GRID A100B
Graphics Card GRID A100B — Specifications, Tests & Benchmarks
GRID A100B
GeForce Ax
•
GA100
•
14.05.2020
•
48 GB
•
HBM2e
•
6144 bit
•
400W
🏆 Top #12536
Performance
Synthetic Score
729
Efficiency
entry
The GRID A100B (Ampere) model joined the NVIDIA lineup of graphics accelerators on 14.05.2020. Thanks to modern manufacturing technology, engineers managed to place 54.2 billion transistorsistors transistors on a 826 mm² die. The core configuration includes 6912 shader units, 0 RT Cores for ray tracing, and 432 Tensor Cores for AI tasks. The graphics card is equipped with 48 GB of HBM2e memory. A 6144 bit-bit bus is used, providing bandwidth at 1914.88 GB/s. The GPU's base clock speed is 900 MHz, and in Boost mode, it can automatically increase up to 1005 MHz. The nominal thermal design power is 400 W. For stable system operation, a PSU of 800 W or more will be required.
General Information
| Release Date | 14.05.2020 |
| Usage Class | gpu.usage_class.values.light_gaming |
| Market Segment | Server / Workstation |
| Series | GeForce Ax |
| Integrated Graphics | Yes |
GPU
| Architecture | Ampere |
| Codename | GA100 |
| Process Node | 7 nm (TSMC) |
| Transistor Count | 54,200,000.0 million |
| Die Size | 826 mm² |
| Transistor Density | 65,599,999 million / mm² |
Rendering Configuration
| Shading Units (ALU) | 6912 |
| Texture Mapping Units (TMU) | 432 |
| Render Output Units (ROP) | 192 |
| Tensor Cores | 432 |
| Compute Units (SM/CU) | 108 |
Clock Speeds
| Base Clock | 900 MHz |
| Boost Clock | 1005 MHz |
| Memory Clock (Actual) | 1215 MHz |
| Memory Clock (Effective) | 2400 MHz |
Memory
| Memory Capacity | 48 GB |
| Memory Type | HBM2e |
| Memory Bus | 6144 bit bit |
| Memory Bandwidth | 1914.88 GB/s |
Cache Memory
| L1 Cache | 20736 KB |
| L2 Cache | 48 MB |
Board Design & Power
| Interface | PCIe 4.0 x16 |
| Slot Width | 0 |
| Power Consumption (TDP) | 400 W |
| Recommended PSU | 800 W |
API Support
| OpenCL | |
| CUDA |
Theoretical Performance
| Pixel Fill Rate | 193000 GPixel/s |
| Texture Fill Rate | 434200 GTexel/s |
| FP16 Performance (Half) | 55.57 TFLOPS (4:1) |
| FP32 Performance (Float) | 13.89 TFLOPS |
| FP64 Performance (Double) | 6.95 TFLOPS (1:2) |
Synthetic Tests & Benchmarks
| TPU Index Score | 21.12% |
| Performance Tier | entry |
Всі тести
Тестів поки немає.
Comments
0Please log in to your account to leave a comment.
Log In or RegisterNo comments yet. Be the first to comment!
Are you sure you want to delete this comment? This action cannot be undone.