Product Compare

NVIDIA A100 SXM4 40 GB vs NVIDIA A40 PCIe

GPU GPU
Graphics Card
Release Date May 2020 Oct 2020
Generation Tesla Ampere Tesla
Type AI GPU Desktop
Bus Interface PCIe 4.0 x16 PCIe 4.0 x16
Clock Speeds
Base Clock 1095 MHz 1305 MHz
Boost Clock 1410 MHz 1740 MHz
Memory Clock 1215 MHz 1812 MHz
Memory
Memory Size 40GB 48GB
Memory Type HBM2e GDDR6
Memory Bus 5120bit 384bit
Bandwidth 1555GB/s 695.8GB/s
Render Config
SM Count 108 84
Shading Units 6912 10752
TMUs 432 336
ROPs 160 112
Tensor Cores 432 336
L1 Cache 192 KB (per SM) 128 KB (per SM)
L2 Cache 40 MB 6 MB
RT Cores 84
Theoretical Performance
Pixel Rate 225.6 GPixel/s 194.9 GPixel/s
Texture Rate 609.1 GTexel/s 584.6 GTexel/s
FP16 (half) 77.97 TFLOPS 37.42 TFLOPS
FP32 (float) 19.49 TFLOPS 37.42 TFLOPS
FP64 (double) 9.746 TFLOPS 584.6 GFLOPS
Graphics Processor
GPU Name GA100 GA102
Architecture Ampere Ampere
Foundry TSMC Samsung
Process Size 7 nm 8 nm
Transistors 54.2 billion 28.3 billion
Die Size 826 mm² 628 mm²
Board Design
TDP 400W 300W
Suggested PSU 800 W 700 W
Outputs No outputs 3x DisplayPort 1.4a
Power Connectors None 8-pin EPS
Graphics Features
DirectX N/A 12 Ultimate (12_2)
OpenGL N/A 4.6
OpenCL 3.0 3.0
Vulkan N/A 1.3
CUDA 8.0 8.6
Shader Model N/A 6.6