Once upon a time, gaming laptops were more of a meme than a serious option. Huge, hot, noisy — they tried to pack desktop ...

When GeForce RTX 50 appeared in laptops, the Internet froze. And rightly so: NVIDIA promises RTX 4080-level mobile ...

NVIDIA is reportedly working on a second Chinese-only variant of the RTX 5090 GPU, and this time around it'll lose some VRAM ...

The launch of the RTX 3080 Ti, of course! This high-end card launched with nearly as many CUDA cores, RT cores, and Tensor cores as the RTX 3090 for a little less money, but it had one downside: less ...

FEVM’s FNGT5 Pro offers powerful RTX 4090 graphics in a portable Thunderbolt 5 eGPU, but lack of Apple support renders it useless for MacBook users.

Meanwhile, the RTX 4090 offers 16,384 CUDA cores, 128 third-gen RT cores, 512 fourth-gen Tensor cores, a max frequency of 2.5GHz, 24 GB of GDDR6X memory across a 384-bit bus, and 21 Gbps memory speed.

The RTX 5090 mobile doesn't have the same uptick in specifications, only offering 8% more CUDA cores, though it does have 50% more VRAM and 55% more memory bandwidth. But that's not enough to ...

The 5070's CUDA core count falls right in between the RTX 4070 and the 4070 Super's, where the 5080 and 5070 Ti both included small increases from their previous-generation counterparts.

The RTX 5080 is, on paper, a weaker card than the RTX 4090, with around 50% fewer CUDA cores, fewer RT and Tensor cores (albeit from a later generation), 50% less memory, and a narrower memory bus.

To put those specs into the perspective, the GeForce RTX 4090 is built around NVIDIA's AD102-300-A1 GPU with 16,834 CUDA cores, 24GB of GDDR6X memory, and a 384-bit bus resulting in just over 1TB ...

Even running on a more modern architecture and process node, the RTX 5070's 6,144 CUDA cores are no match for the RTX 4090's 16,384 - meaning a similar disparity in terms of RT and tensor cores.