AMD FirePro D700 vs AMD Radeon E8950
Graphics Processor
| GPU model | Tahiti | Amethyst |
|---|---|---|
| GPU variant | Tahiti XT GL | E8950 |
| Architecture | GCN 1.0 | GCN 3.0 |
| Foundry | TSMC | TSMC |
| Manufacturing process | 28 nm | 28 nm |
| Number of transistors | 4,313 million | 5,000 million |
| Die size | 352 mm² | 366 mm² |
Graphics Card
| Release date | Jan 18th, 2014 | — |
|---|---|---|
| Generation | FirePro | — |
| Production | End-of-life | — |
| Interface | PCIe 3.0 x16 | — |
Clocks
| GPU clock | 850 MHz | — |
|---|---|---|
| Memory clock | 1370 MHz 5.5 Gbps effective | 1500 MHz 6 Gbps effective |
| Base clock | — | 735 MHz |
| Boost clock | — | 1000 MHz |
Memory Configuration
| Memory size | 6 GB | 8 GB |
|---|---|---|
| Memory type | GDDR5 | GDDR5 |
| Memory bus width | 384 bit | 256 bit |
| Bandwidth | 263.0 GB/s | 192.0 GB/s |
Render Configuration
| Shading units | 2048 | 2048 |
|---|---|---|
| TMUs | 128 | 128 |
| ROPs | 32 | 32 |
| Compute units | 32 | 32 |
| Cache L1 | 16 KB (per CU) | 16 KB (per CU) |
| Cache L2 | 768 KB | 512 KB |
Performance
| Pixel rate | 27.20 GPixel/s | 32.00 GPixel/s |
|---|---|---|
| Texture rate | 108.8 GTexel/s | 128.0 GTexel/s |
| FP32 (float) performance | 3.482 TFLOPS | 4.096 TFLOPS |
| FP64 (double) performance | 870.4 GFLOPS (1:4) | 256.0 GFLOPS (1:16) |
| FP16 (half) performance | — | 4.096 TFLOPS (1:1) |
Dimensions & Outputs
| Slot width | Dual-slot | MXM Module |
|---|---|---|
| Length | 279 mm 11 inches | — |
| TDP | 274 W | 95 W |
| Suggested PSU | 600 W | — |
| Outputs | 6x mini-DisplayPort1x SDI | No outputs |
| Power connectors | — | None |
| Board number | — | C769 |
API Support & Features
| DirectX | 12 (11_1) | 12 (12_0) |
|---|---|---|
| OpenGL | 4.6 | 4.6 |
| OpenCL | 1.2 | 2.0 |
| Vulkan | 1.2 | 1.2 |
| Shader model | 5.1 | 6.3 |
Other Features
Mobile graphics
| Release date | — | Sep 29th, 2015 |
|---|---|---|
| Generation | — | Embedded (8000) |
| Production | — | End-of-life |
| Interface | — | MXM-B (3.0) |
| Reviews | — | 3 in our database |