There could be a ton of thrilling Intel Lunar Lake projects, Lake Strzalkoweand now the Ryzen AI 300 launches and announcements are coming, but it’s impossible not to feel the anticipation for a good, old-fashioned graphics card launch. Nvidia’s RTX 50-series graphics cards are on the way, and now we’re hearing they could be announced soon.
Chinese Tech Side Life on the bench reports that if sources are correct, two next-generation Nvidia GPUs could be coming this month. According to machine translation, Benchlife says, “If all goes well, the GeForce RTX 5090/D and GeForce RTX 5080/D with Blackwell GPU architecture are expected to officially launch in September. If our sources are correct, more information will be available online for reference soon.” (“D” refers to the variants for the Chinese market).
This would be significantly sooner than the original planned launch of overdue 2024 or early 2025.
If Benchlife’s sources are exact and the RTX 5080 and RTX 5090 are announced this month, it would be surprising, but not impossible. While Nvidia’s Blackwell server chips have been delayed, the consumer chips will be completely different than the server ones, meaning the 50-series GPUs could be completely unaffected by the delay.
And while you could argue that an announcement isn’t a launch, it would be odd for Nvidia to announce these graphics cards well before they hit the market. For example, the first RTX 40-series GPUs were announced at GTC in September 2022 and launched less than a month later.
Announcement speculation aside, Benchlife also suggests that, apparently based on factory sources, the RTX 5090 should have a TGP of 550W, while the RTX 5080 should have a TGP of 350W. That would mean the 5090’s TGP would be 100W higher than the 4090, and the 5080’s TGP would be just 30W higher than the 4080. That 550W TGP for the 5090 isn’t that much. 600 W TGP that was recently rumored.
Benchlife further suggests that such power may not refer to TGP (Total Graphics Power) but to the “maximum heat dissipation capabilities of the radiator,” which would make sense. It’s not uncommon for coolers to be tested and rated at higher powers than the GPU’s TGP. That’s why it’s possible to boost the power limits on some cards: because the cooler can handle higher TGPs than the GPU’s default.
Experience tells me to take such energy-use rumors with a grain of salt. If the rumor of an imminent announcement is true, I suspect we won’t have to wait long to find out.
