Nvidia confirms technical data [Update]

The Geforce RTX 2060 12 GB is coming: Nvidia has now confirmed the technical data. It is "a premium version of the RTX 2060 6 GB".

The rumors of the last few days were clear: The Geforce RTX 2060 12 GB is coming and Nvidia itself has now confirmed the technical data. The model will have 2,176 FP32 shaders with a base clock of 1,470 MHz and a boost clock of 1,650 MHz. This actually corresponds to the Geforce RTX 2060 Super. However, there are now 12 GiBytes of memory attached to the 192-bit wide memory interface, which have a maximum bandwidth of 336 GB / s - 112 GB / s less than with the Geforce RTX 2060 Super.

Technically, the card is as old as the name suggests. The TU106 is from the Turing generation and is produced in 12 nm. There is ray tracing and DLSS, but not Resizable BAR and the architectural innovations of Ampere such as PCI Express 4.0 or the more modern and faster ray tracing and tensor cores. The memory increases the TDP by 10 or 25 watts to 185 watts.

Prices have not yet been given; The Verge has a statement that says, "It's a premium version of the RTX 2060 6GB and we expect the price to reflect that." You probably wouldn't have expected anything else. At Nvidia, it has been rather uncommon in recent years to offer discounts during the cycle. Instead, new models are sorted in such a way that they close price gaps between existing models. This is exactly what is expected for the Ampere refresh and should not change significantly as long as there is no pressure from the competition.

With the Geforce RTX 2060 12 GB, we are probably in the range of 300 to 400 US dollars for the US market. Most of the time, the prices are converted one-to-one in euros. What the card will ultimately cost on the market is another topic. At least there should be a Founders Edition that is potentially sold without surcharges.

Update: Nvidia has confirmed to PCGH that there will be no Founders Edition. Any hint is a mistake; the "Another SKU from Partners" card.