(Credit: PCMag)
Nvidia is pushing to bring back its two-generation-old RTX 3060 with 12GB of VRAM to help address ongoing graphics chip shortages and provide gamers with an affordable upgrade option. The cards will drop in July, with board partners receiving their GPU quotas in the near future, WCCFTech reports.
The RTX 3060 was released over five years ago and was entry-level at the time. It's around 10-15% slower than Nvidia's entry-level RTX 5050 from the current generation, but when even that card is around $300, there are precious few options for gamers looking to upgrade on a tight budget. The big advantage the RTX 3060 has is its VRAM capacity, which at 12GB is more than enough for modern gaming, even if its rasterization performance lags modern GPUs.
This latest report on the impending release comes via a leak on the Board Channels forum, where a user claimed most of the new chips will be allocated to Nvidia board partners, so we may not see much in the way of Founders Editions of the card. The leak says Colorful, Asus, MSI, and Galax will get some of the first batch of chips, but it's not clear if other manufacturers will also receive stock. They do suggest overall supply may be limited, however, raising questions over how impactful the launch of this card could be.
Questions remain around pricing. With the RTX 5050 retailing for around $300 right now, any RTX 3060 re-release would need to be substantially cheaper—sub-$250 at least. But that may be a difficult target to meet. According to this latest leak, the RTX 3060 12GB has a high material cost, not least because it has 12GB of VRAM. Even the older GDDR6 that it uses is in high demand and more expensive than it might otherwise be due to the ongoing memory shortage.
But Nvidia and its partners will need to get the cost under the RTX 5050. Although the RTX 3060 does have more VRAM than the 8GB RTX 5050 and even the RTX 5060, it is slower, and its older RT and Tensor cores mean its DLSS and ray-tracing performance is weaker still. Plus, it lacks support for newer technologies like frame generation, let alone multi-frame generation.
From Nvidia's perspective though, this is a card based on older 8nm silicon, so it can manufacture it without eating into the supply lines of its newer, higher-margin GPUs and hardware for data centers. Whether gamers buy it remains to be seen.