | 16 KB (per CU) | — |
| Cache L2 | 256 KB | — |
| SMM count | — | 2 |
Performance
| Pixel rate | 6.240 GPixel/s | 12.29 GPixel/s |
|---|---|---|
| Texture rate | 18.72 GTexel/s | 12.29 GTexel/s |
| FP32 (float) performance | 599.0 GFLOPS | 393.2 GFLOPS |
| FP64 (double) performance | 37.44 GFLOPS (1:16) | 12.29 GFLOPS (1:32) |
| FP16 (half) performance | — | 786.4 GFLOPS (2:1) |
Dimensions & Outputs
| Slot width | Single-slot | IGP |
|---|---|---|
| TDP | 50 W | 15 W |
| Suggested PSU | 250 W | 350 W |
| Outputs | 1x DVI1x HDMI1x VGA | No outputs |
| Power connectors | None | None |
| Length | — | 239 mm 9.4 inches |
| Width | — | 101 mm 4 inches |
| Height | — | 28 mm 1.1 inches |
| Weight | — | 0.4 kg (0.88 lbs) |
| Board number | — | HAC-001 |
API Support & Features
| DirectX | 12 (11_1) | 12 (12_1) |
|---|---|---|
| OpenGL | 4.6 | 4.6 |
| OpenCL | 1.2 | 1.2 |
| Vulkan | 1.2 | 1.1 |
| Shader model | 5.1 | 6.4 |
| CUDA | — | 5.3 |
