Nvidia unveiled a new flagship line of GPUS in the RTX 30 series, starting with the 28-billion transistor chip. They claim it is the fastest GPU ever built for gaming and creative work involving a lot of rendering.
Nvidia is one of the largest and major supplier of chips to data centers. They are also at the top of the graphics food chain in video game GPUs. Nvidia’s gaming business alone netted them about $1.65 billion in the most recent quarterly earnings, with 26% year-over-year growth.
Nvidia’s new series of RTX 30 chips is based on the latest Ampere architecture.
New tech, a giant leap
The Ampere architecture is the same one used in powering the premium A100 artificial intelligence GPU used in data centers. The chips are made using a new fabrication method that is vastly different from the earlier silicon flagships.
Nvidia made the switch from a 16-nanometer manufacturing process to using Samsung’s eight-nanometer method. This yields smaller transistors, allowing the company to fit more of them onto a chip.
Nvidia says that these improvements will increase the performance because they offer up to 2x the performance and 1.9x the power efficiency used in the previous generation of flagship consumer GPUs.
A much-anticipated series
The CEO at Nvidia, Jensen Huang, said in the virtual unveiling event that this is a need for more potent GPUs worldwide. To fulfill that need, they made a giant Ampere. Huang described the series as the greatest generational leap in Nvidia’s history.
This series included two other chips (the 1080 and the 3070), which will become better prices than the RTX 3090. While the 3090 is set to hit the market at a price tag of $1,499, the 3080 and 3070 will be priced at $699 and $499.