NVIDIA RTX PRO 4000 Blackwell Embedded vs NVIDIA GeForce RTX 3090 Ti
GPU · Specs Compare
Graphics Card
Release Date
Unknown
Jan 2022
Generation
Blackwell-MW
GeForce 30
Type
Mobile
Desktop
Bus Interface
PCIe 5.0 x16
PCIe 4.0 x16
Clock Speeds
Base Clock
975 MHz
1560 MHz
Boost Clock
1702 MHz
1860 MHz
Memory Clock
1750 MHz
1313 MHz
Memory
Memory Size
16GB
24GB
Memory Type
GDDR7
GDDR6X
Memory Bus
256bit
384bit
Bandwidth
896.0GB/s
1.01TB/s
Render Config
SM Count
60
84
Shading Units
7680
10752
TMUs
240
336
ROPs
96
112
Tensor Cores
240
336
RT Cores
60
84
L1 Cache
128 KB (per SM)
128 KB (per SM)
L2 Cache
64 MB
6 MB
Theoretical Performance
Pixel Rate
163.4 GPixel/s
208.3 GPixel/s
Texture Rate
408.5 GTexel/s
625.0 GTexel/s
FP16 (half)
26.14 TFLOPS
40.00 TFLOPS
FP32 (float)
26.14 TFLOPS
40.00 TFLOPS
FP64 (double)
408.5 GFLOPS
625.0 GFLOPS
Graphics Processor
GPU Name
GB203
GA102
Architecture
Blackwell 2.0
Ampere
Foundry
TSMC
Samsung
Process Size
5 nm
8 nm
Transistors
45.6 billion
28.3 billion
Die Size
378 mm²
628 mm²
GPU Variant
—
GA102-350-A1
Board Design
TDP
80W
450W
Outputs
Portable Device Dependent
1x HDMI 2.13x DisplayPort 1.4a
Power Connectors
None
1x 16-pin
Suggested PSU
—
850 W
Graphics Features
DirectX
12 Ultimate (12_2)
12 Ultimate (12_2)
OpenGL
4.6
4.6
OpenCL
3.0
3.0
Vulkan
1.4
1.4
CUDA
12.0
8.6
Shader Model
6.8
6.8
Benchmark
3DMark Time Spy
—
21,832
3DMark Time Spy Extreme
—
11,318
3DMark Speed Way
—
5,903
Blender
—
6,217
▲ Specs Compare
Data sourced from official specifications and benchmark tests.