NVIDIA GeForce RTX 3090 vs NVIDIA RTX A5000
We compared a Desktop platform GPU: 24GB VRAM GeForce RTX 3090 and a Professional market GPU: 24GB VRAM RTX A5000 to see which GPU has better performance in key specifications, benchmark tests, power consumption, etc.
Main Differences
NVIDIA GeForce RTX 3090 's Advantages
Larger VRAM bandwidth (936.2GB/s vs 768.0GB/s)
2304 additional rendering cores
NVIDIA RTX A5000 's Advantages
Released 7 months late
Lower TDP (230W vs 350W)
Score
Benchmark
FP32 (float)
GeForce RTX 3090
+28%
35.58 TFLOPS
RTX A5000
27.77 TFLOPS
3DMark Time Spy
GeForce RTX 3090
+36%
19893
RTX A5000
14600
3DMark Time Spy Extreme
GeForce RTX 3090
+40%
10258
RTX A5000
7285
3DMark Speed Way
GeForce RTX 3090
+33%
5383
RTX A5000
4022
Shadow of the Tomb Raider 2160p
GeForce RTX 3090
+24%
93
RTX A5000
75
Shadow of the Tomb Raider 1440p
GeForce RTX 3090
+14%
154
RTX A5000
135
Shadow of the Tomb Raider 1080p
GeForce RTX 3090
+5%
200
RTX A5000
190
Graphics Card
Sep 2020
Release Date
Apr 2021
GeForce 30
Generation
Quadro Ampere
Desktop
Type
Professional
PCIe 4.0 x16
Bus Interface
PCIe 4.0 x16
Clock Speeds
1395 MHz
Base Clock
1170 MHz
1695 MHz
Boost Clock
1695 MHz
1219 MHz
Memory Clock
2000 MHz
Memory
24GB
Memory Size
24GB
GDDR6X
Memory Type
GDDR6
384bit
Memory Bus
384bit
936.2GB/s
Bandwidth
768.0GB/s
Render Config
-
-
-
82
SM Count
64
10496
Shading Units
8192
328
TMUs
256
112
ROPs
96
328
Tensor Cores
256
82
RT Cores
64
128 KB (per SM)
L1 Cache
128 KB (per SM)
6 MB
L2 Cache
6 MB
-
-
-
Theoretical Performance
189.8 GPixel/s
Pixel Rate
162.7 GPixel/s
556.0 GTexel/s
Texture Rate
433.9 GTexel/s
35.58 TFLOPS
FP16 (half)
27.77 TFLOPS
35.58 TFLOPS
FP32 (float)
27.77 TFLOPS
556.0 GFLOPS
FP64 (double)
433.9 GFLOPS
Board Design
350W
TDP
230W
750 W
Suggested PSU
550 W
1x HDMI 2.1
3x DisplayPort 1.4a
Outputs
4x DisplayPort 1.4a
1x 12-pin
Power Connectors
1x 8-pin
Graphics Processor
GA102
GPU Name
GA102
GA102-300-A1
GPU Variant
-
Ampere
Architecture
Ampere
Samsung
Foundry
Samsung
8 nm
Process Size
8 nm
28.3 billion
Transistors
28.3 billion
628 mm²
Die Size
628 mm²
Graphics Features
12 Ultimate (12_2)
DirectX
12 Ultimate (12_2)
4.6
OpenGL
4.6
3.0
OpenCL
3.0
1.3
Vulkan
1.3
8.6
CUDA
8.6
6.6
Shader Model
6.7
