Performance Rating
NVIDIA GeForce RTX 4090 48 GB outperforms NVIDIA A100 PCIe 80 GB by 38.82% in the overall GPU ARK performance rating
NVIDIA GeForce RTX 4090 48 GB
Expert Comparison
NVIDIA A100 PCIe 80 GB лучше для вычислительных задач, научных исследований и обработки больших данных благодаря высокой пропускной способности памяти (1.94 TB/s) и большему объему памяти (80 ГБ HBM2e). NVIDIA GeForce RTX 4090 48 GB более мощный для игровых задач и графики, с более высоким показателем производительности (82.58 TFLOPS) и быстрой частотой boost (2520 МГц), хотя и потребляет больше энергии (450 Вт).
Contents:
Memory ML Performance Compute Power Architecture & Compatibility ML Software Support Clocks & Performance Power Consumption Rendering AdditionalMemory
Memory Size
+67%
80
ГБ
Memory Type
Memory Bandwidth
1.01 TB/s
Memory Bus Width
ML Performance
FP16 (Half Precision)
+6%
82.58
TFLOPS
BF16 (Brain Float)
311.84
TFLOPS
TF32 (TensorFloat)
155.92
Compute Power
FP32 (Single Precision)
+324%
82.58
TFLOPS
FP64 (Double Precision)
9.746
TFLOPS
CUDA Cores
+137%
16,384
RT Cores
128
Architecture & Compatibility
GPU Architecture
SM (Streaming Multiprocessor)
+19%
128
PCIe Version
ML Software Support
CUDA Version
8.9
CUDA Toolkit (first supported)
CUDA Toolkit status
Clocks & Performance
Base Clock
+110%
2,235
Boost Clock
+79%
2,520
Memory Clock
+15%
1,512
Power Consumption
Recommended PSU
-18%
700 W
Power Connector
TDP/TGP
-33%
300 W
Rendering
Texture Units (TMU)
+19%
512
ROP
128
L2 Cache
+11%
80 MB
Additional
Slots
Release Date
Display Outputs
Rent on Intelion Cloud
Available for rent on Intelion Cloud
Available for rent on Intelion Cloud
