April 9 news - Nvidia appears to be ready to roll, planning to launch its next-generation GeForce RTX 5090 and RTX 5080 in the fourth quarter of this year. This move is poised to boost shipments for Nvidia's board partners, particularly as more gamers upgrade to high-end options, thereby also helping to increase the average selling price (ASP) of GPUs.
According to the latest details leaked, the flagship product codenamed GB202, the RTX 5090, will see a 50% increase in the number of CUDA cores, reaching 24,576. This is more than eight times the core count of the Turing-based RTX 2080.
Moreover, the flagship card's memory bandwidth will also see a 52% boost through the use of 32Gbps GDDR7, exceeding the current RTX 4090's GDDR6X.
Clock speeds are expected to rise by 15%, marking one of the biggest upgrades. This would push the RTX 5090's frequency to 2.9GHz, easily surpassing 3GHz in gaming workloads. In contrast, the existing RTX 4090's peak frequency is 2.52GHz.
However, with the performance increase comes a corresponding rise in price. The RTX 5090's price is rumored to exceed $2,500, approximately over 18,000 RMB.
