NVIDIA SUPER Lineup Will Consist Of Three Cards: Faster RTX 2060, 2070 and 2080 Variants With $100 Off MSRP And Souped Up GDDR6
Looks like Tweaktown has some more details on top of the leak we posted yesterday (read: rumor, not confirmed!), according to them, we should expect not two but three new variants in the NVIDIA SUPER lineup. The new cards will slot in the positioning of the current RTX 2060, 2070 and 2080 and will feature considerably faster GDDR6 memory as well as faster clock rates.
NVIDIA's SUPER Graphics Cards: Cheaper, Faster GeForce RTX 2060, 2070 and 2080s
Considering these will almost certainly be higher binned dies, we can also expect clock speeds to improve for the core clocks. The memory clocks will also be higher and GDDR6 will now be be standard for all three of the cards - something that is not the case right now. The higher clocked memory as well as core clocks should allow the cards to be faster than their predecessors. Depending on how much of a clock delta NVIDIA is able to achieve, it could be anywhere from 5% to 15%. Keep in mind, considering this post is tagged with the rumor tag, there is always still the chance that NVIDIA could blindside is completely and introduce something completely different.
The current launch date according to rumors, is going to be at E3 2019, but there is no official (or unofficial) confirmation of release date as of yet so it would be wise to take all this with a grain of salt. AMD is expected to reveal the full and final details for its RX 5000 series Navi GPUs there as well, and it would present quite an opportunity for NVIDIA to upset its rival.
We do know however that the new cards will be based on the Turing GPU architecture and that the RTX 2080 Ti will remain the flagship for the series. It will also be interesting to see how NVIDIA chooses to integrate the nomenclature. They already have the "Ti" suffix and if they are introducing another to the mix, it could get confusing fast. Possible combinations of the word include " GeForce RTX 2060 SUPER", "GeForce RTX 2070 SUPER" and "GeForce RTX 2080" SUPER". We are also not sure right now what would happen to the current lineup. Will NVIDIA continue to sell the non-SUPER counterparts alongside their siblings or will they be EOL-ed.
Chances are, considering this is NVIDIA taking advantage of the maturing process and yields, that their older siblings will be EOL-ed. We have also heard that NVIDIA is going to be handing out some pretty impressive price discounts which would make them instantly competitive with AMD's Navi lineup. Considering current pricing of the lineup, the RTX 2060 SUPER, RTX 2070 SUPER and RTX 2080 SUPER could retail for as low as $249, $399 and $599 respectively. The NVIDIA Turing lineup is based on the special NVIDIA FinFET process from TSMC and should be getting into the golden yield territory. The company still has to adopt 7nm process for its architectures and is unlikely to do so until the yields (and TSMC's ability to handle the volume) increase significantly.
Knowing NVIDIA however, I am sure they may still have some surprises up their sleeves considering this is turning out to be one of the best kept secrets so far. We are also hearing conflicting reports on the lineup itself so only time will tell what turns out to be right. If everything goes well, NVIDIA is set to make a comeback in the next couple of weeks. Depending on whether it turns out to be a preview, announcement or actual launch, it could make things very interesting for our users.