CES 2023 | Nvidia GeForce RTX 4090 and RTX 4080 laptop graphics cards announced with significant performance and efficiency gains over Ampere
Laptop versions of the Nvidia GeForce RTX put up quite a show in leaked benchmarks, with the GeForce RTX 4090 trading blows with Ampere's best, the GeForce RTX 4090. Even the RTX 4080 was over 40% faster than its predecessor, which is nothing short of impressive. Nvidia claims that the high-end Ada Lovelace SKUs are up to 4x faster than Ampere. Both graphics cards are official now, along with the lower-end RTX 4070, RTX 4060 and RTX 4050.
Starting with the flagship GeForce RTX 4090, one gets 16 GB of GDDR6 VRAM on a 128-bit bus, an AD103 GPU with 9,728 CUDA cores, a boost clock of 2,040 MHz and a TGP of up to 150 Watts. A report from earlier got most of the specs right, except for the TGP. Depending on the OEM, that figure could go as high as 175 Watts with Dynamic Boost enabled. The GeForce RTX 4080 uses a cut-down AD104 GPU with 7,424 CUDA cores. It features 12 GB of GDDR6 memory on a 128-bit bus and its TGP value ranges between 60 and 150 Watts. Furthermore, its boost clock is a tad higher at 2,280 MHz.
Nvidia has also refined its Max-Q technology it first introduced in 2016. Fifth-gen MaxQ is designed to reduce the power usage of laptops greatly when idle, thus enhancing battery life. It takes advantage of the low-voltage GDDR6 memory on the RTX 4000 graphics cards and uses technologies like Tri-speed memory control to keep power usage in check. Fifth-gen MaxQ also uses DLSS 3.0 to boost efficiency wherever possible. Nvidia claims that the tech has made laptops 22 times more power-efficient; a claim one should take with a grain of salt. Laptops featuring the Nvidia GeForce RTX 4090 and RTX 4080 will be available from February 8 starting at US$1,999.