AMD Radeon Pro Vega 56 vs NVIDIA GeForce RTX 3070 Mobile
Graphics Processor
| GPU model | Vega 10 | GA104 |
|---|---|---|
| GPU variant | Vega 10 XL (215-0894216) | GA104-770-A1 |
| Architecture | GCN 5.0 | Ampere |
| Foundry | GlobalFoundries | Samsung |
| Manufacturing process | 14 nm | 8 nm |
| Number of transistors | 12,500 million | 17,400 million |
| Die size | 495 mm² | 392 mm² |
Mobile graphics
| Release date | Aug 14th, 2017 | Jan 12th, 2021 |
|---|---|---|
| Generation | Radeon Pro Mac (Vega Series) | GeForce 30 Mobile |
| Production | Active | Active |
| Interface | PCIe 3.0 x16 | PCIe 4.0 x16 |
| gpu.details.availability | — | Feb 2nd, 2021 |
| Predecessor | — | GeForce 20 Mobile |
| Reviews | — | 1 in our database |
Clocks
| Base clock | 1138 MHz | 1110 MHz |
|---|---|---|
| Boost clock | 1250 MHz | 1560 MHz |
| Memory clock | 786 MHz 1572 Mbps effective | 1750 MHz 14 Gbps effective |
Memory Configuration
| Memory size | 8 GB | 8 GB |
|---|---|---|
| Memory type | HBM2 | GDDR6 |
| Memory bus width | 2048 bit | 256 bit |
| Bandwidth | 402.4 GB/s | 448.0 GB/s |
Render Configuration
| Shading units | 3584 | 5120 |
|---|---|---|
| TMUs | 224 | 160 |
| ROPs | 64 | 80 |
| Compute units | 56 | — |
| Cache L1 | 16 KB (per CU) | 128 KB (per SM) |
| Cache L2 | 4 MB | 4 MB |
| SM count | — | 40 |
| Tensor cores | — | 160 |
| RT cores | — | 40 |
Performance
| Pixel rate | 80.00 GPixel/s | 124.8 GPixel/s |
|---|---|---|
| Texture rate | 280.0 GTexel/s | 249.6 GTexel/s |
| FP16 (half) performance | 17.92 TFLOPS (2:1) | 15.97 TFLOPS (1:1) |
| FP32 (float) performance | 8.960 TFLOPS | 15.97 TFLOPS |
| FP64 (double) performance | 560.0 GFLOPS (1:16) | 249.6 GFLOPS (1:64) |
Dimensions & Outputs
| Slot width | IGP | — |
|---|---|---|
| TDP | 210 W | 115 W |
| Outputs | 1x HDMI3x DisplayPort | No outputs |
| Power connectors | None | None |
API Support & Features
| DirectX | 12 (12_1) | 12 Ultimate (12_2) |
|---|---|---|
| OpenGL | 4.6 | 4.6 |
| OpenCL | 2.1 | 3.0 |
| Vulkan | 1.2 | 1.2 |
| Shader model | 6.4 | 6.6 |
| CUDA | — | 8.6 |
