NVIDIA Ampere GPUs general header

Nvidia’s unannounced Geforce RTX 3060 Ti Ampere GPU listed for pre-order in ASIAN retail market

It appears the Chinese online shopping website Taobao has listed Nvidia’s yet to be officially announced RTX 3060 Ti Ampere GPU. Retailers have started taking pre-orders for this upcoming GPU entry in the RTX 30-series lineup, as spotted by @Avery78.

The RTX 3060 Ti is going to be based on the 392 mm², 17.4 billion transistor GA104-200 GPU silicon chip. This GPU is not expected to launch before the mid of November this year. The RTX 3060 Ti should be the new mid-range/mainstream graphics card in the Ampere lineup, and the price is expected to be around $400 USD (rough estimate).

This upcoming SKU has also appeared on the roadmap according to previous rumors. Even GALAX’s roadmap and Gigabyte’s Watch Dogs Legion code redeeming website also hinted that Nvidia plans to release more SKUs based on the 8nm Ampere architecture, including this new RTX 3060 Ti variant GPU.

The card is currently listed at Taobao for 2049 to 2999 Chinese Yuans, which translates to roughly 305 to 446 USD range. We expect this GPU to be faster than the previous gen Turing-based GeForce RTX 2080 card.

From one of the listings we can also see a non-Ti variant being mentioned as well, the RTX 3060. This is the first time we are getting some confirmation on the existence of the RTX 3060/3060 Ti SKUs.

Nvidia has not yet announced this X60 SKU, though we have been hearing rumors that Nvidia might release two variants, based on the GA106 and GA104 GPU core/die.

The GA104 GPU would likely adopt the Ti or Super branding. And we expect the GA106 silicon to be featured on the plain RTX 3060 GPU (the non-Ti/Super variant), and rumor has it that this card might feature 6 GB of GDDR6 memory over a 192-bit wide bus interface.

The RTX 3060 Ti on the other hand will sport 8 GB GDDR6 VRAM over a 256-bit wide memory bus interface, and the memory clock speed would be 14 Gbps.

According to @Kopite7kimi, who is a well known leaker, the RTX 3060 Ti GPU would allegedly feature a total of 4864 CUDA cores, in a total of 38 SMs, which is 1024 CUDA cores less than the RTX 3070’s 5888 count. The max bandwidth should be around 448 GB/s.

Moreover, the card will have 152 Tensor Cores, and 38 RT Cores, respectively, and the power consumption or TDP is expected to be around 180 Watts. This GPU seems to be more of a successor to the GeForce RTX 2060 SUPER SKU.

Someone just posted a GPU-Z screenshot confirming most of the specs of this GPU, as I’ve outlined above. This appears to be a custom card from board partner HP. Most of the specs are accurate except for the TMU/texture mapping unit count.

ROP or raster operations pipeline count is 80. Also, the RTX 3060 Ti will have a base clock of 1410 MHz and a boost clock value of 1665 MHz, as reported by the software. This screenshot further confirms the existence of this new Ampere GPU.

RTX3060Ti leak

Also, don’t expect similar pricing for these variants. These Ti/Super variants aren’t meant to replace the existing RTX 30-series lineup either, but they rather offer a step-up and upgrade in terms of memory and better core specs, but at slightly higher prices.

This all depends on how AMD responds back with their NAVI 2X lineup based on the new RDNA2 architecture.

Stay tuned for more!

13 thoughts on “Nvidia’s unannounced Geforce RTX 3060 Ti Ampere GPU listed for pre-order in ASIAN retail market”

  1. I hope Nvidia has learned something from these Amperes for the next generation (Hopper). You can’t do a release with way, way too small product on hand and release the Hoppers with a limit of 1 per household until demand lessens.

    Releasing a GPU without enough stock on hand is normal but Amperes have been ridiculously short in supply. The last GPU that I ordered immediately on release was the GTX 970 and I had to wait 3 weeks to get it. The wait was not unexpected though and I was fine with it.

    Hopper will be my next upgrade. I saw an interesting rumor about Hopper that it may be the first Nvidia GPU with multi-chip modules. I have seen this as the future for GPUs at least until silicon can be replaced by a newer material.

      1. I have no idea. Years ago there was some news about carbon nano tubes being a possibility. It held the promise of being 5 times faster than using silicon while using less power at the same time but there were big hurdles to overcome. I haven’t seen anything on it since.

        I believe there are some very smart scientists and engineers working hard to find a solution with some kind of material because eventually we will hit a wall with silicon where it’s either impossible to continue going to smaller nodes or the costs to do so will be prohibitive.

        Whatever company finds the solution and patents it will make a fortune on it. That’s probably why they don’t talk a lot about what they are working on. They don’t want to help their competition get to a solution first by sharing info.

      1. If the 180W TDP is correct, a single 8-pin connector is enough, although higher-tier versions of the card (like the Asus Strix) could go for a 2 connectors approach to give more leeway for overclocking

  2. Specs seem to be quite oaky, assuming this card is going to be faster or near in performance than the previous gen Turing-based GeForce RTX 2080 card.

    But about Availability ? Zero on day 1.

    1. It’s not just the frame rate. The thing is 1080Ti doesn’t support Ray Tracing + DLSS 2.0. That’s why ppl are looking into the RTX line-up.

      1. DLSS would be a good reason if it didn’t suffer from artifacts in motion and few games supported it so far. I have high hopes for it but you have to buy hardware for software that exists, not software that hopefully might come out in the future. The 3000 series GPUs might become a great value in the future but today we don’t know that. Consoles might stay “traditional” in their rendering when it turns out the ray-tracing hardware in them can’t do more than improved reflections. There’s also the fact that the ray-tracing performance is NOT a huge step up over the 2000 series, at least not in existing games, unfortunately. The 3070 is head-to-head with the 2080 Ti when you activate all kinds of RTX features in Control and some other games. Logically the 3060 Ti will be worse. If you didn’t buy into RTX 2000 why buy 3000 now? Nothing has changed.

  3. If this performs like 2070 Super (hopefully higher than that) then it’s nothing more, but a refresh under different SKU. Because 2070S is already like $450-500, so $50-100 discount doesn’t do justice. This should be max $350….

Leave a Reply

Your email address will not be published. Required fields are marked *