According to various sources, NVIDIA is currently developing the next generation
graphics cardsRTX 5000 based on the Blackwell architecture with superior advantages. Previously, we knew that Blackwell is the code name for NVIDIA's next-generation GPU architecture, expected to be released in 2024 and may first appear on graphics cards for servers before reaching regular users.

Join the S-Games gamer community on Discord, a new playground where gamers can discuss hot games on the market, share gaming experiences, team up to climb ranks, and chat lively together. At S-Games Discord, you'll also receive exclusive codes for the newest and most exciting games. What are you waiting for, JOIN S-GAMES DISCORD NOW!
Revealed information suggests that NVIDIA may use 3nm process on its
new NVIDIA graphics cardsto continue its trend of shrinking transistors. Over the years, they've consistently shrunk transistors on graphics chips – Pascal architecture was 16nm, Turing was 12nm, Ampere was 8nm, Ada (current generation) is 4nm. However, there are conflicting reports stating that the RTX 5000 won't utilize the 3nm architecture due to cost concerns - 3nm wafer costs more than 25% compared to 5nm, potentially driving up the card prices.

In terms of specs, the Blackwell GPU is also expected to undergo significant changes, including a redesigned CUDA core architecture to optimize Ray Tracing and Path Tracing processing speeds. They will also support GDDR7 RAM chips for even higher performance. An insider revealed that these CPUs will deliver twice the performance of the current RTX 4000 generation – a figure that gamers will surely covet. However, even if the RTX 5000 cards aren't manufactured on the pricey 3nm platform, their prices could still pose a major barrier as both AMD and NVIDIA are following the trend of increasing their graphics card prices.
