The first of Nvidia’s next-generation graphics cards arrive on Oct. 12 with the GeForce RTX 4090, but it comes with an obscenely high price tag of A$2,959.
The company is then unleashing the GeForce RTX 4080 in November through two models: a 16GB model that’ll go for A$2,219 and up, and a 12GB version that’ll start at A$1,659, but features a decrease in the number of total CUDA cores.
Nvidia says both products will deliver a two- to four-times performance increase over the RTX 3090 Ti and 3080 Ti, respectively. However, the company is increasing the starting prices for its next-generation GPUs.
The RTX 4080 will cost around A$500 more than the original starting price for the RTX 3080, which only had 10GB of video memory. Meanwhile, the RTX 4090 will similarly cost A$500 more than the RTX 3090.
The company has improved the RTX 4000 series with a new GPU architecture called Ada Lovelace, which uses a 4-nanometer manufacturing process from TSMC, an improvement from the 8-nanometer process from Samsung in the RTX 3000 series.
"The total processing power of Ada is a massive leap over the Ampere generation, and the performance shows,” said Nvidia CEO Jensen Huang in introducing the technology at the company’s GTC developer event.
“For rasterized games, Ada is up to two times faster, and it's four times faster for ray-traced games. Ada is incredibly efficient, over twice the performance of the same power compared to Ampere,” he added. “And you can really push Ada. We’ve overclocked Ada past 3GHz in our labs.”
It's also important to note the 12GB RTX 4080 has different specs than the 16GB version beyond the memory. As you see, it also features a decrease in the CUDA core count that's lower than the older RTX 3080.
The same technology boasts improved ray-tracing cores that can produce realistic lighting and shadow effects two to three times better than the previous generation. In addition, the RTX 4000 series will feature a new 3.0 version of DLSS, which can substantially increase the frame rates on a PC game even more.
“DLSS 3 generates entire new frames, without processing the graphics pipeline, boosting game performance up to four times over brute-force rendering,” Huang said. Over 35 games and apps plan to add DLSS 3 support with the first games set to arrive next month.
To demonstrate this, Huang showed several short demos about how DLSS 3 can double the frame rates of games including Cyberpunk 2077 and Microsoft Flight Simulator while maintaining the ray-tracing effects.
The big question is whether the RTX 4000 series will face much demand. Currently, Nvidia is facing an oversupply situation with older RTX 3000 GPUs, which has prompted the company to initiate price cuts. As a result, the high cost of the RTX 4000 models may cause consumers to opt for RTX 3000 series, especially as many crypto-miners are considering selling off their GPUs following Ethereum's decision to phase out mining.
Those who want a new GPU can sign up for an Nvidia email notification with details about when the products go on sale. Customers can also find full specs for the RTX 4000 GPUs over the company's website. Vendors including Asus, MSI, Gigabyte, Zotac, among others, all plan on producing the graphics cards.
