NVIDIA GeForce RTX 3050
2 laptops with NVIDIA GeForce RTX 3050 in comparion
NVIDIA GeForce RTX 3050 - RTX for everyone
The NVIDIA GeForce RTX 3050 was released in May 2021 and is the smallest variant of the NVIDIA GeForce RTX 3000 series. It is based on the GA 107 Ampere chip and offers 4 GB of vRAM with a memory bandwidth of 128 bits. It also has 16 ray tracing and 64 tensor cores. The TGP is between 35 and 80 watts, depending on the variant installed. This also makes it suitable for installation in thinner notebooks. Like all graphics cards in the NVIDIA GeForce RTX 3000 series, the NVIDIA GeForce RTX 3050 also supports ray tracing, DLSS and G-Sync.
The graphics card is connected via PCIe 4.0 and the laptop's display is connected via eDP 1.4b. As usual, the DirectX 12.2 API is also supported.
As the smallest graphics card in the NVIDIA GeForce RTX 3000 series, the NVIDIA GeForce RTX 3050 is naturally behind its big brothers, such as the NVIDIA GeForce RTX 3060 and the NVIDIA GeForce RTX 3070, in terms of performance, placing it just below an NVIDIA GeForce GTX 1660 Ti Max-Q.
Excellent price-performance ratio
Due to the limited performance of the NVIDIA GeForce RTX 3050, it is best suited for games with high graphics settings. However, ray tracing should be avoided as the performance is barely sufficient for this. In some games, the Tensor cores can provide higher performance using DLSS. For highly demanding games, you will have to reduce the graphics details.
Due to its performance potential, the NVIDIA GeForce RTX 3050 is particularly suitable for entry-level and mid-range gaming notebooks. MSI and Lenovo in particular have recognized this and offer the NVIDIA GeForce RTX 3050 in just such notebooks.
Core Information
GPU Name
GA107
GPU Variant
GN20-P0-A1
Model Series
NVIDIA GeForce RTX 30
Release Date
11.05.2021
GPU Specifications
GPU Architecture
Ampere
Base Clock
1065 MHz
Boost Clock
1343 MHz
Integrated GPU
No
MXM Module
No
Soldered
Yes
Compute Units
Render Output Units
32
Texture Mapping Units
64
Shader Units
2048
Tensor Cores
64
SM Count
16
Hardware Ray Tracing
Yes
Ray Tracing Units
16
Memory
Memory Type
GDDR6
Memory Clock
1500 MHz
Effective Memory Data Rate
12 Gbps
Memory Bandwidth
192 GB/s
System-Dependent Memory Bandwidth
No
Memory Size
4 GB
Memory Interface
128 Bit
Shared Memory
No
Dedicated Graphics Memory
Yes
Unified Memory
No
Cache
L2 Cache
2 MB
Theoretical Performance
Pixel Fill Rate
42.98 GPixel/s
Texture Fill Rate
85.95 GTexel/s
FP16 Compute Performance
5.501 TFLOPS
FP16 Ratio
1:1
FP32 Compute Performance
5.501 TFLOPS
FP64 Compute Performance
85.95 GFLOPS
FP64 Ratio
1:64
Manufacturing
Official Process Name
Samsung 8N
Foundry
Samsung
Process Node
8 nm
Chip Specifications
Die Area
200 mm²
Transistors
8700 Millionen
Transistor Density
43.5 M / mm²
Chip Package
FCBGA-1358
Power Consumption
Base Power Consumption
45 Watt
Base Power Type
TDP
Maximum Power Consumption
80 Watt
Maximum Power Type
TGP
Supported Standards
DirectX
12_2
OpenGL
4.6
OpenCL
3.0
Vulkan
1.4
CUDA
8.6
Shader Model
6.9
Connectivity
Host Interface
PCIe 4.0 x8
Intel Arc B390
Intel Iris Xe Graphics G7 80 EUs
AMD Radeon RX Vega 8 (4000/5000)
NVIDIA GeForce RTX 4070
Intel Graphics 4 Xe3 2.45 GHz (Panther Lake)
NVIDIA RTX PRO 500 Blackwell
Intel Graphics 4 Xe3 2.5 GHz (Panther Lake)
Intel Graphics 2 Xe3 (Wildcat Lake)
Qualcomm Adreno X2-90 1.7GHz
Qualcomm Adreno X2-85
NVIDIA GeForce RTX 5060
NVIDIA GeForce RTX 5070
Start Laptop Configurator
Start Interactive Buying Guide