NVIDIA GeForce RTX 4080 Laptop GPU

The NVIDIA GeForce RTX 4080 Laptop GPU, codenamed GN21-X9, stands as a formidable high-end graphics solution tailored for laptops. Anchored in the Ada Lovelace architecture and based on the AD104 chip, akin to its desktop counterpart, the RTX 4070 Ti. The AD104 chip offers a robust configuration with a maximum of 7,680 shaders, of which the RTX 4080 Laptop utilizes 7,424, coupled with a 192-bit memory bus. The graphics memory comprises 12 GB of GDDR6, operating at a clock speed of either 16 or 20 Gbps (effective).

The Total Graphics Power (TGP) is customizable within a range of 60 to 150 Watts, with an additional 25 Watts dynamically boosted from the CPU, leading to distinct performance levels. For comparison, the desktop RTX 4070 Ti utilizes 285 Watts. The boost clock speed of the RTX 4080 Laptop spans from 1350 MHz (at 60 Watt TGP) to 2280 MHz (at 150W). Consequently, the performance is heavily influenced by the power settings configured for the RTX 4080 within the laptop.

In our tests, the synthetic performance of the RTX 4080 surpasses that of the mobile RTX 3080 Ti, falling just behind the mobile RTX 4090. In real-world gaming scenarios, even demanding titles with Raytracing enabled, such as F1 23, can be enjoyed in Quad HD (QHD) at maximum settings. Less demanding games like Baldur’s Gate 3 can be played smoothly in 4K, as detailed in the results below.

The RTX 4080 is equipped with 232 tensor cores (4th gen) that facilitate DLSS 3, including frame generation for supported games. Additionally, it features 58 raytracing cores (4rd gen) to enhance raytracing effects in compatible games.

Manufactured using the 5nm process at TSMC (4N process), the AD103 chip exemplifies advanced technology.

Graphics Processor
Chip (AD104)
ArchitectureAda Lovelace
Process Size5 nm
Transistor Count35.8 Billion
Density121.8M / mm²
Die Size294 mm²
Memory Size12 GB
Memory TypeGDDR6
Memory Bus192 bit
Bandwidth432.0 GB/s
Graphics Config
Shading Units7424
Base Clock/Boost Clock1290 MHz/1665 MHz
Memory Clock2250 MHz = 18 Gbps effective
SM Count58
RT Cores/Tensor Cores58/232
L1 Cache/L2 Cache128 KB (per SM)/48 MB
DirectX/OpenGL12 Ultimate (12_2)/4.6
Theoretical Performance
FP32 (float)24.72 TFLOPS
Pixel Rate133.2 GPixel/s
Texture Rate386.3 GTexel/s
3DMark Time Spy Graphics17879
3DM11 Performance GPU57598
3DMark Time Spy Graphics + 3DM11 Performance GPU / 1000 = Score
Scroll to top